The Power of a Tweet: An Exploratory Study Measuring the Female Perception of Celebrity Endorsements...

2012 Summer Educators’ Proceedings

VOLUME 23

MARKETING IN THE SOCIALLY-NETWORKED WORLDChallenges of Emerging, Stagnant & Resurgent Markets

EDITORS TODD J. ARNOLD LISA K. SCHEERAND

2012 AMA WINTER/SUMMER EDUCATORS’ CONFERENCE PROCEEDINGS

Search THE PROCEEDINGS

HelpADOBE® ACROBAT® HELP TOPICS

2012 AMA Educators’ Proceedings

Marketing in the Socially-NetworkedWorld: Challenges of Emerging,

Stagnant, and Resurgent Markets

EditorsTodd J. Arnold, Oklahoma State UniversityLisa K. Scheer, University of Missouri

Track ChairsBrad D. Carlson, Saint Louis UniversityJune Cotte, University of Western OntarioD. Todd Donavan, Colorado State UniversityAndreas Eggert, University of PaderbornTomas Falk, EBS Business SchoolRuud T. Frambach, VU University AmsterdamDwayne D. Gremler, Bowling Green State UniversityKevin P. Gwinner, Kansas State UniversityMaik Hammerschmidt, University of GoettingenKelly L. Haws, Texas A&M UniversityChristian Hinsch, Grand Valley StateMark B. Houston, Texas Christian UniversityDouglas E. Hughes, Michigan State UniversityRobert V. Kozinets, York UniversityRussell N. Laczniak, Iowa State UniversityCait Poynor Lamberton, University of PittsburghMurali K. Mantrala, University of MissouriDetelina Marinova, University of MissouriKelly D. Martin, Colorado State UniversityMaría Pilar Martínez-Ruiz, University of Castilla-La ManchaDarrel D. Muehling, Washington State UniversityRobert W. Palmatier, University of WashingtonDoreén Pick, Freie Universitaet BerlinKeith A. Richards, University of Tennessee at ChattanoogaEdward E. Rigdon, Georgia State UniversityMatthew J. Robson, University of LeedsSaeed Samiee, University of TulsaRonn J. Smith, University of ArkansasAlina Sorescu, Texas A&M UniversityShrihari (Hari) Sridha, Pennsylvania State UniversityTracy A. Suter, Oklahoma State UniversityMichael F. Walsh, West Virginia UniversityKaren Page Winterich, Pennsylvania State UniversityAlex R. Zablah, Oklahoma State University

Volume 23

311 S. Wacker Drive • Chicago, IL 60606

© Copyright 2012, American Marketing AssociationAll rights reserved.

Printed in the United States of America

Publication Director: Andy SeagramCover Design: Kristina WaltonCompositor: Marie Steinhoff, Southeast Missouri State UniversityISSN: 0888-1839ISBN: 0-87757-350-6

No part of the material protected by this copyright notice may bereproduced or used in any form or by any means, including elec-tronic, mechanical, photocopying, recording, Web distribution,information storage and retrieval systems, or any other means,without the written permission of the American MarketingAssociation

iii

Preface and Acknowledgments

Welcome to the 2012 American Marketing Association Summer Educators’ Conferenceand to Chicago, the home of American Marketing Association. This is AMA’s 75th anniversary,so don’t miss the festivities at our special Saturday evening reception, which is co-sponsoredby three exemplars of centers of marketing thought 75, 50, and 25 years ago.

The University of Illinois represents the early years, after the AMA was founded there in1937. Representing 1962 and the subsequent era is The Ohio State University, which sawdevelopments such as the meeting which laid the foundation for the Association for ConsumerResearch. Northwestern University represents 1987 and marketing’s maturation as a rigorousdiscipline encompassing diverse research methodologies and innovations in education. Heart-felt thanks to these co-sponsors for their contributions to the reception and to the video that willdebut Saturday evening.

What is our exemplar of a center of marketing thought for 2012? Centers of marketingthought have become more diffuse and virtual. You and your network of co-researchers are, orcan be, a center of marketing thought! More than ever before, cutting-edge researchers are foundin all types of universities and in numerous countries, collaborating across vast distances andtime zones. We marketing academics thrive in the socially-networked world!

Thus, we are pleased that the theme of this conference is “Marketing in the Socially-Networked World: Challenges of Emerging, Stagnant, and Resurgent Markets.” We encourageyou to examine the range of intriguing papers exploring the implications of the rise of socialmedia. Please also note the special sessions and competitive paper sessions that address thechallenges and opportunities of emerging markets and economies in other stages of develop-ment. We have a great program featuring quality research papers and special sessions spanningdiverse topic areas and methodologies, presented by researchers from a plethora of countriesaround the globe.

We are honored to have had the opportunity to organize this conference. Our deepest thanksto our indefatigable track chairs, who were the key players in assembling the participants andcontent of the sessions. To those who submitted work and are featured at this conference, weappreciate your willingness to share your ideas in this venue. We particularly appreciate thereviewers, the unsung heroes and heroines, who are critical to ensuring quality presentations.Thanks to all others who contribute to this conference, including session chairs, sponsors,exhibitors, AMA’s academic leadership, and the many AMA staff members without whom thisconference would not occur. We must make special note of the role program manager JessicaThurmond-Pohlonski has played; thank you, Jessica, for your positive, can-do attitude andchampioning some of our crazy ideas behind the scenes at AMA.

In contrast to specialty conferences, a broad-based conference such as this one offers avenue in which all research areas and methodologies are welcome. We encourage you to takeadvantage of this characteristic and expose yourself to diverse ideas. We hope that you find youractivities here intellectually stimulating, meet old friends, form new connections, leave herewith new ideas, and have a great time in the process!

Lisa K. ScheerUniversity of Missouri

Todd J. ArnoldOklahoma State University

iv

Best of Conference Award“The Impact of Proactive Marketing at the Fuzzy Front End of Innovation”

Fiona Schweitzer

Best Paper Awards by Track

Advertising and Promotion“Ad Strategy for Multi-Ethnic Markets: The Influenceof Cosmopolitanism”

Enrique M. Becerra, Sindy Chapa,and Delonia O.Cooley

Branding and Brand Management“Do Personality Traits Influence Brand RelatedActivities in Social Networks? An Empirical Study”

Philipp A. Rauschnabel,Björn S. Ivens, and Gunnar Mau

Consumer Psychology and Behavior“How Close Brands Are Included in the Self:Psychological and Neural Processes”

Martin Reimann, Raquel Castaño,Judith L. Zaichkowsky, and Antoine Bechara

Emerging Markets“MNCs and Food Security in Emerging Markets:Provocations from India”

Susan M. Mudambi, Thomas Reardon,and Bart Minten

Ethical, Legal, Social, and Public Policy Issues“Does Corporate Social Responsibility Save Firms? AnExploration of Corporate Social Responsibility, FirmCapability, Environmental Influences, and Firm DefaultRisk”

Wenbin Sun

Global and Cross-Cultural Marketing Issues“The Effects of Product Diversification and Globaliza-ion on the Performance of Large International Firms”

Tianjiao Qiu

Innovative Marketing Technology“The Effects of Product Diversification and Globaliza-tion on the Performance of Large International Firms”

Tianjiao Qiu

Interorganizational Issues in Marketing“Managing Exclusive Channels for RelationshipEffectiveness”

Alberto Sa Vinhas

Marketing Education and Teaching Innovation“Using the RFM Model to Rank Doctoral MarketingPrograms”

Matt Elbeck and Brian A. Vander Schee

Marketing Strategy and Marketing Management“When Customers Show Divided Attitudinal Loyalty:Using Channel Intermediaries to Increase BehavioralLoyalty”

Sridhar N. Ramaswami, S. Arunachalam,and Kirti Rajagopalan

New Product Design and Development, Product Man-agement, and Entrepreneurship“The Impact of Proactive Marketing at the Fuzzy FrontEnd of Innovation”

Fiona Schweitzer

Personal Selling and Sales Management“Managing Salesforce Selling Behaviors andPerformance: The Interactive Effects of Sales ControlSystems”

C. Fred Miao and Kenneth R. Evans

Research Methods and Analytics“Effects of Store Manager Climate upon FLECommitment, Customer Loyalty and Store FinancialPerformance”

George D. Deitz, John D. Hansen,Thomas E. DeCarlo, Emin Babakus,and Kristopher J. Preacher

Retailing and Pricing“Accepting or Fighting Piracy: Can Firms ReducePiracy for Digital Media Products by OptimizingTheir Marketing?”

Felix Eggers, Alexa Burmester,Michel Clement, and Tim Prostka

Services Marketing“Does Technology Orientation Matter in TechnologyServices Organizations?”

Nacef Mouri, Maheshkumar P. Joshi,and Sidhartha R. Das

Sports Marketing“Segmenting Fans of a New Team: A Typology ofEarly Adopters”

Heath McDonald, Civilai Leckie,and Adam Karg

v

2012 AMA Summer Educators’ ConferenceList of Reviewers

ACarmen Abril, Complutense

UniversityManoj Agarwal, State University of

New York, BinghamtonUniversity

Raj Agnihotri, William PatersonUniversity

Billur Akdeniz, University of NewHampshire

Bob Akin Jr., Texas ChristianUniversity

Ulun Akturan, GalatasarayUniversity

Lumina Albert, Colorado StateUniversity

Thomas Allard, University ofBritish Columbia

Alexis Allen, Florida StateUniversity

Cesare Amatulli, University ofSalento (Lecce)

Michelle Andrews, University ofTexas at Arlington

Martin Artz, University ofMannheim

S. Arunachalam, Iowa StateUniversity

Anthony Kwame Asare,Quinnipiac University

Ioannis Assiouras, ESC ToulouseBusiness School

BImad B. Baalbaki, American

University of BeirutSally Baalbaki, University of North

TexasChristof Backhaus, Technische

Universitaet BraunschweigAndrew Michael Baker, San Diego

State UniversityCarmen Balan, Academy of

Economic Studies fromBucharest

Brian Baldus, Michigan StateUniversity

George Baltas, Athens Universityof Economics & Business

Soumava Bandyopadhyay, LamarUniversity

Gloria Barczak, NortheasternUniversity

Paul G. Barretta, University ofTexas–Pan American

M. Jesus Barroso-Mendez,Universidad de Extremadura

Darrell Bartholomew, OklahomaState University

Christopher Bartl, EBS BusinessSchool

Paige Beal, Point Park UniversityEnrique P. Becerra, Texas State

UniversityNina Belei, University of

MaastrichtAronte Bennett, Villanova

UniversityRam Bezawada, State University of

New York, BuffaloPelin Bicen, Pennsylvania State

University, ErieJochen Binder, University of St.

GallenMelissa Bishop, University of New

HampshireVera Blazevic, University of

AachenJohn Charles Blewitt, Saint Louis

UniversityChristopher P. Blocker, Baylor

UniversityMartin Boehm, IE Business SchoolElmira Bogoviyeva, KIMEP

UniversityWilly Bolander, Florida State

UniversitySterling A. Bone, Brigham Young

UniversityLeff Bonney, Florida State

UniversityTorsten Bornemann, University of

MannheimMousumi Bose Godbole, Fairfield

UniversityBrian L. Bourdeau, Auburn

UniversityPhilip Boutin, University of

TennesseeMichael K. Brady, Florida State

UniversityMichael Breazeale, Indiana

University Southeast

Tim Oliver Brexendorf, WHU–Otto Beisheim School ofManagement

Christian Brock, ZeppelinUniversity

Jacob Brower, Queen’s UniversityBrian P. Brown, Virginia

Commonwealth UniversityElisabeth Christine Brüggen,

Maastricht University

CMeltem Cakici, Gediz UniversityBrad D. Carlson, Saint Louis

UniversityBrian R. Chabowski, University of

TulsaDeepa Chandrasekaran, Lehigh

UniversityWei-Lun Chang, Tamkang

UniversityLan Chaplin, Villanova UniversitySimos Chari, University of LeedsSharmila Chatterjee, Massachusetts

Institute of TechnologyChien-Chung Chen, Stillman

CollegeI-Shuo Chen, Trinity College

DublinSteven Chen, California State

University, FullertonViviane Chen, IÉSEG School of

ManagementPilsik Choi, Clark UniversityWoo Jin Choi, Texas A&M

UniversityTilottama G. Chowdhury,

Quinnipiac UniversityGeorge Chryssochoidis, University

of East AngliaJohn E. Cicala, Texas A&M

University–KingsvilleBart Claus, IÉSEG School of

ManagementJoe Cobbs, Northern Kentucky

UniversityLaurel Aynne Cook, University of

ArkansasDelonia O. Cooley, Texas Southern

University

vi

June Cotte, University of WesternOntario

Birton Cowden, Saint LouisUniversity

Adam Craig, University of SouthFlorida

David Crockett, University ofSouth Carolina

Julia Marie Cronin-Gilmore,Bellevue University

Annie Cui, West VirginiaUniversity

Kevin J. Cumiskey, EasternKentucky University

DAntónio Pimenta da Gama,

Instituto de Artes Visuais,Design e Marketing

Steven Dahlquist, Michigan StateUniversity

Prakash Das, University of CalgaryMeredith David, Florida State

UniversityBrad Davis, Wilfrid Laurier

UniversityCassandra Davis, University of

ArkansasScott Davis, Texas A&M

UniversityAd de Jong, Eindhoven University

of TechnologyClaudio Dell’Era, Politecnico di

MilanoMbaye Fall Diallo, University Paul

CézanneAdamantios Diamantopoulos,

University of ViennaBarry Dickinson, Holy Family

UniversityJohn Bowman Dinsmore,

University of CincinnatiAndrea Dixon, Baylor UniversitySara Loughran Dommer,

University of PittsburghTodd Donavan, Colorado State

UniversityThomas Dotzel, McGill UniversityJames D. Doyle, Carleton

UniversityCourtney Droms, Butler UniversityShuili Du, Simmons CollegeLea Dunn, University of British

ColumbiaTurkan Dursun, West Texas A&M

University

Hristina Rumenova Dzhogleva,University of Pittsburgh

EPeter Eberle, Pennsylvania State

UniversityCornelia Ebertin, University of

MannheimFelix Eggers, University of

HamburgAndreas Eggert, University of

PaderbornMajid Eghbali-Zarch, Western

University CanadaA. Meike Eilert, University of

South CarolinaMaik Eisenbeiss, University of

CologneMartin Eisend, European

University ViadrinaHounaida El Jurdi, American

University of BeirutBeatrice Ermer, HHL–Leipzig

Graduate School of Management

FMartin Fassnacht, WHU–Otto

Beisheim School of ManagementMonica J. Favia, Bloomsburg

UniversityBradley Fay, Arizona State

UniversitySven Feurer, University of

MannheimMonica Fine, Coastal Carolina

UniversityAdam Finn, University of AlbertaPeter-Matthias Fischer, University

of St. GallenSabine Fliess, University of HagenJohn Ford, Old Dominion

UniversityGavin Fox, Texas Tech UniversityJohn Paul Fraedrich, Southern

Illinois University–CarbondaleGeorge Franke, University of

AlabamaElisa Fredericks, Northern Illinois

UniversityRegina Frey, German Graduate

School of Management and LawJohanna Frösén, Aalto UniversityChristoph Fuchs, Erasmus

University Rotterdam

Andreas Fuerst, University ofErlangen-Nürnberg

GLili Gai, University of North TexasAndrew Gallan, Case Western

Reserve UniversityJason Garrett, Bradley UniversityLi-Shiue Gau, Asia University,

TaiwanAnja Geigenmueller, Ilmenau

University of TechnologyKatja Gelbrich, Catholic University

Eichstaett-IngolstadtDominik Georgi, Frankfurt School

of Finance and ManagementFrank Germann, Pennsylvania

State UniversityAndreas Giese, EBS Business

SchoolErin Gillespie, University of

MississippiDavid A. Gilliam, University of

Arkansas at Little RockJohn T. Gironda, Florida Atlantic

UniversityDaniel Goebel, Illinois State

UniversityMatthias Gouthier, EBS Business

SchoolStacy Landreth Grau, Texas

Christian UniversityChristopher John Groening,

University of MissouriStephen J. Grove, Clemson

UniversityMark D. Groza, Northern Illinois

UniversityRichard Lars Gruner, University of

MelbourneStephan Grzeskowiak, Rouen

Business SchoolAditya Gupta, Pennsylvania State

UniversityShipra Gupta, University of

Nebraska, LincolnEda Gurel Atay, University of

Puget SoundAnders Gustafsson, Karlstad

University

HPerry Haan, Tiffin UniversityAlexander Hahn, University of

Mannheim

vii

Laura Hainle, University ofMannheim

Maik Hammerschmidt, GeorgAugust University Goettingen

Stephen A. Hampton, University ofMissouri

Jared M. Hansen, University ofNorth Carolina at Charlotte

Savita Hanspal, Kamala NehruCollege

Colleen Harmeling, Saint LouisUniversity

Angela Hausman, HowardUniversity

Kelly Haws, Texas A&MUniversity

Babak Hayati, University ofHouston

Sven Heidenreich, EBS BusinessSchool

Wibke Heidig, University of St.Gallen

Conor M. Henderson, University ofWashington

Stephan Henneberg, ManchesterBusiness School

Nadine Hennigs, LeibnizUniversity of Hannover

Dennis Herhausen, University ofSt. Gallen

Steffen Herm, TechnischeUniversität Berlin

Carlos Hernández-Carrión,University of Valladolid

Marc F. Herz, University of ViennaAndreas Christian Hildesheim,

University of MannheimAndrew Christian Hinsch, Grand

Valley State UniversityTanawat Hirunyawipada,

University of DaytonNga Nguyen Ho-Dac, University of

UtahCharles F. Hofacker, Florida State

UniversityJens Hogreve, University of

IngolstadtLinda Hollebeek, University of

AucklandJessica J. Hoppner, George Mason

UniversityChris Horbel, University of

BayreuthRonald Hoverstad, University of

the PacificChe-Hao Hsu, National Cheng

Kung University

Douglas E. Hughes, Michigan StateUniversity

Magnus Hultman, University ofLeeds

Shane Hunt, Arkansas StateUniversity

Gary Hunter, Case WesternReserve University

Katie Hybnerova, University ofMississippi

ISajna Ibrahim, State University of

New York, BinghamtonMonika Hannelore Imschloss,

University of MannheimPaul Ingenbleek, Wageningen

UniversityCharles Ingene, University of

MississippiChiharu Ishida, Illinois State

UniversityAlicia Izquierdo-Yusta, University

of Burgos

JBenedikt Josef Jahn, Ludwig-

Maximilians-Universität MunichClaudia Jasmand, Imperial College

LondonHongyan Jiang, Sun Yat-sen

UniversityAna I. Jimenez-Zarco, Open

University of CatalunyaYao Jin, University of ArkansasMark Johlke, Bradley UniversityCarol Johnson, University of

DenverBabu Lourduraj John Mariadoss,

Washington State UniversityChristopher Joiner, George Mason

UniversityWilliam Jones, Wayne State

UniversityYogesh Joshi, University of

MarylandJae Min Jung, California State

Polytechnic University, Pomona

KUlrike Kaiser, WU Vienna

University of Economics andBusiness

Sudhir Kale, Bond UniversityAnna Kaleka, Cardiff University

Vamsi Krishna Kanuri, Universityof Missouri

Fahri Karakaya, University ofMassachusetts Dartmouth

Constantine Katsikeas, Universityof Leeds

Eleni Kevork, Athens University ofEconomics and Business

Ceyhan Kilic, MarketingConsultant

Ilse Klanner, Vienna University ofEconomics and Business

Christiane Klarmann, LeibnizUniversity of Hanover

Martin Klarmann, KarlsruheInstitute of Technology

Michael Kleinaltenkamp, FreieUniversitaet Berlin

Susan Bardi Kleiser, TexasChristian University

Richard R. Klink, LoyolaUniversity Maryland

Peter Knight, University ofWisconsin Parkside

Yuliya Komarova, FordhamUniversity

Steven W. Kopp, University ofArkansas

Nicole Koschate-Fischer,University Erlangen–Nuremberg

Christine Kowalczyk, EastCarolina University

Alexander Krasnikov, GeorgeWashington University

Henning Kreis, FU BerlinKirk Kristofferson, University of

British ColumbiaChristina Kuehnl, University of

MannheimSabine Kuester, University of

MannheimWerner H. Kunz, University of

MassachusettsDidem Kurt, University of

Pittsburgh

LVishal Lala, Pace UniversityDesmond Lam, University of

MacauSon K. Lam, University of GeorgiaTak Ming Lam, Hong Kong

Polytechnic UniversityCatherine Lamberton, University of

PittsburghJoel Le Bon, University of Houston

viii

Kenneth Le Meunier-FitzHugh,University of East Anglia

Ju-Yeon Lee, University ofWashington

Ruby Lee, Florida State UniversitySae Rom Lee, Pennsylvania State

UniversityS.H. (Mark) Lee, Colorado State

UniversityKevin Lehnert, Grand Valley State

UniversityAlexander Leischnig, Freiberg

University of TechnologyConstantinos N. Leonidou,

University of LeedsAda Leung, Pennsylvania State

University, BerksMichael Levin, Otterbein

UniversityTheo L. Lieven, University of St.

GallenVeronica Liljander, Hanken School

of EconomicsLily Lin, University of British

ColumbiaBirgit Loehndorf, University of

ViennaDonald J. Lund, University of

Alabama at BirminghamLorman L. Lundsten, University of

St. Thomas

MMelih Madanoglu, Florida Atlantic

UniversityRamana Kumar Madupalli,

Southern Illinois University–Edwardsville

Peter Magnusson, Northern IllinoisUniversity

Igor Makienko, University ofNevada

Lucia Malär, University of BernNaresh K. Malhotra, Nanyang

Technological UniversityAneela Malik, Lahore University of

Management SciencesAvinash Malshe, University of St

ThomasRujirutana Mandhachitara,

Pennsylvania State UniversityMurali Krishna Mantrala,

University of MissouriAndre Marchand, University of

Muenster

Melissa Markley, DePaulUniversity

Kelly D. Martin, Colorado StateUniversity

Noemi Martinez-Caraballo,University of Zaragoza

María Pilar Martínez-Ruiz,University of Castilla–LaMancha

Mercedes Martos-Partal,Salamanca University

Daniele Mathras, Arizona StateUniversity

James Mark Mayer, IndianaUniversity

Regina McNally, University ofLimerick

Cathy Ferris McPherson, MaryBaldwin College

Michael Merz, San Jose StateUniversity

Frederik Meyer, JohannesGutenberg-University of Mainz

Fred Miao, Clarkson UniversityClaudia Mich, Purdue University

CalumetStefan Michel, IMDJoseph Calvin Miller, Rochester

Institute of TechnologyKlaus M. Miller, University of

BernJunhong Min, Michigan

Technological UniversityKyeong Sam Min, University of

New OrleansAnubha Mishra, University of

IdahoSaurabh Mishra, McGill UniversityShashi Shekhar Mishra, Indian

Institute of TechnologyMayoor Mohan, Oklahoma State

UniversityGina S. Mohr, Colorado State

UniversityBruce Money, Brigham Young

UniversityFred Morgan, Wayne State

UniversityAlexander Josef Mrozek, EBS

Business SchoolJifeng Mu, Alabama A&M

UniversityDarrel Muehling, Washington State

UniversityEva Muenkhoff, University of

Paderborn

Michael R. Mullen, FloridaAtlantic University

Ryan Mullins, Clemson UniversityJames Muncy, Valdosta State

UniversityMatthew Bowman Myers,

University of Tennessee

NMohammed M. Nadeem, National

UniversityIman Naderi, University of North

TexasJacques Nel, University of the Free

StateGergana Nenkov, Boston CollegeChristopher Newman, University

of ArkansasJeananne Nicholls, Slippery Rock

UniversityTim Nicolas Nierobisch, University

of Göttingen, GermanyPatricia Norberg, Quinnipiac

University

OMike W. Obal, Temple UniversityJoon-Hee Oh, Georgia State

UniversityDouglas Olsen, Arizona State

UniversityFred Oswald, Rice UniversityJan Owens, Carthage CollegePeren Ozturan, Koç University

PKadayam H. Padmanabhan,

University of Michigan–Dearborn

Nicolas Papadopoulos, CarletonUniversity

Dominik Papies, University ofHamburg

Jeong Eun Park, Ewha WomansUniversity

Jieun Park, Cleveland StateUniversity

Audhesh K. Paswan, University ofNorth Texas

Iryna Pentina, University of ToledoMaria Petrescu, Barry UniversityAdrienne Hall Phillips, Worcester

Polytechnic Institute

ix

Joan M. Phillips, Loyola UniversityChicago

Doreen Pick, Freie UniversitaetBerlin

Wesley Pollitte, University ofSouthern Mississippi

Sonja Martin Poole, University ofSan Francisco

Jana Prigge, University ofMannheim

Elizabeth F. Purinton, MaristCollege

QTianjiao Qiu, California State

University, Long BeachVictor Quinones, University of

Puerto Rico

RBrendan G. Rafferty, Sacred Heart

UniversitySascha Raithel, Ludwig-

Maximilians-UniversitätMünchen

Arilova Randrianasolo, Saint LouisUniversity

Deva Rangarajan, Vlerick LeuvenGent Management School

Adam Rapp, University ofAlabama–Tuscaloosa

Justine Rapp, University ofNebraska–Lincoln

David Raska, Northern KentuckyUniversity

Pradeep A. Rau, GeorgeWashington University

Philipp Rauschnabel, University ofBamberg

Steven W. Rayburn, OklahomaState University

Petra Riefler, University of ViennaChristine Ringler, Arizona State

UniversityTerri Rittenburg, University of

WyomingThomas Ritter, Copenhagen

Business SchoolCarlos M. Rodriguez, Delaware

State UniversityMichael Rodriguez, Elon

UniversityWilliam Taylor Ross Jr., University

of Connecticut

Marty Roth, University of SouthCarolina

Dominique Rouzies, HEC ParisAbhik Roy, Quinnipiac UniversityRosalyn J. Rufer, State University

of New York, Empire StateCollege

Pablo Ruiz-Palomino, Universityof Castilla–La Mancha

SAlok Saboo, Pennsylvania State

UniversityTakisha Salley-Toler, Saint Louis

UniversitySaeed Samiee, University of TulsaMatthew Sarkees, West Virginia

UniversityMarko Sarstedt, Ludwig-

Maximilians-University MunichMichael Anthony Sartor,

University of Western OntarioCinthia B. Satornino, Florida State

UniversityDaniela B. Schäfer, University of

BaselTobias Schäfers, EBS Business

SchoolTammy Schakett, Ohio Northern

UniversityDebbie Schallock, University of

North Carolina at GreensboroJeroen Schepers, Eindhoven

University of TechnologyKristina Schmidt, WHU–Otto

Beisheim School of ManagementPhilipp Schmitt, Goethe University

FrankfurtMonika Schuhmacher, University

of MannheimJan Hendrik Schumann,

Technische Universität MünchenMarko Schwertfeger, Freiberg

University of TechnologyKristin Scott, Minnesota State

University, ManaktoAlexey V. Semenov, Saint Louis

UniversityLaura Serviere-Munoz, University

of North Texas at DallasArun Sharma, University of MiamiPiyush Sharma, Hong Kong

Polytechnic UniversitySohyoun Shin, Eastern Washington

University

Christina Sichtmann, University ofVienna

Shir Way Siew, University ofTexas–Pan American

Debika Sihi, University of Texas atAustin

Bonnie Simpson, University ofCalgary

Ardhendu Shekhar Singh, Instituteof Rural Management, Anand

Ramendra Singh, IIM CalcuttaHotniar Siringoringo, Gunadarma

UniversitySanjay Sisodiya, University of

IdahoEugene Sivadas, University of

Washington, TacomaDionysios Skarmeas, Athens

University of Economics &Business

Jeffery Smith, Florida StateUniversity

Michael Smith, Temple UniversityRonn J. Smith, University of

ArkansasRobin Soster, University of

ArkansasAnne Souchon, Loughborough

UniversityCarlos M.P. Sousa, Durham

UniversityNicola Spiller, Politecnico di

MilanoAmanda Catherine Spry,

University of Melbourne,Australia

Srinivas Sridharan, MonashUniversity

Axel Stock, University of CentralFlorida

Barbara Stoettinger, WU ViennaNicola Stokburger-Sauer,

University of InnsbruckAmy Stokes, Oregon State

UniversityAndreas Strebinger, York

UniversityNadine Stutz, Institution for

Communication and MarketingLishan Su, Iowa State UniversityUrsula Y. Sullivan, Northern

Illinois UniversityWenbin Sun, Rockhurst UniversityRetno Tanding Suryandari,

University of North Texas

x

Esther Swilley, Kansas StateUniversity

TM. Berk Talay, University of

Massachusetts LowellYihui (Elina) Tang, University of

Illinois at ChicagoEmily Tanner, Oklahoma State

UniversityCrina O. Tarasi, Central Michigan

UniversityDavid G. Taylor, Sacred Heart

UniversityKarin Teichmann, University of

InnsbruckSahure Gonca Telli, Okan

UniversityHarri Terho, University of TurkuMarios Theodosiou, University of

CyprusAristeidis Theotokis, University of

LeedsChristoph Thiesbrummel,

University of PaderbornPeter Thirkell, Victoria UniversityCarri R. Tolmie, Saint Louis

UniversityDirk Totzek, University of

MannheimClaudia Townsend, University of

MiamiGina A. Tran, University of North

TexasRodoula H. Tsiotsou, University of

Macedonia, GreeceAnna M. Turri, Texas State

University–San MarcosSven Tuzovic, Pacific Lutheran

University

UWolfgang Ulaga, IMD Lausanne

VBrian A. Vander Schee, Aurora

UniversityEric Van Steenburg, University of

North TexasDaniel J. Veit, University of

Mannheim

Alladi Venkatesh, University ofCalifornia

Nicole Verrochi, University ofPittsburgh

Handan Vicdan, EMLYONBusiness School

Akshaya Vijayalakshmi, Iowa StateUniversity

Jorge Villegas, University ofIllinois at Springfield

Goran Vlasic, University ofZagreb/University of Sussex

Arnd Vomberg, University ofMannheim

Kevin Voss, Oklahoma StateUniversity

WMichael Wachter, Cleveland State

UniversityL. Jean Walker, University of

Houston–Clear LakeMichael Francis Walsh, West

Virginia UniversityChen Wang, University of British

ColumbiaDanielle Way, Woodbury

UniversityJack Wei, University of West

GeorgiaYinghong (Susan) Wei, Oklahoma

State UniversityBert Weijters, Vlerick Leuven Gent

Management SchoolHauke A. Wetzel, Georg August

University GoettingenAllyn White, Eastern Kentucky

UniversityD. Steven White, University of

Massachusetts DartmouthRyan White, University of

Wisconsin–La CrosseAnne Wiese, University of

GoettingenJennifer Wiggins Johnson, Kent

State UniversityMichael Wiles, Arizona State

UniversityRicky Wilke, Copenhagen

Business SchoolRobert Wilken, ESCP Europe

Business School Berlin

Karen Page Winterich,Pennsylvania State University

Kristina Wittkowski, EBS BusinessSchool

David M. Woisetschläger,Technische UniversitätBraunschweig

Jeremy Wolter, Florida StateUniversity

Charles Wood, University of TulsaJohn Andy Wood, West Virginia

UniversityScott Wright, University of

CincinnatiNancy Wuenderlich, Technische

Universität München

XGuang-Xin Xie, University of

Massachusetts Boston

YChun-Ming Yang, Ming Chuan

UniversityJun Yang, University of Houston

VictoriaAtefeh Yazdanparast, University of

EvansvilleMujde Yuksel, University of

Massachusetts Amherst

ZAlex Ricardo Zablah, George

Mason UniversityMuhammad Abrahim Zaka,

Nyenrode Business SchoolLia Zarantonello, Catholic

University of LilleHaisu Zhang, University of Illinois

at ChicagoShuoyang Zhang, Colorado State

UniversityChen Zhou, Pennsylvania State

UniversityYimin Zhu, Sun Yat-sen UniversityStephan Zielke, Aarhus UniversityMohammadali Zolfagharian,

University of Texas–PanAmerican

Yueming Zou, Old DominionUniversity

xi

TABLE OF CONTENTS

PREFACE AND ACKNOWLEDGMENTS iii

BEST PAPERS BY TRACK iv

LIST OF REVIEWERS v

TABLE OF CONTENTS xi

RESEARCH METHODS AND ANALYTICS

The Application of Structural Equations to the Attributes in Discrete ChoiceModels

Cam M. Rungie 1

Anchoring in Choice Experiments: Influence of the Experimental Design onWillingness-to-Pay

Felix Eggers, Franziska Völckner 10

The Dynamics Underlying Performance Signaling: Insights from the ProfessionalSports Market

Stefan Hattula, Hauke A. Wetzel, Maik Hammerschmidt, Hans H. Bauer 12

Effects of Store Manager Climate Upon FLE Commitment, Customer Loyalty, andStore Financial Performance

George D. Deitz, John D. Hansen, Thomas E. DeCarlo, Emin Babakus,Kristopher J. Preacher 14

Do Group-Buying Deals Induce More Coupon Regret?Xueming Luo, Yiping Song, Pengdong Fan 16

Distribution of Changes in Consumer Purchasing BehaviorGiang Trinh, Cam Rungie, Malcolm Wright, Carl Driesener, John Dawes 17

INTERORGANIZATIONAL ISSUES IN MARKETING

Are Suppliers Inaccurate in Judging Their Price Fairness in Buyer-SellerRelationships?

Christian Homburg, Jan Allmann, Dirk Totzek 19

Gratitude Versus Entitlement: An Antagonistic Process Model of theProfitability Impact of Customer Prioritization

Hauke A. Wetzel, Maik Hammerschmidt, Alex R. Zablah 21

Do Seller Perceptions of Fairness Lead to Sales Growth? A Latent Growth CurveAnalysis

Ghasem Zaefarian, Zhaleh Najafi Tavani, Stephan C. Henneberg,Peter Naudé 23

Reciprocity in Interfirm Relationships: A Review and ExtensionJessica J. Hoppner, David A. Griffith 25

Sales and Value Creation: A Synthesis and Directions for Future ResearchAlexander Haas, Nina Stuebiger 27

xii

Implementing Service Growth Strategies at the Industrial Sales Force Level: KeyChallenges in Selecting and Managing the Service-Savvy Sales Force

James M. Loveland, Wolfgang Ulaga 29

Ready to Pitch? Proposal and Vendor Attractiveness as Mediators of Vendor Successin Competitive Tenders

Eva K. Steinbacher, Christian Schmitz, Dirk Zupancic 31

Managing Exclusive Channels for Relationship EffectivenessAlberto Sa Vinhas 33

EMERGING MARKETS

Are Chinese State-Owned Enterprises Lagging Behind in Product Innovation?Xiaomin Zhao, Ping Lan 34

From Innovative and Marketing Capabilities to Firm Performance: EmpiricalComparison on Different Paths in Two Emerging Nations

Xina Yuan, Sohyoun Shin, Sang Yong Kim 36

MNCs and Food Security in Emerging Markets: Provocations from IndiaSusan M. Mudambi, Thomas Reardon, Bart Minten 37

Adding to the “Missing Link” Perspective in Emerging Economy: The Role ofProduct Innovation in Russia

Maria Smirnova, Vera Rebiazina, Alexander Krasnikov, Sergey Kusch 39

Evaluation of Retail Services: A Developed vs. Emerging Markets PerspectivePiyush Sharma, Sherriff T.K. Luk, Ivy S.N. Chen 40

Factors Influencing Growth Potential of E-Commerce in Emerging Economies:A Multi-Theoretical Approach and Research Propositions

James Agarwal, Terry Wu 42

Driving the Value Premium Through Cultural, Symbolic, Economic and SocialCapital Management

Goran Vlasic, Josef Langer, Zoran Krupka 43

Microfinance Market Failures in Emerging MarketsJoseph Hansen-Addy, Esi Abbam Elliot, Joseph Cherian 45

Reactance Vs. Acceptance: Emerging Market Consumers’ Perceptions of LocalBrands After an MNC’s Acquisition

Martin Heinberg, Markus Taube 46

Rethinking Marketing Practices in an Emerging Market Context: An EmpiricalEvaluation of Competing Conceptual Frameworks in Ghana

Kofi Q. Dadzie, Charlene A. Dadzie, Evelyn M. Winston 47

Expanding “Brotherhood” in Emerging Markets: Methodological Approach andCultural Value Analysis

Richard Michon 49

Still on the Road to Capitalism? Weighing the Visible Hand of GovernmentIntervention in the Chinese Property Market

Tao Zhu, Killian J. McCarthy 51

xiii

INTERORGANIZATIONAL ISSUES IN MARKETING

Studying Emotions in Business RelationshipsAndreas Zehetner 63

Interorganizational Favor Exchange as a Cooperative BehaviorAdam Nguyen 72

Using Dyadic Agent-Based Simulations to Model Strategic Decision Making inBusiness Relationships

Sebastian Forkmann, Di Wang, Stephan C. Henneberg, Peter Naudé,Alistair Sutcliffe 74

Reinstitutionalizing Surveillance Through Social Media: Healthcare Market CaseHandan Vicdan 76

The Mediating Role of Marketing Process Improvement in the Market-Based AssetFramework

Anthony K. Asare, Thomas G. Brashear, Jing Yang, Jun Kang 78

How Integration Enables Marketing to Become More AccountableAndreas Waschto, Malte Brettel 79

Why Do Manufacturers Engage in Private Labels Production? Market Strategyand Channel Relationship Perspectives

Ho-Taek Yi, Chae-Un Lim 81

ADVERTISING, PROMOTION AND MARKETING COMMUNICATIONS

Quantifying the Sales Impact of Location-Targeted Mobile AdsXueming Luo, Zheng Fang, Megan E. Keith 90

How Advertising Works Embedded in New Media: Consumer Media ExperienceModel

Mark Yi-Cheon Yim, Minette E. Drumwright, Vincent J. Cicchirillo 92

Viral Advertising and Its Place in the Advertising FrameworkMaria Petrescu, Pradeep Korgaonkar, Tamara Mangleburg, Ann Root 94

The Role of Facebook for Advertising: Advertising Effectiveness of Social NetworksCompared to Traditional Online Advertising, Including Synergies and Time Lags

Jens-Christian Reich, Malte Brettel 96

Buying the Forthcoming: A Prelaunch Information and Value Congruence ModelKyung-Ah Byun, Junghwan Kim, Gavin L. Fox 98

The Role of Financial Gift Formats on the Persuasiveness of the AdvertisingMessages

Qing Yao, Rong Chen 100

A Framework to Understand Customer Data Quality in CRM Systems for FinancialServices Firms

Debra Zahay-Blatz, James Peltier, Anjala Krishen 102

The Role of Fit and Similarity in Social Sponsorship CommunicationsRavi Pappu, T. Bettina Cornwell 103

xiv

Should Firms Name Competitors in the Ads?Chun-Kai Tommy Hsu, Leona Tam 105

Regulatory Focus and Daily Deal Message Framing: Are We Saving or Gainingwith Groupon?

Iryna Pentina, David G. Taylor 107

An Investigation of the Attitude Resistance Process on Negatively FramedComparative Ads

Russell Laczniak, Kristine Ehrich, Darrel Muehling, Akshaya Vijayalakshmi 109

The Effect of Analytic and Holistic Thinking on Consumers’ Attitudes’ TowardHolistic or Attribute Advertising

Beichen Liang 111

Using Self- Versus Other-Benefit Messages in Ads for Green Products: TheModerating Role of Perceived Consumer Effectiveness and Consumer Guilt

Ceren Ekebas, Kiran Karande 113

The Impact of Violent Humor on Advertising Success: A Gender PerspectiveKunal Swani, Marc Weinberger 115

Ad Strategy for Multi-Ethnic Markets: The Influence of CosmopolitanismEnrique P. Becerra, Sindy Chapa, Delonia O. Cooley 117

Stronger Environmental Norms Increase “Green” Buying Intentions but NotBehavior: Implications for Advertisers

Melissa M. Bishop, Nelson A. Barber 119

BRANDING AND BRAND MANAGEMENT

#Loweshatesmuslims: Consumer Ethical Decision Making Related to CorporateAdvertising Choices

R. Nicholas Gerlich, Kristina Drumheller, Emily Kinsky, Meagan Brock,Marc Sollosy 121

Brand Heritage in the Luxury Industry: Creating and Delivering Continuous Valueto Consumers

Nadine Hennigs, Steffen Schmidt, Thomas Wuestefeld, Klaus-Peter Wiedmann 129

Loss of Brand Equity in Crises: The Impact of Emotions and Attributions in Productand Non-Product Negative Publicity

Haodong Gu, Ashish Sinha 131

Investigating How Word-of-Mouth Drives Information Search for a BrandAndrew M. Baker, Naveen Donthu, V. Kumar 133

Employees as Internal Brand Consumers: the Nature of RelationshipsLucy Gill-Simmen, Andreas B. Eisingerich 135

To Be Familiar or to Be There? Brand Familiarity, Social Presence, and OnlineRetailing

Xiaoping Fan, Feng Liu, Jia Zhang 137

Do Personality Traits Influence Brand Related Activities in Social Networks? AnEmpirical Study

Philipp A. Rauschnabel, Björn S. Ivens, Gunnar Mau 139

xv

Identity Presentation and Conflict in a Social Network: Implications for Brandingin Social Games

Ginger Killian, John Hulland 141

Risk Reducing Portfolio Effects: Uncertainty Avoidance in Brand ExtensionsMatthew A. Hawkins, Jatinder Jit Singh 143

Sub-Branding Affect Transfer: The Role of Product Category Crowdedness andBrand Loyalty

Yi He, Qimei Chen, Ruby P. Lee, Leona Tam 150

Explicit and Implicit Measures of Brand Information Processing and Its Impacton Brand Perception and Brand-Related Behavior

Sascha Langner, Nadine Hennigs, Steffen Schmidt, Klaus-Peter Wiedmann 152

Diminishing Effects of Perceived Fit on Vertical ExtensionsNicolas Pontes, Mauricio Palmeira, Colin Jevons 154

Managing Brand Architecture: State-of-the-Art and Decision-MakingChristopher Kanitz, Michael Schade, Christoph Burmann 156

Branding with Political Theories: An Interdisciplinary ApproachTony Yan 158

Constructing Bayesian Network and Nomological Network of Performance-BasedUsability of Mobile Devices

Taewon Suh, John Ford, Jang-Ho Park, Kyungdoc Kim, Chi-Hyuck Jun 165

Investigating the Effects of Vagueness in Advertising Slogans on Brand PreferenceWidyarso Roswinanto 167

CONSUMER PSYCHOLOGY AND BEHAVIOR

Consuming Ethically: The Role of EmotionsDiana Gregory-Smith, Andrew Smith, Heidi Winklhofer 169

Does In-Store Marketing Reduce the Attitude-Behavior Gap for Ethical Products?Gulberk Keysan, Vishal Talwar, Amitav Chakravarti 170

Understanding Teenage Poker Gambling: Policy and Consumer BehaviorImplications

Sudhir H.Kale, Natalina Zlatevska 172

“It’s Not My Fault” and “Money Can Buy Anything”: Recovering from EthicalFailures

Sekar Raju, Priyali Rajagopal 174

Word-of-Mouth Behavior, Online Activity and Company’s Communication StrategyDanny P. Claro, Silvio A. Laban Neto, Priscila B.O. Claro 176

The Role of Resistance and Positive Coping in Consumption of Mobile AppsAnubha Mishra, Mary Ann Eastlick 178

Strategies for Increasing Online Survey ParticipationMary Foster, Anne Warner, Adam Froman 180

xvi

Too Much Facebook: An Exploratory Examination of Social Media FatigueStacy Landreth-Grau, Laura F. Bright, Susan Bardi Kleiser 182

On Materialism, Coping and Consumer Behaviors Under Traumatic StressAyalla A. Ruvio, Eli Somer 184

The Green Side of MaterialismPia Furchheim, Steffen Jahn, Cornelia Zanger 186

Subjective Well-Being and its Influence on Consumer Purchase Behavior of LuxuryItems Do Materialism and Affective Autonomy Have an Impact?

La Toya M. Russell 188

The Impact of Esoteric Product Specifications on ChoiceChelsea Wise, Paul F. Burke, Sandra J. Burke 194

Conformity, Uniqueness, and Social Class Mobility Consumer Self Identity in theYard

Elizabeth C. Hirschman, David L. Kendall 195

How Close Brands Are Included in the Self: Psychological and NeuralProcesses

Martin Reimann, Raquel Castaño, Judith L. Zaichkowsky, Antoine Bechara 205

The Referral Backfire Effect: The Identity Threatening Nature of Referral FailureBart Claus, Kelly Geyskens, Kobe Millet, Siegfried Dewitte 206

Self-Other Differences in Purchase Uncertainty and Contingent Decision StrategiesDemetra Andrews, Stephanie Oneto 208

Corporate Social Responsibility and Consumer Relationships: An Application ofthe Motivator-Hygiene Theory

Russell Lacey, Pamela Kennett-Hensel 210

A New Tool for Customer Segmentation: Defining and Measuring RelationshipProneness

Stephanie M. Mangus, Jacob L. Hiler, Benjamin D. McLarty 212

Determinants of Customer Loyalty: The Role of Relational Benefits in the Contextof Customer Club

Kevin Yu, Song Yang 213

Turning a Lion into a Kitten: The Role of Apology Timing and RelationalExpectation

Kyeong Sam Min, Jae Min Jung, Kisang Ryu 222

The Impact of Parent-Child Buying Communication Strategies on Self-BrandConnections

Farrah Arif, Wayne D. Hoyer, Omar Merlo 224

Making Brand Associations and Brand Elements Memorable in Elderly ConsumersPraggyan (Pam) Mohanty, S. Ratneshwar, Moshe Naveh-Benjamin 226

On the Interpersonal Transmission of LuckChun-Ming Yang, Edward Ku, Chung-Chi Shen 228

xvii

Personal and Historical Nostalgia: Investigating Consumption Consequences ofTwo Different Types of Nostalgia

Tina Kiessling, Steffen Jahn, Cornelia Zanger 230

When I Can’t Touch: An Examination of the Role of Purchase Involvement andRegulatory Orientation in Online Shopping Decisions

Atefeh Yazdanparast, Nancy Spears 231

Differences in the Influence of Choice Confidence and Outcome Quality onSatisfaction as a Function of Information Diagnosticity

Demetra Andrews, Alexis M. Allen, Edward Blair 233

Self-Efficacy and Self-Prophecy Effects on Preventive Health BehaviorCarmen-Maria Albrecht, Hans H. Bauer, Kai Bergner, Tamara Gogia,Daniel Heinrich, David E. Sprott 235

The Role of Goal Specificity in Initial and Subsequent Decisions RegardingConsumption of Healthy or Unhealthy Foods

Meredith E. David, William O. Bearden 236

Checkout Frustration: Investigating Consumer Reactions Using FrustrationTheory

Eric Van Steenburg, Nancy Spears, Robert O. Fabrize 238

Antecedents and Consequences of Situated Learning in Stressful ServiceExperiences: A Cross-Cultural Perspective

Mousumi Bose, Lei Ye 240

The Pursuit of Extraordinary Experiences: An Exploration of TriggersColleen Harmeling, Mark Arnold 242

Consumer Goal Dynamics: An Empirical IllustrationShilpa Iyanna 244

ETHICAL, LEGAL, SOCIAL, AND PUBLIC POLICY ISSUES

CSR-Related Communication in Different Industries: A Qualitative andQuantitative Study Based on Corporate Annual Reports

Thomas Kilian, Nadine Hennigs 245

Does Corporate Social Responsibility Save Firms? An Exploration of CorporateSocial Responsibility, Firm Capability, Environmental Influences, and FirmDefault Risk

Wenbin Sun 247

Communicating Corporate Social Responsibility Using Social Media: Implicationsfor Marketing Strategists

Amy Lyes, Nitha Palakshappa, Sandy Bulmer 249

Customer Outcomes of Corporate Social Responsibility in Supplier-CustomerRelationships

Christian Homburg, Marcel Stierl, Torsten Bornemann 257

Measuring Consumer Confusion Evoked by Design Copycats: The Influence ofEvaluation Mode

Steffen Herm, Jana Möller 259

xviii

Consumers View Nutrition Facts Through Cultural LensesAnnie Cui, M. Paula Fitzgerald, Karen Russo France 260

Willing to Pay for a Better Brand: Consumer Responses to CSR PerformanceScores

Laurel Aynne Cook, Ronn J. Smith, Yao Jin 262

Environmentally Friendly Consumer Behavior: A Scale Review, Modification, andValidation

Lynn Sudbury-Riley, Florian Kohlbacher, Agnes Hofmeister 264

Afro-Descendents in Children’s Television Advertisements: A Content AnalysisStudy in the Brazilian Context

Claudia Rosa Acevedo, Marcello Muniz, Jouliana Jordan Nohara 266

How Companies Should React on Social Misconducts: The Role of In- and Extra-Domain Compensation

Isabell Lenz, Hauke Wetzel, Maik Hammerschmidt 268

How Big a Concern? Privacy, Targeted Advertising and Social Media: TheCanadian Corporate and Consumer View

Avner Levin, Bettina West, Mary Foster 270

Developing a Corporate Social Responsibility Process Scale of IndividualStakeholder’s Perception

Seongjin Kim, Clarissa Chaiy, Seoil Chaiy 271

NEW PRODUCT DESIGN AND DEVELOPMENT, PRODUCTMANAGEMENT AND ENTREPRENEURSHIP

The Roles of Marketing and Sales in New Product DevelopmentWim G. Biemans, Avinash Malshe 281

Sales Integration and Its Performance Effects in New Product Development:The Moderating Role of Innovativeness

Malte Brettel, Andreas Waschto 283

Resources in New Product Development: An Investigation of Resource Flexibility,Augmentation, and Leveraging

Sanjay R. Sisodiya, Jean L. Johnson, Stephen M. Wagner 285

Incremental Service Innovation, Service Employees’ Innovative Activities, and theEnabling Factors

Jun Ye 287

Examination of the Link Between New Product Preannouncements and FirmValue: The Case of the U.S. Automotive Industry

M. Billur Akdeniz, M. Berk Talay 289

Learning Marketing Management and Leadership Practices from Peers: The Caseof Industry Peer Networks

Ada Leung, Huimin Xu, Kyle Luthans, Susan Jensen 290

The Impact of Proactive Marketing at the Fuzzy Front End of InnovationFiona Schweitzer 292

xix

I Don’t Need It and I Don’t Want to – Examining Effects of Situational andCognitive Resistance to Innovations

Sven Heidenreich, Matthias Handrich 294

RETAILING AND PRICING

The Pricing of Soft and Hard Information: Lessons from Screenplay SalesMilton Harris, S. Abraham Ravid, Suman Basuroy 296

The Entrant’s Pricing Dilemma: Linking Low Price Entry Strategies to AggressivePrice Responses by Incumbents

Alexander J. Mrozek, Tomas Falk 297

Accepting or Fighting Piracy: Can Firms Reduce Piracy for Digital Media Productsby Optimizing Their Marketing?

Felix Eggers, Alexa Burmester, Michel Clement, Tim Prostka 299

Framing Effects on Evaluation of Coupon Offers: Assessing the Performance ofAlternative Face Value Frames

George Baltas, Grigorios Painesis, Paraskevas Argouslidis 301

I Thought It Was All Over and Now It Is Back Again: Customer Reactions to TimeExtensions of Sales Promotions

Ina Garnefeld, Eva Muenkhoff, Andrea Bruns 303

Purchase Behavior and Psychophysiological Responses to Different Price LevelsOuti Somervuori, Niklas Ravaja 305

Social Commonalities and Subjective DiscountingMark S. Rosenbaum, Richard Wozniak, Carolyn Massiah 307

Price-Quality Relationships and the Price HeuristicMark J. Kay 309

Product Masking: Effects of Consumer Embarrassment on Shopping Basket Sizeand Value

Bridget Satinover Nichols, David Raska, Daniel J. Flint 311

Examining the Role of Cognitive Dissonance After Purchase on Product ReturnIntentions

Devdeep Maity 313

Assessing Customer Satisfaction and Unplanned Buying Behavior in GroceryRetailing

María Pilar Martínez-Ruiz, Alicia Izquierdo-Yusta, Phil Megicks 315

Flow Experience in Physical Retail StoresLiz C. Wang, Daniel F.J. Hsiao 317

Effects of Illumination on Store Atmosphere, Price, and Quality Perception, andShopping Intention

Stephan Zielke, Thomas Schielke 318

The Hybrid Online Shopping EnvironmentJeffrey Hu Xie 320

xx

Temporary Stores: Consumer Responses to a Limited Assortment and LimitedTime Horizon

Doreén Pick, Marko Schwertfeger 322

Do Customer Experiences Enhance Retail Brand Equity?Alexander Leischnig, Marko Schwertfeger 323

Antecedents of Primary Store Disloyalty in a Stagnant MarketPeter J. McGoldrick, Daniel P. Hampson 325

Signaling E-Tailer Trust: Exploring Source Influences of Internet TrustmarksK. Damon Aiken, Vincent Pascal, Sohyoun Shin 326

A Comparison of Brand Extensions for Private Labels and National BrandsMonica Grosso, Sandro Castaldo 328

MARKETING STRATEGY AND MARKETING MANAGEMENT

An Experimental Examination of Buyers’ Responses to Relationship FailuresDonald J. Lund, John D. Hansen, Thomas E. DeCarlo, Lisa K. Scheer 330

When Customers Show Divided Attitudinal Loyalty: Using Channel Intermediariesto Increase Behavioral Loyalty

Sridhar N. Ramaswami, S. Arunachalam, Kirti Rajagopalan 332

Customer Reactions to Massive Workforce Reductions: When Is SatisfactionAffected?

Martin Klarmann, Christian Homburg, Johannes Habel 334

A Customer-Centric Approach to B2B Market SegmentationYihui (Elina) Tang, Murali Mantrala, Esther Thorson 336

A Stratified Examination of the Drivers and Outcomes of Marketing Cross-Pollination

Jeff S. Johnson, Amit Saini 338

The Role of Franchisee Marketing Commitment on Promotion Execution andPerformance

Joseph Matthes, Amit Saini 340

Determinants of the Adoption of a Mobile Commerce Strategy: The PerceivedBenefits to the Firm

Esther Swilley 342

The Moderating Effect of Creativity on Consensus and Performance: A ModeratedPolynomial Regression Model

Linda M. Orr, Frederik Beuk, Hyeong-Gyu Choi 343

Does CEO Tenure Really Matter? The Mediating Role of Employee and CustomerRelations

Xueming Luo, Vamsi K. Kauri, Michelle Andrews 345

When the Fog Dissipates: The Choice of Strategic Emphasis in a PartnerTodd Morgan, Sergey Anokhin 347

xxi

Why Quality May Not Always Win: The Impact of Product Generation Life-Cycleson Quality and Network Effects in High-Tech Markets

Richard T. Gretz, Suman Basuroy 349

GLOBAL AND CROSS-CULTURAL MARKETING ISSUES

Foreign Direct Investment Impact on Retail StructureBoryana V. Dimitrova, Bert Rosenbloom 351

The Differential Effect of Internet Trustmark Source: An Exploratory Study ofKorean and U.S. Consumers

Sohyoun Shin, K. Damon Aiken, Vincent J. Pascal 353

Market Orientation and Business Performance in MNC Foreign Subsidiaries: TheModerating Effects of Integration and Responsiveness

Riliang Qu 354

Emergence of Online Shopping in India: Shopping Orientation SegmentsKenneth C. Gehrt, Mahesh N. Rajan, G. Shainesh, David Czerwinski,Matthew O’Brien 355

It’s Nice to Be Important, But It’s More Important to Be Nice: Country-of-OriginEffects in Product Failures

Huimin Xu, Ada Leung, Terry Yan 357

The Effects of Product Diversification and Globalization on the Performance ofLarge International Firms

Tianjiao Qiu 359

What Happens to “Brand Japan” When Toyota Suddenly Accelerates? TheSpillover Effects of Brand Transgressions on Country Image and Related Brands

Peter Magnusson, Vijaykumar Krishnan, Srdan Zdravkovic,Stanford A. Westjohn 360

The Real Mccoy: Product Ethnicity, Authenticity, and Cosmopolitanism inEvaluations of Search and Experience Goods

Michael Wachter, Jieun Park 362

The Hofstede Paradox: Fact or Fiction?Terry Clark, Songpol Kulviwat, Juan (Gloria) Meng 364

Ethnocentrism, Consumer Ethnocentrism, Religion and Religiosity: A ConceptualReformation

Daniel Friesen, Attila Yaprak 365

Consumers’ Global Identity and National Identity: An Empirical StudyJoyce X. Zhou, Nitish Singh, Jun Yu 367

INNOVATIVE MARKETING TECHNOLOGY

The Effect of Blog Interactivity and Perceived Trust on Visitor Response: TheModerating Role of Blogger Expertise and Consumer Involvement

Yueming Zou, Kira Karande 368

How Quality of Life Affects Social Networking Site Use Intention: Role ofPerceived Self Disclosure and Social Tie Quality

Chung-Chi Shen, Jyh-Shen Chiou, Chun-Ming Yang, Szu-Yu Chou 370

xxii

Factors Affecting Online Trust in Online Shopping: The Role of NetworkExternality and Internet Skill

Kazuhiro Kishiya, Nao Yamamoto 371

Effects of Personalized E-Mail Messages on Perceived Risk: Moderating Roles ofControl and Intimacy

Sung-Won Lee, Ji Hee Song, Hye Young Kim 379

Linking the Virtual World with the Real World: How to Build Online RelationshipsThat Lead to Offline Interactions

Sukanya Seshadria, Werner H. Kunz 381

101 People Like This: Evaluating the Facebook Message Strategy Effectiveness ofFortune 500 Companies

Kunal Swani, George R. Milne, Brian P. Brown 382

Does Social Media Matter for Marketing? The Effects of Social Media Engagementon the Consumer-Brand Relationship

Benedikt Jahn, Werner H. Kunz 384

Evaluation and Authentication of Music Shared Through Social Networking:eWOM of Cultural Products

Paul G. Barretta, Michael S. Minor 385

Social Networking Sites and Planned BehaviorJohn T. Gironda, Pradeep K. Korgaonkar 387

What Consumers Really Think of Targeted Online Advertising: A SegmentedApproach

Bettina West, Avner Levin, Mary Foster 389

Engaging the Facebook User to Foster Sales: Drivers of Advertising Effectivenessin Social Networks, Incorporating Synergies and Time Lags

Jens-Christian Reich, Malte Brettel 391

Social Media Marketing Intensity and Its Impact on the Relation Between DynamicCapabilities and Business Performance

Felipe Uribe, Josep Rialp, Joan Llonch, Henry Robben 393

Netnography and Metaphysical BrandingSteve Oakes, Noel Dennis, Helen Oakes 395

Integration vs. Regulation: What Really Drivesg User-Generated Content in SocialMedia Channels?

Welf Weiger, Hauke Wetzel, Maik Hammerschmidt 397

Is There a Dark Side to Customer Co-Creation? Exploring Consequences of FailedCo-Created Services

Matthias Handrich, Sven Heidenreich 399

The Role of Moral Identity in Online Consumer Review BehaviorNan Zhang, Mavis T. Adjei 401

Who Cares About Crowdsourcing from a Virtual Brand Community? The Case ofMarvel.com

Paul G. Barretta 403

xxiii

The Omnichannel Luxury Retail Experience: Building Mobile Trust and TechnologyAcceptance Through Symbolic Self-Completion

Charles Aaron Lawry, Laee Choi 405

Can You Hear Me Now? How Product Attributes Influence Online Phone ReviewsElliott Manzon, Richard Gonzalez, Colleen M. Seifert 407

Design of Reward Systems in Customer Referral ProgramsChristoph Look 409

The Power of a Tweet: An Exploratory Study Measuring the Female Perception ofCelebrity Endorsements on Twitter

Nicole Cunningham, Laura Bright 416

Consumers’ Online Responses to the Death of a CelebrityScott K. Radford, Peter H. Bloch 424

Extending the Electronic Technology Acceptance Model: Consumer Adoption ofAugmented Reality-Based Marketing Tool

Mark Yi-Cheon Yim, Shu-Chuan Chu 426

MARKETING EDUCATION AND TEACHING INNOVATION

Do Personal Response Systems (Clickers) Enhance Learning and Retention ofKnowledge in Higher Education: An Empirical Investigation

Erin Cavusgil 428

Using the RFM Model to Rank Doctoral Marketing ProgramsMatt Elbeck, Brian A. Vander Schee 430

Impact Dynamics of Marketing Scholarship: Going Beyond Journal QualityShibo Li, Eugene Sivadas, Mark S. Johnson 439

The Effects of Individual and Team Characteristics on Simulation-enhancedCritical Thinking: A Multilevel Analysis

George D. Deitz, Alexa K. Sullivan, Robert Evans Jr. 440

SPORTS MARKETING

Without the Volunteers the Event Cannot Go On: Examining Retention with aBehavioral Reasoning Theory Approach

Mya Pronschinske, Mark D. Groza, Mark Peterson 442

Motivating Factors for Participation in National and Olympic Sports in KazakhstanElmira Bogoviyeva 444

Segmenting Fans of a New Team: A Typology of Early AdoptersHeath McDonald, Civilai Leckie, Adam Karg 446

The Service Profit Chain in a Professional Sports SettingAdrien Bouchet, James J. Zboja 448

SERVICES MARKETING

The Effect of Formal and Informal Marketing Controls on Customer ContactEmployee Performance

Ryan C. White, Roger J. Calantone, Clay M. Voorhees 450

xxiv

Bad Customers or Bad Management? An Empirical Investigation of What DrivesService Employees’ Deviant Behavior

Gianfranco Walsh, Simon Brach, Arne K. Albrecht, David Dose, Patrick Hille 452

Linking IMO with Different Fit Types and Willingness to Report Service ComplaintsAchilleas Boukis, Spiros Gounaris, Kostas Kaminakis 454

Firm Self-Service Technology Readiness: A Socio-technical Systems PerspectiveB. Ramaseshan, Russel P.J. Kingshott 462

Does Technology Orientation Matter in Technology Services Organizations?Nacef Mouri, Maheshkumar P. Joshi, Sidhartha R. Das 464

Adapting Principles of Service Quality to Personal Media Marketing CommunicationTodd J. Bacile, Charles F. Hofacker 466

Consumer Reactions to Aesthetic Incongruity: Investigating Aesthetic Incongruityin the Service Consumption Context

Seonjeong (Ally) Lee, Miyoung Jeong 468

Enhanced Regret: Ruminative Thinking, Mood Regulation and Service FailureKevin Lehnert, Mark Arnold 470

Serving in an Online World: How to React on Negative Electronic Word-of-Mouth?Werner H. Kunz, Andreas Munzel, Benedikt Jahn 472

Relinquishing the Moment of Truth: A Model of Firm and Customer-Led ServiceRecovery

Sara K. Bahnson 474

The Impact of Perceived Service Recovery Justice on Customer Affection, Loyalty,and Word-of-Mouth

Beomjoon Choi 476

Caught by Surprise: The Behavioral Effects of Surprise and Delight on Consumersin Different Industries

Tobias Kraemer, Andreas Giese, Christopher Bartl, Nadine Ludwig,Matthias Gouthier 478

Forging Relationships to Coproduce: A Consumer Commitment Model in anExtended Service Encounter

Lin Guo, Cuiping Chen, Chuanyi Tang 480

You’re Such an Embarrassment! A Qualitative Study of the Determinants andConsequences of Vicarious Embarrassment in Customer-to-Customer Interactionsin the Service Context

Thomas Kilian, Kathrin Greuling, Eva Hammes 482

Tempted by Another: How Customer-Perceived Competitive Advantage InfluencesRepurchase Intentions in Service Relationships?

Martin Mende, Scott Thompson, Christian Coenen 484

The Service Level/Gratitude/Reciprocation Relationship and the ModeratingImpact of Reciprocation Wariness

Yihui (Elina) Tang, Chris Hinsch 486

xxv

Give-and-Take in Loyalty Programs: The Asymmetric Effects of MediumMagnitude

Sören Köcher, Markus Blut 488

The Bright and Dark Side of Endowed Status in Hierarchical Loyalty ProgramsAndreas Eggert, Ina Garnefeld, Lena Steinhoff 490

New Insights in the Moderating Effect of Switching Costs on the Satisfaction-Loyalty Link

Thomas Rudolph, Liane Nagengast, Heiner Evanschitzky, Markus Blut 492

PERSONAL SELLING AND SALES MANAGEMENT

Managing Salesforce Selling Behaviors and Performance: the Interactive Effectsof Sales Control Systems

C. Fred Miao, Kenneth R. Evans 494

Do Salesperson Perceptions of Management-Directed Technology-EnabledTransparency Influence Their Behavioral Ethicality?

John E. Cicala, Alan J. Bush, Daniel L. Sherrell, George D. Deitz 496

A New Customer Typology for Adaptive SellingJeffrey S. Larson, Sterling A. Bone 498

Salesperson’s Acculturation Behavior and its Impact on Buyers’ CommitmentHalimin Herjanto, Sanjaya S. Gaur 500

Solution Selling Teams: A Multi-Perspective Review of the Impact of Cross-Functionality

Doreen Wienhold, Michael Nippa 502

The Strategic Alignment of Organizational Development Interventions forSalesperson Value Management with a Salesperson Lifecycle Management Model

Joon-Hee Oh 504

Demystifying Network Strategies: Increasing Product Quality, CustomerSatisfaction, and Profitability Through the Strategic Deployment of InfluentialHubs

Cinthia B. Satornino, Michael K. Brady, Michael Brusco, Clay Voorhees 506

A Nuanced View of the Marketing-Sales “Activity” Interface: A Case of Small B2BFirms

Avinash Malshe, Wim G. Biemans 508

Investments in Customer Relationships and Relationship Strength: Evidence fromInsurance Industry in China

Guicheng Shi, Yuan Ping, Yonggui Wang, Matthew T. Liu 510

Salesperson’s Personality and Relationship Quality: Are You a Friend or aCustomer?

Kaveh Peighambari, Setayesh Sattari, Maria Ek Styvén, Lars Bäckström 511

Knowledge Transfer Antecedents and Consequences: A Conceptual ModelNicholas Kolenda, Lee McGinnis, Brian Glibkowski 519

xxvi

Intrafirm Information Advantage and Brokerage: Effects on SalespersonPerformance

Gabriel R. Gonzalez, Danny P. Claro 527

SPORTS MARKETING

Befriending Sport Celebrities Through Mediated Relationships: ParasocialInteractions and Relationships with Athletes in Social Media

Mujde Yuksel 529

The Effect of Sport Event Advertising on Brand AttitudeWonseok Jang, Yong Jae Ko, Songhyun Cho 530

Branding Higher Education Through PSAS: Producing Global Citizens for the21st Century

Michael J. Clayton, Kevin V. Cavanagh 532

A Cutting Edge Approach to Achieve Vivid Destination for Sport SponsorshipAlireza Faed, Afsaneh Ashouri, Morteza Saberi 534

AUTHOR INDEX 545

American Marketing Association / Summer 2012 1

THE APPLICATION OF STRUCTURAL EQUATIONS TO THEATTRIBUTES IN DISCRETE CHOICE MODELS

Cam M. Rungie, University of South Australia

ABSTRACT

The paper demonstrates the integration of the meth-ods of structural equation modeling into the modeling ofthe attributes in discrete choice models and in particulardiscrete choice experiments. The approach generates out-comes well known to research and to structural equationmodeling but previously not available within discretechoice models.

Key Words: discrete choice, structural equation, la-tent variable, attribute, choice set.

INTRODUCTION

Recently, methods known as structural choice mod-eling (SCM) have been developed for including latentvariables and structural equations in discrete choice mod-els (DCMs). The methods allow for the modeling ofassociations in the utilities for the attributes, and theirlevels, over one or more DCMs. SCM gives structure tothe variance covariance matrix of random coefficientmodels and allows the analyst to operationalise a prioriknowledge and theory. This paper reviews the outcomes,and what has been learnt so far, from eight studies usingSCM undertaken in five countries. The first observation inapplying SCM is of a technical nature. Parsimoniousmodels fit the data better than the traditional fixed andrandom coefficient models. There is always structure inthe unobserved heterogeneity of DCMs. However, themore interesting and useful observations relate to theinterpretation and nature of the outcomes. SCM leads tosegmentation, hypothesis testing, modeling across prod-uct category, analyst of joint decision making, evaluationof the association between attitudes and behavior, statedependent and temporal models. These outcomes are notnew to research but are new to DCMs.

The marketing literature over the past 30 years hasseen substantial growth in the use of structural equationmodeling (SEM) and DCM but as relatively separatefields. SEM has offered an outstanding ability tooperationalize and evaluate theory particularly with re-gards to consumer behavior. In contrast DCM has offeredoutstanding ability in measurement and demand estima-tion, and is used especially in those fields where predictivevalidity is at a premium including econometrics, transportplanning and non-market valuation. In DCMs, and inparticular in discrete choice experiments (DCEs), therespondent makes a selection from each of several differ-

ent choice sets containing a discrete number of alterna-tives in a manner similar to how consumers make choiceson a daily basis. From the pattern of selections over therespondents it is possible, using random utility theory(RUT), to estimate and evaluate the impact and elasticityof the attributes of the alternatives, and the variability ofthis impact over respondents. In DCEs the analyst ma-nipulates the choice sets so as to optimize identification,estimation and causal inference. SCM integrates SEMinto DCM and so combines the better operationalizationof theory of SEM with the better observation, measure-ment, causal inference, and predictive validity of DCM.SCM is old, it draws on two well-established modelingapproaches, and it is new, it integrates the two andgenerates outcomes previously not available. The out-comes are demonstrated below.

The paper reviews the literature and the outcomesfrom the eight studies and then finishes with a discussionemphasizing the new contributions and expanded role ofDCMs.

LITERATURE

In their paper, evolving from the Choice Symposium,Ben-Akiva et al. (2002) develop the hybrid choice modelscontaining latent class (LCM) and latent variable (LVM)models. They present a diagram, which is reproduced inFigure 1 but to which has been added a third form ofmodel, structural choice (SCM) (Walker 2001; Rungie2011; Rungie, Coote, and Louviere 2011).

LVM and SCM are examples of the techniques de-signed to incorporate latent variables and structural equa-tions into the analyses of discrete choice models (DCM)and more generally into choice processes and RUT(McFadden 1974, 2001). There are indeed several impor-tant precursors and examples. Firstly, factor analyticchoice models have been applied to the study of brandspreferences using revealed preference (RP) data. This isas if “brand” is an attribute with the individual brands aslevels. One or more factors have been applied across thebrands and other attributes (Elrod 1988; Elrod and Keane1995; Keane 1997; Walker 2001). Secondly, factor ana-lytic models have been applied to the characteristics ofrespondents using indicator variables based on RP and SPdata (Walker 2001; Ashok, Dillon, and Yuan 2002;Morikawa, Ben-Akiva, and McFadden 2002; Temme,Paulssen, and Dannewald 2008; Bolduc and Daziano

2 American Marketing Association / Summer 2012

2010; Yáñez, Raveau, and de Dios Ortúzar 2010; Hessand Stathopoulos 2011). Thirdly, methods using latentvariables have been developed for combining RP and SPdata (Ben-Akiva and Morikawa 1990; Hensher, Louviere,and Swait 1999; Louviere et al. 1999; Ben-Akiva et al.2002; Louviere et al. 2002; Morikawa, Ben-Akiva, andMcFadden 2002). The approaches differ in the nature ofthe covariates. In the first the covariates are the attributesof the alternatives and in the second the characteristics ofthe respondents. However, the approaches are similar intheir mathematics and in their use of factor analytics.SCM adapts this mathematics and, as in SEM (Jöreskog1970, 1973; Bollen 1989; Jöreskog and Sörbom 1996),adds the capacity to specify autoregressions, structuralequations and correlations for the factors.

In the traditional random coefficient model (e.g.,Ben-Akiva et al. 1997; McFadden and Train 2000; Dubeet al. 2002; Train 2003, 2009) the coefficients for eachcovariate are independent random variables with meansand variances estimated from the data. That is, the vari-ance covariance matrix, ΣΣΣΣΣ, for the random coefficient isdiagonal. Various approaches have been applied to allow

Σ Σ Σ Σ Σ to be other than diagonal including adding the correla-tions to the random coefficient model and latent classmodels LCM (Kamakura and Russell 1989). In SCM, thecoefficients have a multivariate distribution where, throughthe prudent and parsimonious use of factor analytics,autoregressions, structural equations, and specific corre-lations, Σ Σ Σ Σ Σ can be other than diagonal. The number ofparameters need not be excessive and the number ofdisturbances can be less than the number of randomcoefficients. SCM allows the analyst to specify structurefor ΣΣΣΣΣ. A priori knowledge, hypotheses and theory can beoperationalized and competing specifications for ΣΣΣΣΣ can betested empirically.

The covariates for a DCM can record the attributes ofthe alternatives and the characteristics of the respondents.LVM and SEM both conceptualize latent variables asbeing the process driving the observations of indicatorvariables which are characteristics of the respondents. Incontrast, SCM conceptualizes latent variables as prefer-ences driving the utilities for the attributes and their levels.The latent variables are higher order utilities for over-arching and meta attributes or constructs. By-and-large,

FIGURE 1Hybrid Choice Modeling and Structural Choice Modeling; Modified from

Ben-Akiva et al. 2002. Marketing Letters

Indicators Indicators

Choice Indicators:

Stated

Latent

Variables

Latent

Classes

Decision

Process

Latent

Variables

SP and RP disturbances and Kernel

Disturbances Disturbances

Disturbances

Choice Indicators:

Revealed

Latent

Variable

Model

Choice Model

Covariates.

Explanatory Variables.

Latent

Class

Model

Structural

Choice

Model Factor Analytics

Autoregression Latent

Variables

DecisionProcess

LatentClasses

LatentVariables


LVM concentrates on the characteristics of the respon-dents while SCM concentrates on the attributes of thealternatives.

As a general tendency revealed preference (RP),while being excellent data, is constrained to the alterna-tives commercial suppliers make available and does notalways record the choice set from which each selection isdrawn. By comparison stated preference (SP) data, par-ticularly when collected through the use of DCEs(Louviere, Hensher, and Swait 2000), can provide moreinformation on preferences for attributes and better iden-tification, estimation and causal inference. Thus, of theeight studies using SCM reviewed below seven use datafrom DCEs and only one from RP.

From RUT (McFadden 1974, 2001), let the utilitiesfor the alternatives in a DCM be u comprising of asystematic components v and idiosyncratic disturbances ewhere u = v + e. The vector and matrix notation for SCMis presented in detail by Rungie, Coote, and Louviere(2011) and in summary here. Let the covariates be x withrandom coefficients η where v = ηx. The variance cova-riance matrix for η is Σ. SCM specifies that has a factoranalytic structure based on the latent variables ξ whereη = γξ. The γ are constants to be estimated from the dataand can be referred to as regression coefficients, factorloadings or weights. The ξ have a distribution over respon-dents but for each respondent are latent and fixed. SCMspecifies that the factors ξ can be autocorrelated where ξ= ßξ + δ. The ß are constants to be estimated from the dataand can be referred to as autoregression or structuralregression coefficients. Over respondents, the disturbanceshave a distribution, such as a multivariate Gaussian withselected correlations. For each respondent the δ are latentand fixed. Solving gives:

η = γ(1- ß)-1δ where v = ηx and u = v + e

The scope for DCMs has been expanded throughrecognition of the potential to simultaneously model morethan one discrete choice data set recorded from the samerespondents (Rungie, Coote, and Louviere 2011). Latentvariables have a new role to play in linking the utilitymodels specified for each data set. The potential tooperationalize constructs and to generate new outcomes isgreatly enhanced through allowing the different data setsto reflect different choice tasks but with similar attributes.Of the eight studies reviewed below six combine data setswith different choice tasks. Latent variables are modeledacross product categories, across people in joint decisionmaking, to model the associations between attitudes andbehavior and across time periods.

The more technical aspects of SCM are not repre-sented here. A lot is now known about validity andidentification of SCMs, as discussed by Walker (2001)

and Rungie, Coote, and Louviere (2011). The syntax forSCM is described in detail by Rungie (2011). Software,known as DisCoS (discrete choice software), is availablefor beta testing in academic research from the author’shome page (http://people.unisa.edu.au/Cam.Rungie).Rather than discuss technical issues, the motivation of thispaper is to review the new outcomes and contributions ofSCM. The paper shows what can be achieved usingdiscrete choice data and SCM.

THE STUDIES

In each of the eight studies a comparison is drawnbetween an SCM model and the traditional models usingfixed coefficients, where ΣΣΣΣΣ = 0, and random coefficient,where ΣΣΣΣΣ is diagonal. Not surprisingly, the SCMs fit thedata better as assessed using log likelihood values, AICand BIC, as seen in the Tables 3 and 4. The SCMs haveslightly more parameters than the random coefficientmodel – ranging, over the eight studies, from only 5percent more to 25 percent more – as shown in Table 3.However, most of the SCM have fewer disturbances, asshown in Table 5. In this trade-off where there are fewersources of randomness but more parameters it can beargued that the SCMs are parsimonious. They simplify ΣΣΣΣΣ.

However, a low P value in a likelihood ratio test anda better fit to the data is a relatively inconsequentialoutcome. Of far greater relevance are the theoretical andpractical outcomes of the new (and better fitting) models.Over eight studies SCM leads to segmentation, hypoth-esis testing, evaluations of antecedents, and modelingacross product categories, time and people in joint deci-sions. The review that now follows emphasizes these newoutcomes and contributions for DCMs.

1. Postgraduate research candidates completed a DCEon preferences for post doctorial employment (Rungie,Coote, and Louviere 2010). With candidates tradingoff status, security and life style, high levels ofcolinearity were expected in the utilities for the attri-butes and their levels. A separate single factor wasapplied to the levels within each attribute and then anoverarching factor was applied to these attributefactors. The result was a model that confirmed theexistence of the colinearity and demonstrated theconsiderable segmentation in the market; reputationof university department versus family life style.When making post doctoral appointments, schoolswithin one segment could see from the study howmuch, and on which attributes, they competed withother schools in the same segment and the schools inthe other segments.

2. Consumers completed a DCE on preferences for beefwhere the attributes included price, brand, feed con-ditions, fat content and marbling (Umberger and


Mueller 2010; Rungie 2012). Propositions wereoperationalized as six hypotheses specifying associa-tions between utilities for attributes or levels, such asthe preference for higher marbling would correlatewith high tolerance for fat trim and for higher prices.Models with latent variables were specified to for-mally test each of the hypotheses. This form ofanalysis of DCMs to test hypothesis regarding con-sumer behavior is relatively unknown but with SCMis now relatively straight forward.

3. In a study of an Islamic market two attributes ofpackaging and labeling were used in three separateDCEs across different product categories; soup,chocolate and detergent (Abou Bakar, Lee, and Rungie2011; Rungie, Abou Bakar, and Lee 2012). The twopackaging and labeling attributes were the crescentshaped moon, which was either present or absent, andthe background color, which was red or green. Thestudy was undertaken in Pakistan, an Islamic marketwhere culturally symbols are of importance, particu-larly the crescent moon and the specific color ofgreen. The aim was to measure the impact of thesymbols and to explore if this impact can be attributedto religiosity or nationalism. Both symbols havereligious importance and both are part of the Pakistannational flag. The study showed different preferencesfor symbol and color. The preferences varied be-tween respondents. Most respondents were relativelyconsistent over the three categories. The requirementfor religious considerations and Halal processing arenot the same for the three product categories. Thus,given the consistency, for individual respondents, ofthe utility for the attributes over the three categories,the study concluded that the variation between respon-dents in preference was more appropriately attributedto nationalism than religiosity. The study demon-strates the additional contribution from simultaneouslymodeling the same attributes over multiple productcategories.

4. In a study in China of country of origin (COO) effectstwo separate DCEs measured the preferences forwine and for seafood. For the COO attribute, twolevels – China and Australia – were common acrossboth DCEs as was the price attribute. Thus, threecorrelations across the two DCEs were specified. Itwas proposed that the utility for price (i.e., money)would have a high correlation over the two DCEs, asthe same respondents would not mind paying a littlemore. It was also proposed that COO-China would becorrelated but not as highly, because there would bedifferences between respondents in their preferencefor goods produced in China, which would tend to begeneric and consistent across categories. For COO-Australia it was proposed that the correlation would

be less. Preference would be experience and stimulusbased reflecting specific categories. A respondent,positively influenced regarding Australian wine maydevelop a preference for it, but may not transfer thepreference to seafood, and vice versa. Hence, it wasproposed that the three correlations between the twoDCEs would be positive with price the highest, fol-lowed by COO-China and then COO-Australia. Theactual variances for the coefficients were all large andthe correlations were price 0.42 (p = .0000), COO-China 0.22 (p = .0025) and COO-Australia 0.09 (p =.35). This style of choice analysis of consumer behav-ior using DCEs across product categories is quitenew.

5. SCM was applied to the study of water quality inwhich the impact of individuals on joint decisionmaking by couples was evaluated (Rungie, Scarpa,and Thiene 2011). Three DCEs were used, one com-pleted by each of two adults in a household and thenthe third completed jointly by the couple. Waterquality was assessed using six attributes. In the SCMthe attributed were specified to be latent variables andthe structure linked the three experiments. The SCMmeasured how much each individual influenced thejoint decisions. The influence was measured overalland separately for each attribute. Women in generalhad greater influence but not on all attributes. This isthe first formal statistical method for fully evaluatingjoint decision making using DCEs.

6. One of the most pressing public policy issues today isto understand consumers’ preferences for carbonmitigation. Two DCSs were undertaken by the samerespondents but with different choice tasks (Coote,Rungie, and Louviere 2011; Rungie, Coote, andLouviere 2011). The first recorded preferences forgeneric mitigation strategies of a government policynature and the second recorded the selection of prod-uct that embedded the strategies. The DCEs showedwhich strategies were preferred and which wouldhave actual impact. The latent variables linked thesimilar constructs across the two DCEs through regres-sions the contribution of which was evaluated usingR2 goodness-of-fit measures. The study evaluated theassociations between attitudes and behavior.

7. SCM was applied to the study of the antecedents forthe selection of brands of airlines (Rungie, Coote, andLouviere 2011). Two DCEs were completed by thesame respondents; the first recorded the qualities ofeach of the major brands available in the market andthe second recorded the selection of a journey wherebrand was one of the attributes. The brands weremodeled as latent variables which linked the twoDCEs. The SCM showed how the brands influencedthe choice of journey and which qualities influenced


the perception of the brand. Thus, the role of thequalities on the selection of brand was assessed.

8. Of the eight studies only one analyzed RP data. Itexamined changes over time in the purchases of wineclassified by three attributes, price, format and denom-ination (regional quality and production standards)(Corsi and Rungie 2011; Rungie, Coote, and Louviere2011). A separate random coefficient model was

specified for each year. The models were linked andfitted to the data simultaneously using latent vari-ables and regressions. The outcome was a process forseparating the various forms of state dependence anddemonstrated how to validly model changes in util-ity, rather than the specific choices, over time. Themethod, applied in the study to RP data, is equallyapplicable to DCEs with a temporal component suchas before/after treatment/control designs.

TABLE 1Studies

Seven of the Eight Studies Used Stated Preference DCEs

Combination of Data Sets

NumberStudy N Country Data of data sets Purpose

1 Post Doc 797 USA DCE2 Beef 1846 Australia DCE3 Islamic 241 Pakistan DCE 3 Across categories4 COO 1033 China DCE 2 Across categories5 Water 80 Italy DCE 3 Joint decisions6 Carbon 1204 Australia DCE 2 Attitude v behaviour7 Airlines 200 Australia DCE 2 Attitude v behaviour8 Wine 693 Italy RP 3 Across time (years)

TABLE 2Choice Tasks

Six of the Studies Combined DCM with Differing Choice Tasks

Number of Number of Choice SetStudy Choice Task Attributes Choice Sets Size

1 Post Doc Select employment 7 8 22 Beef Selection of beef 9 16 53 Islamic Selection of soup 2 6 2

Selection of chocolate 2 6 2Selection of detergent 2 6 2

4 COO Selection of wine 3 8 4Selection of seafood 3 8 4

5 Water Selection by female 6 8 4Selection by male 6 8 4Selection by couple 6 8 4

6 Carbon Mitigation program 5 5 4Select Refrigerator 9 8 9

7 Airlines Corporate qualities 6 16 4Selection of trip 7 16 5

8 Wine Purchase wine 2007 3 10 12Purchase wine 2008 3 10 12


TABLE 3Log Likelihood Values

For Each Study the SCM Fitted the Data Better

Fixed RandomStudy Coefficient Model Coefficient Model SCM

# par LL # par LL # par LL

1 Pos Doc 19 -2,326 38 -2,245 45 -2,1762 Beef 31 -40,390 62 -29,150 70 -26,5183 Islamic 6 -2759 12 -2311 13 -22924 COO 12 -21,128 18 -19,606 21 -19,5575 Water 36 -1,250 72 -1,090 84 -9686 Carbon 33 -26,847 68 -23,913 77 -21,6537 Airlines 29 -13,253 57 -11,930 60 -11,6878 Wine 8 -31,512 16 -23,557 20 -22,372

TABLE 4AIC and BIC

As in Table 3, for Each Study the SCM Fitted the Data Better

Fixed RandomStudy Coefficient Model Coefficient Model SCM

AIC BIC AIC BIC AIC BIC

1 Post Doc 4690 4818 4566 4823 4442 47462 Beef 80842 81099 58424 58938 53176 537573 Islamic 5530 5568 4646 4723 4610 46934 COO 42280 42373 39248 39387 39156 393185 Water 2572 2772 2324 2724 2104 25716 Carbon 53760 54013 47962 48483 43460 440507 Airlines 26564 26760 23974 24360 23494 239008 Wine 63040 63100 47146 47267 44784 44935

TABLE 5Number of Disturbances*

Given the Ability of SCM to Specify Structure for the Unobserved Heterogeneity They Often Fit the DataBetter While Having Fewer Disturbances than the Traditional Random Coefficient Model

Disturbances in Random DisturbancesStudy Coefficient Model in SCM δδδδδ

1 Post Doc 19 82 Beef 31 93 Islamic 6 24 COO 6 65 Water 18 126 Carbon 35 147 Airlines 29 148 Wine 8 8

*Excluding the RUT idiosyncratic disturbances, e = u – v.)


DISCUSSION

The eight studies demonstrate that applying latentvariables to attributes has new, useful and relevant out-comes that contribute to the subject area to which theDCMs and their choice tasks are applied. The outcomesare not new to research; they are just new to DCMs. Study1 used factors to identify segments. Study 2 tested hypoth-eses. Studies 3 and 4 examined patterns over multipleproduct categories. Study 5 examined patterns over peoplein the same household and modeled joint decision mak-ing. Studies 6 and 7 modeled the associations betweenattitudes and behavior. Study 8 modeled changes overtime. All eight studies deliver statistical process andoutcomes prevalent in research but generally not in theanalysis of DCMs.

Through applying latent variables to the utilities ofattributes SCM has been demonstrated to be a useful toolin generating a range of new outcomes for DCMs. Thereare other approaches that make contributions in the sameareas, most notably LCM and random coefficient modelswith correlations. Statistically speaking, to DCMs (i)LCMs bring cluster analysis, (ii) traditional random coef-ficient models with correlations bring mixture models,and (iii) LVMs and SCMs bring factor analysis and SEM.The formal comparison between the approaches, cover-ing issues such as the nature of the outcomes,operationalization of constructs and theories, parsimony,fit and predictive validity are left to further research.

Turning now to the nature of the covariates, latentvariables are conceptualized by (i) SCM as influencingthe utilities for the attributes, and (ii) by LVM and SEM

as influencing the observed characteristics of the respon-dent such as demographics, attitudes and indicator vari-ables. There appears to be a chasm separating the twotypes of covariates but there is not. Characteristics can beexplored using DCEs. Studies 6 and 7 both use a DCE tomeasure attitudes. Constructs which might initially beseen as being only operationalized as observed character-istics of the respondent can be operationalized through thealternatives in DCEs with specifically selected choicetasks. Furthermore, the characteristics of the respondentsand the attributes of the products they select can bemodeled simultaneously by combining separate DCEs,one eliciting the respondents and the other eliciting theproducts, as in studies 6 and 7. For covariates, the demar-cation between attributes and characteristics is now not asclear as it once was.

CONCLUSION

SCM is a new method for modeling the attributes,levels and covariance matrix in discrete choice models.The paper has documented briefly the outcomes fromeight studies to which SCM has been applied. The topicsfor the studies were post doctorial employment, beef,Islamic markets, country of origin, water quality, carbonemissions, brands of airlines, and wine. The subject areaswere conceptual and quite varied including segmentation,hypothesis testing, across category analysis, joint deci-sion making, the impact of attitudes on behavior, anteced-ents to brand preference, and state dependence. In eachstudy latent variables were specified for the attributes andthen structures were specified linking the latent variables.The outcomes from this form of modeling are not new toresearch, but are new to discrete choice modeling.

REFERENCES

Abou Bakar, Lee R. and C. Rungie (2011), “The Use ofReligious Symbols on Packaging to Influence Prod-uct Preference and Rating,” Paper presented at theAustralia and New Zealand Marketing AcademyConference, Perth, Australia.

Ashok, K., W.R. Dillon, and S. Yuan (2002), “ExtendingDiscrete Choice Models to Incorporate Attitudinaland Other Latent Variables,” Journal of MarketingResearch, 39 (February), 31–46.

Ben-Akiva, M., D. McFadden, M. Abe, U. Böckenholt, D.Bolduc, D. Gopinath, T. Morikawa, V. Ramaswamy,V. Rao, D. Revelt, and D. Steinberg (1997), “Model-ing Methods for Discrete Choice Analysis,” Market-ing Letters, 8 (3), 273–86.

____________, ____________, K. Train, J. Walker, C.Bhat, M. Bierlaire, D. Bolduc, Z. Boersch-Supan, D.Brownstone, D.S. Bunch, A. Daly, A. De Palma, D.

Gopinath, A. Karlstrom, and M.A. Munizaga (2002),“Hybrid Choice Models: Progress and Challenges,”Marketing Letters, 13 (3), 163–75.

____________ and T. Morikawa (1990), “Estimation ofSwitching Models from Revealed Preference andStated Intentions,” Transportation Research, A 24A(6), 485–95.

Bolduc, D. and R. Alvarez-Daziano (2010), “On Estima-tion of Hybrid Choice Models,” in Choice Modeling:The State-of-the-Art and the State-of-Practice, S.Hess and A. Daly eds. Emerald, England.Bollen,K.A. (1989), Structural Equations with Latent Vari-ables. New York: John Wiley & Sons.

Coote, L.V., C.M. Rungie, and J.J. Louviere (2011),“Consumers’ Preferences for Carbon Mitigation Strat-egies,” UQ Business School, University ofQueensland, Brisbane, Australia.

Corsi, A.M. and C. M. Rungie (2011), “A LongitudinalAnalysis of Consistency in Repeated Choice.” Paper


Presented at the Australia and New Zealand Market-ing Academy Conference, Perth, Australia.

Dube, J-P., P. Chintagunta, A. Petrin, B. Bronenberg, R.Goettler, P.B. Seetharaman, K. Sudhir, R. Thomadsen,and Y. Zhao (2002), “Structural Applications of theDiscrete Choice Model.” Marketing Letters, 13 (3),207–20.

Elrod, T. (1988), “Choice Map: Inferring a Product-Market Map from Panel Data,” Marketing Science, 7(1), 21–40.

____________ and M.P. Keane (1995), “A Factor-Ana-lytic Probit Model for Representing the Market Struc-ture in Panel Data,” Journal of Marketing Research,32, 1–16.

Hensher, D., J. Louviere, and J. Swait (1999), “Combin-ing Sources of Preference Data,” Journal of Econo-metrics, 89, 197–221.

Hess, S. and A. Stathopoulos (2011), “Linking ResponseQuality to Survey Engagement: A Combined Ran-dom Scale and Latent Variable Approach,” Leads,Institute for Transport Studies, University of Leeds.

Jöreskog, K.G. (1970), “A General Method for Analysisof Covariance Structures,” Biometrika, 57, 239–51.

___________ (1973), “A General Method for Estimatinga Linear Structural Equation System,” in StructuralEquation Models in the Social Sciences, A.S.Goldberger and O.D. Duncan, eds. New York: Semi-nar Press, 85–112.

____________ and D. Sörbom (1996), LISREL 8: User’sReference Guide. Lincolnwood, IL: Scientific Soft-ware International.

Kamakura, W.A. and G.J. Russell (1989), “A Probabilis-tic Choice Model for Market Segmentation and Elas-ticity Structure,” Journal of Marketing Research, 36(November), 379–90.

Keane, M.P. (1997), “Modeling Heterogeneity and StateDependence in Consumer Choice Behavior,” Jour-nal of Business and Economic Statistics, 15 (3), 310–27.

Louviere, J.J., R.J. Meyer, D.S. Bunch, R. Carson, B.Dellarert, W.M. Hanemann, D. Hensher, and J. Irwin(1999), “Combining Sources of Preference Data forModeling Complex Decision Processes” MarketingLetters, 10 (3), 205–17.

____________, D.A. Hensher, and J.D. Swait (2000),Stated Choice Methods: Analysis and Application.Cambridge: Cambridge University Press.

____________, D. Street, R. Carson, A. Ainslie, J.R.Deshazo, R. Cameron, D. Hensher, R. Kohn, and A.Marley (2002), “Dissecting the Random Componentof Utility,” Marketing Letters, 13 (3), 177–93.

McFadden, D. (1974), “Conditional Logit Analysis ofQualitative Choice Behavior,” Frontiers in Econo-metrics, P. Zarembka, ed. New York: AcademicPress, 105–42.

____________ and K. Train (2000), “Mixed MNL Mod-els for Discrete Response,” Journal of Applied Econo-

metrics, 15, 447–70.____________ (2001), “Disaggregate Behavioral Travel

Demand’s RUM Side: A 30–Year Retrospective,”The Leading Edge of Travel Behavior Research, D.Hensher, ed. Oxford: Pergamon Press.

Morikawa, T., M. Ben-Akiva, and D. McFadden (2002),“Discrete Choice Models Incorporating RevealedPreference and Psychometric Data,” EconometricModels in Marketing, 16, 29–55.

Rungie, C.M., L.V. Coote, J.J. Louviere (2010), “LatentVariables in Discrete Choice Experiments,” Schoolof Marketing, University of South Australia, Adelaide,Australia.

____________, ____________, ____________ (2011),“Structural Choice Modeling: Theory and Applica-tions to Combining Choice Experiments,” Journal ofChoice Modeling, 4 (3), 1–29.

____________, R. Scarpa, and M. Thiene (2011), “SingleVersus Joint Preference for Tap Water: A StructuralChoice Modeling Investigation,” CenSoC, Univer-sity of Technology, Sydney, Australia.

____________ (2011), “Structural Choice Modeling Soft-ware Manual,” School of Marketing, University ofSouth Australia, Adelaide, Australia.

____________ (2012), “Structural Choice Modeling CaseStudy: The Beef Discrete Choice Experiment,” Schoolof Marketing, University of South Australia,Adelaide, Australia.

____________, Abou Bakar, and R. Lee (2012), “Consis-tency in Preference Across Product Categories,”School of Marketing, University of South Australia.

Temme, D., M. Paulssen, and T. Dannewald (2008),“Incorporating Latent Variables into Discrete ChoiceModels: A Simultaneous Estimation Approach UsingSEM Software,” BuR – Business Research, 1 (2),230–37.

Train, K. (2003), Discrete Choice Methods with Simula-tion. New York: Cambridge University Press.

____________ (2009), Discrete Choice Methods withSimulation. New York: Cambridge University Press.

Umberger, W. and S. Mueller (2010), “Using DiscreteChoice Experiments to Determine the Relative Valueof Intrinsic and Extrinsic Meat Attributes: A CaseStudy of Australian Beef Consumers,” ACSPRI Soc-ial Science Methodology Conference 2010, The Aus-tralian Consortium for Social and Political Research,December 2010, Sydney, 1–3.

Walker, J.L. (2001), “Extended Discrete Choice Models:Integrated Framework, Flexible Error Structures, andLatent Variables,” Department of Civil and Environ-mental Engineering, Massachusetts Institute of Tech-nology. PhD.

Yáñez, M., S. Raveau, and J. de Dios Ortúzar (2010),“Inclusion of Latent Variables in Mixed Logit Mod-els: Modeling and Forecasting,” TransportationResearch Part A: Policy and Practice, 44 (9), 744–53.


For further information contact:Cam M. Rungie

University of South AustraliaNorth Terrace

Adelaide, SA 5000Australia

Phone: +61(0)8.83020768E-Mail: [email protected].

WWW home page: http://people.unisa.edu.au/Cam.Rungie


ANCHORING IN CHOICE EXPERIMENTS: INFLUENCE OF THEEXPERIMENTAL DESIGN ON WILLINGNESS-TO-PAY

Felix Eggers, University of Hamburg, GermanyFranziska Völckner, University of Cologne, Germany

SUMMARY

In the consumer behavior and marketing researchliterature there is increasing evidence that, to some extent,consumer preferences are constructed in the decisioncontext rather than retrieved from memory (Amir andLevav 2008; Dhar 1997). This notion can explain whyconsumers’ decisions are susceptible to context effectsand external anchors, such as external reference prices(ERP; see Mazumdar et al. 2005 for a review). That is, inorder to make a purchase decision consumers judge aspecific product relative to their prior knowledge andother cues that are externally available. For example, listprices or price promotions provide an external input thatconsumers compare with their memory of prior purchaseexperiences, i.e., their internal reference price (IRP). If theinformation differs consumers may update their knowl-edge which leads to IRP shifts and a modified willingness-to-pay (WTP; Adaval and Monroe 2002; Kalyanaram andWiner 1995; Mazumdar et al. 2005; Simonson and Drolet2004; Suk et al. 2010; Tversky and Kahneman 1981).Consumers’ (nonconscious) need to rely on external infor-mation to make a purchase decision can be very strong sothat they use external cues as an anchor even if these cuesare unrelated to the choice context (Adaval and Wyer2011; Ariely et al. 2003; Nunes and Boatwright 2004).

Knowledge of the interplay between ERP and IRP isimportant for managers when settings prices but also forresearchers when measuring WTP. Regarding the latter,there is a research gap to what extent the experimentaldesign of WTP elicitation methods serves as an anchor forconsumers and how this anchor may bias the WTP esti-mates. For example, conjoint analysis as one of the mostpopular experimental methods for measuring consumers’WTP presents stimuli to respondents that differ in prede-termined attribute levels, including prices. Based on priorresearch about reference prices (Adaval and Monroe2002; Adaval and Wyer 2011; Park et al. 2011) we arguethat respondents may use these prices as an externalanchor so that their IRP shifts into the direction of the pricelevels of the experimental design. If this is the case, WTPestimates are biased and the results are only internally butnot externally valid.

In an empirical study with four conditions and 137respondents we examine if the results of conjoint choiceexperiments may be influenced by the experimentaldesign. For the conjoint section, we selected brand and

price as attributes, each having three levels, and manipu-lated the price levels across the experimental conditions,i.e., the overall magnitude of the price levels as well as theorder of the price information across choice sets. As anindicator of the consumer’s IRP we used a directly statedWTP measure for each of the brands of the conjointdesign, which we elicited after the choice section. Simi-larly, we asked for quality ratings for each brand rangingfrom 1 (very high) to 5 (bad).

The results show that there is a general referenceprice effect in choice experiments caused by the pricelevels used in the experimental design that influencesconsumers’ IRP and derived WTP measures, such thatconsumers have a higher (lower) absolute WTP for brandswhen higher (lower) prices are provided during the con-joint procedure. Apart from shifting the IRP, externalprices also influence the quality perception such thatconsumers have higher (lower) quality ratings for brandswhen higher (lower) prices are provided during the con-joint procedure (Wathieu and Bertini 2007). Apart fromthis general reference price effect, we have found noevidence of a brand-specific primacy of recency effect,i.e., the initial price of the brand in the procedure does notprovide an anchor for subsequent questions and overallWTP (Aprahamian et al. 2007; Bhatia 2005; Flachaire andHollard 2002; Whitehead 2002).

These results stress the importance of setting pricelevels in conjoint methods according to a realistic marketscenario so that the ERP in experiment and market match.However, if no market prices exist, e.g., when researchinginnovations, our results show that the implications ofchoice experiments can be biased by the experimentaldesign. In this case, marketing researchers should system-atically test different price ranges to determine the extentto which consumers’ WTP might be influenced by theprice levels used in the study. A bias can also occur whenmarket prices are distributed heterogeneously across theproduct category, e.g., when researching strong (national)brands in association with (relatively weak) store brands.To account for this problem, brand-specific prices shouldbe implemented into the experimental design to increaserealism.

This study presents a first of several planned applica-tions to explore the extent to which methods for measur-ing consumers’ WTP may be biased by their experimentaldesign. Moreover, it provides the basis for further research


that examines the formation of internal reference pricesand mechanisms that are able to diminish the biasingeffect of the experimental design in marketing researchapplications, e.g., incentive-aligned procedures which

relate decisions to a real outcome, i.e., reward product(Ding et al. 2005; Ding 2007; Dong et al. 2010). Refer-ences are available upon request.

For further information contact:Felix Eggers

Institute of Marketing and MediaUniversity of Hamburg

Welckerstraße 820354 Hamburg

GermanyPhone: +49.40.42838.8714

Fax: +49.40.42838.8715E-Mail: [email protected]


THE DYNAMICS UNDERLYING PERFORMANCE SIGNALING:INSIGHTS FROM THE PROFESSIONAL SPORTS MARKET

Stefan Hattula, University of Mannheim, GermanyHauke A. Wetzel, Georg August University Goettingen, Germany

Maik Hammerschmidt, Georg August University Goettingen, GermanyHans H. Bauer, University of Mannheim, Germany

SUMMARY

Over the past decades, professional sports have be-come an important and top-selling industry. The $414billion professional sports market accounts for 2.8 percentof the GDP of the United States and is growing rapidly at3.8 percent per year (Price Waterhouse Coopers 2010).Characterized by information asymmetry between theproducer (i.e., sports clubs) and consumers (i.e., fans,spectators), the entertainment services the sports marketprovides in the form of matches are a typical example ofan experience good (Sainam, Balasubramanian, and Bayus2010). Sports are live events and the “people factor” is ofmajor importance. As compared to other experience goodssuch as motion pictures (Basuroy, Desai, and Talukdar2006), technological mediation is virtually infeasible.Owing to its high degree of intangibility, the performanceof sports clubs that consumers evaluate in terms of enjoy-ment value cannot be standardized, is difficult to foresee,and is characterized by high variation. Hence, consumersinterpret multiple signals that improve their assessment offuture performance and determine their purchase behav-ior (Funk et al. 2009).

Team expenditures and brand equity are two power-ful signals available to consumers for overcoming uncer-tainty about future performance. For sports clubs, theteam and the brand represent two major assets that act aseffective levers for influencing consumer purchase be-havior (Yang, Shi, and Goldfarb 2009). However, theneed to master more than one signal discloses several gapsthat prior research has left unaddressed.

In closing these gaps, our investigation contributes inseveral ways. First, prior research has widely neglectedthe role of dynamics in a signaling context (Erdem,Keane, and Baohong 2008), and we address this void byexploring temporal differences in the effectiveness ofperformance signals. Organizational life cycle (OLC)theory implies that the availability of a club’s specificsignals changes over time (James 1974). Building on priorresearch in other domains, we propose that in earlierstages of the OLC only the team expenditure signal isestablished, whereas in later stages the brand enters thesignal portfolio (Shankar, Carpenter, and Krishnamurthi1999). The potential for change in consumers’ signal use

patterns raises the question of whether consumers (a) stickto the initial signal, (b) adopt a partial shift and makecomplementary use of both signals, or (c) shift signalscompletely and replace one signal once another becomesavailable.

We find that the signaling power of both signalsexhibits a cyclical sensitivity, as proposed by the cuescope framework. Specifically, for team expenditures, wefind a countercyclical movement in that the relevance ofthis signal is highest in the nascent stage and declines asa club moves through the life cycle until its role as a signalis lost in the mature stage. In contrast, the impact of brandequity is subject to a pro-cyclical movement as the effectof brand equity on consumer purchase behavior is estab-lished in the growth stage and bolstered in the maturestage. Thus, team expenditures as a low-scope signal arecompletely replaced by the high-scope brand equity sig-nal across the OLC. These results suggest that much priorresearch on signaling effects may be subject to a maturitybias, as previous studies have mostly analyzed establishedfirms that may be in the mature stage (Joshi and Hanssens2010).

Second, prior research has not investigated interrela-tionships (e.g., bi-directional causality) between multiplesignals (Kirmani and Rao 2000). Consideration of inter-relationships is important because one signal could be aprerequisite for another signal to unfold in the future. Asa consequence, simultaneous investments into both sig-nals or, even worse, sequential investments in the wrongorder would harm efficiency (Engers 1987). Thus, wesimultaneously consider two signals – team expendituresand brand equity – and explore how the causalitiesbetween them evolve over a club’s life cycle.

We find that reverse causality between our signalsexists. The results show that team expenditures help inestablishing the brand equity signal, implying that playersare a powerful medium for building brand equity. Further,a strong brand serves as a performance claim, and sportsclubs are eager to fulfill this promise by increasing teamexpenditures as a vehicle to secure superior on-fieldperformance. However, the findings suggest that the twocausalities that form the interrelationships are temporarilyasymmetric. That is, the reinforcement pattern of both


signals follows a sequential trajectory, where the causal-ity between the signals is fully reversed over time. Acrossthe OLC stages, team expenditures lose their brand-building capability while the brand evolves as a strongpredictor of team expenditures, showing that inter-signalrelationships strongly depend on the OLC stage. Theseresults encourage future studies with multiple signal set-tings to consider interrelationships between different sig-nals.

Third, this study offers one of the first large-scaleempirical tests of OLC theory, which has not been testedcomprehensively owing to a lack of data and methodol-ogy (Jap and Anderson 2007). This undertaking is note-worthy. While the literature on the OLC proposes thecycle to be a fairly rigid sequence of stages with sharpboundaries between the stages, it does not provide empiri-cal support for the number and length of stages (Jawaharand McLaughlin 2001). In this investigation, we takeadvantage of the strength and flexibility of a mixedanalysis methodology by employing vector autoregressive(VAR) modeling and nonlinear regressions. VAR model-

ing is a powerful technique that accounts for the richdynamic structure of longitudinal settings, because inVAR models the effects between variables are not restric-ted a priori in direction, time, sign, or magnitude (Dekimpeand Hanssens 1999). Nonlinear regressions are helpful inidentifying the stages of the OLC in a scientifically soundmanner. This identification allows us to fit stage-specificVAR models and to reveal shifts in the importance of thesignals across different life cycle stages.

The results imply a very slow movement through thelife cycle with relatively long-lasting stages. Thus, theprofessional soccer market can be characterized as a slow-growth and stable industry where clubs pass throughpredictable stages.

In summary, our goal is to shed light on the dynamicsof signaling – a vital but poorly understood issue of thisimportant marketing instrument. We hope that our paperprovides an impetus to other researchers to make thedynamics of signaling the focus of their work. Referencesare available upon request.

For further information contact:Stefan Hattula

Marketing DepartmentUniversity of Mannheim

L 5, 168131 Mannheim

GermanyPhone: +49.621.181.1575



EFFECTS OF STORE MANAGER CLIMATE UPON FLE COMMITMENT,CUSTOMER LOYALTY, AND STORE FINANCIAL PERFORMANCE

George D. Deitz, The University of MemphisJohn D. Hansen, University of Alabama – Birmingham

Thomas E. DeCarlo, University of Alabama – BirminghamEmin Babakus, The University of Memphis

Kristopher J. Preacher, Vanderbilt University, Nashville

SUMMARY

Because frontline employees (FLEs) can be an im-portant source of competitive advantage (Beatty et al.1996), establishing a favorable work climate thatenhances FLE attitudes and behaviors has been posited ascritical factor in improving firm performance. Indeed, it isnow commonplace for firms to explicitly to formalize“employee-first” corporate strategies within their missionstatements and annual reports. In recent years, however,top managers seeking to sustain employee-focused strat-egies have been faced with the difficult act of balancingthe interests of FLEs with those of investors and other keystakeholders. Such disconnects between employee-centered strategies and FLE perceptions and attitudestoward the organization are widespread, suggesting apervasive failure in implementation.

Several scholars have pointed to the importance ofworkplace climate in seeking to resolve questions relatingto human resource strategy implementation gaps. Organi-zational climate is a key element in firm efforts to influ-ence FLE attitudes and behaviors because it represents theshared perceptions of the activities and practices that aresupported and rewarded by the organization (Reichersand Schneider 1990). To date, however, the role of theretail store manager in enacting favorable climates andfostering higher FLE commitment has received insuffi-cient attention. This study presents findings from a mul-tilevel model linking store managers’ psychological cli-mate of the firm’s concern for employees with strongerFLE commitment, which in turn, affects customers’ will-ingness to promote the store and sales growth.

Open systems theory suggests that what happens toone individual affects other individuals within the samenetwork (Gharajedaghi 1999). With respect to spillovereffects deriving from managers’ perceptions, the presentstudy focuses upon store manager climate pertaining totop management’s concern for employees (CFE). Climatetheory suggests that individuals’ climate perceptions areformed based upon cognitive evaluations of employeeexperiences with organizational policies and practices.For most boundary-spanning workers, the content ofthese interpretations as well as the degree to which shared

perceptions of climate emerge are shaped by their interac-tions with managers. Supervisory support behaviors onthe part of a store manager signal a credible commitmentto FLEs that top management’s “people-first” message isnot simply a slogan. In turn, high levels of supervisorysupport may create feelings of obligation, whereby em-ployees feel an obligation to engage in behaviors thatsupport organizational goals. Greater FLE commitment,in turn, should favorably influence customer attitudes andpromote sales growth. Based on this brief discussion, wepropose:

H1: Store manager supervisory support positively medi-ates the relationship between store manager psycho-logical climate-CFE and FLE commitment.

H2: FLE commitment will be positively associated withsame store sales.

H3: FLE commitment will be positively associated withstore NPS ratings.

H4: Store NPS ratings will be positively associated withgrowth in same store sales.

H5: Store manager climate-CFE will be positively associ-ated with store NPS ratings.

The extensiveness of a focal store’s competitiveintensity (CI) is an especially salient factor within retail-ing contexts. We reason the extra demands placed uponstore managers in such environments may make it moredifficult for them to devote as much time to coaching andmentoring individual FLEs. Thus:

H6: The indirect effect of store manager psychologicalclimate-CFE upon FLE commitment through super-visory support will be negatively moderated by CI.

Findings

Hypotheses were tested using multi-source data col-lected as part of a comprehensive employee and customersurvey sponsored by a national retail chain supplementedby corporate archival data. The data followed a two-level


framework and were analyzed using MPlus 6.2. The maineffects model showed adequate fit to the data (Satorra-Bentler χ2(26) = 37.35, RMSEA = 0.012). Using the prod-uct of coefficients method for testing mediation, we foundthe indirect path proposed by H1 was significant (β =0.02, p < .01). The relationship between commitment andsame store sales advanced in H2 found mild support (β =0.02, p = .06), but the proposed link between commitmentand customers’ store NPS ratings (H3) was not signifi-cant. Store NPS was positively associated with store salesgrowth (β = 0.01, p < .01), supporting H4. Store managerclimate would be positively associated with store NPSratings (β = 0.02, p < .05), bolstering H5.

H6 proposed that the Store Manager Climateà-Supervisor SupportàFLE Commitment mediation sequ-

ence supported in H1 was a conditional indirect effect,moderated by the level of CI facing the focal store. Asinitial results provided but moderate support (β = -0.03,p < .10), we proceeded to probe the significance of thispath at varying levels of CI (Bauer and Curran 2003). Wefound the effect was significant at low (β = 0.036, p < .01)and moderate (β = 0.023, p < .01) levels, but not at high(+1 SD; β = 0.011, p = .19) levels of CI. Thus, weconcluded H6 was supported. Further, we noted the Com-mitmentàSales Growth parameter was significant afteraccounting for the interaction effect, providing insight tothe more tepid support for H2 in the main effects model.The full paper concludes with a discussion of the theoret-ical, methodological and managerial implications of thesefindings. References are available upon request.

For further information contact:George D. Deitz

The University of Memphis302 Fogelman College Admin Building

Memphis, TN 38152Phone: 901.678.2667

E-Mail: [email protected]


DO GROUP-BUYING DEALS INDUCE MORE COUPON REGRET?

Xueming Luo, University of Texas at Arlington, TexasYiping Song, Fudan University, China

Pengdong Fan, East China University of Science and Technology, China

SUMMARY

Recently, online daily group-buying deals have seizedincreasing attention among marketers. According to theLocal Offer Network’s study, in the U.S. market, rev-enues from online daily deal and discounts sites such asLivingSocial, topped $2.67 billion in 2011, up 138 per-cent from 2010. The leading company, Groupon, wasvalued recently at over $15 billion and has expanded itsbusiness to 88 cities in the U.S. and 230 markets of 29countries. In China, the online group-buying was launchedin January 2010, and more than 1,200 companies hadentered into this promising area by August 2010. Lashou,the leading group buying site in China, covers more than100 cities such as Shanghai. It successfully launched 659deals with daily revenues of about 4 million RMB (1US$ =6.3RMB) in its first six months.

An interesting phenomenon of online group-buyingis that while consumers prepay the online coupon, not allof them will redeem it. Industry surveys suggest thatapproximately 20 percent to 32 percent of the purchasedonline group-buying coupons are not redeemed. Millionsof deals, savings, and even the pre-payments for consum-ers are lost eventually each year. From traditional eco-nomic theory, coupon price is a sunk cost because con-sumers have already paid for the group-buying coupon.So, failing to redeem it would cause consumers to feelregret due to wasted spending. Additionally, prospecttheory suggests that people are loss-aversion biased. Thus,consumers’ loss perception would motivate them toredeem the coupon. In other words, paying for the couponwould not only enhance consumers’ rational thinkingtoward their coupon purchases, but also motivate them toredeem their coupons to avoid the sunk cost. As such, ifnot redeemed, the group-buying coupons are a net loss,and coupon regret is inevitable.

Given the popularity of online group-buying deals onthe one hand and the large amount of unredeemed deals on

the other, this research seeks to model the coupon redemp-tion of group-buying and accounts for the price sensitivityof coupon regret. A key difference between online group-buying coupons and offline traditional deals is that con-sumers get traditional coupons for free, but must pay forgroup-buying coupons. Prior studies have modeledredemption of traditional free coupons. Particularly, Wardand Davis (1978) developed a coupon response model inwhich the redemption rate monotonically decays overtime. Later, Inman and McAlister (1994) extend themodel and find that after the initial decay, the couponredemption rate climbs back up until the expiration date,the effect they dubbed as “coupon regret.” However, theseprior models address traditional coupons for which thecoupon price is not considered. Yet, price is always acrucial factor influencing consumer behaviors. Thus, with-out explicitly accounting for coupon price, prior modelsmay not adequately explain the redemption rates for therecent business platform of group-buying coupons.

This study proposes a new coupon redemption modelfor the online group-buying deals. Compared to the exist-ing models, its advantage is three-fold: It fixes the redemp-tion “data overflow” problem. Its model fitness is signifi-cantly better in terms of in-sample and out-of-sampleindices, along with better predictions of redemption rates.It includes coupon price as a key variable. Price cannotonly affect group-buying coupon redemption rates anddecay, but also affect coupon regret. Substantively, thenew model reveals that coupon prices influence couponregret: low prices reduce it, and high prices enhance it, butmedian prices have no impact. Follow-up analyses sug-gest that the price sensitivity of coupon regret is driven bycoupon values, business maturity, and deal popularity.We also find that deal popularity of group-buying is notdriven by coupon price or face value, but rather bybusiness maturity, deal debut time, and deal selling dura-tion. References are available upon request.

For further information contact:Xueming Luo

Eunice & James L. West Distinguished Professor of MarketingThe University of Texas at Arlington

701 S West StArlington, TX 76019Phone: 817.272.2279,

E-Mail: [email protected]://www3.uta.edu/faculty/luoxm/


DISTRIBUTION OF CHANGES IN CONSUMERPURCHASING BEHAVIOR

Giang Trinh, University of South AustraliaCam Rungie, University of South Australia

Malcolm Wright, Massey University, New ZealandCarl Driesener, University of South Australia

John Dawes, University of South Australia

SUMMARY

Marketers spend substantial sums of money to changeconsumer behavior toward their brands, for example, byincreasing consumers’ propensity to buy the brand. Muchresearch has been done to explain how different marketingstimuli (e.g., sales promotion, advertising, distributionbreath, and product innovation) lead to such changes,however, there has been little research on the form thatsuch changes take. That is, how are increases in salesdistributed among brand buyers? Theoretically, as previ-ous research has not identified the form that sales changestake, it has not comprehensively explained the relation-ship between marketing stimulus and changes in purchas-ing behavior.

Several authors have proposed models for bench-marking changes in consumer purchasing behavior. Thetwo best known such models are the negative binomialdistribution (NBD) and the Pareto/NBD. Both have beenused with considerable success in benchmarking changesto buyer behavior. The NBD model is used for brands inpackaged goods context to identify whether an overallsales change is accounted for by the previous non buyers,light buyers or heavier buyers of the brand, whereas, thePareto/NBD model is used at the organizational context toidentify inactive customers (those who change from activeto inactive purchasing). Yet a critical question of analyz-ing changes in consumer behavior has not been addressed:when there is a sales trend, what is the distribution ofchanges – are the changes due to nudging or radical con-version? This is an important question with obvious mana-gerial applications.

Suppose a brand manager would like to increasebrand sales. Essentially, the two main options are to directmarketing efforts toward making a small change in pur-chase propensity among a large group of buyers, oralternatively toward making a big change in the purchasepropensity of a small group of buyers. The question ofwhich option to take, and how sales increases will then bemanifested in the distribution of purchases will inform thecorrect choice of marketing strategy. For example, if salesincreases mainly come from a small group of buyers then

segmentation and targeting play a potentially importantrole in brand growth. On the other hand, if sales increasescome from a large group of buyers, increasing reachbecomes crucial, and so consequently mass marketing ismore important.

However, typically brand managers cannot answerthis question due to the stochastic nature of actual pur-chasing behavior. For example in a two-year period,comparing year two to year one, some buyers would haveincreased their brand purchasing by one unit and someother buyers would have increased by five units, andsome no longer buy, even if the overall brand sales areunchanged. This may all be normal stochastic fluctuationin purchases in a stationary market, including the normalregression to the mean expected from each buyer class inthe subsequent period. In order to find out which group iscausing sales increase or decline, a benchmark that quan-tifies this stationary behavior, and stationary market re-gression to the mean, is needed. Unfortunately, there hasnot been any method available to benchmark changes inbuyer purchasing behavior at this disaggregate level,particularly as empirical benchmarks will confound regres-sion to the mean with dynamic change.

This paper therefore proposes a method that helpsbrand managers answer the question of how sales changesare distributed across the population of buyers. The methodis based on the distribution of changes that is generatedfrom a special case of a bivariate compound Poissondistribution for the analysis of consumer behavior in twosequential time periods, where the compound Poissondistribution for the second period is assumed to have thesame parameters as that of the distribution for the firstperiod (stationary condition).

The distribution of changes describes the stochasticnature of changes in purchase frequency. Some buyerspurchase the brand one, two, three…x times more, somebuyers purchase less, and some others purchase the sameamount from one time period to another. The methodallows us to identify those classes of buyers who havechanged their buying behavior more than is expected, bycomparing the changes against the bivariate compound


Poisson benchmarks that describe the expected stochasticregression to the mean. The method creates theoreticallyexpected benchmarks of the extent of changes in purchas-ing behavior. Importantly, comparison between the obser-ved data and the predicted distribution of changes allows

brand managers to determine if a sales increase or declineis due to a small shift in purchase propensity of a largegroup of buyers, or a big change in purchase propensity ofa small group of buyers. Practical examples of the methodare also presented. References are available upon request.

For further information contact:Giang Trinh

School of MarketingEhrenberg-Bass Institute

University of South AustraliaGPO Box 2471

Adelaide SA 5001Australia

Phone: +61.8.830.20600Fax: +61.8.830.20442



ARE SUPPLIERS INACCURATE IN JUDGING THEIR PRICE FAIRNESSIN BUYER-SELLER RELATIONSHIPS?

Christian Homburg, University of Mannheim, GermanyJan Allmann, University of Mannheim, GermanyDirk Totzek, University of Mannheim, Germany

SUMMARY

In buyer–seller relationships, perceptions of suppli-ers and customers can differ on many issues. Such percep-tional differences are frequently inevitable. In particular,price fairness – a key concept for understanding priceevaluations – can be seen from two different points ofview. First, the customer firm has a certain perception ofthe supplier’s price fairness. Second, the supplier has hisown idea of price fairness. If the supplier believes hisprices to be fairer than actually perceived by the customer,a judgment inaccuracy on the side of the supplier occurs.Prior work implies that such inaccuracies could leadsuppliers to diminished relational efforts or investmentsin the business relationship.

In the B2B context, there are only few studies onjudgment inaccuracies or perceptional discrepancies inbuyer-seller relationships and none with respect to pricefairness. In particular, prior research highlights the needto identify factors driving judgment inaccuracies, ideallywith dyadic and cross-sectional data. This study thusaddresses three questions: Are suppliers in B2B marketsinaccurate in judging their price fairness? What are theconsequences of price fairness judgment inaccuracies forsuppliers? Which supplier pricing elements promote orweaken price fairness judgment inaccuracies?

Addressing these issues, the basic idea of our study isto analyze both the antecedents and outcomes of supplierprice fairness judgment inaccuracies. The inaccuracy ofprice fairness judgment is modeled as the differencebetween the supplier’s and the customer’s price fairnessjudgments. Price fairness is defined as the perception thatprices are right, just, or legitimate. We consider two keyoutcome variables of buyer-seller relationships account-ing for both relational (customer satisfaction) and price-related success (customer’s willingness to pay a pricepremium) with a customer firm. As perceptual inaccura-cies are very difficult to detect, identifying conditions thatfavor inaccuracies can support suppliers in becomingaware of the potential pitfalls of certain pricing param-eters. This study focuses on central supplier pricing pa-rameters in B2B settings and considers, in particular, thesupplier’s emphasis on cost-plus pricing, the emphasis ona low price positioning, and the degree of price differen-tiation. Furthermore, we address price interaction vari-ables, particularly, the seller’s emphasis on benefit selling

as well as the frequency and intensity of price negotiationsas potential preconditions impacting the supplier’s pricefairness judgment inaccuracy.

We empirically test our framework with a cross-industry sample of 150 matched buyer-seller dyads fromdifferent industries (e.g., manufacturing, metal process-ing, electronics, logistics, chemicals). Dyadic data camefrom a survey of purchasing managers (customer firm)and their respective sales contact person (supplier firm).An initial examination of our data reveals that supplierssubstantially self-enhance in their price fairness judg-ment: 119 (~79%) of the 150 suppliers overestimate theirprice fairness; 61 (~51%) of those do so by more than onescale point (on a seven-point scale). Twelve suppliersevaluate their price fairness equally to the customer; 19suppliers underestimate their price fairness.

To test our model and hypotheses we use an innova-tive technique: latent congruence modeling. The model isable to consider both the level and congruence of twointerdependent price fairness measures from suppliersand customers. Since latent congruence modeling allowsfor considering the general level of price fairness in thesame model, we are also able to examine the antecedentsand consequences of the level of price fairness in buyer–seller relationships.

Results show that all hypothesized relationships aresupported. In particular, supplier price fairness judgmentinaccuracies negatively affect the customer firm’s will-ingness to pay a price premium and satisfaction. Lookingat the supplier’s pricing preconditions, the use of cost-pluspricing and the degree of price differentiation negativelyimpact price fairness judgment inaccuracies. An empha-sis on low price positioning, however, has no effect. Theemphasis on benefit selling and the frequency of pricenegotiations reduce price fairness judgment inaccuracies.The intensity of price negotiations increases inaccurateprice fairness judgments. Finally, the results highlight thatthe level of price fairness is a positive determinant ofcustomer’s willingness to pay a price premium and satis-faction.

The findings of this study advance academic knowl-edge in four ways. First, suppliers are indeed highlyinaccurate in judging their own price fairness: They aresubject to a self-enhancement bias. Second, price fairness


judgment inaccuracies negatively affect the customer’swillingness to pay a price premium and satisfaction.Third, there are major pricing conditions under whichinaccuracies are more or less likely to occur. Fourth, latentcongruence modeling overcomes major weaknesses in-herent in other approaches to analyze congruence.

One practical implication of this study is that suppli-ers in business markets need to be aware of their tendency

to overestimate their price fairness. More importantly,both monetary (customer willingness to pay a price pre-mium) and relational (customer satisfaction) goals withina buyer-seller relationship are negatively impacted bybeing inaccurate with respect to price fairness judgments.Moreover, our findings provide suppliers in B2B marketswith guidelines as to which pricing setups are likely tolead to accurate versus inaccurate judgments. Referencesare available upon request.

For further information contact:Christian Homburg


68131 MannheimGermany

Phone: +49.621.181.1555Fax: +49.621.181.1556



GRATITUDE VERSUS ENTITLEMENT: AN ANTAGONISTIC PROCESSMODEL OF THE PROFITABILITY IMPACT OF CUSTOMER

PRIORITIZATION

Hauke A. Wetzel, Georg August University Goettingen, GermanyMaik Hammerschmidt, Georg August University Goettingen, Germany

Alex R. Zablah, George Mason University, Fairfax

SUMMARY

Customer prioritization, the allocation of organiza-tional resources to individual accounts based on theirimportance to the firm, is a critical element of relationshipmanagement practices in business-to-business (B2B)markets. Several studies have affirmed that customerprioritization initiatives lead to “bright side” or perfor-mance-enhancing outcomes such as customer gratitude(e.g., Homburg, Droll, and Totzek 2008; Palmatier et al.2009).

While these findings are encouraging, the literaturealso suggests that the desirable effects of customerprioritization programs on high-priority accounts may betempered by a “dark side” (Anderson and Jap 2005).Specifically, concerns are growing that prioritizationefforts may undermine account profitability by encourag-ing customers to expect more for less (Boyd and Helms2005; Lacey, Suh, and Morgan 2007). Surprisingly, empi-rical attention has not been devoted to understanding howand why this dark side undermines the performance-enhancing outcomes of prioritization, what triggers thisdark side as compared to the bright side, and when itprevails.

In an attempt to redress this critical knowledge gap,we extend the literature in several ways. First, we presentan initial study that empirically contrasts bright side anddark side phenomena of prioritization efforts. Specifi-cally, we consider customer gratitude as the key processvariable that accounts for the proposed bright side ofprioritization (Palmatier et al. 2009). We also introducecustomer entitlement (Boyd and Helms 2005) as a rel-evant construct to the literature on prioritization and arguethat it plays an important intervening role in the proposeddark side process of prioritization. This study thus impro-ves understanding of why prioritization efforts might alsoresult in undesirable (i.e., dark side) outcomes.

Second, this study improves our understanding ofwhy prioritization-induced responses influence customerprofitability (Drèze and Nunes 2009). While priorresearch finds that significant variation exists in the rev-

enue-to-service-cost ratios of top-tier customers (Niraj,Gupta, and Narasimhan 2001), a substantive explanationas to why this occurs is still lacking. We propose and findthat customer gratitude enhances account profitability byincreasing sales, while customer entitlement has the oppo-site effect by increasing the costs necessary to servicecustomer accounts. The findings thus suggest that varia-tion in the revenue-to-service-cost ratios of top-tier cus-tomers may be attributed to differences in the levels ofgratitude and entitlement each experiences when dealingwith a particular provider.

Third, by isolating bright side and dark side processeswe also offer critical insights regarding the differentiallevers that are likely to “trigger” both processes. In a cleardeparture from prior research, we disentangle prioritizationbenefits in order to answer the call for research to simul-taneously assess how different prioritization benefits com-bine to influence relationship outcomes (Henderson, Beck,and Palmatier 2011; Lacey, Sue, and Morgan 2007).Specifically, this research demonstrates that non-socialand social prioritization benefits operate in different waysto “activate” the competing processes that account for thebright side and dark side consequences of prioritizationefforts. Both non-social and social benefits are posited toincrease gratitude. In contrast, we find that entitlement –the intervening construct of dark side prioritization out-comes – is only triggered by social benefits.

Finally, we consider how contextual factors alter therelative impact of perceived prioritization benefits ongratitude versus entitlement. Therefore, our results offervaluable knowledge regarding the conditions under whichthe different types of perceived prioritization benefits aremore or less likely to have a positive or negative effect onaccount profitability.

In sum, our study results offer managers clear guid-ance as to the factors that trigger both desired and undes-ired prioritization outcomes and how context affects theprofits derived from prioritization efforts. Managers canutilize this knowledge to design prioritization programsthat emphasize the right benefits and maximize profitgrowth. References are available upon request.


For further information contact:Hauke A. Wetzel

Economic SciencesGeorg August University Goettingen

Platz der Goettinger Sieben 337073 Goettingen




DO SELLER PERCEPTIONS OF FAIRNESS LEAD TO SALES GROWTH?A LATENT GROWTH CURVE ANALYSIS

Ghasem Zaefarian, University of Leeds, United KingdomZhaleh Najafi Tavani, University of Leeds, United Kingdom

Stephan C. Henneberg, University of Manchester, United KingdomPeter Naudé, University of Manchester, United Kingdom

SUMMARY

Today’s competitive environment has increased theimportance of building and maintaining effective rela-tionships with supplying companies. The fundamentalassumption of supply chain management is that long-lasting relationships between a manufacturer and its sup-pliers can provide significant opportunities for gainingcompetitive advantages and achieving superior perfor-mance (Choi and Hartley 1996; Jap 1999; Johnston et al.2004). Perceptions of fairness in such relationships be-tween manufacturers and suppliers become a vital factorin the sustainability and quality of these long-term rela-tionships (Gassenheimer et al. 1998; Kumar et al. 1995b).

Despite the growing research interest in the issue ofperceived fairness (justice) in business relationships, areview of the extant literature reveals that the fairnessliterature is skewed toward research studies conducted inthe business-to-consumer. The fairness research in busi-ness-to-business settings in general and buyer-supplierrelationships in particular has been somewhat neglected.In addition, although some studies have investigated theimpact of fairness on different aspects of business rela-tionships, including trust and commitment, the literaturefails to provide empirical evidence for examining thelong-term effects of fairness on both relationship qualityand sales growth, specifically with regard to the seller’sperceptions of fairness. Finally, it has been argued that thedegree of dependence may lead to contradictory results injustice perception studies (Grégoire and Fisher 2008;Kumar 1996). Therefore, while fairness perceptions aswell as relationship quality may impact a suppliers’ salesgrowth, it is not clear whether such associations areequally relevant in situations of different degrees ofdependency within the business relationship. As such, weposit that the lack of attention regarding the issue ofdependency in examining the impact of justice percep-tions in manufacturer-supplier relationships limits ourcurrent understanding.

We aim to address these gaps by investigating thedirect and indirect impact of justice perceptions on rela-tionship quality and on sales growth, based on a sellerperspective. To do this, we report the results of a longitu-dinal study based on both subjective data collected from212 automotive suppliers in 2009 using where possibletwo key informants from the supplier side of the businessrelationship, and objective sales data for these suppliersfrom an automotive manufacturer, i.e., the buyer side ofthe dyadic buyer-supplier relationship over a three-yearperiod after 2009.

We employ a latent growth curve model, whichreveals that seller’s organizational justice perceptionsconsiderably enhance the quality of the relationshipbetween the automotive manufacturer and its AutomotiveParts Suppliers (APSs). Our findings also show a strongpositive effect of relationship quality on sales growth, inline with previous results (Kumar et al. 1995a). However,there was no direct impact of seller’s fairness perceptionson sales growth, i.e., the effect of seller fairness assess-ments are fully mediated (by relationship quality). Never-theless, this indirect effect of fairness perceptions on salesgrowth is considerable.

In addition, a multi-group moderation analysis sug-gests that whilst for suppliers with low levels of depen-dency, relationship quality significantly increases long-term sales growth, for those with high levels of depen-dency this effect becomes insignificant. One interpreta-tion is that when the APS is highly dependent on the carmanufacturer, it cannot easily switch to other manufactur-ers since such changes are considered costly and risky.Therefore, the APS maintains its partnership (and mayeven increase its collaboration activities) not because ithas a reliable long-lasting relationship but because it hasno other options.

Keywords: Relationship Quality, Organizational Jus-tice, Dependency, Fairness (Justice) Theory, Sales Growth.References are available upon request.


For further information contact:Ghasem Zaefarian

Leeds University Business SchoolUniversity of Leeds

Clarendon RoadLeeds, LS2 9JTUnited Kingdom

Phone: +44(0)113.343.3233Fax: +44(0)113.343.4885



RECIPROCITY IN INTERFIRM RELATIONSHIPS: A REVIEWAND EXTENSION

Jessica J. Hoppner, George Mason University, FairfaxDavid A. Griffith, Michigan State University, East Lansing

SUMMARY

The influence of reciprocity on interfirm relation-ships is well-documented within the extant literature (e.g.,Anderson and Weitz 1992; Jap and Ganesan 2000; Hoppnerand Griffith 2011). However, as the number of studiesexamining reciprocity has increased, our understandingof what reciprocity is and how it operates within interfirmrelationships has become increasingly removed from theoriginal conceptualization of reciprocity. Thus, thisresearch reexamines seminal statement on reciprocityunderlying the majority of interfirm research (i.e., Gouldner1960) to discuss the dimensionality and its universality ofreciprocity, and empirically evaluate how these elementsinfluence the relationship quality of interfirm relation-ships (e.g., a characteristic argued to develop from afoundation of reciprocity).

A review of the extant literature demonstrates thatdescriptions of the general reciprocation process remainconsistent within interfirm relationships (e.g., Campbellet al. 1988; Frazier and Rody 1991); that is, reciprocity asa norm prescribes that benefits should be exchanged inreturn for benefits previously received. However, incon-sistencies emerge in the description and recommenda-tions regarding reciprocity once the specifics of the recip-rocation process are considered, such as what isexchanged, when the exchange occurs, and how cultureinfluences these exchanges.

The dimensionality of reciprocity, via its dimensionsof equivalence and immediacy, specifies what isexchanged and when it is exchanged, respectively. Equiva-lence ranges from homeomorphic (i.e., identical in form)to heteromorphic (i.e., concretely different). Immediacyranges from short-term (i.e., minimal time betweenexchanges) to long-term (i.e., extended time betweenexchanges). Within the extant literature, the equivalencedimension of reciprocity has received the most attention;recommending that homeomorphic equivalence shouldbe used due to its theorized positive effect on relationshipquality. Comparatively, the immediacy dimension hasbeen virtually ignored by the extant literature, with nostrong recommendation for the effect of immediacy onrelationship quality have been proposed.

The universality element of the conceptualization ofreciprocity posits that reciprocity is universally appli-cable and pervasive in all exchange relationships, but is

not necessarily unconditional. Within the extant litera-ture, neither of the two forms that the universality ofreciprocity may take, as an unconditional universal normor a culturally conditioned universal norm, has beenstrongly advocated for.

The data from this study comes questionnaires com-pleted by managers of interfirm relationships between asupplier and their primary buyer based in the UnitedStates (n = 284) and in Japan (n = 296). The questionnaireincluded items to measure the dependent variable ofrelationship quality (i.e., a second order factor of satisfac-tion, conflict, and willingness to invest), the independentvariables of reciprocity (i.e., equivalence and immediacy)and the control variables of relationship length and per-cent of business conducted within the interfirm relation-ship. A two-group structural equation model was used totest the hypotheses.

The results of this research demonstrate that thedimensionality and the universality of reciprocity influ-ence how relationship quality develops within interfirmrelationships. For the dimension of equivalence, the like-for-like exchanges required by homeomorphic equiva-lence were found to be beneficial to the quality of therelationship. It appears that the simplicity in recognizinghomeomorphic equivalence reduces the uncertainty sur-rounding the possibility of one’s partner failing to recip-rocate (or failing to recognize when one’s partner hasreciprocated).

For immediacy, the results demonstrate a more com-plicated relationship. In certain interfirm relationships(i.e., United States), short-term immediacy has a negativeeffect on relationship quality; whereas in other interfirmrelationships (i.e., Japan), it has no effect. It appears that,when the negative effect exists, the quickness ofexchanges required by short-term immediacy is harmsrelationship quality by eliminating the constraintsimposed by the social norm to continue performing coop-erative behaviors.

As for the universality of reciprocity, the dimensionsof reciprocity were found to exist consistently acrosscultures, whereas the effects of reciprocity demonstratesome cultural peculiarities (i.e., equivalence is consistent;immediacy varies). This finding suggests that reciprocityis more appropriately described as a culturally condi-tioned universal norm. This initial insight into the nature


of the universality of reciprocity is consistent with theoriginal conceptualization that noted while reciprocityshould exist in all cultures; the norm may function differ-

ently in some degree in different cultures as its concreteformulations may vary with time and place. Referencesare available upon request.

For further information contact:Jessica J. Hoppner

School of ManagementGeorge Mason University

4400 University Drive, MS 5F4Fairfax, VA 22030

Phone: 703.993.1796Fax: 703.993.1809



SALES AND VALUE CREATION: A SYNTHESIS AND DIRECTIONS FORFUTURE RESEARCH

Alexander Haas, Giessen University, GermanyNina Stuebiger, Giessen University, Germany

SUMMARY

Understanding the role of sales in the creation ofvalue to firms and their customers has been a long-standing goal of researchers and managers alike (Lindgreenand Wynstra 2005). This comes as no surprise as sales-people are thought to play a pivotal role in the creation ofcustomer value (Weitz and Bradford 1999) and scholarswidely recognize the creation of customer value as the keyto firms’ long-term survival and success and increasinglysee creating customer value as the next source of competi-tive advantage (Ulaga and Eggert 2006; Woodruff 1997).

How does the sales organization contribute to thecreation of value to a firm and its customers? In theirreview of relevant literature, Haas, Snehota, and Corsaro(2011) convincingly argue that prior research on businessrelationships has dealt with sales’ value-creating rolemostly per assumption and not systematically. Researchhas focused on selling approaches, exploring issues suchas consultative selling (e.g., Liu and Leach 2001), rela-tionship selling (e.g., Frankwick, Porter, and Crosby2001), and, particularly, customer-oriented selling (e.g.,Saxe and Weitz 1982; Stock and Hoyer 2005) and adap-tive selling (e.g., McFarland, Challagalla, and Shervani2006; Spiro and Weitz 1990). Despite scholars’ extensivefocus on sales’ performance outcomes, the two mostprominent salesperson behaviors under investigation inthe sales literature (i.e., adaptive selling and customer-oriented selling) have been shown to account for only 9%or less of the variance in salesperson performance (Frankeand Park 2006). And research that explicitly addresses thequestion of how the sales function adds value to thecustomer is still in its infancy. Accordingly, Singh andKoshy (2010, p. 2) observe that “we do not yet know ifbusiness-to-business salespersons actually create value intheir relationship with customers.” As it is still unclearhow sales creates value to firm and customer, scholarstried to intensify sales-related research on value creationby calling for research on “how specific ways of imple-menting customer value orientation relate to organiza-tional performance” (Woodruff 1997, p. 151) and identi-fying sales’ value-creating role as an important avenue forfuture research (Marshall and Michaels 2001).

The objective of this paper is to enhance the under-standing of sales’ role by examining the question of howvalue originates and is enhanced through sales in transac-tional and relational processes between business partners.Specifically, the paper investigates two main research

questions: (1) What is the mechanism through which salesmay contribute to the creation of value to the firm and itscustomers? (2) What are the organizational capabilitiescritical for sales to implement the value-creating mecha-nism?

To answer these questions, this paper (a) develops aninteraction-based framework of sales’ role in the creationof value; (b) suggests the concept of sales’ value-creatingcapabilities as key mechanism for the creation of value;and, based on this framework, (c) identifies fruitful areasfor future research which may guide systematic investiga-tion of sales’ value-creating role in business relationships.

The paper contributes to previous research by synthe-sizing current empirical macro sales research and provid-ing an interaction-based framework of the value-creatingrole of sales. It integrates extant value and sales literature,proposes two perspectives of sales’ role as value creator,and defines value creating capabilities as a set of organi-zational abilities related to a firm’s sales function. Thisresearch is rooted in the concept of interaction and in linewith current research on firms’ capabilities (e.g., Morganet al. 2009). To reflect the four intertwined characteristicsof value-creating interactions (Haas, Snehota, and Corsaro2011), sales functions’ value-creating capabilities areconceptualized as consisting of four intertwined facets:mediating, initiating, realizing, and sense-making. Thesefour dimensions of sales value creating capabilities serveas a key mechanism, which enhances our understandingof sales’ role in the creation of value to the firm and itscustomers. As such, the present research provides a foun-dation for the systematic development of an interaction-based theory of value creation and sales’ role in it.

In managerial terms, this paper highlights importantareas sales managers should devote attention to (e.g., interms of processes, sales training, etc.) in their efforts tocreate customer value and enhance sales performance.The proposed capabilities provide starting points foranalyzing failures in the creation of customer value. Ourresults also inform firms about ways to design the salesfunction so that it will create customer value. Thus, ourresearch supports firms and sales managers in their effortsto increase performance and customer value, and tostrengthen the firms’ competitive positions. Referencesare available upon request.

Acknowledgement: We thank the Swiss NationalScience Foundation (SNSF) for a grant supporting thisresearch.


For further information contact:Alexander Haas

Giessen UniversityLicher Strasse 66

35394 GiessenGermany

Phone: +49.641.99.22401Fax: +49.641.99.22409



IMPLEMENTING SERVICE GROWTH STRATEGIES AT THEINDUSTRIAL SALES FORCE LEVEL: KEY CHALLENGES

IN SELECTING AND MANAGING THESERVICE-SAVVY SALES FORCE

James M. Loveland, HEC Montreal, MontrealWolfgang Ulaga, IMD, Lausanne

SUMMARY

Manufacturing companies are increasingly seekingservice-led growth to secure their existing positions and toexpand in competitive and commoditized markets.Although this trend is well noted in the literature (Ostromet al. 2010), managers are still reporting problems withtheir service transitioning strategies. For example, Stanleyand Wojcik (2005) note that half of all solution providersrealize only modest profitability, and 25 percent actuallylose money with their value-added service offerings.Clearly, there is considerable risk for firms not cognizantof the challenges in transitioning to a service-centricbusiness model in B2b markets, and we still know littleabout what exactly drives the success or failure of servicegrowth strategies in B2B firms (Bolton et al. 2007). Someevidence suggests that the sales force represents a majorhurdle in moving toward a service-centric business model.For example, Reinartz and Ulaga (2008) discuss severalcases of firms experiencing strong resistance to changefrom within the sales organization. They note that, evenafter extensive training, firms experienced high levels ofchurn among sales people and had “little choice but to fireand hire; a few in our study replaced 80 percent of theirexisting sales forces.” Similarly, Ulaga and Reinartz (2011)suggest that only one third of industrial sales people easilytransition to hybrid offering sales, while two thirds eitherrequire considerable retraining or prefer to be reassignedto goods-centric sales. Given the key role of salespeoplewhen shifting to a service-centric business model inmanufacturing firms, research is needed to generate adeeper understanding of the distinctive sales capabilitiesneeded for selling hybrid offerings. While this touches onissues in organizational and managerial domains, wefocus on the individual salesperson level. Executivesconsistently point to the HR challenges of hiring andmotivating the “right” service sales reps above and be-yond their existing goods-centric sales force. Against thisbackdrop, we propose our research questions:

1. What capabilities are needed to successfully sellhybrid offerings in B2B markets?

2. Which personality traits resonate in effective hybrid-offering salespeople?

Method and Discussion

To address our research questions we conducted aseries of focus groups and depth interviews (18 and 20,respectively) with carefully chosen C-level executivesand senior sales managers in charge of growing servicerevenues in traditionally goods-centric organizations withstrong records in goods-dominant sales but that have beenmoving toward a service-led business model. Using anabductive approach, we developed several importantinsights from both an academic and a managerial perspec-tive.

First, our findings emphasize the pivotal role of theindustrial sales force in successfully mastering servicetransitioning strategies. Firms often underestimate themagnitude of change required at the sales organizationlevel; even with a dominant market position, a successfulgoods-centered sales force can fail when venturing intohybrid offering sales. Our study also reveals severaldistinctive characteristics regarding the nature of thehybrid offering sales process, as opposed to traditionalgoods-centric sales. These specificities refer to (a) a salesmodel firmly grounded in a co-creation perspective (b) anemphasis on specifying hybrid offering requirements incooperation with the customer (c) a broader and deepernetwork of stakeholders involved, both in the customer’sand vendor’s organizations, and (d) a focus on growingcustomer share throughout the vendor’s installed base.

We also identify four key capabilities of salespeople:(1) an ability to gain understanding of a customer’sbusiness model and operations and a capacity to leveragethis intimate knowledge for identifying opportunities ofselling hybrid offerings; (2) an ability to reach beyond hisor her comfort zone of established contacts and manage acomplex network of relationships in the customer andvendor organizations; (3) a capacity to proactively man-age customer expectations to ensure profitability overtime for contracts attached to hybrid offering sales, and (4)an ability to practice value selling in the context of hybridofferings, i.e., a capacity to help customers understand thevalue of the intangible service elements as part of indus-trial good-service combinations. Fourth, we identify per-sonality traits that are potentially relevant in a hybrid


offering sales context: learning orientation, customer ser-vice orientation, intrinsic motivation, general intelligence(“g”), emotional stability, teamwork orientation, and intro-version. Most interestingly, we find that some traits thathave been identified in the sales literature as highlyrelevant in goods-dominant sales appear as non-relevant(or even detrimental) in a hybrid offering sales context. Insum, our findings suggest that the profiles of a high-performing goods sales person and a “stellar performer”in hybrid offering sales diverge, confirming that goodssales people may indeed be “from Mars,” while hybridoffering sales people appear to be “from Venus” (Ulagaand Reinartz, 2011).

Academic Implications. Prior studies have called forfurther research on the specific resources and capabilitiesneeded to master service transitioning strategies in busi-ness markets (Ulaga and Reinartz 2011). Our researchsheds light on the specific resources and capabilitiesneeded in the sales arena. Scholars have also argued thatexisting sales models do not account for the increasingcomplexity of B2B sales contexts (Plouffe et al. 2008).Here again, our findings provide important insights intothe nature of the hybrid offering sales process, the capa-bilities required, and the personality traits that enablesales in such a complex context. Our findings also helpexplain some of the discrepancies found in the salesliterature on the linkages between job performance andpersonality traits of industrial sales people. Prior researchhas tended to view all sales as essentially being the same:our results highlight the importance of the sales context

when determining the traits necessary for a particularsales job.

Managerial Implications. Hybrid offering sales arenot simply an extension of goods sales. While companiesmay be able to grow sales of standard product life cycleservices, such as extended warranties attached to equip-ment sales, with their existing industrial sales force, itwould be detrimental to expect competent and experi-enced goods-centric salespeople to go about “business asusual” when selling complex combinations of goods andservices. In particular, our data suggest that there is aserious gap between the demands being placed on indus-trial sales forces and their capability to effectively sellhybrid offerings, evidenced by the high rates of turnoveramong (otherwise highly efficient and effective) salesreps allocated from goods sales to hybrid offering sales.Our findings also suggest that top management mustactively manage the transformation process. A company’sindustrial sales force plays a pivotal role in succeeding aservice transitioning strategy, and steering the industrialsales organization to a service-centric sales modelrequires full attention from C-level management. Ourresults not only show that this move touches on how salespersons are recruited and allocated to sales jobs, they alsoshow that firms need to reconsider their sales organiza-tion, develop coordination mechanisms for specializedsales forces, or redesign incentive structures to align theirsales organizations with overall corporate strategies. Ref-erences are available upon request.

For further information contact:James M. Loveland

HEC MontrealService de l’enseignement du Marketing

3000, chemin de la Côte-Sainte-CatherineMontréal, QC H3T 2A7

CanadaPhone: 514.340.6824

Fax: 514.340.5631E-Mail: [email protected]


READY TO PITCH? PROPOSAL AND VENDOR ATTRACTIVENESS ASMEDIATORS OF VENDOR SUCCESS IN COMPETITIVE TENDERS

Eva K. Steinbacher, University of St. Gallen, SwitzerlandChristian Schmitz, University of St. Gallen, Switzerland

Dirk Zupancic, German Graduate School of Management and Law, Heilbronn

SUMMARY

Over the last decade, organizational purchasers haveincreasingly employed competitive tenders when pur-chasing products or services. These tenders are initiatedand controlled by professional purchasing agents. Inorder to acquire profitable deals, vendors need to adapt toindividual tender rules and tailor their offerings to uniquelyfit a specific customer’s needs. During the last fiscal year,the United States Army, as a single institution, purchasedgoods and services worth 200 billion USD via competi-tive tenders. In February 2011, Boeing announced that ithad won a 35 billion USD deal in a historical tender of theUnited States Air Force. The other participants, includingNorthrop Grumman and EADS, left this particular com-petition empty-handed. Tenders are not a phenomenonexclusive to public institutions or the airline industry. Dueto rapidly developing Internet technology and highlytransparent national and international markets, the rise oftenders may be observed in many other industries (Fuller2004; Jap 2007; Jap and Haruvy 2008). Customers intendto enhance comparability among vendors’ proposed solu-tions while establishing objective supplier selection crite-ria. By preparing a detailed requirement specification,customers avoid unneeded functions and components ofthe purchased objects, thereby reducing overall procure-ment costs (Chen 2008).

This change on the buying side poses major chal-lenges to vendors’ salespeople. To successfully operate intheir market and participate in tenders, salespeople needto understand the customer’s buying process and signalsuperior attractiveness in the transaction (Johnston andLewin 1996). Vendors with traditional sales forces faceparticularly high failure rates and waste considerableresources on unsuccessful participation (Jap and Haruvy2008). If buying companies prepare requirement specifi-cations and make decisions based solely on proposals thatcan be developed in a back office without any personalcontact between vendor and customer, do vendors needsalespeople in tenders at all? Unsurprisingly, there ismajor theoretical and managerial interest in the concep-tualization of the role of salespeople in tenders and theirremaining influence on vendor success. Despite its rel-evance, current research lacks a systematic concep-tualization and empirical investigation of this issue.

The tender literature originates predominantly fromthe research field of supply chain management, with anincreasing number of publications found in the marketingliterature. These contributions focus on selected elementsof the tender rather than customer responses and vendorsuccess. The majority of publications addresses instru-ments of vendor selection used in tenders, such as biddingprocedures and their effects on buyer-supplier-relation-ships and firm performance (Burguet and Perry 2009;Carter and Kaufmann 2007; Chen-Ritzo et al. 2005;Donaldson 1996; Engelbrecht-Wiggans and Katok 2006;Jap 2007; Jap and Haruvy 2008; Schmeltzer and Carr2003; Smart and Harrison 2003; Tassabehji et al. 2006;Taylor 2005). In disregarding the importance of an appro-priate sales organization in tenders, the research has failedto address salespeople’s influence on vendor success in aworld of aspiring tenders. Accordingly, the existingresearch does not provide implications for salespeople’sfunctional behavior when facing tenders.

In this study, we examine how salespeople’s engage-ment during the buying process may influence (a) therelational attractiveness of the vendor, (b) the transac-tional attractiveness of the submitted proposal, and (c)vendor success in competitive tenders. Our study contrib-utes to existing research in two major ways. First, wedevelop a conceptual model explaining the effect ofvendor engagement during the buying process on rela-tional and transactional attractiveness as well as on ven-dor success. Second, we specifically address the moderat-ing influence of the proposed price in a context of almostidentical proposals, which has raised major questionsconcerning the role of the sales force. Thus, we generateinsights into salespeople’s value in crafting successfulinterorganizational exchange in an increasingly impor-tant purchasing approach. The model was tested empiri-cally on a cross-industry data set of 170 salespeople andshowed good fit to the data. The analyses confirmed directeffects of both relational and transactional attractivenesson vendor target achievement. We further identified directeffects on relational attractiveness of vendor engagementin the specification, execution, and post-tender phases.Finally, we confirmed a direct effect and a moderatingeffect on transactional attractiveness of vendor engage-ment in the specification phase and the relative price.


These findings are highly relevant to the tenderliterature because we provide not only a conceptualizationof vendor success and the buying process in competitivetenders but also a comprehensive explanation of theinfluence of vendor engagement in those tenders. Thisstudy also contributes to cooperation research. We showthe importance of vendors’ relational attractiveness incompetitive tenders. Our findings indicate that buyingfirms fail in their effort to eliminate subjective decisioncriteria. As long as there are interaction opportunities,vendors will produce different relational attractivenessdepending on their engagement in the buying process.This attractiveness affects the buying firm’s decisionmaking. Furthermore, this study contributes to the nascentcustomization literature by investigating the effect onsupplier target achievement of transactional attractive-ness brought about by a vendor’s customization efforts.We find that customization is particularly relevant incompetitive tenders. Vendors engaging in the specifica-tion phase to learn about the customer’s needs and deve-lop customized solutions based on this information gener-ate higher transactional attractiveness and commensu-rately higher vendor target achievement.

Moreover, the findings have implications for ven-dors’ management. We provide insights into how sales-people, by producing both relational and transactionalattractiveness, affect vendor target achievement through

their engagement in the tender-specific buying process.Our findings show that even though the proposed priceplays a prominent role in tenders, there is a remaining needfor discerning salespeople to seize opportunities duringthe buying process that increase transactional and rela-tional attractiveness. Rather than simply proposing a lowprice, transactional attractiveness increases when a com-petitive price is coupled with engagement in the specifica-tion phase. Furthermore, close interaction helps to reducethe customer’s perceived transaction risk, producing rela-tional attractiveness regardless of the proposed price. Ifvendors simply await a specification of requirements andhave a back office prepare a proposal without sending asales force to engage in the buying process, the vendor isunable to produce superior relational and transactionalattractiveness to influence vendor success in the tender.

This study has limitations that provide avenues forfuture research. We expect buying firms to employ differ-ent types of tenders depending on the product or servicethey wish to purchase. The range of types and theirvarying implications for salespeople call for further inves-tigation. Because engagement in the specification phaseappears to be particularly critical based on its relationaland transactional value, further investigation of this par-ticular phase of the buying process seems promising.References are available upon request.

For further information contact:Eva K. Steinbacher

Institute of MarketingUniversity of St. Gallen

Dufourstrasse 40aCH–9000 St. Gallen

SwitzerlandPhone: +41.71.224.7180



MANAGING EXCLUSIVE CHANNELS FOR RELATIONSHIPEFFECTIVENESS

Alberto Sa Vinhas, Washington State University, Vancouver

SUMMARY

The focus of this study is the effectiveness of channelrelationship management strategies or initiatives acrossdifferent channel types in business-to-business markets.In particular, we look at channel relationship managementin exclusive versus non-exclusive channels. In general,the channels literature provides limited insights into howsuch channel design strategies (i.e., usage of differentchannel types) impact channel relationship management.Exclusive dealing (ED) can be defined as a contractualrequirement by which resellers promise a supplier thatthey will not handle the goods of competing producers(Marvel 1982). ED is a common contractual arrangementacross industries (Heide, Dutta, and Bergen; Li and Dant1997; Marvel 1982). Despite its importance, there hasbeen a general lack of empirical evidence on how thesechannels can be managed effectively. We consider tworelationship efforts or characteristics: the level of infor-mation exchange and the level of conflict in a givenrelationship between the supplier and a distribution chan-nel. Consistent with previous literature, we find thathigher levels of information exchange and lower levels ofrelationship conflict lead to improvements in relationshipquality in both ED and non-ED channels. We then buildon existing channel management and economics litera-ture on exclusive dealing to show that the relative orcomparative effectiveness of these two channel relation-ship management efforts across channel types depends onthe specific initiative being considered. While the mar-ginal benefits of improvements in the quality of informa-tion exchange are higher in non-ED channels (than in EDchannels), the opposite is true for improvements (de-creases) in relationship conflict. In addition, and consis-tent with previous research (e.g., Li and Dant 1998), wefind that relationship quality tends, ceteris paribus, to be

higher in ED channels. We also consider the simultaneoususe of both channel types in a territory and show how asupplier’s usage of exclusive channels influences thequality of its relationships with non-exclusive resellers inthat territory. ED channels in a territory perform animportant demand generation function, benefitting allchannels in the vicinity. Better levels of informationexchange between the supplier and its ED channels, andlower levels of conflict in these relationships, improve theability of these channels to effectively perform this func-tion. We find support for our hypotheses in a cross-sectional sample of 847 observations corresponding todifferent exclusive and non-exclusive channels selling asupplier’s product-line across several European coun-tries. The supplier was a major manufacturer of Informa-tion Technology products with annual sales in Europe inexcess of $1.5 billion, selling to approximately 6,500resellers. Both ED and non-ED channel arrangementswere common in this industry. Form a theoretical perspec-tive, we contribute to the existing channels literature by(1) illustrating the interaction between channel designdecisions, or choice among different channel forms, andthe management of individual channel relationships, and(2) taking an extra-dyadic perspective, or going beyond asingle supplier-reseller relationship. From a managerialperspective, our results underscore (1) the need to adaptrelationship management to the individual characteristicsof each channel type, and (2) the importance of managinga supplier’s multiple channels as an integrated systemconsidering the interactions among the different chan-nels.

Keywords: distribution channels, exclusive chan-nels, multiple channels, channel design and relationshipmanagement, extra-dyadic effects. References are avail-able upon request.

For further information contact:Alberto Sa Vinhas

Washington State University14204 NE Salmon Creek Avenue

Vancouver, WA 98686Phone: 360.546.9146



ARE CHINESE STATE-OWNED ENTERPRISES LAGGING BEHIND INPRODUCT INNOVATION?

Xiaomin Zhao, Jilin University, ChinaPing Lan, University of Alaska Fairbanks

SUMMARY

The Chinese government has paid particular atten-tion to innovation in the last decade. It has increased thecapital investment in new product research, accompaniedby a series of innovative programs and incentive policies.Currently, Chinese innovation has been moving forwardat an accelerating speed. Comparing the changes of newproducts in all industries, the average growth rate for newproduct development between 2000 and 2004 is 33.3percent, while from 2004 to 2009 the same rate is 88.5percent from 2004 to 2009.

What force is accelerating Chinese innovation? Somestudies suggest that it is non-state-owned enterprises(Non-SOEs) that promoted the progress of Chinese inno-vation (Tan 2001; Jefferson, Hu, Guan, and Yu 2003),along with the continuous decline of overall contributionof Chinese state-owned enterprises (SOEs) to economicgrowth (Jing and Tylecote’s 2005; Tan 2001). Besidesthat, Chinese SOEs’ low efficiency and lack of innovation(Yang 2008; Yao 2006; Perotti, Sun, and Zou 1999),negative operating effectiveness (Yang 2008) and heavysocial burden (Jing and Tylecote 2005; Chen 1998; Wang,Zhao, Ning, and Yu 2009) have been repeatedly dis-cussed. We do not deny the role of a non-state ownedeconomy. However, we want to know whether ChineseSOEs still hinder the development of innovation in China,after various reforming measures implemented in Chinasince the late 1970s. Therefore, the goals of this researchis to reveal (1) what role have SOEs played in China’sinnovation catch up? (2) How different is this role amongvaried industries? And (3) what is an ideal industryenvironment for maximizing the role of Chinese SOEs?

This research is an empirical analysis. We employdifferent qualitative and quantitative analytical methodssuch as intuitive comparison, panel data analysis, T testand analysis of variance (ANOVA), and regression tomeasure the innovation contribution, innovation effi-ciency and innovation’s industrial influence of ChineseSOEs. Judging from overall growth of new product inno-vation, this research finds that non-SOEs’ contributiondemonstrates an increasing trend, which is consistent withprevious studies. However, judging from individual out-puts of new product innovation, it finds that each SOEgenerates more new products and capitalizes more valuefor each new product than non-SOEs. It means that there

is a gap between Chinese SOEs and non-SOEs measuredcollectively or individually. This gap indicates that usingonly dark color to paint the role of Chinese SOEs inChina’s innovation development is not accurate. Theresults of this study clearly show that Chinese SOEs are nolonger hindering the development of innovation at thecurrent stage. They have been playing an important role inthe recent innovation catch up in China.

In addition to directly answering the research ques-tion, this study makes it useful in several ways for theoreti-cal exploration and practical use. The revealing of theinnovation gap could stimulate theoretical exploration inseveral areas. One area is in the research about the reformof Chinese SOEs. Through further exploring the innova-tion gap, studies on Chinese SOEs’ reform could betterlink reform contents, procedures, policy requirements etc.with innovation catch up. Undoubtedly, it will be a keyconcern for any research dealing with China’ reform anddevelopment. Another area is in research about innova-tion measurement. The innovation gap confirms thatdifferent type of enterprises play a different role in inno-vation, and the innovative manner and learning abilities ofSOEs is different from those of non-SOEs. Studies havebeen conducted to show the differences between differenttypes of enterprises, However, a clear picture about theinnovation division and integration between SOEs andnon-SOEs in different industries in China has not emerged.Based on the innovation gap, the patterns of innovationdivision and integration between SOEs and non-SOEsand factors associated to those patterns could be betterconsolidated and understood.

The establishment of an inverted “U” shape relation-ship between innovation input of SOEs and industries’total innovation output indicates that SOEs’ 15–30 per-cent ownership is a magic range. When SOEs’ ownershipfalls into this range, an industry’s total innovation outputis maximized. The inverted “U” shape relationship illumi-nates a nonlinear relationship between SOEs’ share in anindustry and the total innovation performance of theindustry. It also challenges the property right theory or itsapplication in China. According to the property righttheory, SOEs should not have enough incentive to investin innovation, such as the cases in the group of nationalbasic and restricted industries. However, the inverted “U”shape relationship confirmed other observations that state


ownership has a positive effect on innovation perfor-mance (Choi, Lee & Williams 2011, Bin 2005). What arethe reasons behind this?

Understanding and dealing with China’s innovationwill be either a continual hot topic in academic research,

or an ongoing agenda in managerial practice. Exploring itfrom the standpoint of Chinese SOEs undoubtedly offersa solid base, because the uniqueness of China’s develop-ment will continue to be tied to the distinctiveness ofChinese SOEs, particularly in the innovation domain.References are available upon request.

For further information contact:Ping Lan

University of Alaska Fairbanks303 Tanana Loop

Fairbanks, AK 99775Phone: 907.474.7688



FROM INNOVATIVE AND MARKETING CAPABILITIES TO FIRMPERFORMANCE: EMPIRICAL COMPARISON ON DIFFERENT

PATHS IN TWO EMERGING NATIONS

Xina Yuan, Xiamen University, ChinaSohyoun Shin, Eastern Washington University, Spokane

Sang Yong Kim, Korea University, Korea

SUMMARY

In response to the growing research needs in emerg-ing markets and caution on the gap in relevant studies, thisresearch empirically tested distinct influence of two criti-cal capabilities: innovative capability (IC) and marketingcapability (MC) in two emerging nations at differentdevelopment level. By investigating the detailed pathsfrom strategic orientations through both or either of ICand/or MC to firm performance in China and Korea, theauthors attempt to provide a longitudinal guideline onhow firms transform resource set requirements to obtaincompetitive advantages as their economies progress.

Based on the combined data of 665 companies inemerging markets; 385 in China and 280 in Korea, weproved IC and MC successfully transform into organiza-tional performance within the broad territories of emerg-ing market, portraying the results from the industrializedmarket. In addition, as critical drivers both market orien-tation and learning orientation were proven to have anaffirmative impact on both IC and MC. From each data setof two, IC is found to be more critical in fast-growing andmanufacturing-focused market like China while MC isfound to be more vital to an advanced emerging countrysuch as Korea. The findings suggest that organizationsshould develop a set of capabilities required for each level

of development in order to achieve a competitive advan-tage, answering to the needs in cross-cultural comparisoneven between emerging nations.

Further analysis based on different industries in eachregion: manufacturing vs. non-manufacturing sector, pro-vided detailed differentiated paths to firm performance ineach cases. IC significantly influenced firm performancefor both sectors in China, whereas MC was influentialonly in the non-manufacturing sector of China. In con-trast, MC was found to be equally important for firm rentsin both industries within Korea while IC failed to prove itsinfluence in both Korean industries. Thus, as the majorindustry of an economy evolves from manufacturingbusiness to more service-oriented industry, the findingssuggested that firms should be equipped with an IC-centered resource set at first and then reconfigure abalanced set with IC and MC.

Overall findings of the study provide managerialimplications on how to allocate limited organizationalcapitals to strategically better fit and to compete in theever-changing emerging market environments. The limi-tations of the study are also provided along with thefurther research opportunities. References are availableupon request.

For further information contact:Sohyoun Shin

Eastern Washington University668 N. Riverpoint Blvd.

Spokane, WA 99202–1677Phone: 509.828.1243



MNCS AND FOOD SECURITY IN EMERGING MARKETS:PROVOCATIONS FROM INDIA

Susan M. Mudambi, Temple University, PhiladelphiaThomas Reardon, Michigan State University, East Lansing

Bart Minten, International Food Policy Research Institute, Ethiopia

SUMMARY

Chronic hunger remains a troubling concern in emerg-ing markets. To the World Bank, rising food prices and thefood insecurity of world’s poor renders the global economy“one shock away from a full-blown crisis” (Martin 2011).Hunger is the world’s number one health problem (FAO2010), the reality for an estimated 925 million people.Although government and non-governmental organiza-tions (NGOs) play critical roles in food emergencies, thereduction of chronic hunger depends on private enter-prise. For example, in India, the government accounts foronly seven percent of the food economy (Reardon andMinten 2011). Private sector retailers and wholesalersshape the day-to-day food security of the world’s poor,and the rise in multinational investment has generatedpolarizing views. Some have argued that foreign directinvestment (FDI) improves the capability of emergingmarket producers and food security, while others arguethat FDI bypasses local producers and hurts poor con-sumers.

This motivates the central research questions of thispaper:

How does FDI in food supply chains affect subsis-tence consumers and producers? What factors affectthe development of food supply intermediaries?

To address these questions, we provide backgrounddetails on modern food retailing in emerging markets, andexamine past empirical research. Since past research inmarketing is limited, we draw on multiple studies inagricultural economics, development economics, and eco-nomic geography, and take a particular look at the uniquecase of India. Despite considerable economic growth andsuccess, nearly one-fifth of the Indian population faceschronic hunger, and India has made little or no progresstoward the UN’s hunger reduction target since 1990 (FAO2010).

An emerging market “supermarket revolution”(Reardon and Hopkins 2006; Humphrey 2007) started inthe early to mid-1990s (Reardon, Henson, and Berdegue2007) in China, Thailand, Kenya, and other emergingmarkets. The revolution was driven by government endor-

sed FDI (Coe and Wrigley 2007; Reardon, Timmer,Barrett, and Berdegue 2003) and the pro-active strategiesof supermarkets (Reardon, Henson, and Berdegue 2007).India is the exception, as FDI in retailing is prohibited. Inthe mid-1990s, India began allowing multinational corpo-ration (MNC) supermarkets such as Tesco, Walmart, andCarrefour to invest in food wholesaling. In 2006, in partdue to the expectation that FDI in food retailing wouldsoon be permitted, investment by large Indian firms suchas Reliance spurred a huge growth in modern food retail-ing. The India government granted permission for FDI infood retailing late in 2011, but widespread public protestsresulted in a quick reinstatement of the ban in January2012. This turbulent political environment exemplifiesthe highly complex and sensitive nature of food securityin India and other emerging markets.

Our analysis of past empirical studies (e.g., Mintenand Reardon 2008) indicates that supermarkets sell foodmore cheaply than do traditional retailers, and are gradu-ally reaching the poorer consumers, who are most affectedby food prices. Supermarkets have generally benefitedpoor consumers in terms of food access, quality andprices, but results have been mixed for the smallestproducers and retailers, with the varied results related tothe food sector, time frame, and other institutional andgeographic factors. Intermediaries also matter. In India, a“quiet revolution” (Reardon and Minten 2011) has takenplace outside the supermarket channels by domestic firmsengaged in wholesale, cold store and processing. Theunexpected innovation of intermediary firms has quietlycomplemented the more noticeable modern supermarketrevolution. Of special note are the rural agricultural hubsand supermarkets such as Hariyali Kisaan Bazaar, as thesehelp to connect the rural poor to food supply chains, bothas producers and as consumers.

We conclude by highlighting the connections be-tween hunger and food security research and research onmarketing at the base of the pyramid. With a goal ofstimulating new research on the food insecurity of theworld’s poor, we offer a series of propositions or provo-cations for future research on food retailing and foodintermediaries in emerging markets. References are avail-able upon request.


For further information contact:Susan M. MudambiTemple University

1801 Liacouras Walk, 524 Alter HallPhiladelphia, PA 19422Phone: 215.204.3561



ADDING TO THE “MISSING LINK” PERSPECTIVE IN EMERGINGECONOMY: THE ROLE OF PRODUCT INNOVATION IN RUSSIA

Maria Smirnova, Saint Petersburg State University, RussiaVera Rebiazina, National Research University, Russia

Alexander Krasnikov, George Washington University, WashingtonSergey Kusch, Saint Petersburg State University, Russia

SUMMARY

The ability to introduce innovations in emergingmarkets drive success of firms from such markets. Prod-uct innovation has been extensively studied in differentcontexts (including emerging and developed markets);however, it was noticed that such innovations are notintroduced similarly within BRIC (Brazil, Russia, India,and China) countries. Moreover, Russia is rather losing itscompetitive positions in comparison to other BRIC econo-mies. We study potential reasons for that in this paper.

Following framework developed by Sawhney et al.2006 we explore the role of key offering dimensions –platform and solution innovations – in influencing perfor-mance outcomes.

A platform represents a set of common components,assembly methods or technologies that serve as buildingblocks for a portfolio of derivative products or services(Sawhney and Chen 2011). For successful application ofproduct platform strategy and creation of platform inno-vation firms needs to understand core and differentiatedcustomer needs and be aware of firm’s target group (Stoneet al. 2008). Platforms are based on technological advan-cements and understanding customer expectations andrepresent both a requirement and opportunity for develop-ing new products and services (Kumar and Allada 2007).

Solution innovation represents complex bundle ofproducts and services that solve particular customer’sproblem. Solution may be successful only when a supplier

truly understands customer needs, which may be achievedthrough collaboration between buyers and suppliers andestablishing strong relationships (Roegner, Seifert, andSwinford 2001).

Following existing theory on market orientation –innovation – performance relationship, direct and indirecteffects of market orientation on firm performance weretested. Since there is still no substantial research evidenceon the mediating effect of innovation on market orienta-tion – performance link in emerging economies, our studyaims to close this gap. Moreover, the model includesmoderating effect of product innovativeness. The study isbased on empirical survey of 204 Russian innovativefirms with multiple respondents approach, resulting in331 qualified respondents.

There are positive and significant effects of bothplatform and solution innovation on firm performance.While platform innovation affects performance strongly(0.542, p < 0.001), the effect of solution innovation ismarginal (p < 0.1). Among market orientation compo-nents, only customer orientation has direct positive impacton firm performance (0.259, p < 0.005). Effects of com-petitor orientation and interfunctional coordination aremediated by platform and solution innovation constructs.

The results confirm existing link between marketorientation and performance, as well as illustrate differentimpact of market orientation dimensions and dimensionsof product innovation on firm performance. Referencesare available upon request.

For further information contact:Maria Smirnova

Graduate School of ManagementSaint Petersburg State UniversityVolkhovsky Per. 3 St. Petersburg

Russia 199004Phone: +7(812)323.8456

Fax: +7(812)329.3234E-Mail: [email protected]


EVALUATION OF RETAIL SERVICES: A DEVELOPED VS. EMERGINGMARKETS PERSPECTIVE

Piyush Sharma, Hong Kong Polytechnic University, Hong KongSherriff T.K. Luk, Hong Kong Polytechnic University, Hong Kong

Ivy S.N. Chen, Hong Kong Polytechnic University, Hong Kong

SUMMARY

Customers from different countries have variedexperiences and expectations due to diverse socio-economic, cultural, and environmental factors (Douglasand Craig 2006). They also differ in their service experi-ence in terms of customer perceived value, customersatisfaction, and perceived service quality (Brady et al.2005). However, prior research on the evaluation of retailservices focuses mostly on the shoppers from developedmarkets with little attention to shoppers from emergingmarkets such as China (Uncles 2010) and India (Sengupta2008).

Theoretical Background and Hypotheses

We address the above gap by comparing the influ-ence of the antecedents of customer perceived value(CPV) for shoppers from developed and emerging mar-kets. Shoppers from developed and emerging marketsdiffer in their knowledge of various retail services due todifferences in their prior experience (Jin and Sternquist2003; Shukla 2010) and this may moderate the influenceof antecedents (e.g., effort, lifestyle, price, risk, and ser-vice quality) on perceived value of the retail service.Specifically, we hypothesize that product quality (PQ),perceived effort (EFF), perceived risk (RSK), store envi-ronment (ENV) and value-for-money (VFM) have a stron-ger influence on customer perceived value for shoppersfrom emerging vs. developed markets, whereas servicequality (SQ) and lifestyle congruence (LSC) have a stron-ger influence on customer perceived value for shoppersfrom developed vs. emerging markets. We also hypoth-esize that CPV has a greater impact on word-of-mouthintentions for shoppers from emerging markets.

Methodology

We test our hypotheses with data from foreign shop-pers in Hong Kong, a popular tourist destination attractingmillions of shoppers from around the world every year.The sample (N = 802, response rate = 16%) consists of 490shoppers from an emerging market (China) and 312shoppers from three developed markets (Australia, U.K.,and North America). We adapted existing scales to mea-sure product quality, perceived risk and value-for-money(Sweeney and Soutar 2001); service quality (Brady andCronin 2001), store environment (Baker et al. 2002),lifestyle congruence (Johnson et al. 2006), perceived

effort (Zeithaml et al. 1988), word-of-mouth (Anderson1998) and customer perceived value (Sharma et al. 2012).

Data Analysis

Since we use data collected from shoppers fromdifferent countries, we first assessed cross-cultural mea-surement invariance for all the scales using a multi-stepapproach (Steenkamp and Baumgartner 1998). We thenused confirmatory factor analysis and internal consis-tency reliability tests to assess the psychometric proper-ties of all the scales (Bagozzi and Yi 1988), and thediscriminant and convergent validity of all the scales(Anderson and Gerbing 1988).

Next, we tested all the hypotheses in this study bycomparing the differences between regression coeffi-cients for the participants from developed and emergingmarkets, coded as home country (HC), using moderatedmultiple regression analysis (Baron and Kenny 1986).The regression model provided a good fit to the data(adj.R2 = .47, F = 36.38, p < .001. As expected, SQ (ß =.28, p < .001), RSK (ß = .11, p < .01), LSC (ß = .13, p <.001), and VFM (ß = .16, p < .001) all have a significantpositive effect on CPV, however, PQ (ß = .08, p > .10),ENV (ß = .04, p > .10), EFF (ß = .01, p > .10), and HC (ß =-.04, p > .01) have no significant effect on CPV.

Finally, interactions between HC and PQ (ß = .13, p <.01), RSK (ß = .13, p < .001), and VFM (ß = .11, p < .01)are significant and positive; and interactions between HCand SQ (ß = -.11, p < .01), and LSC (ß = -.10, p < .01) aresignificant and negative. Hence, hypotheses 1, 2, 3, 5, and7 are supported. However, the interactions between HCand ENV (ß = -.04, p > .10), and HC and EFF (ß = -.02, p >.10) are not significant, hence hypotheses 4 and 6 are notsupported. Among the covariates (gender, age, income,education, and retail category) only income has a signifi-cant but small influence on CPV (ß = .06, p < .05).

Discussion and Contribution

This study adds to growing body of work exploringthe differences between developed and emerging marketshoppers (e.g., Sharma 2011; Shukla 2010). We alsoextend existing research on customer perceived value byincorporating both tangible (product quality, perceivedeffort, store environment, and value-for-money) and intan-gible (service quality, perceived risk, and lifestyle con-gruence) antecedents of customer perceived value toprovide a more complete picture of how shoppers form


their value perceptions in the retail context. Finally, weuse WOM as an outcome of customer perceived value,whereas prior research focuses on other behavioral inten-tions such as repeat purchase and customer loyalty.

Limitations and Future Research

We examine intentions rather than actual WOM inthis study due to its cross-sectional design. Future

research may use a longitudinal design to overcome thislimitation. Future research may also include other mod-erators such as variety-seeking tendency, value-conscious-ness, prior experience, product knowledge, and personalcultural orientations on the retail service evaluation pro-cess (Sharma 2010. Finally, the study can be extended toshoppers from other emerging markets such as Brazil,India, and Russia to assess its generalizability. Referencesare available upon request.

For further information contact:Piyush Sharma

Hong Kong Polytechnic UniversityHung Hom, Kowloon, Hong Kong

Phone: 852.2766.7367Fax: 852.2765.0611



FACTORS INFLUENCING GROWTH POTENTIAL OF E-COMMERCE INEMERGING ECONOMIES: A MULTI-THEORETICAL APPROACH

AND RESEARCH PROPOSITIONS

James Agarwal, University of CalgaryTerry Wu, University of Ontario Institute of Technology, Oshawa

SUMMARY

E-commerce is widely accepted as the platform forconducting business all over the world. Given the tremen-dous success of Internet and e-commerce in developedcountries, emerging economies are quickly embracing theinformation technology as well. Although there is a greatdeal of research on e-commerce development, most of theresearch on e-commerce focuses on developed countries,with relatively little research on emerging economies. Thepurpose of the study is to examine factors influencing(both determinants and deterrents) the growth potential ofe-commerce in emerging economies from a multi-theo-retical perspective (namely, institution-based Network-Ownership, Location, Internalization, i.e., i-based N-OLIframework). Several theoretical strands from Institutionaltheory, Network theory, and the Eclectic OLI paradigmare utilized for explaining the adoption and expansion ofe-commerce in emerging economies. Factors are identi-fied at three levels. At the global level, we identify

multilateral agreements, strategic behavior of multina-tional enterprises (MNEs), and technological innovationas the key factors. At the national level institutionalenvironment, infrastructure and culture are identified.The transactional level examines the role of integrity oftransactions, online intermediaries, and network exter-nalities and value clustering as the key factors for growthof e-commerce in emerging economies. While we providea multi-layered approach to understanding the growthpotential of ecommerce, the salience of each factor will, toa large degree, depend on the emerging economies them-selves. For example, we find that while China has rela-tively well-developed infrastructure compared to institu-tional factors, India lags behind considerably in infra-structure. We hope our multi-layered and multi-theoreti-cal approach will spark interest among scholars to furtherexamine these factors and their interactions for futureresearch and serve as a template to guide practitioners intheir e-commerce activities in emerging economies. Ref-erences are available upon request.

For further information contact:Terry Wu

University of Ontario Institute of Technology2000 Simcoe Street North

Oshawa, OntarioCanada L1H 7K4

Phone: 905.721.8668, Ext. 2623Fax: 905.721.3372



DRIVING THE VALUE PREMIUM THROUGH CULTURAL, SYMBOLIC,ECONOMIC AND SOCIAL CAPITAL MANAGEMENT

Goran Vlasic, University of Zagreb / University of SuxxexJosef Langer, Alpen-Adria Universitat Klagenfurt

Zoran Krupka, University of Zagreb

SUMMARY

Getting consumers to be willing to pay the pricepremium for one product over another has long been thegoal of marketing experts. In discussing different sourcesof such price premium, focus was mainly on the reputationand brand image. In this contribution we (a) link theseconcepts to their origins of impression management insociology; (b) analyze different forms of capital (eco-nomic, social, symbolic, and cultural) that can be man-aged to drive reputation and thus value premium; (c)analyze the mechanism through which Bourdieu’s differ-ent forms of capital impact reputation and value premium.In doing this, we also delineate the different capital formsfrom the concept of reputation, both theoretically andempirically, leaving reputation to be the stakeholders’evaluation of entity’s observable outputs. We discussthese ideas from the perspective of consumer, since allthese categories exist only in relation to the context inwhich they are observed (Bourdieu 1990).

We start from Bourdieu’s theory of inequality, wheremarkets are described as multidimensional social spacewhere “the different forms of capital . . . serve as thebuilding principles of these social spaces, enjoying higheror lower ratings in the different markets.” (Gergs 2003,pp. 40–41). These capital forms ecompass: (a) economic– defined as accumulated financial resources and assets(Bourdieu 1990), also seen as “productive capital”; (b)social capital – defined as resources, which are controlledon the basis of existing and potential social networks (Burt1992; Bourdieu and Wacquant 1992); (c) cultural capital– defined as the favorable cultural traits that one indi-vidual entity has, reflected in embodied, objectified andinstitutionalized advantages that provide higher status insociety (Bourdieu 1990); and (d) symbolic capital –defined (and measured) as the resources available to anentity on the basis of honor, prestige and/or recognition,resulting from investments of time, energy and wealthinto activities which do not yield a short-term economicreturn for the entity (Calhoun 2002; Bird and Smith 2005;Bourdieu 1990). These three are also termed as “fictivecapital” by Marx. All these forms of capital have a certainlevel of liquidity – changing one capital form into another.

These capital forms are managed by entities (Bourdieu1990) to enhance the value premium by changing per-ceived entity characteristics of: (a) charisma, defined as

the trait characterized by extreme charm and a “magnetic”quality of personality and/or appearance along with in-nate and powerfully sophisticated communicability andpersuasiveness (Conger, Kanungo, and Menon 2000); (b)trust, i.e., calculus-based trust, seen as a rational choicethat an actor has intentions and competences to behave ina certain manner and arises from signals creating anindirect expectation of trustworthy behavior (Doney,Cannon, and Mullen, 1998); and (c) reputation which,after empirical analyses of items and conceptual consid-erations, was defined as stakeholders’ perceptions of anentity’s current and potential performance on observabledimensions. The mechanism therefore starts with man-ageable capital forms which change entity’s perceptualproperties in order to impact the value premium that anentity can expect for its outputs.

Research was carried out in one European country intwo steps: (a) identification of the most salient entitiesselecting those that had the most/least admirable reputa-tion the “top of the mind” entities; (b) convenient sampleof consumers indicated by selected entities who are expec-ted to be well informed about different activities of theseentities (Bartikowski and Walsh 2009). All measuresexhibit Crombach alpha above .7, AVE above .5 and CRabove .7 with measurement model that exhibits needed fitwithin the limits suggested in social sciences with RMSEAbeing 0.08 and CMIN/DF = 2.189 (Hair, Black, Babin,Anderson, and Tatham 2005; Camines and McIver 1981;Steiger 1990). Data also satisfy the condition of multivari-ate normality.

For robustness reasons the model was tested usingboth the structural equation modeling and regressionanalysis, testing the capital interactions and controllingfor various potential determinants of expected addedvalue. Final model was analyzed taking into account onlyrelationships that were persistently shown to be signifi-cant (ROSEA = 0.08, CMIN/DF = 2.197).

Results indicate that economic capital has a negativeimpact on perceived variables but a strong positive impacton value premium consumers are expecting to pay for theentity’s outputs. All the “fictive capitals” exhibit onlyindirect impact on the value premium through their impacton perceptual aspects. Symbolic capital positively impactscharisma, as symbolic capital displays extraordinary sig-nals, and trust (Klapwijk and Van Lange 2009). Interest-


ingly, cultural capital has impact on all three perceptualvalues: charisma, trust and reputation, and the strongestoverall impact on expected value premium. Also, socialcapital, is the only capital that exhibits negative impact onexpected value premium, through its negative impact ontrust – as all entities with great social capital do not dependas much on each particular stakeholder or stakeholdergroup and therefore can behave opportunistically.

The impact of charisma on trust is positive, as char-ismatic entities “inspire high levels of . . . trust in theirfollowers” (Zhou, Gao, and Zhou 2005, p. 1052). Trusthas a positive impact on reputation, as it implies an abilityto rely on the other party to achieve expectations (Rotter1967) and deliver high quality output (Moorman, Zaltman,and Deshpande 1992; Anderson and Narus 1990). Thegreater the reputation, i.e., the entity’s output quality, themore an entity can charge for these outputs.

Thus, results indicate that different forms of capitalthat any entity possesses (social, cultural, symbolic) donot have a direct impact on the extra value that it canexpect to get for its products/services but rather an indirecteffect through such abstract concepts as charisma andtrust and a more applied concept of reputation. Thestrongest overall impact on expected value premium hasthe cultural capital of an entity thus justifying investmentsin it. The importance of social capital a bit more indirectsince it impacts the value premium by being convertedinto other forms of capital (with especially high convert-ibility into economic and cultural capital). Besides this,we stress the importance of the “divine” concept ofcharisma which has a strong influence on trust and is amanageable category, both directly and indirectly. It is notjust a characteristic of a person but can also be ascribed toan organization. References are available upon request.

For further information contact:Goran Vlasic

University of Zagreb/University of SussexPhone: +38598271512



MICROFINANCE MARKET FAILURES IN EMERGING MARKETS

Joseph Hansen-Addy, University of Ghana, AccraEsi Abbam Elliot, University of Illinois, ChicagoJoseph Cherian, University of Illinois, Chicago

SUMMARY

Market failures such as lack of market-underpinninginstitutions have been the cause of markets malfunction,especially in a way that excludes the poor. Market failuresare viewed as the lack of facilitating structures and keyexchange mechanisms that cause business malfunction(Fligstein 2001). In this study, we define market failuresas the failure of a more or less idealized set of price-marketinstitutions to sustain “desirable” activities or to impede“undesirable” activities. In the case of microfinance,market failures such as lack of available credit to the poorreflects the non-inclusiveness of markets (Mendoza 2008),which represents the inability to sustain desirable activi-ties.

The four levels of market failures detailed by Barton(1958) are failure by signal, failure by incentive, failure bystructure and failure by enforcement. Our study extendsthis model with the consideration of failure by default.Failure by default refers to the interplay of institutionalfailures and market failures and the survival behavior ofactors demonstrated in the exploitation of market failures.

Micro and Small Enterprises (MSEs) are often clas-sified among the poor in subsistence markets due to theirlimited capital. It is an undeniable fact that Micro andSmall Enterprises (MSEs) are important agents for eco-nomic growth in emerging markets. They also help infinancial intermediation. These businesses however facemyriad challenges in accessing credit from financial insti-tutions. Even though there are loanable funds available,most MSEs cannot access these funds because banks findit difficult to assist MSEs while MSEs find it difficult tomeet the requirements of the banks. By examining indetail, specific market failures that inhibit the efforts of

MSEs to access microfinance in emerging markets, themain contributions of this paper are to: (1) bridge theliteratures on microfinance and market failures relevant toentrepreneurship in emerging markets; (2) contribute toliterature by identifying additional market failures rel-evant to emerging markets (3) inform policies for entre-preneurial development in emerging markets. In integrat-ing microfinance and market failure arguments, this paperalso represents an attempt to unpack the reasons whymarket failures in emerging markets exclude the poorfrom accessing microfinance.

In this paper we address these questions through acase study of a local bank in Ghana. More specifically wefocus on the perspectives of the officials of the bank andtheir MSE customers about their market interactions. Wechose this specific setting because Ghana is an emergingmarket, the fastest growing economy in the world as at theend of 2011 (IMF Report 2011). In Ghana, due to liberal-ization, there is improving relationships betweenmicroenterprises and financial services firms. However,several required institutions that characterize modernmarket economies are lacking. Our findings revealed fourfundamental aspects of failure by default as a dimensionof market failures that limit access to microfinance byMSEs in emerging markets: (i) Insufficient informationand timing issues based on limited infrastructure; (ii)Rigorous lending requirements due to the need for highrisk mitigation; and (iii) Difficulties in loan repaymentsdue to market volatility (iii) Power imbalance due toresource constraints. These findings extend Barton’s(1958) model on market failures with the additional con-sideration of failure by default. The findings have practi-cal implications for microfinance initiatives in emergingmarket economies. References are available uponrequest.

For further information contact:Esi Abbam Elliot

University of Illinois at Chicago601 South Morgan Street, MC 243

Chicago, IL 60607Phone: 773.991.0160



REACTANCE VS. ACCEPTANCE: EMERGING MARKET CONSUMERS’PERCEPTIONS OF LOCAL BRANDS AFTER AN MNC’S ACQUISITION

Martin Heinberg, University of Duisburg – Essen, GermanyMarkus Taube, University of Duisburg – Essen, Germany

SUMMARY

Multinational companies frequently use acquisitionsof local brands as a way to extend their brand portfolio inemerging markets or as a market entry strategy. However,to the authors’ knowledge the marketing literature offersonly little general advice on cross-border acquisitions ina developing country by a developed country brand.When considering both the sensitiveness of developingcountry consumers to foreign brands (Tian and Dong2011) and the common practice of western/global compa-nies to enrich their brand portfolios through local acqui-sitions, research in this field appears necessary. We makea first attempt to shed light into this field, by drawing onsignaling theory, the resource-based view and the theoryof consumer reactance. Using the hierarchical structuralequation modeling approach, we analyze data from astudy encompassing 36 consumer good brands in China.

The study offers two important contributions, namelythe definition of a new brand category and the integrationof conflicting theory streams. We define the acquireddeveloping country brands in our study as internationalportfolio acquisition brands (IPA brands). We expectthem to fulfill two premises: A foreign dominated strat-egy, and a preserved local brand heritage. We prove thatconsumer behavior and brand attribute evaluations to-ward IPA brand significantly differentiates these brandsfrom foreign and local brands alike. Second, our studycontributes to the scarce literature that deals with con-sumer responses toward a brand after an acquisition, by

integrating two conflicting theory streams. On the onehand, the theory of psychological reactance (Brehm 1966;Thorbjørnsen and Dahlén 2011) supposes that an acquiredbrand is not appreciated by consumers, since consumersmight interpret the takeover as a threat to their freedom ofchoice and might try to restore their freedom by devaluat-ing the appeal of the forced alternative. On the other hand,the resource-based view and signaling theory (Wernerfeld1984, 1988; Swaminathan, Murshed, and Hulland 2008)propose, that consumers might view an acquisition of alocal brand more favorably, since the international brandname of the acquirer acts as a bond for quality promisesand consumers might expect an upgrade of the brand dueto better access to foreign resources. We argue that thosetheories can be integrated into one model and thus candemonstrate the interaction of both theoretical streams.

Our results show that IPA brands are faced with threekey challenges: First, emerging market consumers indeedshow a feeling of reactance by expressing a lower overalllevel of loyalty intentions for IPA brands compared toforeign and local brands in general. Second, consumersexpect an upgrade of IPA brands’ quality. Third, wediscover that consumers’ rise of quality expectations isnot matched by an increase in willingness to pay. In lightof these three key challenges we argue that multinationalsshould carefully consider their growth strategy for emerg-ing markets instead of hastily implementing the temptingIPA strategy as a means for easy market entry or portfolioenrichment. References are available upon request.

For further information contact:Martin Heinberg

Mercator School of Management, In-East InstituteUniversity of Duisburg – EssenForsthausweg 2, Room LE 735b

47057 DuisburgGermany

Phone: +49(0)203.379.4190Fax: +49(0)203.379.4157



RETHINKING MARKETING PRACTICES IN AN EMERGING MARKETCONTEXT: AN EMPIRICAL EVALUATION OF COMPETING

CONCEPTUAL FRAMEWORKS IN GHANA

Kofi Q. Dadzie, Georgia State UniversityCharlene A. Dadzie, University of North TexasEvelyn M. Winston, Clark Atlanta University

SUMMARY

Recent literature on marketing in emerging markets(EMs) has focused on the development of theoretical andconceptual frameworks for understanding the impact ofcontextual factors on marketing activities. One such con-struct that has received renewed attention is the 4Psmarketing mix framework. However, a number of limita-tions have been observed in the 4Ps framework, includ-ing: oversimplification of the marketing tasks; overlapbetween some of the mix elements, namely, price andpromotion; the static nature of the framework; the trans-actional nature of the framework, and the lack of consid-eration of relationship and long-term relational exchanges.

Despite the concerns with developing new constructsto extend the marketing mix, little attention has been givento incorporating the unique characteristics of emergingmarkets. One exception is the 4As – acceptability,affordability, accessibility and awareness (Sheth andSisodia 2011). Proponents argue that the 4As frameworkis more pertinent to consumption environments of scar-city, where consumers make tradeoffs with a focus onsurvival, where they simply cannot consume what is notaffordable, and where convenience is valued less than ina surplus-oriented consumer economy (Sheth and Sisodia2011). These arguments are consistent with early researchby Dadzie (1989), who concluded that concerns abouteconomic scarcity lead to strong regulatory policies andultimately “demarketing” of the marketing mix activitiesby firms in African EM countries. Consequently, we seekto move beyond the conceptual phase regarding the 4Asand evaluate the adequacy of this recently proposedframework relative to the traditional 4Ps approach in theenvironmental context of an African emerging economy,Ghana. Thus, we explore the validity of the 4As againstthe 4Ps in Ghana’s business environment.

Specifically, this study addresses the following ques-tions:

1. Which marketing mix activities do managers con-sider to be more adequate for planning the marketingconcept under the 4As versus the 4Ps approach?

2. Which mix activities are linked to performance undereach framework?

3. To what extent are the mix activity – performancerelationships mediated/ moderated by the type ofbusiness strategy?

4. What are the theoretical and managerial implicationsof using the 4As in an African emerging marketcontext?

To address these research questions, the data werecollected in the industrial and political capital city ofGhana, Accra, by trained interviewers. Four focal con-structs were operationalized and measured in the presentstudy including (1) the 4Ps of the marketing mix, (2) the4As of the marketing mix, and (3) the competitive busi-ness strategies (differentiation and low cost), and marketperformance. Next, analysis was conducted in three incre-mental stages including: (1) scale accuracy analysis, (2)common method bias assessment and scale validation and(3) structural model analysis.

The results suggest three insights into the relativeimportance of both competing marketing managementframeworks. First, although both models exhibit accept-able measurement properties in Ghana’s corporate con-text, the 4As framework explains market performanceimpact more adequately than does the 4Ps framework.Second, the study demonstrates that low cost strategy hasboth mediating and moderating influences on the core4As activities of acceptability, affordability, awarenessand accessibility. Third, the 4As framework is morerobust for firms that pursue standardization strategieswhile the 4Ps is less robust for firms pursuing differentia-tion strategies. These results are interpreted as lendingsupport for academicians who advocate a paradigm shiftfrom the traditional demand differentiation strategies todemand aggregation strategies based on the 4As foremerging market contexts.


For further information contact:Charlene A. Dadzie

Department of Marketing and LogisticsUniversity of North Texas

Denton, TX 76203Phone: 678.325.8486



EXPANDING “BROTHERHOOD” IN EMERGING MARKETS:METHODOLOGICAL APPROACH AND CULTURAL

VALUE ANALYSIS

Richard Michon, Ryerson University, Toronto

SUMMARY

Entrepreneurs are not the only one to be attracted atthe call of emerging countries. Major non-profit organiza-tions raising money in high income OECD countries forredistribution in developing countries are now looking fornew sources of funds. Some of the countries that benefitedfrom private philanthropy are now in a position to contrib-ute themselves. The popular press is full of horror storiesof firms’ that were lured by emerging market size, growthand new wealth. Their initial research was often limited toeconomic data and failed to identify firms’ specific suc-cess factors associated with product-market specificity(Arnold 2003).

This paper examines a methodology to identify prom-ising potential markets for charitable donations. It looks attwo major dimensions that charities should consider beforemaking any move. The empirical research reviews thephilanthropic market selection process. It looks at coun-tries’ (1) capacity and (2) propensity to give.

Methodology

Capacity to Give and Country Infrastructure:Research on philanthropic market size and capacity togive is only a prerequisite for the most important issue, thepropensity to give or to support a private non-profit’sorganization. The study is based on the experience of amajor NGO with fundraising activities in some 25 coun-tries on five continents. The initial research followedCavusgil’s (1997, 2004) xxx methodology. Economicdata were extracted from the World Bank DevelopmentIndicators at the urban household level. Income decilesand quintiles were used to estimate potential annual dona-tions.

Nonprofits must be able to reach people and engagein advocacy action. Country infrastructure data wereavailable from Michigan State University’s Market Poten-tial Indicators for Emerging Markets (Global Edge 2011)and from the World Bank that publishes statistics oncountries’ communication infrastructure: telephone lines,mobile telephone subscribers, PCs, Internet servers, TVsets, and daily newspapers per 1000 inhabitants.

Major fundraising operations rely on efficient bank-ing systems to process donations, credit card and bankingpayments, and to transfer money in and out of countries

without currency controls. Freedom and human rightindices were provided by the Heritage Foundation andFreedom House. Euromoney rates countries on political,social, and financial stability.

Propensity to give: A large gap has been observed inOECD countries between official development assistance(ODA) and private philanthropy. The Hudson Institute’sGlobal Philanthropy Index compares private donationsand immigrants’ remittances to ODA. Countries thatscore relatively higher on ODA do rather “poorly” onprivate donations. Australia, Canada, New Zealand, andthe United Sates are among them. The Johns Hopkins’Comparative Nonprofit Sector (Salamon et al. 2004)explains some of these variations by cultural clusters.Anglo Saxon countries are more inclined toward privatedonations while continental European countries favorpublic interventions.

Brooks (2006) looked primarily at donations andhelping behavior in the U.S. but also in other countries,mainly from those in the International Social SurveyProgram that now includes some 40 plus nations. Brooksobserved that “generous” donors are politically to theright (conservative) and have strong family values. Theyvalue personal entrepreneurship and are skeptical aboutgovernments’ interventions into economic life and incomeredistribution. Finally, they are regular churchgoers, irrel-evant of denominations. Values identified by Brooks arealso available for most emerging countries in Inglehart’s(2005) various World Value Survey waves.

Research Findings

Privately funded NGOs already raise substantialincome from Australia, Canada, Germany, New Zealand,South Korea, Switzerland, Taiwan, the United Kingdom,and the United States. These countries were identified asreferenced targets. Xxx Regression analysis isolated themost significant predictors for charitable donations inthose countries. The regression function was used to scorethe remaining developed and emerging countries. Themodel identified other markets with similar profiles toreferenced countries.

Countries were ranked according to their capacity togive (potential market) and their propensity for privatephilanthropy, based on values defined by Brooks (2006)and the World Value Survey. A joint space map was


constructed that underscored markets’ capacity to give ison the abscissa and propensity to give is on the ordinate.

Markets located in the 1st quadrant scored well onboth counts and should receive priority attention. Most ofthem are considered as high income developed or emerg-ing countries. Countries like China, South Korea, andTaiwan are also among the top ones. Nations in the 2nd

quadrant have smaller potential markets but display agood predisposition for private donations. Lithuania,Croatia, South-Africa, Estonia, and Hungary well posi-tioned emerging economies.

Findings from the World Value Survey obtain theo-retical support in Max Weber’s Protestant Ethic and the

Spirit of Capitalism (1904). Top countries in privatephilanthropy are of Protestant tradition and at the origin ofcapitalism as we know it today. The Protestant ethic gavebirth to a renewed form of capitalistic entrepreneurshipand economic growth. Many capitalists are reluctant togive to the poor or public charities that they did notcontrol. They often associate poverty with laziness and“divine disfavor.” They also reject the idea that govern-ments are responsible for income distribution and univer-sal social safety net. They have a net preference for privatecharities (e.g., compassionate conservatives), and a pro-pensity to support religious causes. Emerging countriesopting for capitalistic economies may be sharing some ofthese values. References are available upon request.

For further information contact:Richard Michon

Ted Rogers School of ManagementRyerson University350 Victoria Street

Toronto, Ontario M5B 2K3Canada

Phone: 416.979.5000, Ext. 7454E-Mail: [email protected]


STILL ON THE ROAD TO CAPITALISM? WEIGHING THE VISIBLEHAND OF GOVERNMENT INTERVENTION IN THE CHINESE

PROPERTY MARKET

Tao Zhu, University of Groningen, The NetherlandsKillian J. McCarthy, University of Groningen, The Netherlands

ABSTRACT

This paper introduces a model to describe the Chi-nese property market. In doing so, the aim is to commenton the applicability of the current property “market mod-els,” but also to discuss China, and the current state of itsjourney to capitalism. We use a straightforward empiricalmodel, and employ data, collected in the period 2000–2010, on the market performance of the real estate indus-try. We find that “market models” explain as little as onepercent of the variance in property sales. Adding a vari-able that accounts for government intervention to thespecification creates a “regulated market model. And withthis specification, the predictive power of the model soarsto 87 percent. In the “regulated market model,” however,the significance of the price system disintegrates. Mean-ing that government supply, and not price, dominates. Theimplications of our study are significant. Because, notonly do we evidence the important role of government’s“visible hand” in the Chinese property market but, byimplication, illustrate that China remains a controlledeconomy.

Keywords: Property markets; China; Capitalism;Socialism; Government Intervention.

INTRODUCTION

There has been a long-standing interest in regulatedeconomies, and especially in emerging markets, amongwhich, in particular, we found the unfolding story ofChina’s road to capitalism (e.g., Li et al. 2000). One of themost dramatic stories – that of the property market –remains relatively untold.

In 1949, the communist party came to power inChina. And soon after the state embarked on a policy ofnationalizing all private property (Kuang 1992; Wang1990). Between then and 1976, housing was treated as awelfare right, to which everybody was entitled. It wasdistributed according to the formal definitions of mini-mum and maximum per person space entitlements(Andrusz 1984). Property could not be bought or sold(Zhou and Logan 1996). And rental rates were set suchthat “ownership” of rental property yielded no returns(Deng et al. 2009).

The minimum cost of maintenance was more thanthree times as much as the rent collected. And the actual

cost of providing housing was almost twenty-five times ashigh (World Bank 1992). There was thus little incentive,during this period, for housing investment and/orimprovement. And consequently, China suffered bothsevere shortages in housing, and a deterioration of itsexisting housing stock. By 1978, the norm was 3.6m2 perperson (Li 1998). Sixty percent of the households were inhousing without exclusive use of running water, 72 per-cent without sanitary facilities and 71 percent withouttheir own kitchens (cited in the World Bank 1992).

Recognizing the limits of a centrally planned publichousing system, the state embarked upon a period ofeconomic reform in 1979. Property rights were reintro-duced, and the freedom to buy, and to sell, and to reap thebenefits of improvement turned what was once a drain ongovernment resources, and a non-productive investment,into an engine of economic growth. Huge increases ininvestment, a massive public housing privatization project,and strong government incentives for home purchase, sawhome ownership rates rise to 80 percent. And with thisliving conditions increased dramatically; average per per-son living spaces increased from 3.6m2 in 1978 to 26.11m2 in 2005 (Ye and Wu 2008). So profound have thechanges been, in fact, that prominent scholars suggestthat, at least in terms of the property market, China hascompletely moved from a centrally planned public hous-ing system to a market-based system, with competition inboth production and consumption (Ye and Wu 2008;Deng et al. 2009).

Others point out that this new market-based systemhas its own particularly Chinese characteristics. Localgovernment continues to play a strong role in the marketbecause, according to the Chinese Constitution, all landstill belongs to the state. Local governments are permittedto expropriate rural land for urban uses. And by control-ling both the land supply, and the zoning regulations, localgovernments can decide what can be built, when, and bywhom.

So how market-based is China’s new market-basedproperty system? Fortunately, a number of models havebeen created to explain the behavior of the real propertymarket. And most of these point to factors like price,interest rate and income as being the more importantexplanatory variables (cf., Quigley and Redfearn 1997;Gottlieb 1976; Quigley 1999; Mankiw and Weil 1989;


Hendershott 1991; Englehardt and Poterba 1991;Aelpeovich 1995; Woodward 1991; Dipasquale andWheaton 1992; Case et al. 2005; Campbell 2006). To thebest of our knowledge, however, few have applied suchmodels to the Chinese property market.

Using data on the Chinese market – in the period 2000to 2010 – this paper aims to fill that gap. We present andtest the applicability of two models: a standard marketmodel, which includes all the factors that presently domi-nates the analysis of property markets, and a new modelfor the regulated economy, which explicitly accounts forthe role of government intervention. By differencing thetwo, we attempt to gauge the scale and effect of govern-ment involvements. And, by doing so, we aim not only todiscover how market-based China’s new market-basedproperty system is, but also to comment on China’s roadto capitalism.

We find, in the base case, and using a standard marketmodel – specified according to suggestions of the currentliterature on the analysis of property markets – that inChina markets work. The standard model behaves asexpected. And price is significant in explaining marketmovements. From this, one might conclude that Chinaoffers an open and transparent market. But the marketmodel explains as little as one percent of the variance inChinese property sales. Consequently we next move toconsider the application of a regulated market model,which explicitly includes a measure of governmentalintervention; the “government” or “G” Factor. By doingso, we find that the predictive power of our models soar to87 percent. By accounting for the effects of governmentintervention, however, we see that the price system disin-tegrates: in models that include the G-Factor only the G-Factor significantly impacts property sales. We thus con-clude that, despite the reforms, the Chinese propertymarket remains heavily regulated market. And one that isincomparable with the Western markets that the currentmodels are based upon. In terms of the Chinese road tocapitalism, our findings suggest that while progress hasbeen made, the property market remains firmly under thecontrol of the government.

The remainder of this paper is structured in the usualway. Section 2 reviews the literature on the propertymarket and the role of government intervention. Section3 discusses the data, with which we test our hypotheses,and the models used to analyses it. Section 4 presents theresult, Section 5 discusses the implications, and Section 6concludes on the nature of the market.

LITERATURE

The Chinese Property Market

The story of the Chinese property market is one of afew key dates and policy documents. Together, these form

three distinct phases in the evolution of the market:Nationalization (1949–1976), followed by two distinctperiods of economic reform 1979–1994 and 1994–Date.

Nationalization: 1949–1976

In 1949 the communist party came to power in China.And the state quickly moved to nationalize land and todismantle the system of private housing. The governmentconfiscated all properties that had belonged to formerofficials, “anti-communist reactionaries” and foreign capi-talists (Zhou and Logan 1996). The Municipal Bureau ofHousing Management (MBHM) was set up in 1953, withthe purpose of nationalizing: (1) all residential housingthat was 100m2 or larger (housing owned by overseasChinese and by the remaining petty bourgeoisie, such asmerchants and small business owners, was usually withinthis target range); (2) all privately owned rental housing;and (3) all privately owned commercial buildings. Someproperty remained outside this net. But during the Cul-tural Revolution (1966–1976), all remaining private prop-erty was completely taken over by the socialist state(Kuang 1992; Wang 1990).

After the Cultural Revolution, the sale of land anddevelopment of real properties for speculative purposeswere prohibited (Wang 1990). Housing was treated as awelfare right, to which everybody was entitled and wasideally intended to be distributed according to a formaldefinition of a minimum requirement and a maximumentitlement of space per person (Andrusz 1984). Rentalrates were set such that “ownership” of rental propertyyielded no returns (Deng et al. 2009; Lee 1988).1 In fact,the minimum cost of housing maintenance was more thanthree times as much as the rent collected. And the actualcost of providing housing was almost twenty-five times ashigh (World Bank 1992). There was thus little incentivefor housing investment and improvement. And as a pro-portion of Gross Domestic Product (GDP) housing in-vestment averaged only 1.5 percent in the period from1949 to 1978 (World Bank 1992). Consequently, thePeoples Republic suffered both severe shortages in hous-ing and deterioration of its existing housing stock. TheChinese norm was the assignment of two persons to a one-room unit, three to five persons to a two-room unit and sixto eight persons to a three-room unit (Friedman 1983). Butin reality housing was not available for such a distribution(Zhou and Logan 1996). And as the population grew –from 541 million in 1949 to 961 million in 1978 – theeffect was that average per capita dwelling space dropped;from 4.5 m2 in 1949, to 3.6m2 in 1978 (Li 1998).

Economic Reform: 1979–1994

Recognizing these limits, the state embarked upon aperiod of economic reform in 1979. The State Councilformed the Housing Reform Task Force in 1982. And four


cities (later extended to 80 other cities and towns in 23provinces) were designated for experimentation in hous-ing reform (Wang and Murie 2000; Zhou and Logan1996). Rental rates were adjusted, the existing housingstock began to be privatized, and the state began a processof confirming and registering ownership titles to proper-ties that had been seized during the Cultural Revolution.

In 1988 the Chinese central government issued animportant document: Implementation Plan for a GradualHousing System Reform in Cities and Towns. It aimedexplicitly to transform housing from a welfare good intoa quasi-commodity, with the purpose of raising funds torecover the costs of construction and maintenance, and oflimiting the demand for housing (Zhou and Logan 1996).As a result, public housing units throughout the countrystarted to be sold to their sitting tenants, at heavily dis-counted prices. And when units could not be sold, effortswere made to raise the rent; even if the increase was stilllargely symbolic and still fell far short of what wasnecessary to cover basic maintenance costs. Rentincreases were intended to stimulate investments by resi-dents (by making purchase more economical than renting)and by public and private investors (by increasing theirreturns). And so policy-makers began to rely on marketforces to undo the scarcities and inequalities that charac-terized the socialist housing system.

Prior to this, there was no “market for property.”Although there was some private housing in the cities, andexchanges of public housing were allowed, sales wereuncommon. All land was (and still is) owned by the state.Land requisition for development projects was handledthrough the central planning process. And the central,provincial and municipal governments had the power toappropriate urban land gratis, to acquire buildings withminimum payment to owners, and to acquire rural landwith payment for fair compensation to owners (the people’scommunes) for government approved developmentprojects. In 1988, however, the Chinese National People’sCongress amended the Constitution, stipulating for thefirst time that “land-use rights can be transferred.” For thefirst time, the government legally recognized propertyrights.

Property rights remained weak, however. And thegovernment – particularly the municipal government –continued to control the market. In 1992 “The TemporaryMeasure for Pricing Commodified Housing” was jointlyissued by the Ministry of Commodity Pricing, the Minis-try of Construction and the Ministry of Finance. Thismeasure became the legal basis for price setting, anddictated, through legislative means, how open the pricingsystem should be. In practice, however, municipal gov-ernments set prices arbitrarily, with land prices as low as10 yuan per m2 or as high as 1500 yuan per m2, andcommodified housing selling for as low as 500 yuan per

m2 or as high as 3000 yuan per m2 (Zhou and Logan1996). Clearly therefore, while some steps had been madein the right direction, much work needed to be done tocomplete the task.

Economic Reform: 1994–Date

Another milestone in the move toward a marketsystem came in 1994, when the Chinese central govern-ment issued The Decision on Deepening the Urban Hous-ing Reform. Within this, supply-side and demand-sideprograms were created to facilitate the development of ahousing market. On the supply side, the governmentdecided to build a multi-layer housing provision systemfor different income groups. And on the demand side, adual housing finance system was established to combineboth social saving and private saving (Wang and Murie2000; Deng et al. 2009). Families who paid market pricesfor their units got full property rights – including the rightto resell their units on the open market – while familieswho pay subsidized prices would have partial ownershipbut face restrictions regarding resale (State Council 1994).

Immediately after the 1994 reform, the country sawthe rapid growth of a professional housing developmentindustry and an unprecedented housing construction boom.Yet instead of being sold to individual urban families,most of the housing units were purchased by work units,2

which then resold them at deeply discounted prices totheir employees (Wang and Murie 1996). Since many ofthe work units were state owned and were not subject tohard budget constraints, their purchase behaviors signifi-cantly distorted the emerging housing market (Zhou andLogan 1996). In 1990, and in urban China as a whole, 59percent of housing was owned and managed by work units(Yang and Wang 1992). And even in Shanghai, whereonly 12 percent of housing was owned and managed bywork units in 1990, some 86 percent of new investmentcapital for public housing construction was raised bywork units (Bian et al. 1995; Zhou and Logan 1996).

Recognizing this, the Chinese central governmentdecided, in 1988, to cut the link between work units andhousing provision. Specifically, it issued A Notificationfrom the State Council on Further Deepening the Reformof the Urban Housing System and Accelerating HousingConstruction, which prohibited work units from buildingor buying new housing units for their employees. This,effectively, terminated the distribution of housing by thewelfare principle, which had been implemented for morethan 40 years (Ye and Wu 2008). Under this housingstrategy, urban residents were encouraged to seek housesin the market – according to their own saving and incomeconditions – and could no longer rely on governments, orany other kind of institutional support, in acquiring prop-erty. As such, the 1998 reform marked a turning point ofChina’s housing reform, and with it China was said to


have established a real market mechanism in both housingproduction and consumption (Deng et al. 2009; Ye andWu 2008).

The effect of this reform was certainly tangible. Thepolicy was fully implemented from January 2000. Andfrom 1997 to 2005, the annual housing investment amountincreased by about six times (Ye et al. 2006). Livingstandards soared; the per capita floor space in urban areasrose from 3.6 m2 in 1978 to 17.8 m2 in 1997, to 24.97 m2in 2004 (Ye and Wu 2008). Thanks to both massive publichousing privatization, and strong government incentivesfor home purchase, the levels of homeownership in Chinasoared too, to reach 80 percent in 2004 (Feng 2003).

A True Chinese Property Market?

According to prominent scholars, the 1988 reformcompletely transformed the Chinese market, and com-pleting the move from a centrally planned public housingsystem to a market-orientated one, in both production andconsumption (Ye and Wu 2008; Deng et al. 2009).

Stronger property rights, freedom to buy, and to sell,and to reap the benefits of improvement turned what wasonce a drain on government resources, and a non-produc-tive investment, into an engine of economic growth. As apercentage of GDP, investment in housing rose from anaverage of 1.5 percent in the period from 1949 to 1978(World Bank 1992), to a level of 7 percent in 2005. This,a massive public housing privatization project, and stronggovernment incentives for home purchase, saw home-ownership rates rise, to reach 80 percent in 2004. With thisliving conditions increased dramatically; from an averageof 3.6 m2 per capita in 1978 (Li 1998), average livingspaces increased to 26.11 m2 in 2005 (Ye and Wu 2008).And as demand remains high – since 2003, housing priceshave grown at an average of over 10 percent (Feng2003) – it appears that the transformation of the propertymarket has been a complete success.

This new market-based system, however, has its ownhistoric footprint and particularly Chinese characteristics(Deng et al. 2008). Local governments play a strong rolein the market because, according to the Chinese Constitu-tion, all land belongs to the state. Local governments arethe representatives of the state and are in charge ofexpropriating rural land for urban uses and allocating itamong different users. And by controlling both the landsupply, and the zoning regulations, local governments candecide what can be built, when, and by whom. This leadsto the obvious question: how market-based is China’smarket-based property system?

Modeling the Market

To answer this question, we will test the applicabilityof two property market models: the standard marketmodel, which includes all the factors that presently domi-nates the analysis of market-based property systems, anda new model for the regulated economy, which attemptsto explicitly account for the role of government interven-tion in the property market system.

The Standard Model

In a market economy, the market – guided by the self-interested desire of individuals to maximize their privateprofits – is the “invisible hand” (Smith 1776) of “sponta-neous order” (Hayek 1949). In a market economy, price,supply and demand, and individual economic decisions,allocate resources. And government enforces contracts(Cooter and Ulen 2004).

In a market economy, market forces can therefore beexpected to explain most of the variation in the propertymarket. And several scholars have attempted to specify amarket model, using such market forces, that accuratelydescribes the behavior of the property market in a marketeconomy (cf., Quigley and Redfearn 1997; Gottlieb 1976;Quigley 1999; Mankiw and Weil 1989; Hendershott 1991;Englehardt and Poterba 1991; Aelpeovich 1995; andWoodward 1991; Dipasquale and Wheaton 1992; Case etal. 2005; Campbell 2006). Most of these point to approxi-mately the same significant list of factors (see Table 1).Household income, national economic indicators (mea-sured in terms of gross domestic product), house prices,and interest rates together explain the lions-share thevariance. Although employment rates (as a precursor tohousehold incomes) also plays a part, as does housingsupply, investor expectations and a willingness to spend,vacancy rates, mortgage availability and demographicchange.

These models are far from perfect. Nevertheless, andbased on this literature, one could expect that, in a marketeconomy, a model which accounts for price, and controlsfor income and interest rates, should explain a significantproportion of the variance in sales. We refer to such amodel as the standard model. And if China’s marketreforms created a truly / largely market-based system,then we would expect that such a model should explain asignificant proportion of the variance in Chinese propertysales. Thus, we hypothesize:

H1: The Standard Model will best explain the variance inChinese property sales.


The Regulated Market Model

At the other end of the spectrum, and in a plannedeconomy, a central government – guided by the objectiveof satisfying human need – is the visible hand that controlsindustry (von Mises 1920). And in a planned economy,government decides on the allocation of resources.

Few economies today are planned. Although thelevels of intervention varies substantially. In some cases,intervention can be a positive influence on the marketsoperation; in fact, market failures – which occur when theprivate pursuits of the individual lead to an inefficientpublic outcome (Mueller 2003) – and externalities –which occur when a cost or benefit is not transmittedthrough the price system (Arrow 1969) – are two oftquoted justifications for the existence of the state in amarket economy. But not all intervention can be seen inthe same light.

In some cases, the government intervenes not tosupport the market, but to control it. To the best of ourknowledge, few – if any – studies have attempted tomodify the standard model to accurately describe varia-tion in the property market of a heavily regulated economy.But clearly, the free market models, which the literaturehas focused on, cannot be used in such a setting. Arestriction in supply will stifle demand, alter prices, andwill drastically alter the normal operation of the market.Arguably, and depending on the level of regulation,intervention may so distort the price system that the pricesystem collapses, and only government matters.

Either way, and because government intervention inthe Chinese property restricts supply, and upwardly dis-torts prices, we suggest that intervention will be nega-tively related to sales. And so, to better understand theChinese property market, it will be necessary to supple-

ment the existing literature, and the current market modeldescribed in The Standard Model. Thus:

H2: The Regulated Model will best explain the variancein Chinese property sales.

The G-Factor

If the Chinese property market were a free market,and subject only to market forces, then a model based onHypothesis 1 would explain the vast majority of thevariation in the Chinese property sales. If the Chineseproperty market is free in nothing but name, however, itwill be necessary to account for the role of governmentalintervention. Because in the extreme, government canblot out the effects of normal market forces. In this case,a model based on Hypothesis 2 will better explain varia-tions in the market. The difference between the twomodels is a measure of government intervention, whichwe refer to as the G-Factor. The significance of this factorwill not only serve to supplement the property marketmodels described in The Standard Model Section, andwill help us to better understand the role of regulation inChina, but can be used to chart the so-called “socialistmarket economy of China’s” progress on the road tocapitalism.

METHODOLOGY

Data

The data, against which our hypotheses are tested,was made available by an independent consulting firm,which operates in the international arena. It includes data– collected at the monthly level in the period 2000–2010– on a number of general economic indicators, as well asinformation on the performance of the Chinese propertymarket, at both the city and national levels.3 Table 2

TABLE 1Market Models: A Review of the Literature

+ +

+ +

+

+

++

+

+ +

+–

–

–

–

–


identifies these and the other variables employed in ourresearch. And Appendix A provides a descriptive over-view of the dataset, and reports the correlation coeffi-cients.

Dependent Variable

We use property sales as the dependent performancevariable. Three measures can be employed to measuresales in the market: cumulative real estate investment(IOH), cumulative real estate sales volume (HSV) andcurrent real estate sales in square meter (HS). The first twoof these reflect investments (IOH and HSV) in real estate,in financial terms, and include all niche market investmentlevels as a whole. HSV can, in additions, also be used asa proxy for total current market demand for real estate.Current real estate sales in square meter (HS), by contrast,describes the total supply of real estate, in real terms, andbecause of the nature of the questions asked in thisresearch, we move forward with HS as our dependentvariable of choice.

Independent Variable

We use the G-Factor, or the Government-Factor, toproxy real estate business response to government inter-vention and regulation with current new housing projectin square meter area (NHP), and government interventionwith licensed amount of land in square meter for realestate development by the government (RP). That is:

This ratio projects, mathematically, how many squaremeters the Governments will use to leverage the real-estate business in the next round, as a result of its currentlevels of intervention. Thus, the interaction between gov-ernment and business can be described numerically. Thegame played can be either a larger or lesser ratio: theformer displays a stronger market-oriented power by realestate business and the latter the strength of government’sdirect intervention. The trend, illustrated in Figure 1, tellsus that, every year the business plotting showed a clearhigh market momentum in comparison with a downturntoward the end of the year, reconciling with the strength-ening Government control; the trends also seem to changeas the average slope of the curve was changing – the mostrecent year, i.e., 2010 appeared to show the strongestGovernment determination to bring down the marketoriented real estate business projecting efforts by lesslicensed Government owned land for the business.

We co-integrated the ratio when entering this factorinto the model with all model fit indicators as shownbelow. An Augmented Dicky-Fuller (ADF) test showed aunit root for RP, but NHP was found to be stationary.

Similarly, we construct a variable for the total supply-and-demand of the real estate market at time (t). We definethis variable as the log ratio of the current level of

TABLE 2Data Descriptions

Variable Description Details

HS Housing Sales Cumulative Real Estate Sales in Square Meters

INC Income Levels National Average Income

HSV Housing Sales Volumes Cumulative Real Estate Sales Volume

IOH Investment in Housing Monthly Investment in Real Estate Development

NHP New Housing Projects Cumulative new Real Estate Project in Square Meters

SINTR Short Term Interest Rates Current Short Term Interest Rates

LINTR Long Term Interest Rates Current Long Term Interest Rates

HP Housing Price Current Real Estate Prices

RP Real Property Supply Current Cumulative Amount of Land for Real Estate Development,Under License by the Government, in Square Meters

NHPt

RPt

Gt = In ( ) Eq(1 )


cumulative sales volume of real estate, to the level ofmonthly investment in real estate development. That is:

Finally, we control for the short- and long-terminterest rates (SINTR and LINTR respectively), houseprice (HP), and average income levels (INC).

Models

Next, and following from Hypothesis 1, we model theChinese property market as Equation 3, which suggeststhat the variation in real estate sales (HP) can be explainedby the current real estate price (HP), the total supply-and-demand of the real estate market (DS), the current short-term (SINTR) and long-terms (LINTR) interest rates, aswell as the average income levels (INC). To this, and inline with the discussions in Section 2.2 on the role of theregulation, we add the levels of government intervention(G) to test Hypothesis 2, with Equation 4.

Equation 3:δ(In(HS

t)) = β

0 + β

1 In(HP

t) + β

2 δ(DS

t) + β

3 In(SINTR

t) + β

4 In(LINTR

t)

+ β5 In(INC

t) + ∈

Equation 4δ(In(HS

t)) = β

0 + β

1 In(HP

t) + β

2 δ(DS

t) + β

3 δ(G

t) + β

4 In(SINTR

t) + β

5 In(LINTR

t)

+ β6 In(INC

t) + ∈

We estimate Equation 3 and 4 using both the OLS andGARCH (General Autoregressive Conditional Hetero-skedasticity) methods. After comparing the results (notreported), we find that the former behaves better in termsof fit: the GARCH provides a higher R-squared, aninformation criteria value (AIC) closer to zero, and a

Drubin-Watson closer to 2. Consequently, we move for-ward with the GARCH, and report the results of ouranalysis below.

RESULTS

To test our hypotheses, we estimate two variants ofEquations 3 and 4. Models 1 and 2 are estimated testEquation 3, in accordance with the discussion in TheChinese Property Market Section, and as such considerthe appropriateness of a standard market specificationwhich does not include the effect of government interven-tion. Models 3 and 4, by contrast, are estimated in accor-dance with Equation 4, and in line with the discussion inModeling the Market Section, and as such explicitlyaccount for government regulation in a controlledeconomy. Models 1 and 2, and Models 3 and 4 differ onlyin their inclusion/exclusion of the DS variable. The resultsof all four models are presented on Table 3.

On the Invisible Hand of the Price System

Of the results reported on Table 3, Models 1 and 2 –the standard market models described in The StandardModel Section – consider the impact of price (HP),supply-and-demand (DS), short- and long-term interestrates (SINTR and LINTR), as well as average incomes(INC) on sales.

Model 1 shows, effectively, that the invisible hand ofthe price system works: price is negatively and signifi-cantly related (at the 1\% level) to real estate sales, to thepoint that a 1\ percent rise in price results in almost a 6percent reduction in the level of real-estate sales. Long-term interest rates are positively and significantly related(at the 1% level) with property rates, while short-term

NSVt

IOHt

DSt = In ( ) Eq( 2)

0 . 0

0 . 2

0 . 4

0 . 6

0 . 8

1 . 0

1 . 2

1 . 4

1 . 6

1 . 8

5 12 3 1 0 2 9 2 8 2 7 2 5 1 2 4

M 1 M 1 M 1 M 1 M 1 M 1 M 1

G

FIGURE 1The G-Factor


interest rates have a negative and significant (at the 1%level) effect. This is because of the fact that short-terminterest rates reflect the current price of money, whereaslong-term interest rates can be viewed as an expectation ofgrowth (Hirschleifer 1958). An increase in short-terminterest rates thus makes borrowing more expensive, andlowers the rates of investment in property, while a rise inlong-term interest rates suggests a positive outlook, risingproperty prices, and economic growth, and this attractive-ness of current investment. Finally, we observe that theaverage income level is positively related with propertysale, and that a one percent rise in the former causes a 3.5percent expansion in the latter. This, again, is an intuitivefinding, and taken together the results of Model 1 providesevidence in support of Hypothesis 1, on the natural andfamiliar operation of a free-market price system, in thepresence of Hayekian “spontaneous order.” Finally, andto account for the relatedness between price, and supply-and-demand, Model 2 estimates Equation 4, without con-trolling for supply-and-demand. In it, the effects observedin Model 1 hold, and are shown to be robust. The pricesystem in China, it seems, works.

The “goodness of fit” of Models 1 and 2 is, however,extremely low: an R-Squared of 0.06 and 0.10, and an AdjR-Squared of 0.01 and 0.06 respectively, suggests that ourmodels explain as little one percent of the variance inproperty sales. The free market models therefore, wesuggest, quite clearly misses some important explanatoryvariables in understanding the market.

On the Visible Hand of the Chinese Government

Of the results reported on Table 3, Models 3 and 4 –the regulated market models described in The RegulatedMarket Model – attempt to correct for the obvious omis-sion of governmental intervention in Models 1 and 2, byadding the G-Factor to the free market specification. Indoing so, we move to explicitly consider the effects of the“visible hand” on the variance in the Chinese propertymarket.

Model 3 shows that governmental action is nega-tively and significantly related (at the 1% level) withproperty sales, and suggests that a one percent rise in

TABLE 3Regression Results

Market Model Regulated Market Models

Variable Model 1 Model 2 Model 3 Model 4

lnHP -5.64*** -5.72*** -0.39 -0.60[-26197.80] [-3.24] [-0.48] [-0.70]

G – – -4.66*** -4.61***– – [-31.53] [-10.73]

DS – 0.33 – 0.89– [0.31] – [2.05]

LINTR 34.87*** 34.61*** 2.90*** 1.93[4.79] [225.94] [137.10] [0.30]

SINTR -31.75*** -31.53*** -3.03*** -1.99[-4.29] [-443.10] [-1572.36] [-0.34]

INC 3.44*** 3.55* 0.17 -0.45[41.68] [1.65] [0.20] [0.45]

C 4.31*** 3.96 1.66 0.44[3.58] [0.45] [0.66] [0.12]

No. Obs 1050 1050 1050 1050R2 0.06 0.10 0.88 0.87

Adj-R2 0.01 0.06 0.87 0.86AIC 1.99 1.95 0.36 1.95DW 2.09 2.09 2.14 2.27***.

Significant at the 0.005 level (2-tailed). **. significant at the 0.01 level (2-tailed). *. significant at the0.05 level (2-tailed).


government activities causes a 4.6 percent drop in thelevel of property sales. This, we suggest, supports Hypot-hesis 2, on the existence of a significantly negative rela-tion between governmental intervention and propertysales. The other variables – namely supply and demand,short- and long-term interest rates, housing price, andaverage income – are not significantly related anymore.Again, to account for the possible relatedness of thesupply-and-demand variable (DS) with price (HP), Model4 re-estimates Model 3, but this time without controllingfor the effects of supply-and-demand. In it, the insignifi-cant effects of price – observed in Model 3 – hold, and areshown to be consistent – thus supporting our conclusionson the effects of governmental intervention on the market– although then the significance of short- and long-terminterest rates are re-established.

Finally, and turning to the “goodness of fit,” we seethat adding government intervention to the equation, inModels 3 and 4 drastically improves our understanding ofthe variance in the market. An R-squared of 0.88 and 0.87,and an Adj R-squared of 0.87 and 0.86 respectivelysuggest that property models which include the G-Factorexplain as much as 88 percent of the variance in sales. Incontrast to the results of Models 1 and 2, this is animpressive finding.

On the Significance of the G-Factor

Taken together these results suggest that, in theabsence of government intervention, the price systemworks, and the market functions as well. Low R-Squaredin both Models 1 and 2, however, suggest that the freemarket models fail to explain much of the variation in theChinese market. Adding government intervention to theequation in Models 3 and 4 improve the predictive powerof the model; an R-Squared of 0.88 and 0.87, and an AdjR-Squared of 0.87 and 0.86 respectively suggest that theG-Factor models explain as much as 88 percent of thevariance in sales. We can therefore suggest that not onlycan the results of the G-Factor models be relied upon, butsee that intervention plays a significant – of not the mostsignificant – role in the Chinese market. In fact, becauseprice (HP) becomes insignificant in Models 3 and 4, wecan suggest that, when utilized, the “visible hand” of aninterventionist government can neutralize the “invisiblehand” of the market mechanism. This is an importantfinding, we suggest, with a number of important implica-tions, both in general, and for China, which are discussedbelow.

DISCUSSION

Key Findings

1. The property market is ruled by the invisible hand. . . .

The results presented on Table 3 suggest that, in the“base case,” and using standard market models, the prop-erty market operates under normal market conditions. Ourresults show that property price negatively impacts sales,as do short-term interest rates. Long-term interest rates,which we interpret as an expectation of future growth, andaverage incomes positively, impact sales. All are, wesuggest, intuitive results, and evidence the natural opera-tion of the “invisible hand” in the property market. Indoing so, we replicate the findings of numerous authors onthe operation of the property market; papers and findingswhich we summarize on Table 1.

2. . . .until the visible hand of government intervenes.

The results of Models 3 and 4 – on the significance ofthe G-Factor – suggest, however, that when we accountfor the “visible hand” of an interventionist government,the “invisible hand” of the market mechanism dissolves.Our results show that the G-Factor negatively and signifi-cantly impacts the volume of sales, and that in the pres-ence of the G-Factor the price system fails to significantlyimpact the volume of sales. Long- and short-term interestrates, and average incomes, report the expected results.But the significance of the G-Factor evidences the domi-nance of the visible hand in the property market. Forproperty market models in general, and for the case ofChina road to capitalism in particular, this is an importantfinding. It suggests not only that scholars looking toexplain Chinese market movements should account forthe role of the government. But also suggests that,although much progress has been made, the Chineseproperty market is not free, or market-oriented, but stillvery much state controlled.

3. Property market models must include governmentintervention.

Low R-Squares in both “base case” models suggestthat the free-market price system fails to explain much ofthe variation in the Chinese property market. An R-Squareof 0.06 and 0.10, and an Adj R of 0.01 and 0.06 respec-tively, suggests that these models explain as little onepercent of the variance in property sales. Adding govern-ment intervention to the equation in Models 3 and 4,however, drastically improve the predictive power of themodels. An R-Square of 0.88 and 0.87, and an Adj R of0.87 and 0.86 respectively suggests that the G-Factormodels explain as much as 88 percent of the variance insales. We can therefore suggest that not only can theresults of the G-Factor models be relied upon, but thatintervention plays a significant role in our data-set, andmust be included in future analyses of the Chinese prop-erty market. This is an important finding for authorsworking to better describe property market movements, ingeneral.


4. Chinese property market depends on governmentsupply.

In our model, government intervention is measuredin terms of the volume of land that it adds or withdrawsfrom circulation. The implication from our findings isthat, from the supply side, it the Government, and thedecisions that they make on the quantity and quality ofland available which effects market demand, and not theprice system. Government intervention is therefore thebottle-neck in property sales, we suggest, and this raisesquestions as to what level an equilibrium would have beenachieved in the absence of an interventionist government.We can speculate that a free-market would lower prices,and fuel a boom, and that perhaps government interven-tion may be rationalized as a measure to slow the propertyboom. The literature on transaction costs economics,however, draws the logic of this conclusion into question:if the purpose of intervention is to slow demand, thecreation of a monopoly on the supply of land is not theoptimal approach for achieving these ends.

5. The Chinese economy remains a controlled economy.

Finally, and in terms of classification, it is clear thatwhile the “socialist market economy of China” has mademuch progress on the road to capitalism, much work stillneeds to be done. It is clear from our analysis in this paperthat China remains a controlled economy. In spite of therecent discussions on the opening up of China, the market-based economic reforms it has instigated since the 1970s,and the emergence of a consumerist middle-class, it isclear from our results that the government remains the firstand foremost power, at least in the Chinese propertymarket. Whether this is to be welcomed, or worried aboutremains unclear. But for now our results suggest that the“visible hand” of government remains predominant.

Future Research

Owing to the richness of the data, and the fact that wehave observations too at the city level, we invested the roleof government regulation in each of the 10 cities; namely,Shanghai, Beijing, Guangzhou, Shenzheng, Tianjing,Wuhan, Hangzhou, Chongqing, and Nanjing. In eachcase we estimated Equation 4, and our results (not re-ported) show variation in the significance of governmentintervention across regions. In eight of the ten casesgovernment influence had a negative and significantimpact (at the 1% level), with beta estimates ranging from-0.24 to -1.18. The cause of the variance is, however,outside the scope of this particular project, and so the

picture of government intervention that emerges from ourstudy – at least at the local level – remains unclear. In afuture research project, it might be interesting to considerpolicy differences, at the city level, not only in order tobetter explain variance across regions, but in an effort tobetter profile the differences in the visible hands of theChinese government.

CONCLUSIONS

The Chinese property market has undergone majorreforms in recent years. And so profound were the changesthat prominent scholars suggest that, at least in terms ofthe property market, China has completely moved from acentrally planned public housing system to a market-based system, with competition in both production andconsumption (Ye and Wu 2008; Deng et al. 2009). At thesame time, however, it is said that the new market-basedsystem has its own particularly Chinese characteristics.With the local governments decided what can be built,when, and by whom. The aim of this research is to describehow market-based China’s new market-based propertysystem really is. And to do so, we test the applicability ofa number of standard, property market models, calabratedto market-orientated property systems.

We used a straightforward empirical model, andemploy data, collected in the period 2000–2010, on themarket performance of the real estate industry, along witha number of important macroeconomic controls variables.We tested the standard “market model,” and in doing sofound that markets work, in general. However, suchmodels explain as little as 1 percent of the variance inproperty sales. And clearly therefore, a huge amount ofvariance remains unexplained. By adding a variable thataccounts for government intervention to the “marketmodel,” we create a “regulated market model. And withthis specification, the predictive power of the model soarsto 87 percent. In the “regulated market model,” however,the significance of the price system disintegrates. In thesemodels, it is government supply, and not price, whichdominate sales.

The implications of our study are significant. In termsof property market models, we show that those modelscalibrated to Western markets do not necessarily fit east-ern ones. Secondly, and more importantly, we not onlyevidence the important role of government’s “visiblehand” in the Chinese property market but, by implication,illustrate that, despite the progress that has been made,China remains a controlled economy. This leads us towonder if China is still on the road of capitalism. Or if thegoal is to have a free market in nothing but name.


ENDNOTES

1 On average, an urban worker spent less than 5 percentof his or her family income on housing includingutilities in the 1970s, down from 8 percent in 1957and from 14 percent in 1950 (Lee 1988).

2 A work unit or danwei is the name given to a place ofemployment in the People’s Republic of China.

3 At the city level, we include details on the market ofperformance of Shanghai, Beijing, Guangzhou,Shenzheng, Tianjing, Wuhan, Hangzhou, Chongqing,and Nanjing

REFERENCES

Aelpeovich, G. (1995), “The Baby Boom, the Baby Bustand the Housing Market a Further Look at the De-bate,” The Annals of Regional Science, 111–16.

Andrusz, G.D. (1984), “Housing and Urban Develop-ment in the USSR.” Albany, NY: State University ofNew York Press.

Arrow, K.J. (1969), “Classificatory Notes on the Produc-tion and Transmission of Technological Knowledge,”The American Economic Review, 59 (2), 29–35.

Bian, Y. and J.R. Logan (1995), Market Reform and thePersistence of Power: The Changing StratificationSystem in Urban China. Unpublished manuscript.

Campbell, J.Y. (2006), “Household Finance,” The Jour-nal of Finance, LXI (4).

Case, K.E., J.M. Quigley, R.J. Shiller (2005), “Compar-ing Wealth Effects: the Stock Market Versus theHousing Market,” Advanced in Macroeconomics, 5(1).

Cooter, R. and T. Ulen (2004), Law and Economics, 4th

Edition. Addison Wesley Publishing Company.Deng, L., Q. Shen, and L. Wang (2009), Housing Policy

and Finance in China: A Literature Review. Preparedfor U.S. Department of Housing and Urban Planning.

Dipasquale and Wheaton (1992), “The Market for RealEstate Agents and Space, a Conceptual Framework,”Journal of the American Real Estate and UrbanEconomics Association, 20, 181–97

Englehardt, G. and J. Poterba (1991), “House Prices andDemographic Changes: Canadian Evidence,” Re-gional Science and Urban Economics, 539–46.

Feng, Wang (2003), “Housing Improvement and Distri-bution in Urban China: Initial Evidence from China’s2000 Census,” The China Review, 3 (2), 121–43.

Friedman, B.S. (1983), “Public Housing in China: Poli-cies and Practices,” Journal of Housing, 40 (3), 82–85.

Gottlieb, M. (1976), Long Swings in Urban Development.Columbia University Press for National Bureau ofEconomic Research.

Hayek, F.A. (1949), “The Intellectuals and Socialism,”The University of Chicago Law Review, 16 (3), 417–33.

Hendershott, P. (1991), “Are Real Prices Likely to De-cline by 47%,” Regional Science and Urban Eco-nomics, 553–63

Hirshleifer, J. (1958), “On the Theory of Optimal Invest-ment Decision,” Journal of Political Economy, 66(August), 329–52.

Kuang, C. (1 992), Two-Hundred Points in Buying, Sell-ing, Renting, Renovating, or Constructing. Bejing:The Central University of Ethnicity Press.

Lee, Y.F. (1988), “The Urban Housing Problem in China,”China Quarterly, 115, 387–407.

Li, S., Shuhe Li, and Weiying Zhang (2000), “The Roadto Capitalism: Competition and Institutional Changein China,” Journal of Comparative Economics, 28(2), (June), 269–92.

Li, Rongxia (1998), “Residential Houses Going to theMarket,” Beijing Review, 41(20), 14–17.

Mankiw, N.G. and D.N.Weil (1989), “The Baby Boom,the Baby Bust, and the Housing Market,” RegionalScience and Urban Economics, 19, 235–58.

Mueller, D.C. (2003), Public Choice, 3rd Edition. Cam-bridge University Press (February 17, 2003).

Quigley, J.M. and C.L. Redfearn (1997), Are Variationsin Regional Housing Prices Predictable? Mimeo,August 23.

____________ (1999), “Real Estate Prices and EconomicCycles,” International Real Estate Review, 2 (1), 1–20.

Smith, A. (1776), An Inquiry into the Nature & Causes ofthe Wealth of Nations.

State Council of People’s Republic of China (1994), “TheDecision of the State Council on Deepening UrbanHousing System Reform” (Guan Yu Shen Hua ChengZhen Zhu Fang Zhi Du Gai Ge de Jue Ding). Re-trieved from http://www.bjjs.gov.cn/publish/portal0/tab932/info3639.htm.

Wang, Ya Ping and Alan Murie (2000), “Social andSpatial Implications of Housing Reform in China,”International Journal of Urban and Regional Re-search, 24 (2), 397–417.

Wang, Y. (1990), “Private Sector Housing in UrbanChina Since 1949: The Case of Xian,” HousingStudies 7.2, 119–37.

Woodward, S. (1991), “Economists Prejudices: Why theMankiw Weil Story is not Credible,” Regional Sci-ence and Urban Economics, 531–37.

World Bank (1992), China: Implementation Options forUrban Housing Reform. A World Bank CountryStudy. Washington, D.C.: The World Bank.

Ye, Jian-Ping and Zheng-Hong Wu (2008), “Urban Hous-


For further information contact:Killian J. McCarthy

University of GroningenSchool of Economics and Business

Nettelbosje 2, 9747 AEGroningen, the Netherlands.Phone: +0031.50.363.6810


APPENDIX A

TABLE A1Data Descriptives

TABLE A2Correlations

ing Policy in China in the Macroregulation Period2004–2007,” Urban Policy and Research, 26, 283–95.

Ye, Jianping, Deye Wu, and Jian Wu (2006), “A Study onthe Chinese Housing Policy During Social Transi-

tion: Practice and Development,” Housing FinanceInternational, 20 (3), 50–58.

Zhou, M. and J.R. Logan (1996), “Market Transition andthe Commodification of Housing in China,” Journalof Urban and Regional Research, 26, 223–35.


STUDYING EMOTIONS IN BUSINESS RELATIONSHIPS

Andreas Zehetner, University of Applied Sciences, Austria

ABSTRACT

This paper studies the nature of emotions in profes-sional purchase decisions. The findings of an exploratorystudy show a variety of emotions which are present inbusiness buying decisions. Based on this, an extendedconceptual model is proposed which includes emotions asvariables in organizational buying behavior.

INTRODUCTION

In the light of current trends in global marketing (e.g.,global sourcing, cross-linked buyers and sellers, net-works and alliances, information transparency . . .) busi-ness to business (B2B) buyer-seller relationships arebecoming increasingly more important and complex.Knowledge about what shapes a buyer’s decision is essen-tial for business success in a highly competitive environ-ment. There is strong evidence that, besides rationalaspects, emotional factors play a pivotal role within pur-chasing decisions in general (Andersen and Kumar 2006).However, many professional buyers and sellers still ne-glect the dual nature of human consciousness: the indi-vidual as an objective, rational problem-solver versus anemotional, sensual being with mystical properties(Hirschman 1985). Business purchasers base their deci-sions primarily on rigorous cognitive analysis (Erevelles1998), straightforward processes and a cognitive chain offunction-based arguments (Ho et al. 2010). It can beassumed that neglecting other than cognitive antecedentsto business decision making reduces the potential ofbeneficial results and performance as only “one side of thebrain” is used. Scientific contributions regarding the roleof emotions in a professional vendor-customer relation-ship are found only occasionally. Some surveys are basedon artificial lab settings (e.g., Kopelman et al. 2006;Overbeck et al. 2010) whose authors admit the limitedexplanatory power of such methods for understandingemotional processes (Erevelles 1998). Other findings areof limited generalizability due to the fact that only specialaspects such as advertising (Poels and Dewitte 2006) orbranding (Lynch 2004), or specific business areas such asstock markets (Myeong-Gu and Barrett Feldmann 2007)are analyzed. A third group of contributions remain in thestage of conceptual models which still await empiricalconfirmation (Andersen and Kumar 2006; Bagozzi 2006).This gap in literature gives reason to empirically investi-gate the role of emotions in business relationships in amore general way in order to provide better preconditionsfor B2B sales and marketing activities.

LITERATURE REVIEW

Prior to empirical analysis, a literature review wasconducted on sell-side motivated issues as well as on buy-side-related matters. The following overview starts withrather generic articles and gradually increases the degreeof specialization. Emotions are short-lived psychologicalreactions that respond to a specific situation, triggered bya certain event or used as a means to achieve an expectedgoal (Lazarus 1991; Kopelman et al. 2006). From acognitive perspective, emotionality in managerial deci-sions is considered to be the opposite of rationality oreffectiveness and thus should be regulated and “normal-ized” (Bagozzi 2006; Ashforth and Humphrey 1995;Ashforth and Kreiner 2002). However, if this were thetruth, “firms could computerize much of [. . .] transactionsand do away with most of the human element” (Bagozzi2006, p. 453).

Decision making is supposed to be based on rationalcoordination of beliefs and desires, but additionally con-siders a supporting role of emotions (Lakomski and Evers2010). In contrast to psychological or neurobiologicalresearchers, who firstly acknowledged the impact of emo-tions on decision making (Kuzmina 2010), in a manage-rial context decisions are often supposed to depend on theexpected utility. This becomes manifest in economictheories such as the transaction cost approach (Williamson1981). Following the assumption of bounded rationality(Simon 1990, 1997), however, modifications comparedto purely rational decision models are demanded for tworeasons: first the availability of complete information andperfect foresight is unrealistic, making the calculation ofexpected values difficult. Second the disposability of allthinkable behavior patterns is not granted. Simon (1990)therefore proposes using approximation approaches, inclu-ding rational criteria and other, intuitive components,usually without being able to clearly distinguish betweenthe two (Lakomski and Evers 2010).

Discussions about the role of emotions in marketingare ubiquitous. Wind (2006, p. 474) states that “the buy-ing criteria of organizational buyers and consumers aremultidimensional and involve relational and emotionalcharacteristics, not only a consideration of feature, func-tionality, delivery, and price.” Bagozzi et al. (1999, p. 202)describe the relevance of emotions for the entire market-ing process: “they [emotions] influence information pro-cessing, mediate responses to persuasive appeals, mea-sure the effects of marketing stimuli, initiate goal setting,


enact goal-directed behaviors, and serve as ends andmeasures of consumer welfare.”

In consumer research, the efficiency of exploitingemotional approaches of consumers to stimulate, shapeand reconfirm decisions is widely accepted and wellresearched (e.g., Bagozzi et al. 1999; Laros and Steenkamp2005). In contrast, in B2B settings there is scepticismabout the impact of emotional approaches on marketingsuccess and performance (Lynch and Chernatony 2007).The effect of emotional issues on business operations hasbeen only rarely acknowledged, e.g., associated with teamprojects (Peslak 2005), concerning management(Brotheridge and Lee 2008) or in specific fields, such aspublic relations (Yeomans 2007). Cater and Cater (2009),who investigate customer loyalty in the B2B servicesector, conclude that emotion influences business rela-tionships more than ratio. In many conceptual models,emotions are hypothesized as playing a role in initiating,developing and sustaining relations (e.g., Andersen andKumar 2006).

A body of scholars argues that negotiations are pre-dominantly influenced by cognition (Raiffa 1982; Nealeand Bazerman 1991; Olekalns and Smith 2005). In mana-gerial practice, purchasers also claim to be motivated byintellect alone, whereas sales practitioners assume thattheir counterparts run on both, reason and emotions(Reichard 1985). Recent work on the use of affect inpersonal selling is mostly related to the seller’s point ofview, focusing on sales call anxiety (Belschak et al. 2006),shame/embarrassment (Verbeke and Bagozzi 2002), lis-tening skills (Comer and Drollinger 1999), optimism(Schulman 1999) and positive mood states (George 1998).Interpersonal effects of emotions (van Kleef et al. 2004)and the buyer’s perspective – how industrial purchasersperceive and handle emotions – are not empirically sub-stantiated.

Only a few publications cover the issue of emotionsin buyer-seller relationships from a buy-side point ofview. Hook et al. (2002) emphasize emotional capabilitiesas a necessary supplement to technical capabilities. Theyhowever treat, like many other related articles, emotion ona rather generic level, without specific focus on buyer-seller relations or purchasing decisions. A similar abstrac-tion level is found in Wilding (1999), who proposes moreefficiently leveraging soft skills to obtain time savings andagility improvements.

Contributions that take a closer look at purchasingdecisions show a diverse focus. Giunipero et al. (1999)observed a fair balance between the use of formal data andtacit knowledge (implying behavior-related aspects) in asurvey with purchasing managers and state a clear rela-tionship to the concept of bounded rationality. However,there is no explicit discussion regarding the role of emo-

tions in the course of purchasing decisions. An empiricalapproach is presented by Donada and Nogatchewsky(2009), who study the issue of supplier switching in thehotel industry. They assume a clear influence of emotionscomplementing rational issues (e.g., switching cost), how-ever restricted to the hotel sector, thus the ability togeneralize results is limited.

The literature review has confirmed constitutiveassumptions: emotional influences on professional pur-chasing decisions have not yet been intensively investi-gated from a buy-side perspective, although there isnotable evidence that such decisions do not only rely oncognitive analysis. Conceptual and empirical researchneither exhibits sufficient explanatory power nor a satis-factory generalization capability. Emotions are complexconstructs that were intensively investigated by psycholo-gists, but have barely been a subject among economists(Loewenstein 2000).

METHODOLOGY AND EMPIRICALRESEARCH DESIGN

Being in a very early stage of research, it is importantto explore in-depth in what regard emotions impact pro-fessional purchasing decisions. Prior to selecting researchmethods and deciding on what type of data to collect, a“pre-empirical question development” (Punch 2005, p. 21)was applied. An initially small bundle of questions wassubjected to creative exploration in a discourse with peerscholars, yielding an expanded question list that wassubsequently “disentangled” (Punch 2005, p. 35) into twostable research questions:

RQ1: Do emotions play a role in the organizationalbuying and decision making process?

RQ2: Which emotions can be identified as playing adominant role?

The selection of an applicable research approachdepends on the access to the field (Flick 2006) and the typeof data required answering the RQs (Punch 2005; Patton2002). The main concern was to explore the nature ofemotional aspects of B2B purchasing including possiblyemerging unexpected other “irrational” aspects ratherthan testing predefined hypotheses or quantitatively mea-suring generic emotional categories. The study’s princi-pal focus on how people make decisions including emo-tional aspects can’t be investigated with quantitative meth-ods or experimental lab-settings only, when seeking aholistic understanding of the given social reality (Patton2002). An established methodical approach to conductingqualitative research was introduced by Mayring (2008).As this proceeding notably suits the circumstances of thepresent investigation, it was particularly respected in themethodological progression.


Methodological Mix, Triangulation, and Data Pro-cessing

In order to achieve realistic and thorough insights andto enhance validity through reciprocal verification, thisstudy applies different kinds of triangulation (Denzin1978; Jick 1979): Data triangulation: Interviewing, obser-vation, document analysis and content analysis of the dataderived from the fieldwork were combined. Investigatortriangulation: A multidisciplinary team was employed,including researchers from sales and marketing, supplychain management, sociology, and psychology. Method-ological triangulation: this research used multiple meth-ods to study a single problem to widen the insights and toexclude methodological artefacts (Jick 1979). The quali-tative methods of personal, individual interviews, expertinterviews, and focus groups with two different targetgroups (sell-side/buy-side) have been combined.

The data processing and analysis were done by quali-tative content analysis based on audio/ video records andwritten transcriptions and secondly by a computer assis-ted qualitative data analysis using MAXQDA (Flick 2006;Corbin and Strauss 2008).

Empirical Research Design

The study design was three-tiered. First, narrativeexpert interviews were conducted to ensure a high qualityset-up with a broad perspective and yet conceptual clarity(Punch 2005). The subsequent fieldwork took place in theform of cross-industry personal interviews and focusgroups. At first, three experts, who exhibited a substantivefield experience in (1) sales, (2) SCM and purchasing, and(3) psychology were interviewed concerning their mindset,appraisal and ideas regarding the investigation of emo-tional influences on professional purchasing. Second, aguided interview design was developed and pretestedwith research peers based on the findings from step oneand literature review. Next, ten in-depth face-to-faceinterviews with key purchasing decision makers wereconducted. Core issues during these interviews includedpurchasing decisions, decision strategies, procedures, con-sequences to be expected subsequent to a decision, andcontext factors, e.g., incomplete information, lack of time,complexity, etc. Initially, the intention to investigate therole of emotions was hidden to obtain an unbiased impres-sion regarding cognitive and affective aspects and theirmutual weighting. Step by step, indirect questions wereappended (e.g., asking for examples and asking how arespondent had thought or felt in a reported situation thatimplied emotional aspects). Finally, emotions were expli-citly asked for. In the third and last step, two focus groupsof six participants each were conducted (sales and pur-chasing managers separately) with a new participantsample. This was done to discover collective orientationsand to ensure that differences compared to the interviews

or literature findings would result from the researchsubject but not from the used method (Jick 1979).

Sample Determination

Respondents from Austrian B2B firms were invitedto participate via phone, indicating the intention to inves-tigate purchasing situations without explicitly mention-ing emotional issues. Selection criteria were: consider-able B2B purchasing/sales experience and decision re-sponsibility. The study was conducted across industrysectors, product range, purchasing volume, and companysize. Main industry sectors involved were automotive,electronics, glass, paper, mechanical engineering, metalware, automation, and environmental engineering.

Empirical Findings and Discussion

First, the following rational decision drivers, beingcongruent with those mentioned in the literature (e.g., Hoet al. 2010), were identified: financial aspects: price,follow up costs, terms of payment; product issues: techni-cal attributes, product quality/ service; order characteris-tics: purchase order volume and value, warranty regula-tions; logistic attributes: delivery time, delivery capacity/reliability, replenishment lead time, flexibility; supplierissues: supplier performance/competence, references,transparency; coordinative attributes: internal processes.

Most respondents confirmed that emotions do play arole in the buyer-seller encounter (“feelings are alwaysvital [. . .] not only the rational arguments [. . .] also thefeeling”). However, the hard facts are regarded as para-mount in the decision process, while emotions seem toplay a shifting role during the buying process and weresaid to be neglected in the final decisive stage (“by nomeans are emotions pivotal for the decision [. . .], howeverfeelings may influence the way toward the decision”;“emotions have no influence when it comes to decision,but [. . .] earlier”; “emotions [. . .] are important in thesupplier pre-selection”).

Purchasing managers try to exclude emotions duringsupplier selection. However, they admit they cannot beneutralized (“emotions should be eliminated”; “primarygoal is to exclude emotions”; “it’s people that are work-ing, and people are triggered by emotions [. . .] which mayhave little relation to the factual arguments”). There arepositive associations with emotions such as a “feeling fora supplier” or a “gut instinct.” According to a majority ofrespondents, such intuition originates from long-termexperience (“if your intuition says ‘be careful,’ you shouldbe careful, because mostly it is an appropriate interpreta-tion [. . .] related to experience”).

In regard to the stated cognition-emotion balance(“friendly with humans while hard in business”), a sound


information acquisition and diligent preparation weretaken for granted for sourcing decisions. Environmentalfactors such as complexity, incomplete information orexpected risks were faced with initially rational and thenemotional arguments. Responds from interviewees andgroup participants were inconclusive concerning the emo-tional constituents of such situations: some respondentsemphasized anxiety to avoid emotions; others consideredaffective impetus as constitutional – at times utile, at timesunsettling.

In long-term and strategic supplier relations, feelingsgain a greater significance (“if we are concerned aboutsoft facts [. . .], we would rather pay a cent more”),showing a broad range of emotions (“we are all humanbeings related to emotions”; “purchasers also have emo-tions”; “a positive atmosphere is extremely important”).The respondents perceive emotions to be a trigger for theestablishment and maintenance of relations, assumingpersonal contact as vital. This is in line with Andersen andKumar (2006), who state that negative emotions willincrease transaction costs as well as problems to maintainand repair a buyer-seller relationship. The human compo-nent was specified as sympathy/antipathy, friendship, joy,happiness, excitement or pride, but also anger, fear andtension.

The observed variety of emotions was broad. Thedata in particular yields the following positively per-ceived issues as more dominant than others: Feeling ofpersonal success and motivation, i.e., experiencing suc-

cess results in pleasure, enjoyment and motivation, impli-cating corporate and personal success, responsibility andlong-term orientation are beneficial. Pride (“if you [. . .]show a saving of at least 100,000 EUR [. . .], this is a goodfeeling” or “I become ‘emotional,’ if I complete a gooddeal”). This supports the transferability of existing re-search regarding the central force of pride as a positivesocial emotion in B2B marketing (Lewis 2000; Bagozzi2006). Positive excitement, thrill, fun, joy, and pleasurewere prevalently mentioned;

The following negative emotion bundles have beenindicated most frequently: Anger, annoyance, negativeexcitement, and tension, related to business partnerships,negotiations or company-internal relations. A further occa-sion for anger, fear and tension was seen in case of wrongdecisions, especially if associated with high risk or uncer-tainty. Nervousness, mistrust, fear, or tentativeness wasstated often as well. Power was mentioned in the contextof business partnerships and negotiations, above all con-cerning the issue of pressure on price.

PROPOSED CONCEPTUAL FRAMEWORK

The proposed conceptual model builds on the find-ings of the aforementioned exploratory study. It drawsheavily on existing conceptualizations of the linkagebetween loyalty, satisfaction and its antecedents in differ-ent contextual settings (cf., e.g., Cronin and Taylor 1992;Fornell 1992; Johnson et al. 2001; Ka-Shing and Ennew2005; Lee et al. 2008; Helgesen and Nesset 2007). Based

ILLUSTRATION 1 Proposed Conceptual Model

Cognitiveantecedents of

B2B purchasingdecisions:

•Product quality•Price/cost

•Service quality•Relationship quality

PositiveEmotions:

Joy, pleasure,thrill,

excitement. . .

NegativeEmotions:

Anger,tension,

annoyance. . .

Satisfaction Loyalty

Controlvariables:

Buying situationIndrusty type

Switching costs


on the knowledge gained by the exploratory investigation,positive and negative emotions are added as mediators ofthe relationship between cognitive antecedents and satis-faction and loyalty, both being predictors of businessperformance.

Key Constructs

Referring to social exchange theory, it can be statedthat a business buyer will be most likely to decide for asupplier from which he expects to receive the greatestbenefits (Thibaut and Kelley 1959). The more “key buy-ing criteria” a supplier is able to provide, the better hischance of also being considered for future transactions(Anderson and Narus 1984; Anderson and Narus 1990).Thus, it is important to know about suppliers’ choicefactors and their relative weight for decision making inorder to achieve customers’ satisfaction, loyalty and repur-chase intention.

Loyalty

This model focuses on loyalty as the ultimate depen-dent variable, being explained by cognitive as well asaffective drivers which lead to satisfaction and, conse-quently, to loyalty. Loyalty, defined as a “deeply heldcommitment to rebuy or repatronize a preferred productor service in the future” (Oliver 1999, p. 34) is a pivotalcomponent in B2B relationships which are characterizedby high efforts during the whole acquisition, decision andfulfillment process. Keeping an existing customer is manytimes more efficient than acquiring a new one (Fornell1992).

Satisfaction

According to Fornell (1992), customer satisfaction isan overall evaluation which is formed on the basis ofexperience after purchasing a product or service. Thisexperience is formed by appraisal of cognitive and emo-tional factors that occur during the relationship or buyingprocess (Mano and Oliver 1993; Westbrook 1987).Despite the fact that satisfaction does not universally leadto loyalty (Oliver 1999) and satisfied customers mightdefect (Naumann et al. 2010), satisfaction is widelyacknow-ledged as being an important predictor of cus-tomer loyalty (e.g., Oliver 1980; Zeithaml, Berry, andParasuraman 1996; Johnson et al. 2001).

Cognitive Antecedents

Many potential variables may represent cognitivequality drivers. A main concern is to select variables thatare important in the respective context. There are studiesexamining effects of cognitive factors on satisfaction andloyalty (Rauyruen and Miller 2006). In a B2B setting,Bharadwaj (2004) found product quality, service quality

and order fulfillment to be the most important criteria inthe procurement of electronic components. Ho et al.(2010) found financial aspects, product issues, ordercharacteristics, logistic attributes, supplier issues, andcoordinative attributes. Based on this and within thecontext of B2B buying decisions, product quality (1),price/cost perception (2), service quality (3), and relation-ship quality (4) are selected as main cognitive antecedentswithin the proposed model.

Product quality (1): The principal reason for anorganization to engage in relationship with its suppliers isto source products and services needed in its own produc-tion and transformation processes (Ulaga 2003). In thelong run, customers will not continue the relationship witha supplier unless they receive a product that meets theirstandards (Cater and Cater 2010). Product quality servesas a “conditio sine qua non” and quality standards have tobe met to become or remain a supplier respectively.Following Ulaga (2003), product quality and its sub-constructs are directly and positively related to valueperception of buyers, respectively to purchase intentions(Boulding and Kirmani 1993; Wells et al. 2011).

Price/cost perception (2): As opposed to the expectedbenefits, the customer’s sacrifices have to be included inthe model. Both practitioners and scientists agree that theprice of an offer is a pivotal determinant of purchasingdecisions, satisfaction and future purchase intentions(Anderson et al. 1994; Voss et al. 1998; Zeithaml 1988).

Service quality (3) has been defined by Bitner andHubbert (1994, p. 77) as “the customer’s overall impres-sion of the relative inferiority/superiority of the organiza-tion and its services.” Caused by an ongoing convergenceand homogenization of offered products in B2B markets,differentiation becomes more and more difficult for busi-ness suppliers. The less the product itself can providedifferentiating value, the more service quality plays a rolein providing success (Parasuraman, Zeithaml, and Berry1985; Reichheld and Sasser 1990). Previous researchfound an association of service quality with behavioralintentions (Zeithaml et al. 1996).

Relationship quality (4) and its determining factorshave been investigated in numerous recent studies(Athanasopoulou 2009; Edvardsson et al. 2008; Ka-shingand Ennew 2004). Relationship quality and satisfactionand/or behavioral outcome have also been found to berelated by Rauyruen and Miller (2007) and Gil-Saura et al.(2009).

Affective Variables as Mediators

Helgesen and Nesset (2011) examined positive andnegative affective antecedents of loyalty and satisfactionin the context of a collage library. They found evidence for


a positive relationship between positive affect and satis-faction. Appraising this strong role of emotions in shapingand influencing the appraisal, processing and evaluationof cognitive antecedents and behavioral outcome, it isproposed that in B2B constellations (positive) emotionshave a significant influence on satisfaction and behavioraloutcome. Based on existing empirical evidence, Beeret al. (2006) suggest that negative emotions may biasjudgments and decision making, e.g., by stimulating elabo-rative processing and provoking overestimations of risk.Research in the field of neuroeconomics (e.g., Gutniket al. 2006) shows that offerings which are perceived asunfair activate a region in the brain that is in operationduring negative emotions such as pain and disgust.

The proposed model understands positive and nega-tive emotions as mediators, because it is assumed thatemotions “account for the predictor and the criterion”(Baron and Kenney 1986, p. 1176). According to Baronand Kenney (1986) the following conditions for servingas a mediator have to be fulfilled: variations in levels of theindependent variable significantly account for variationsin the presumed mediator and variations in the mediatorsignificantly account for variations in the dependent vari-able. Following the aforementioned argumentation of theimpossibility to separate cognitive from emotional pro-cesses in general and in decision making (Bechara et al.2003; Lakomski and Evers 2010), it can be argued thatemotions explain the relationship between independentand dependent variables, thus are mediators, not mod-erators.

Proposed Control Variables

In order to reflect on the complexity and multidimen-sionality of organizational buying behavior and tostrengthen the research design, several control variablesare proposed. Including them in the model allows holdingthem constant in order to assess or better clarify therelationship between the main independent and depen-dent variables. Buying situation/buyclass (1), industrytype (2) and switching barriers (3) will serve as controlvariables.

Buying situation (1) refers to the decisive situation inwhich the purchasing party is involved. The degree ofdependency, atmospheric characteristics, and satisfactionvaries significantly among the groups of buying situations(Leonidou 2004). The majority of studies measuring theinfluence of the buyclass (i.e., new task, modified rebuy,

straight rebuy) assume new tasks so to be more important(e.g., Dholakia et al. 1993), more risky (e.g., Andersonet al. 1987), more complex (e.g., Lund 1989), and associ-ated with greater uncertainty (e.g., Bunn 1993) thanstraight-rebuy situations (Jackson et al. 1984). More impor-tant, more risky, more complex and more uncertain deci-sions might affect emotional states of business buyers.

Industry type (2): This variable addresses possiblevariance of decisive factors across different industrycategories. It can be assumed that the decision criteriaapplied by business customers might fluctuate by industryand product category due to industry specific proceduresin the buyer-seller constellation (Bharadwaj 2004; Bennionand Redmond 1994), so it is necessary to control the noisethat may arise from industry specific habits.

Switching barriers (3) make it costly for the customerto terminate the relationship with an existing supplier infavor of a new one (Fornell 1992). Switching barrierscomprise search costs, transaction costs, learning costs,loss of acquired discounts, and increased risk compo-nents. Fornell (1992) argues that loyalty depends on (a)satisfaction and (b) switching barriers. Thus, switchingbarriers should be controlled for in the proposed model.

CONCLUSIONS AND LIMITATIONS

The empirical results of the qualitative investigationstrongly support the assumption that emotions are animportant determinant for professional buying decisions.Numerous positive and negative emotions have beenfound to be present during the day-by-day work of pur-chasing representatives in B2B firms. Altogether thefindings of the study have laid a solid foundation formodeling the mediating influence of emotions on B2Bpurchasing decisions. Subsequently, a conceptual modelhas been worked out and awaits empirical confirmation ordisconfirmation.

It was attempted to apply a robust mix of methods toyield sound and reliable results. However, caution has tobe exercised not to overgeneralize the findings. The studywas conducted among Austrian managers, thus it cannotbe concluded that the findings can be transferred todifferent regional and cultural environments. The concep-tual model should help to continue investigations in thisfield and to reappraise these findings with mixed-methodapproaches.

REFERENCES

Andersen, Poul H. and Rajiv Kumar (2006), “Emotions,Trust and Relationship Development in Business

Relationships: A Conceptual Model for Buyer-SellerDyads,” Industrial Marketing Management, 35 (4),522–35.

Anderson, E., W. Chu, and B. Weitz (1987), “Industrial


Purchasing: An Empirical Exploration of the BuyclassFramework,” Journal of Marketing, 51 (3), 71–86.

Anderson, J.C. and J.A. Narus (1984), “A Model of theDistributor’s Perspective of Distributor-Manufac-turer Working Relationships,” Journal of Marketing,48 (4), 62–74.

____________ and ____________ (1990), “A Model ofDistributor Firm and Manufacturer Firm WorkingPartnerships,” Journal of Marketing, 54 (1), 42–58.

Anderson, E.W., C. Fornell, and D.R. Lehmann (1994),“Customer Satisfaction, Market Share, and Profit-ability: Findings From Sweden. Journal of Market-ing, 58 (3), 53.

Ashforth, Blake E. and Ronald H. Humphrey (1995),“Emotion in the Workplace: A Reappraisal,” HumanRelations, 48 (2), 97–126.

_____________ and Glen E. Kreiner (2002), “Normaliz-ing Emotion in Organizations: Making the Extraordi-nary Seem Ordinary,” Human Resource Manage-ment Review, 12 (2), 215–35.

Athanasopoulou, P. (2009), “Relationship Quality: ACritical Literature Review and Research Agenda,”European Journal of Marketing, 43 (5/6), 583–610

Bagozzi, Richard P., Mahesh Gopinath, and Prashanth U.Nyer (1999), “The Role of Emotions in Marketing,”Journal of the Academy of Marketing Science, 27 (2),184–206.

____________ (2006), “The Role of Social and Self-conscious Emotions in the Regulation of Business-to-Business Relationships in Salesperson-CustomerInteractions,” Journal of Business & Industrial Mar-keting, 21 (7), 453–57.

Baron, R.M. and D.A. Kenny (1986), “The Moderator-Mediator Variable Distinction in Social Psychologi-cal Research: Conceptual, Strategic, and StatisticalConsiderations,” Journal of Personality and SocialPsychology, 51 (6), 1173–82.

Bechara, A., H. Damasio, and A.R. Damasio (2003),“Role of the Amygdala in Decision-Making,” Annalsof the New York Academy of Sciences, 985, TheAmygdale in Brain Function: Basic and ClinicalApproaches, 356–69.

Beer, J.S., R.T. Knight, and M. D’Esposito (2006), “Con-trolling the Integration of Emotion and Cognition,”Psychological Science (Wiley-Blackwell), 17 (5), 448–53.

Belschak, F., W. Verbeke, and R.P. Bagozzi (2006),“Coping With Sales Call Anxiety: The Role of SalePerseverance and Task Concentration Strategies,”Journal of the Academy of Marketing Science, 34 (3),403–18.

Bennion, M. L. and W.H. Redmond (1994), “ModelingCustomer Response in an Industrial CommodityMarket,” Industrial Marketing Management, 23 (5),383–92.

Bharadwaj, N. (2004), “Investigating the Decision Crite-ria Used in Electronic Components Procurement,”

Industrial Marketing Management, 33 (4), 317.Bitner, Mary Jo and Amy R. Hubbert (1994), “Encounter

Satisfaction Versus Overall Satisfaction Versus Qual-ity,” in Service Quality: New Directions in Theoryand Practice, Roland T. Rust and Richard L. Oliver,eds. New York: Sage Publications, Inc., 72–84.

Boulding, W. and A. Kirmani (1993), “A Consumer-SideExperimental Examination of Signaling Theory: DoConsumers Perceive Warranties as Signals of Qual-ity?” Journal of Consumer Research, 20 (1), 111–23.

Brotheridge, C.M. and R.T. Lee (2008), “The Emotions ofManaging: an Introduction to the Special Issue,”Journal of Managerial Psychology, 23 (2), 108–17.

Bunn, Michele D. (1993), “Taxonomy of Buying Deci-sion Approaches,” Journal of Marketing, 57 (Janu-ary), 38–56.

Cater, T. and B. ater (2010), “Product and RelationshipQuality Influence on Customer Commitment andLoyalty in B2B Manufacturing Relationships,” In-dustrial Marketing Management, 39 (8), 1321–33.

Comer, L.B. and T. Drollinger (1999), “Active EmpatheticListening and Selling Success: A Conceptual Frame-work,” Journal of Personal Selling & Sales Manage-ment, 19 (1), 15–29.

Corbin, J. and A. Strauss (2008), Basics of QualitativeResearch: Techniques and Procedures for Develop-ing Grounded Theory, 3rd Edition. Los Angeles: SagePublications Inc.

Cronin, J., Jr. and S.A. Taylor (1992), “Measuring Ser-vice Quality: A Reexamination and Extension,” Jour-nal of Marketing, 56 (3), 55–68.

Denzin, N.K. (1978), The Research Act, 2nd RevisedEdition. New York: McGraw-Hill Inc.

Dholakia, R., J.L. Johnson, A.J. Della Bitta, and N.Dholakia (1993), “Decision-Making Time in Orga-nizational Buying Behavior: An Investigation of ItsAntecedents,” Journal of the Academy of MarketingScience, 21 (4), 281–92.

Donada, C. and G. Nogatchewsky (2009), “Emotions inOutsourcing: An Empirical Study in the Hotel Indus-try,” International Journal of Hospitality Manage-ment, 28 (3), 367–73.

Edvardsson, B., M. Holmlund, and T. Strandvik (2008),“Initiation of Business Relationships in Service-domi-nant Settings,” Industrial Marketing Management,37 (3), 339–50.

Erevelles, Sunil (1998), “The Role of Affect in Market-ing,” Journal of Business Research, 42 (3), 199–215.

Flick, U. (2006), An Introduction to Qualitative Re-search, 3rd Edition. London: Sage Publications Ltd.First published as “Qualitative Forschung.” Reinbeck:Rowohlt.

Fornell, C. (1992), “A National Customer SatisfactionBarometer: The Swedish Experience,” Journal ofMarketing, 56 (1), 6–21.

George, J.M. (1998), “Salesperson Mood at Work: Impli-cations for Helping Customers,” Journal of Personal


Selling & Sales Management, 18 (3), 23–30.Gil-Saura, I., M. Frasquet-Deltoro, and A. Cervera-Taulet

(2009), “The Value of B2B Relationships,” Indus-trial Management & Data Systems, 109 (5), 593–609.

Giunipero, L., D. Dawley, and W.P. Anthony (1999),“The Impact of Tacit Knowledge on PurchasingDecisions,” Journal of Supply Chain Management,35 (1), 42–49.

Gutnik, L.A., A.F. Hakimzada, N.A. Yoskowitz, and V.L.Patel (2006), “The Role of Emotion in Decision-Making: A Cognitive Neuroeconomic ApproachTowards Understanding Sexual Risk Behavior,” Jour-nal of Biomedical Informatics, 39, 720–36

Helgesen, O., J. Ivar Håvold, and E. Nesset (2007),“Images, Satisfaction and Antecedents: Drivers ofStudent Loyalty? A Case Study of a NorwegianUniversity College,” Corporate Reputation Review,10, 38–59.

____________ and E. Nesset (2011), “Does LibQUAL+TM

Account for Student Loyalty to a University CollegeLibrary?” Quality Assurance in Education, 19 (4),413–40.

Hirschman, Elizabeth C. (1985), “Dual Consciousnessand Altered States: Implications for Consumer Re-search,” Journal of Business Research, 13 (3), 223–34.

Ho, William, Xiaowei Xu, and Prasanta K. Dey (2010),“Multi-Criteria Decision Making Approaches forSupplier Evaluation and Selection: A Literature Re-view,” European Journal of Operations Research,202 (1), 16–24.

Hook, van R.I., V. Chatham, and R. Wilding (2002),“Managers in Supply Chain Management, the Criti-cal Dimension,” Supply Chain Management: An In-ternational Journal, 7 (3), 119–25.

Jackson, Jr., D.W., J.E. Keith, and R.K. Burdick (1984),“Purchasing Agents’ Perceptions of Industrial Buy-ing Center Influence: A Situational Approach,” Jour-nal of Marketing, 48 (4), 75–83.

Jick, T.D. (1979), “Mixing Qualitative and QuantitativeMethods: Triangulation in Action,” AdministrativeScience Quarterly, 24 (4), 602–11.

Johnson, M.D., A. Gustafsson, T. Andreassen, L. Lervik,and C. Jaesung (2001), “The Evolution and Future ofNational Customer Satisfaction Index Models,” Jour-nal of Economic Psychology, 22 (2), 217.

Ka-shing, W. and C.T. Ennew (2005), “Measuring Busi-ness-to-Business Professional Service Quality andits Consequences,” Journal of Business Research, 58(9), 1178–85.

Kopelman, Shirli, Shelby R. Ashleigh, and Leigh Thomp-son (2006), “The Three Faces of Eve: StrategicDisplays of Positive, Negative, and Neutral Emo-tions in Negotiations,” Organizational Behavior &Human Decision Processes, 99 (1), 81–101.

Kuzmina, J. (2010), “Emotion’s Component of Expecta-

tions in Financial Decision Making,” Baltic Journalof Management, 5 (3), 295–306.

Lakomski, G. and C.W. Evers (2010), “Passionate Ratio-nalism: The Role of Emotion in Decision Making,”Journal of Educational Administration, 48 (4), 438–50.

Laros, Fleur J.M. and Jan-Benedict E.M. Steenkamp(2005), “Emotions in Consumer Behavior: a Hierar-chical Approach,” Journal of Business Research, 58(10), 1437–45.

Lazarus, Richard (1991), Emotion and Adaptation. NewYork: Oxford University Press.

Lee, Y., C. Lee, S. Lee, and B.J. Babin (2008),“Festivalscapes and Patrons’ Emotions, Satisfaction,and Loyalty,” Journal of Business Research, 61 (1),56–64.

Leonidou, L.C. (2004), “Industrial Manufacturer–Cus-tomer Relationships: The Discriminating Role of theBuying Situation,” Industrial Marketing Manage-ment, 33 (8), 731–42.

Lewis, M. (2000), “Self-Conscious Emotions: Embar-rassment, Pride, Shame, and Guilt,” in Handbook ofEmotions, 2nd ed., M. Lewis and J.M. Haviland-Jones, eds. New York: Guilford, 623–36.

Loewenstein, G. (2000), “Emotions in Economic Theoryand Economic Behavior: The American EconomicReview,” Papers and Proceedings of the One Hun-dred Twelfth Annual Meeting of the American Eco-nomic Association, 90 (2), (May), 426–32.

Lund, D.B. (1989), “Retail Scanner Checkout System:How Buying Committees Functioned,” IndustrialMarketing Management, 18 (3), 179–85.

Lynch, J. and L. Chernatony (2007), “Winning Hearts andMinds: Business-to-Business Branding and the Roleof the Salesperson,” Journal of Marketing Manage-ment, 23 (1/2), 123–35.

Lynch, Joanne (2004), “The Power of Emotion: BrandCommunication in Business-to-Business Markets,”Journal of Brand Management, 11 (5), 403–19.

Mano, H. and R.L. Oliver (1993), “Assessing the Dimen-sionality and Structure of the Consumption Experi-ence: Evaluation, Feeling, and Satisfaction,” Journalof Consumer Research, 20 (3), 451–66.

Mayring, P. (2008), Qualitative Inhaltsanalyse.Grundlagen und Techniken, 2nd Edition, Weinheimand Basel: Beltz.

Myeong-Gu, Seo and Lisa Barrett Feldmann (2007),“Being Emotional During Decision Making – Goodor Bad? An Empirical Investigation,” Academy ofManagement Journal, 50 (4), 923–40.

Naumann, E., M. Havirela, M.S. Khan, P. Williams (2010),“Understanding the Causes of Defection among Sat-isfied B2B Service Customers,” Journal of Market-ing Management, 26 (9/10), 878–900.

Neale, M.A. and M. Bazerman (1991), Cognition andRationality in Negotiation. New York: Free Press.

Olekalns, M. and P.L. Smith (2005), “Cognitive Repre-


sentations of Negotiation,” Australian Journal ofManagement, 30 (1), 57–76.

Oliver, R.L. (1980), “A Cognitive Model of the Anteced-ents and Consequences of Satisfaction Decisions,”Journal of Marketing Research (JMR), 17 (4), 460–69.

____________ (1999), “Whence Consumer Loyalty?”Journal of Marketing, 63 (4), 33–44.

Overbeck, J.R., M.A. Neale, and C.L. Govan (2010), “IFeel, Therefore You Act: Intrapersonal and Interper-sonal Effects of Emotion on Negotiation as a Func-tion of Social Power,” Organizational Behavior &Human Decision Processes, 112 (2), 126–39.

Parasuraman, A.A., V.A. Zeithaml, and L.L. Berry (1985),“A Conceptual Model of Service Quality and ItsImplications for Future Research,” Journal of Mar-keting, 49 (4), 41–50.

Patton, M.Q. (2002), Qualitative Research & EvaluationMethods, 3rd Ed., Thousand Oaks: Sage.

Peslak, A.R. (2005), “Emotions and Team Projects andProcesses,” Team Performance Management, 11 (7/8), 251–62.

Poels, Karolien and Siegfried Dewitte (2006), “How toCapture the Heart? Reviewing 20 Years of EmotionMeasurement in Advertising,” Journal of Advertis-ing Research, 46 (1), 18–37.

Punch, K.F. (2005), Introduction to Social Research –Quantitative and Qualitative Approaches, 2nd Edi-tion, London: Sage Publications Inc.

Raiffa, H. (1982), The Art and Science of Negotiation.Cambridge: Harvard, University Press.

Rauyruen, P. and K.E. Miller (2007), “Relationship Qual-ity as a Predictor of B2B Customer Loyalty,” Journalof Business Research, 60 (1), 21–31.

Reichard, C.J. (1985), “Industrial Selling: Beyond Priceand Persistence,” Harvard Business Review, 63 (2),127–33.

Reichheld, F.F. and W. Sasser Jr. (1990), “Zero Defec-tions: Quality Comes to Services,” Harvard BusinessReview, 68 (5), 105–11.

Schulman, P. (1999), “Applying Learned Optimism toIncrease Sales Productivity,” Journal of PersonalSelling & Sales Management, 19 (1), 31–37.

Simon, H.A. (1990), “Invariants of Human Behavior,”Annual Review of Psychology, 41(1), 1–19.

____________ (1997), “Administrative Behavior: A Study

of Decision-Making Processes in Administrative Or-ganizations, 4th Edition, New York: The Free Press.

Thibaut, J.W. H.H. Kelley (1959), The Social Psychologyof Groups. Oxford, England: Wiley.

Ulaga, W. (2003), “Capturing Value Creation in BusinessRelationships: A Customer Perspective,” IndustrialMarketing Management, 32 (8), 677.

Van Kleef, G.A., C.W. De Dreu, and A.R. Manstead(2004), “The Interpersonal Effects of Emotions inNegotiations: A Motivated Information ProcessingApproach,” Journal of Personality & Social Psy-chology, 87 (4), 510–28.

Verbeke, W. and R.P. Bagozzi (2002), “A SituationalAnalysis on How Salespeople Experience and Copewith Shame and Embarrassment,” Psychology &Marketing, 19 (9), 713–41.

Voss, G., A. Parasuraman, and D. Grewal (1998), “TheRoles of Price, Performance, and Expectations inDetermining Satisfaction in Service Exchanges,”Journal of Marketing, 62 (4), 46–61.

Wells, J.D., J.S. Valacich, and T.J. Hess (2011), “WhatSignal Are You Sending? How Website QualityInfluences Perceptions of Product Quality and Pur-chase Intentions,” MIS Quarterly, 35 (2), 373–418.

Westbrook, R.A. (1987), “Product/Consumption-BasedAffective Responses and Postpurchase Processes,”Journal of Marketing Research, 24 (3), 258–70.

Wilding, R. (1999), “The Role of Time Compression andEmotional Intelligence in Agile Supply Chains,”Supply Chain Practice, 1 (4), 14–24.

Williamson, O.E. (1981), “The Modern Corporation:Origins, Evolution, Attributes,” Journal of EconomicLiterature, 19 (4), 1537–68.

Wind, Y. (2006), “Blurring the Lines: Is There a Need toRethink Industrial Marketing?” Journal of Business& Industrial Marketing, 21 (7), 474–81.

Yeomans, L. (2007), “Emotion in Public Relations: ANeglected Phenomenon,” Journal of Communica-tion Management, 11 (3), 212–21.

Zeithaml, V.A. (1988), “Consumer Perceptions of Price,Quality, and Value: A Means-End Model and Syn-thesis of Evidence,” Journal of Marketing, 52 (3), 2–22.

____________, L.L. Berry, and A.A. Parasuraman (1996),“The Behavioral Consequences of Service Quality,”Journal of Marketing, 60 (2), 31–46.

For further information contact:Andreas Zehetner

University of Applied SciencesWehrgrabengasse 1–3

Upper AustriaPhone: +436649663481



INTERORGANIZATIONAL FAVOR EXCHANGE AS ACOOPERATIVE BEHAVIOR

Adam Nguyen, Siena College, Albany, New York

SUMMARY

This research examines a potentially importantinterorganizational (IO) cooperative behavior, namely,the exchange of preferential treatments or favors betweenboundary personnel across organizations.

Although favor exchange is an age-old concept insociology, the study of favor exchange as an IO coopera-tive behavior is still in its early stage. In the extantliterature, IO cooperative activities have been examinedprimarily as role-based behaviors driven by (economic)rationality. Within this approach, favor exchange activi-ties can at best be seen as a source of (random) deviationfrom the rational and by implication, optimal outcomes ofIO exchanges rather than a topic of study in its own right.The view of transaction cost economics (TCE) exempli-fies this approach: IO cooperation can be parsimoniouslyexplained in terms of calculativeness; favor exchange isirrelevant, and moreover is generally not viable due to theproblem of opportunism.

However, emerging views have pointed to an oppos-ing view. Social networks scholars argue that personal-based cooperation (which includes favor exchange), playsa key and positive role in IO social networks as a meansof coordinated adaptation. Consistent with this view,marketing scholars in the relational exchange traditioninclude favor doing in their measures of IO cooperation.Few of these studies have explicitly examined favorexchange as a distinct IO cooperative behavior. Theseconflicting views reflect diverse conceptualizations of thephenomenon, contradictory assumptions regarding itslogic, and have insufficient empirical substantiation. Thepresence of these gaps indicates that the field needs moretheory development works.

The objective of this research is to explicate the IOfavor exchange concept, with a view to address thesegaps. Via this endeavor, the research aims to advanceunderstanding of the performance implications of per-sonal relationships, and of the scope and logic of IOcooperation.

The research uses a multicase, multi-data sourcequalitative case research design to build an empirical-based theory that is more balanced and has wider analyti-cal generalizability than existing theories. This approachinvolved (1) examining both success and failure accountsof IO favor exchange from various IOR contexts; (2)

utilizing three data sources: (i) secondary data, (ii) ques-tionnaire, and (iii) in-depth interview. Together, the threedata sources yielded about 300 favor exchange incidentsplus practitioners’ viewpoints.

The resulting grounded theoretical frameworkdepicts favor exchange as an extra-role cooperative behav-ior versus role-based cooperation that has been the focusof the extant literature. It compares the nature and logic ofthese two types of IO cooperation, their value as a meansof adaptation, and the incentive required for mutuallybeneficial cooperation to occur. In conducting role-basedcooperative activities boundary personnel act as agentsfor their respective organizations; in exchanging favorsthey act as a friend. While role-based cooperation isgoverned by economic rationality directed toward advan-cing organizational interests, favor exchange is governedby altruistic rationality directed toward advancing per-sonal interests. The value of favor exchange as an IOcooperative mechanism is conditional upon its fit with theneed for adaptation in the specific exchange relationship.Favor exchange enhances IO exchange outcomes beyondrole-based cooperation when the IO exchange involveshighly frequent but lowly consequential adaptation thatrequires coordinated efforts. When the IO exchangerequires autonomous adaptation, favor exchange, if used,will derail exchange outcomes. Whereas IO interest align-ment alone provides sufficient incentive for role-basedcooperation, mutually beneficial favor-based cooperationalso requires organizational-individual interest alignmentin each participating organization.

The research findings extend the scope of IO coop-eration. IO cooperation has been examined primarily as arole-based concept. The fact that relational contractinginvolves the whole person requires investigation of behav-ior brought on by both role requirements and individualattributes. The inclusion of the favor exchange concept inthe study of IO cooperation helps address this overlookedaspect of IO cooperation.

The findings support an intermediate view betweenthe TCE view that IO favor exchange is irrelevant andharmful, versus the social network view that favorexchange plays a key and positive role in IO relationships.IO favor exchange is relevant, but its role is more limitedthan posited by social network scholars. For example, itmay not be the optimal cooperative mechanism when a lotis at stake; thus in most exchange situations favorexchange should play only a supplemental role. More-


over, this mode of cooperation involves a high probabilityof misuse.

The research findings are relevant for the discussionof whether and to what extent IO cooperation is governedby rationality versus heuristics and how these two logicsare related. IO favor exchange, a heuristic-based coopera-tive behavior, is a rational behavior from an individualperspective (altruistic rationality). However, given thatfavor exchange is a qua persona behavior, the rationalityof this cooperative behavior is directed, first and fore-most, toward advancing the interests of the individualsinvolved; these interests may or may not be consistentwith organizational interests. In this regard, the currentresearch adds complexities to scholarly understanding ofthe relations between the two logics of IO cooperation byshowing when and how these logics can be consistent orinconsistent with one another. More generally, the currentresearch demonstrates the value of a micro-analyticapproach to the study of IO cooperation and the impor-tance of exploring assumptions about decision makers.

The findings allow for a more detailed understandingof when and how boundary personnel’s personal relation-ship enhances or derails IO exchange outcomes. In the

extant literature, it has been shown that although friend-ship may have a positive effect on business outcomes, thiseffect is moderated downward by a conflict betweenfriendship-role and business-role expectations. The cur-rent research adds specifications to this finding by linkingbusiness relationship-friendship alignment/conflict toproper versus improper favor exchange.

The research has implications for how organizationscan manage favor exchange, or more broadly, informalcooperative activities, to their benefits. Organizationsneed to develop an alignment between organizational andindividual interest by means of appropriate economic andsocial incentive programs. Organizational-individual inte-rest alignment in both participating organizations is neededto encourage beneficial favor exchange activities betweenthe two organizations and discourage harmful ones. Inselecting and structuring IO favor exchange opportuni-ties, boundary personnel should consider whether a par-ticular favor exchange opportunity would be mutuallybeneficial from both an individual and organizationalperspective, and they should evaluate the value of favorexchange relative to alternative cooperative mechanisms.References are available upon request.

For further information contact:Adam NguyenSiena College

515 Loudon RoadAlbany, NY 12211

Phone: 518.729.5110E-Mail: [email protected]


USING DYADIC AGENT-BASED SIMULATIONS TO MODELSTRATEGIC DECISION MAKING IN

BUSINESS RELATIONSHIPS

Sebastian Forkmann, The University of Manchester, United KingdomDi Wang, The University of Manchester / Khalifa University, UAE

Stephan C. Henneberg, The University of Manchester, United KingdomPeter Naudé, The University of Manchester, United Kingdom

Alistair Sutcliffe, The University of Manchester, United Kingdom

SUMMARY

Understanding how to effectively manage in busi-ness relationships has been a central topic for scholars inthe area of business marketing (Ford et al. 2003a). Animportant aspect of this issue relates to the way managersmake decisions and choose certain strategies to affectbusiness relationships, and in particular their position inthe surrounding business network (Gadde et al. 2003;Harrison et al. 2010; Baraldi et al. 2007). Such strategizingissues are often linked to how actors understand theparticular network in which they are embedded (Holmenand Pedersen 2003). To grasp such aspects, research onsense-making in networks (e.g., using the concept ofnetwork pictures) has recently aimed at gaining insightsinto how managers perceive their surrounding businessnetwork and thereby underpins their understanding oftheir strategic options for managing in relationships aswell as choices in complex systems (Ford et al. 2003b;Henneberg et al. 2006; Ramos and Ford 2011). Accordingto Henneberg et al. (2006, p. 409),

“the notion of network pictures refers to the differentunderstanding that players have of the network. It isbased on their subjective, idiosyncratic sense-mak-ing with regard to the main constituting characteris-tics of the network in which their company is operat-ing. These perceived network pictures form the back-bone of managers’ understanding of relationships,interactions and interdependencies, and thereforeconstitute an important component of their individualdecision-making processes.”

A recent study by Corsaro et al. (2011) empiricallyestablished the connection between managerial cognitionin terms of managers’ perceptions of their surroundingbusiness network, and their subsequent propensity forchoosing specific strategic options regarding how to affectbusiness relationships. The authors found significant assoc-iations between certain network picture characteristics(i.e., different expressions of power, dynamics, broadness,and indirectness of the subjective network pictures) andpreferred networking strategies (understood as activitiesaffecting a company’s network position; Ford et al. 2003a).

However, while this research has been essential inlinking research on subjective perceptions of actors on theone hand, and managerial strategic decision making onthe other, no strategic decision with respect to anorganization’s business relationships is likely to be madein isolation of the current and anticipated relationshipperformance (Hambrick and Snow 1977). In fact, most ofthe time (potential) performance in itself is a primarydriver of strategizing decisions. Furthermore, such per-formance outcomes are invariably dependent on the rela-tionship partner’s actions, and so any consideration ofstrategic decision making needs to be seen in a dyadiccontext (Henneberg et al. 2010). It is therefore importantto expand research on network pictures in a strategycontext by incorporating other well-established drivers ofstrategy decision making, e.g., performance, and toinclude an interactive, or dyadic perspective. Further-more, according to Ford et al. (2003a, 2003b), networking(i.e., choosing and implementing a networking strategy),network pictures, and network outcomes (i.e., perfor-mance) form an important conceptually interlinked tri-angle for firms to do business and navigate in relation-ships and networks.

Therefore, the objective of our research is to bringthese three important elements together and provide abetter understanding of the interrelationships betweenmanagers’ perceptions of their surrounding business net-works, their networking choices as an outcome of theirstrategic decision making, and relationship outcomes,particularly performance. In order to capture these inter-relationships between the constructs, we employ an agent-based dyadic simulation as it allows us to combine previ-ous findings about the focal constructs, and to systemati-cally experiment and study the interaction effects amongthem. Hence, simulation methods are particularly usefulfor researchers in exploring and developing theories (Daviset al. 2007). Although agent-based simulation is aresearch technique that has received increasing attentionin the area of organization, strategy and managementresearch (e.g., Aggarwal et al. 2011; Davis et al. 2007;Lazer and Friedman 2007; Levinthal 1997; Fang et al.2010; March 1991), it is still in its infancy with respect tostudying business relationships and networks.


For the purpose of our study we develop a parsimo-nious evolutionary simulation model of a business rela-tionship that focuses on network pictures with varyingdegrees of perceived power of the focal company withinthe embedding network, as well as the networking strat-egy framework outlined in Hoffmann (2007). We derivecertain performance and power outcomes through thesimulation. The strategy framework by Hoffmann buildson the seminal work of March (1991) in organizationallearning and conceptualizes fundamental approacheswhich firms can adopt to interact in relationships, and thusmanage in their networks. We furthermore single outpower as our focal network picture variable due to itsimportance in affecting business relationships as well asnetworks (Anderson and Weitz 1989; Håkansson andFord 2002; Levinthal and March 1993; Palmatier et al.2006). We test and contrast three simulation models toinitially establish the validity of our computationalapproach, and then study step by step the interactioneffects between the focal constructs as well as the sensi-tivity of the model to key construct changes.

Our dyadic simulation approach contributes to thebusiness marketing and strategy literature in several ways.First, we introduce an agent-based simulation to the studyof business relationships and networks, and thereby dem-onstrate how simulation methods can be utilized to gaininsights into phenomena which are difficult to study withtraditional empirical research methods. Second, we con-trast different networking strategies and demonstrate thattheir success is context dependent, hence providing anextension of existing research on strategic decision mak-ing in business relationships. Finally, we demonstrate theeffects of performance and power- driven managerialdecision making on relationship success (i.e., relationshipcontinuation, relationship performance), thereby reveal-ing essential interaction effects between these two con-structs that suggest that strategic relational decisions,especially the change of an existing strategy, need to bewell justified, as volatility in the strategic direction (i.e.,changing networking choices too often over time) causesrelationships to become unstable, hence negatively affect-ing relationship performance. References are availableupon request.

For further information contact:Sebastian Forkmann

Manchester IMP Research GroupManchester Business School

The University of ManchesterBooth Street West

Manchester M15 6PBUnited Kingdom



REINSTITUTIONALIZING SURVEILLANCE THROUGH SOCIALMEDIA: HEALTHCARE MARKET CASE

Handan Vicdan, EMLYON Business School, France

SUMMARY

This research explores new systems of marketingdeveloped in healthcare through social media, and howthese systems constitute unconventional systems of inter-face that empower patients to participate in productionand distribution of medical knowledge. I offer a newperspective on this phenomenon by exploring recentdevelopments in modes of surveillance through socialmedia. I present discoveries from an empirical investiga-tion of a Medicine 2.0 community organization calledPatientsLikeMe (PLM) regarding the means through whichthese changes emerge, and how surveillance becomes atool for organizing the roles and relations of healthcareactors, thereby creating new markets.

Consumer researchers have conceptualized surveil-lance as a process through which marketers gaze upon theconsumers (Humphreys 2006) to discipline, control and,even, produce them (Zwick and Denegri-Knott 2009), orconsumers want to be gazed upon by marketers withexhibitionist and narcissistic desires as they also gazeupon marketers (Kozinets et al. 2004) or resist surveil-lance by marketers to have control over their privateinformation (Dholakia and Zwick 2001). Surveillance isconsidered as a tool of social control and discipline(Foucault 1977), digital discrimination (Lyon 2003), andcollection and analysis of information about populationsin order to govern their activities (Ericson and Haggerty2006). Others have adopted a more negotiated view ofsurveillance by conceptualizing it as an enabler (Lyon2007), a productive, and a playful practice (Albrechtslundand Dubbeld 2005; Marx 1996). While acknowledgingthese theorizations of surveillance, I argue that surveil-lance becomes an omnipresent phenomenon of everydaylife-giving way to new possibilities of organizing livesand growing rhizomatically, inclusive of multiple stake-holders, rather than only the macro institutions includingthe state and corporations (Deleuze and Guattari 1987).

Findings lead us to further develop Foucault’s con-cept of surveillance as a theoretical construct in order tosignify the complex biosocial and multidirectional con-trol. I conceptualize surveillance as a multidirectional anda biosocial phenomenon, and a social media-enabled toolthrough which new organizational roles and relationshipsare constructed multilaterally, including a multitude ofdiverse market actors. Not only state institutions but alsoother institutional forces in the healthcare market are

involved in surveillance, and begin to have a say in theway private health information is generated and distrib-uted, and roles and relations are organized. I focus onproductive capabilities of consumer surveillance in col-laboration with multiple stakeholders including market-ers and the nation state, and focus on surveillance as thenew means of production, not just consumption, of medi-cal knowledge and creation of potential markets andorganizational roles. Hence, findings deflect from previ-ous sociological models of framework on how surveil-lance works, as the tool for the totalitarian state or as thedisciplinary tool in modern society – the panopticon(Foucault 1977). Surveillance becomes all inclusive anddecentralized with a multitude of market actors, and putsroles and relations on more equal footing among market-ing entities.

I also provide an alternative way of thinking ofsurveillance by focusing on non-state institutions as itsexecutor, rather than viewing it as just an institutionalstate activity (Rose 1996), and drawing attention to its co-constructive nature (Lyon 2004, 2007). The processes oforganization, generation and distribution of health dataare also beginning to be controlled by non-state institu-tions like PLM. Community intervention to organizingbusiness relations is intriguing in that the communitycomes to serve as a new territory for and a means oforganizing relations among diverse actors and institution-alizing new allegiances and responsibilizations (Millerand Rose 2008) through non-dominating discourses (Rose2007). PLM uses responsibilization strategies to engagepatients and other healthcare actors in generation anddistribution of medical knowledge, and facilitate surveil-lance among these market entities both inside and outsideof the community. PLM strives to establish sharing as aright versus privacy (a right not to share) for the discoveryof new medical knowledge and health data sharing. Insti-tutionalizing surveillance through PLM offsets privacyissues, since PLM increases connectedness amongpatients, pharmaceuticals, state, and other healthcare pro-viders, and engages them in constant and real-time sur-veillance for improved patient care. Patient license toproduce and distribute new medical knowledge is recog-nized and encouraged in this process by both PLM andother healthcare actors.

This study also has implications for marketing theory.Scholars have conceptualized “community” as a socialphenomenon outside of the organization or separate from


the organization (Peñaloza and Venkatesh 2006). Withgradual dilution of top-down provisioning in healthcare,multilateral processes of organizing relations amonghealthcare actors through social media also indicate thatthe community is not just a means to support, celebrate, orcritique the organization – it begins to interpenetrate the(commercial and state) organization, and transforms thenexus of connections into a new quasi-organization that isto some extent free from bureaucratic and market disci-pline. The new “organization” created by PLM serves asa community comprised of multiple firms (partnershipwith research organizations, universities, pharmaceuticalfirms), healthcare providers (e.g., physicians, research-ers, caregivers), patients, family members, and PLMfounders and administrators. Furthermore, in modernmarket society, organizations were considered as distinct/detached entities from consumers in the market, provid-ing goods/services to satisfy the needs of their targetmarkets (F rat and Dholakia 2006; Peñaloza and Venkatesh2006). Yet, transformation of social networking (Web

2.0) into a business phenomenon (Tapscott and Williams2008) challenges conventional business systems that treatorganizations and consumers as distinct entities. Organi-zations now serve as systems of real-time processes,whereby performers in the market together discover anddesign their needs in actual or virtual collectivities (F ratand Dholakia 1998). Modern marketing’s response tothese transformations should then center on finding newways to engage consumers of the connected society in theproduction processes through forming communities withorganizations. Furthering the new dominant logic of mar-keting (Vargo and Lusch 2004), marketing’s focus in theWeb 2.0 and subsequent era begins to shift away from top-down provisioning, and toward acting on behalf of or forconsumers (Zwick et al. 2008); to collaboration, enablingpeople to act and empowering them to partake in commu-nities of co-creating life experiences through social me-dia, and ultimately blurring the distinctions between theconsumer and the organization. References are availableupon request.

For further information contact:Handan Vicdan

EMLYON Business School23 avenue Guy de Collogue

69132 Ecully CedexFrance



THE MEDIATING ROLE OF MARKETING PROCESS IMPROVEMENTIN THE MARKET-BASED ASSET FRAMEWORK

Anthony K. Asare, Quinnipiac University, HamdenThomas G. Brashear, University of Massachusetts, Amherst

Jing Yang, University of Massachusetts, AmherstJun Kang, Hunan University, China

SUMMARY

Over the past few decades, the source of a firm’scompetitive advantage has changed from physical assetssuch as plant and machinery to market-based assets (intan-gible assets) such as innovation, brands and supplierrelationships (Day 1994; Eisenhardt and Martin 2000;Wernerfelt 1984). As such, the contribution of market-based assets (MBA) toward the market capitalization offirms has substantially increased (Ramaswami et al. 2009).For example, firms such as Apple and Google are notvaluable because of their buildings and machinery, butrather because of their strong brands and ability to developinnovative products and services. Although MBAs havegrown in importance, it is hard to measure their valuesince they are intangible and typically not recorded on afirm’s balance sheet (Sharp 1995). To help determine thecontribution of MBA to a firm, Srivastava et al. (1999)developed a conceptual framework (MBA framework)that links MBA to performance. They assert that MBAmust be transformed and leveraged as part of an organ-ization’s processes if they are to generate economic valueto the organization. In a nutshell, a firm’s business pro-cesses should mediate the relationship between its MBAand performance. Although the literature recognizes thatMBA contribute to a firm’s performance through improv-ements to its business processes, very few studies haveempirically tested the framework (Ramaswami et al. 2009).Also, those studies focus more on a firm’s relationshipswith its customers and channel members while focusinginadequately on supplier relationships. This paper tests

the MBA framework and also extends it to a firm’s supplychain. The paper focuses on an important MBA, a firm’ssupplier relationships, and examines how a buyer firm’ssupplier development programs contribute to the buyerfirm’s performance. Supplier development refers to aprogram developed by a buyer firm to foster ongoingimprovements and upgrade its supplier’s capabilities(Krause and Handfield 2007). Supplier development pro-grams are created by buyer firms to help their suppliers,particularly their deficient ones, improve their capabili-ties and business processes (Wagner 2006). Numerousfirms including United Technologies and Toyota havedeveloped supplier development programs aimed at help-ing their suppliers.

Drawing from the MBA framework this paper assertsthat a firm’s supplier development programs contribute tothe buyer firm’s performance through its marketing pro-cesses. As such, a buyer firm’s investments in its supplierdevelopment programs will lead to improvements in its(the buyer firm’s) marketing processes which will in turnlead to better performance for the buyer. Srivastava et al.(1999) categorized supply chains as one of a firm’s threecore marketing processes which led us to believe that afirm’s supply chain activities could impact its marketingprocesses. Our results provide support for the market-based asset framework by demonstrating that a buyerfirm’s supplier development programs contribute to thebuyer’s performance through its marketing process improv-ements. References are available upon request.

For further information contact:Anthony K. Asare

MarketingQuinnipiac University

275 Mount Carmel AvenueHamden, CT 06518

Phone: 203.582.3452Fax: 203.582.8664



HOW INTEGRATION ENABLES MARKETING TO BECOMEMORE ACCOUNTABLE

Andreas Waschto, RWTH Aachen University, GermanyMalte Brettel, RWTH Aachen University, Germany

SUMMARY

Marketing departments face increasing pressure tolink the outcome of their decisions, activities, and expen-ditures to firm performance and shareholder value andincreasingly strive to assess and quantify their impact(e.g., Day and Fahey 1988; Srivastava, Shervani, andFahey 1998). Accordingly, research has evolved on per-formance metrics and financial consequences of market-ing strategy and activities (e.g., Kumar and Petersen 2005;Srivastava, Shervani, and Fahey 1999). For example,Rust et al. (2004), building on the work of Srivastava,Shervani, and Fahey (1999), outline a “chain of marketingproductivity” that ranges from tactical marketing actionsvia mostly nonfinancial outcomes to marketing’s finan-cial impact.

Marketing accountability has been identified as onesource of marketing’s credibility (e.g., Rust et al. 2004),a fact especially important as a considerable decline of theinfluence of marketing departments within firms hasrecently been noticed (e.g., Brown et al. 2005; Verhoefand Leeflang 2009). Moreover, O’Sullivan and Abela(2007) found evidence for the positive impact of market-ing performance measurement ability – a marketingdepartment’s ability to measure the performance of spe-cific marketing activities and its ability to provide acomprehensive set of detailed, meaningful marketingperformance metrics –, both on the status of the marketingdepartment as well as overall firm performance, profit-ability, and stock returns.

Despite evolving research, little is known about theantecedents of marketing accountability and marketingperformance measurement ability. The question occurswhich factors facilitate and enable marketing departmentsto actually improve their abilities to link their activities tofinancial outcomes, excel in performance measurementand provide meaningful measurement metrics. Calls fromacademia and practice request marketing departments tofurther enhance these abilities and target additionalresearch on marketing performance measurement – lay-ing promising ground for research on underlying facilita-tors (e.g., Ambler 2003; Bolton 2004; Donthu, Hersh-berger, and Osmonbekov 2005; Rust et al. 2004).

The present study contributes to marketing, innova-tion, and cross-functional integration research by linking

operational marketing-finance and marketing-R&D inte-gration to marketing accountability and marketing perfor-mance measurement ability. First, we analyze whetherclose cross-functional integration facilitates their devel-opment. Marketing-finance integration is expected to beespecially fruitful as finance departments possessadvanced skills in business analysis and financial, cost,and risk management (e.g., de Ruyter and Wetzels 2000)and integration has been revealed to facilitate cross-fertilization and mutual learning (e.g., Griffin and Hauser1996; Troy, Hirunyawipada, and Paswan 2008). We exam-ine integration in the case of new product development onan organizational-level (cf. Nakata, Zhu, and Izberg-Bilgin 2011). Second, we introduce the finance functionto this research stream that has mainly relied on themarketing-R&D relationship (e.g., Troy, Hirunyawipada,and Paswan 2008). Third, we examine the moderatingrole of uncertainty in terms of product innovativeness. Wetest our research model by means of survey data from 318companies, validated by dyadic responses.

We developed our hypotheses based on resourcedependency theory (Pfeffer and Salancik 1978) and con-siderations regarding different thought worlds (e.g., Hom-burg and Jensen 2007) of marketing, R&D, and financedepartments.

H1: Cross-functional marketing-finance integration ispositively related to (a) marketing accountability, (b)marketing’s ability to measure the performance of itsactivities, and (c) marketing’s ability to provide mar-keting performance measurement metrics.

H2: Cross-functional marketing-R&D integration is posi-tively related to (a) marketing accountability. It hasno significant effect on (b) marketing’s ability tomeasure the performance of its activities and (c)marketing’s ability to provide marketing performancemeasurement metrics.

H3: The relationship between cross-functional market-ing-finance integration and (a) marketing account-ability, (b) marketing’s ability to measure the perfor-mance of its activities, and (c) marketing’s ability toprovide marketing performance measurement metricsis stronger when product innovativeness is high ratherthan low.


H4: The relationship between cross-functional market-ing-R&D integration and (a) marketing accountabil-ity is stronger when product innovativeness is highrather than low. The relationship between cross-functional marketing-R&D integration and (b)marketing’s ability to measure the performance of itsactivities and (c) marketing’s ability to provide mar-keting performance measurement metrics is not sig-nificantly affected when product innovativeness ishigh rather than low.

We applied multiple regression analysis with interac-tion terms to test our hypotheses. Our main effects resultsreveal highly significant positive effects of marketing-finance integration on all three dependent variables, con-firming H1a, H1b, and H1c. Regarding marketing-R&Dintegration, we find support for H2a, H2b, and H2c, as asignificant positive relationship with marketing account-ability, but not with marketing performance measurementability is revealed. We further find support for the moder-

ating role of product innovativeness. As suggested, highlevels of marketing-finance integration are especiallyfruitful in highly innovative environments. However, forH3a, we find only weak support in the partial, but not thetotal interaction model. H3b and H3c are confirmed.Product innovativeness shows a positive moderating effecton the relationship between marketing-R&D integrationand marketing accountability, but not toward marketingperformance measurement ability, supporting H4a, H4b,and H4c.

Our results present evidence for the importance ofcross-functional marketing-finance and marketing-R&Dintegration in the case of new product development for theexchange of knowledge, mutual learning, and improve-ment of skills. In particular close integration with financehelps marketing to improve its accountability and perfor-mance measurement ability. This holds in particular trueunder high levels of product innovativeness, representinguncertainty. References are available upon request.

For further information contact:Andreas Waschto

RWTH Aachen UniversityKackertstr. 7

52072 AachenGermany

Phone: +49.241.80.96359Fax: +49.241.80.92371



WHY DO MANUFACTURERS ENGAGE IN PRIVATE LABELSPRODUCTION? MARKET STRATEGY AND CHANNEL

RELATIONSHIP PERSPECTIVES

Ho-Taek Yi, GSM, Sogang University, KoreaChae-Un Lim, GSM, Sogang University, Korea

ABSTRACT

This article investigates the antecedents why themanufacturers engage in retailers’ private labels programbased on market strategy and channel relationship per-spectives. Findings indicate that manufacturer’s PL pro-duction is strongly related to its marketing strategy andchannel relationship factor. Then, the authors draw impli-cations for a theoretical understanding of private labels inmarketing, as well as for the management in practice.

INTRODUCTION

Private labels, generally referred to as own labels,store brands, distributors own private-label, home brandor own label brand have also been performing strongly inevery single local market (De Wulf et al. 2005). Nowa-days, private labels account for one out of every five itemssold every day in U.S. supermarkets, drug chains, andmass merchandisers (Kumar and Steenkamp 2007), andthe market share in Western Europe is even larger(Euromonitor 2007). In the U.K., grocery market share ofprivate labels grew from 39 percent of sales in 2008 to 41percent in 2010 (Marian 2010). Planet Retail (2007, p.1)recently concluded that “[PLs] are set for acceleratedgrowth, with the majority of the world’s leading grocersincreasing their own label penetration.”

Private labels have gained wide attention both in theacademic literature and popular business press and thereis a glowing academic research to the perspective ofmanufacturers and retailers. Empirical research on privatelabels has mainly studied the factors explaining privatelabels market shares across product categories and/orretail chains (Dahr and Hoch 1997; Hoch and Banerji1993), factors influencing the private labels proneness ofconsumers (Baltas and Doyle 1998; Burton et al. 1998;Richardson et al. 1996) and factors how to react brandmanufacturers towards PLs (Dunne and Narasimhan 1999;Hoch 1996; Quelch and Harding 1996; Verhoef et al.2000). Nevertheless, empirical research on factors influ-encing the production in terms of a manufacturer-retaileris rather anecdotal than theory-based.

Goals of Paper

The objective of this paper is to bridge the gap in thesetwo types of research and explore the factors which

influence on manufacturer’s private label production basedon two different theories – market strategy and channelrelationship perspective. In order to do so, the authorsused in-depth interviews with marketing managers, re-viewed retail press and research and presents the concep-tual framework that integrates the major determinants ofprivate labels production. A survey was also conductedwith marketing managers of manufacturers in CPG (con-sumer package goods) industry.

THEORETICAL BACKGROUND ANDHYPOTHESES

Different Perspectives and Research Model

The decision of whether or not a manufacturer shouldengage in private label production is not an easy one, sinceseveral considerations come into play. The arguments forproducing private labels by manufacturers fall into twocategories: to generate additional profits and have a greaterinfluence over the category (Kumar and Steenkamp 2007).

From a manufacturer’s perspective, supplying pri-vate labels often starts on a strategic basis. When amanufacturer engages in private labels, the manufacturerdoes not have to spend on advertising, retailer promotionsor maintain a dedicated sales force. Moreover, if a manu-facturer has weak marketing capabilities, the manufac-turer can make use of retailer’s marketing capability toproduce private labels and lessen its marketing cost andincreases its profit margin.

Another possible perspective is a manufacturer toengage in private label production to manage the cat-egory. It allows the firm to enhance its relationship withthe retailer. The manufacturer demonstrates its coopera-tive behavior by supplying private label products. As PaulLuchsinger, CEO of Ontario Foods, observes “One rea-son manufacturers choose to do this is to cultivate betterrelations with retailers” (Kumar and Steenkamp 2007, p.136). On the basis of these two perspectives, the authorsput firm’s strategic view as market strategy perspectiveand relation view as channel relationship perspective (seeFigure 1).

Market Strategy Perspective

Strategy has been defined as “the match an organiza-tion makes between its internal resources and skills . . . and


the opportunities and risks created by its external environ-ment,” (Hofer and Schendel 1978, p. 12).

In strategy literature, two major perspectives domi-nate the literature offering divergent explanations withrespect to the conditions underlying firm’s performance(Caloghirou et al. 2004). The first, drawing from theIndustrial/Organizational (I/O) literature emphasizesindustry effects on performance, also called “outside-in”perspective. The second, known as the resource-basedview of the firm (RBV), which is also named “inside-out”approach, centers attention on idiosyncratic organiza-tional resources and capabilities (Caloghirou et al. 2004).

The I/O literature, which adopts an “outside-in” per-spective in terms of market structure and its effect onperformance, seeks to attribute performance variance toexternal factors. The basic model follows from the struc-ture-conduct-performance (SCP) paradigm. This para-digm was initially developed by Edward Mason duringthe 1930’s and Joe Bain during the 1950’s, which itsnature of a firm’s performance in the marketplace depend-ing on the characteristics of the environment in which itcompetes (McDermott 2003). It posits that firm perfor-

mance is determined primarily by two fundamental sets ofantecedents. First is the structural characteristics of thefirm’s markets that determine the competitive intensitythe firm faces so a firm’s behavior is essentially treated asa theoretical “black box.” The second antecedent is thefirm’s ability to achieve and sustain positional advantagesthrough the efficient and effective execution of plannedcompetitive strategy (Porter 1980; Scherer and Ross 1990).

In contrast to the I/O theory, the resource-based view(RBV) looks inside the firm for sources of superiorperformance with respect to competition. RBV empha-sizes resources as central to understanding firm perfor-mance (e.g., Amit and Shoemaker 1993; Peteraf 1993).This view is originally developed in the field of strategicmanagement, and then, the concepts have been embracedby marketing academics as potential explanations of mar-keting effects on performance and the routes to sustain-able competitive advantage (Day 1994; Hunt and Morgan1995). This perspective suggests that it is the capabilities,more than the resources, that enable the deployment andleveraging or resources that help some firms performbetter than others (Grant 1996; Teece et al. 1997).

FIGURE 1Conceptual Framework


Consumer Marketing Capabilities and Private LabelProduction

Marketing capability is defined as integrative pro-cesses designed to apply the collective knowledge, skills,and resources of the firm to the market-related needs of thebusiness. This enables the business to add value to itsgoods and services and meet competitive demands (Day1994) via its marketing mix strategy. Distinctive market-ing capabilities enable a firm to have outstanding perfor-mance against their competitors by reaching target mar-kets more effectively. Although rivals may focus onsimilar market needs, the idiosyncratic way in which eachfirm integrates knowledge creates unique and potentiallydifferent ways of solving similar customer needs (Vorhiesand Harker 2000). Based on Vorhies and Morgan’s view,we identified consumer marketing capabilities, marketingcapabilities for consumer promotion, which is the conceptof marketing capabilities, less trade promotion, volumeincentive, and slotting allowance, and firm-specific-ability to communicate directly with its target customer.As it were, this concept includes all of pull marketingrelated activities such as new product development (NPD)capabilities, market information management, pricing,marketing communication, marketing planning and imple-mentation.

In private label production, most of the companies arededicated private label manufacturers. They are exclusiveprivate labels production for a retailer. Such companieshave several characteristics in common (Kumar andSteenkamp 2007). Most of these companies are small tomedium-sized enterprises, specializing in a few productcategories. They are high volume-low margin producersand pursue volume at almost any cost in order to fillcapacity. Their own research and development only focuson spotting, copying, and anticipating new product intro-ductions by big national brand manufacturers. Conse-quently, these companies have lack of product develop-ment, flexibility and marketing support. One of reasonsthe smaller manufacturer produce private label is to lever-age with retailer’s marketing capabilities to enhance itsprofit margin (Kumar and Steenkamp 2007).

Except for the small to medium-sized manufacturers,some of national brand manufacturers produce privatelabel is because they have surplus capacity due to atemporary imbalance between supply and demand. Aprivate label order can be used to fill the spare capacity. Insuch case, retailers sell private labels with their ownmarketing capabilities so any contribution over and abovethe variable costs of production is incremental profit to amanufacturer. In the short run, this makes good businesssense. In market strategy perspective, manufacturer’smarketing capabilities are critical to engage in privatelabels production. Therefore, it can be hypothesized that,

H1: The greater the level of manufacturer’s consumermarketing capabilities, the lower its private labelsproduction engagement.

Channel Relationship Perspective

The marketing literature is replete with perspectivesthat shed light on the underlying characteristics of manu-facturer-retailer’s exchange relationships. Among thesetheories/perspectives, the authors adopted the transactioncost theory and relational contract theory as underlyingtheoretical framework for this study. This is appropriatebecause the variety of trading relationships and gover-nance structures outlined earlier supports the continuumof governance structures proposed by Joskow (1985) andthe market, hybrid, and hierarchical forms as outlined byWilliamson (1985, 1991). The transaction cost perspec-tive continues to be usefully applied in a wide variety ofbusiness-to-business and relationship marketing investi-gations (see Rindfleisch and Heide 1997 for a review).

Central to the transaction cost perspective is theconcept of discriminating alignment, where “firms chosethe organizational arrangements to minimize the expectedcosts of governing the transaction over the life of therelationship” (Masten et al. 1991, p. 2). These transactioncosts emerge because of the behavioral assumptions ofbounded rationality and opportunism, coupled with cer-tain transaction characteristics. These behavioral charac-teristics by themselves would not impose transactioncosts without a number of transaction characteristics,namely asset specificity, uncertainty, and frequency oftransaction (Collins and Burt 2006).

According to relational contract theory, MacNeil(1980) differentiates discrete transactions from relationalcontracts, relational exchange, along several key dimen-sions. Most important is the fact that relational exchangetranspires over time; each transaction must be viewed interms of its history and its anticipated future. The basis tofuture collaboration may be supported by implicit andexplicit as assumptions, trust and planning. In this view,a channel member’s perception of its partner’s fairness isexpected to enhance channel relationship quality. Dwyeret al. (1987) regard fairness in interactions as necessaryfor developing trust between business partners. Andersonand Weitz (1989) observe that suppliers with a reputationfor fairness engender greater trust and expectation ofcontinuity.

Channel Bonding and Private Label Production

The concept of relationship is very important in anykinds of commerce and it can likewise to manufacturersengaging in private label production. Sometimes,manufacturer’s channel bonding, creating durable rela-


tionships with a focal retailer, is the motive of private labelproduction, although the manufacturer may have enoughmarketing capabilities not to depend on retailer’s one. Forexample, Bausch & Lomb, Birds Eye, Del Monte, andHeinz engage in private labels production in spite of highlevel of marketing capabilities.

Successful collaboration with retailers requires ahigh level of purposeful cooperation aimed at maintaininga trading relationship over time (Frazier et al. 1988;Spekman 1988) and the production of private labels canimprove relationships with the retailer and helps tostrengthen its relationship. The hope is that because themanufacturer is also a private label supplier to the retailer,the retailer will reward it with more favorable shelfallocation for the manufacturer’s own brands, joint pro-motions, and so on. The brand manufacturer also learnsmore about the needs and behaviors of private labelbuyers, which, after all, constitute an important segmentin many markets. Proponents of private label productionalso argue that engaging in this activity increases thebrand manufacturer’s channel bonding to the retailer.

H2: The greater the level of manufacturer’s channelbonding, the higher its private labels productionengagement.

Precursors of Private Labels Production

Using structure-conduct-performance paradigm andresource-based view, the authors can identify three differ-ent levels of precursors of private labels production –corporate level (firm’s market position and pull marketinginvestment), brand level (manufacturer’s NB reputation),and product level (product portfolio). In addition, theauthors place manufacturer’s asset-specificity, marketuncertainty and perception of trade unfairness as precur-sors of private label production using transaction costtheory and relational contract theory. Based on thesetheory and previous researches, we offer the followinghypotheses. We will explain each of precursors in detailslater.

H3: The higher the level of manufacturer’s marketposition, the greater its consumer marketing capa-bilities.

H4: The higher the level of manufacturer’s pull market-ing investment, the greater its consumer marketingcapabilities.

H5: The higher the level of manufacturer’s nationalbrand reputation, the greater its consumer market-ing capabilities.

H6: The higher the level of manufacturer’s productportfolio, the greater its consumer marketing capa-bilities.

H7: The higher the level of manufacturer’s asset-speci-ficity, the greater its channel bonding to the retailer.

H8: The higher the level of demand volatility, the lowermanufacturer’s channel bonding to the retailer.

H9: The higher the level of market diversity, the lowermanufacturer’s channel bonding to the retailer.

H10: The higher the level of manufacturer’s perceivedtrade fairness, the greater its channel bonding to theretailer.

Control Variable

We collected data on control variables such asmanufacturer’s sales volume and manufacturer’s depen-dence on a retailer, which factors might be potentiallyinfluenced to the ratio of private label production.

DATA ANALYSIS AND RESULTS

Data Collection

The sample consisted of 209 responses in 191 con-sumer package goods manufacturers in South Korea.Questionnaire is administered based on previous researchand in-depth interview. In cooperation with Korea Cham-ber of Commerce, we can get a list of PL manufacturersand survey is performed both on and offline question-naires. All the data is analyzed to ensure that there is nosystematic error emerging from the use of both on andoffline questionnaires. Slightly over half of the respon-dents were males (56.9%) and average age of respondentswas less than 37 years old. Respondents’ business expe-rience was fairly distributed across two year increments,with those having one to three years of experience beingthe largest group (29.7%), and those with seven to nineyears being the smallest (6.7%). Nearly three-fifths (69.2%)of respondents belonged to food and beverage industry,and rest of respondents are working at consumer productand cosmetic companies.

Measurement

Attributed measures are used to measure most vari-ables using Likert-type scales and previously developedand well-established scales were adapted as measures ofconstructs (Bordley 2003; Collins and Burt 2006; Day1994; Ganesan 1994; Kim 2001; Kim et al. 2011; Klein


et al. 1990; Kumar et al. 1995; Miller 1987; Vorhies andMorgan 2005). Then these measures were modified forthe consumer package goods industry. A pretest will beconducted to refine the research instrument and ensurethat the construct measures displayed the required degreeof reliability prior to the main study.

Data Analysis

To assess each construct’s reliability and validity,Cronbach’s alpha test and exploratory factor analysisusing principal components with varimax rotation areundertaken, and then, structural equation modeling (SEM)is conducted to test the proposed structural equationmodel.

The results of reliability test as show that Cronbach’salpha of all variables exceeds 0.7, which is generallyjudged to be satisfactory (Churchill 1979). An explor-atory factor analysis using principal components withvarimax rotation was undertaken to examine the uni-dimensionality/convergent validity of each predefinedmulti-item construct (Nunnally 1978). Consistent withAnderson and Gerbing (1988), all constructs are evalu-ated using confirmatory factor analysis (CFA), andrequirements for CR and AVE are met, as each multiple-item scale exceeded the recommended cut-off criterion of0.60 and 0.50, respectively (see Table 1). CFA revealed anadequate model fit for the proposed measurement model(χ2 = 623.335, degree of freedom = 423, χ2/df = 1. 74, p =

.000; ROSEA = .048, CFI = 0.959, NFI = 0.884, IFI =0.960, TLI = 0.952).

Structural equation modeling, analyzed using AMOS18.0, was conducted to test the proposed structural equa-tion model. As Table 2 shows, the results of structuralmodel indicated fit indices: χ2 = 932.551, degree of free-dom = 600, χ2/df = 1.554, p = .000; RMSEA = 0.052,CFI = 0.938, NFI = 0.844, IFI = 0.938, TLI = 0.931. Theadequacy of the structural equation model was evaluatedon the criteria of overall fit with the data.

As depicted in Table 2, most hypotheses are sup-ported. As expected, consumer marketing capabilities arenegatively related (p < .01) to private label production,thus confirming H1. The three hypotheses (H4, H5, H6)related to the effects of marketing capabilities are sup-ported. Unexpectedly, firm’s market position (H3) had nosignificant effect (0.024) to marketing capabilities (H3 isrejected). Generally, most of respondent companies werequite small so it seems that the respondents are not familiarwith their market share and competitive position. Anotherpotential explanation is the firm’s market position and thatit matters little to engage in private labels production. Aspreviously mentioned, even high market position compa-nies like Bausch & Lomb, Del Monte, and Heinz doengage in private labeling.

Conversely to H1, channel bonding is positivelyrelated (p < .001) to private label production (H2 is

TABLE 1The Results of Confirmatory Factor Analysis

Cronbach’s St. Factor CompositeConstruct ααααα Loading Reliability AVE C.R.

Channel Bonding .875 .710~.976 .919 .807 15.245***

Demand Volatility .790 .775~1.115 .842 .807 11.535***

Marketing Capabilities .932 .764~.906 .911 .712 13.758***

NB Reputation .883 .802~.956 .907 .746 16.167***

Trade Fairness Manufacturer’s .814 .731~1.239 .898 .809 10.398***

Asset-specificity .956 .707~.915 .901 .729 15.168***

Market Diversity Manufacturer’s .833 .754~1.057 .897 .763 12.429***

Dependence on a Retailer .828 .726~.944 .832 .652 10.338***

Model Fit: χ2 = 623.335, degree of freedom = 423, χ2/df = 1.474, p = .000; RMSEA = .048, CFI = 0.959, NFI =0.884, IFI = 0.960, TLI = 0.952 *** p < .01 ;** p < .05


supported). As anticipated, relationship factors(manufacturer’s asset-specificity, and trade fairness) arepositively related to channel bonding, thus supporting H7and H10. Demand volatility is negatively related to chan-nel bonding (p < .05, H8 is supported). However, marketdiversity had no significant effect to channel bonding andeven positively related to (H9 is rejected). Perhaps, manymanufacturer-retailer relationships are highly asymmetri-cal, especially those smaller retailers that are affiliatedwith larger, more powerful manufacturer. In this situa-tion, although market diversity is getting higher, manu-facturers are more likely to rely on developing relation-ships with a focal retailer than to diversify their channelpartners (Oh et al. 1992).

Common Method Bias

Common method bias has been attracting increasedattention in structural equation modeling studies. Weconducted Harmon’s single factor test (Podsakoff et al.2003) to examine the threat of common method bias. Thefirst factor from the explanatory factor analysis explained

23.555 percent of total variance, which is not large enoughto generate concern about common method bias.

DISCUSSION

Managerial Implications

The private label phenomenon has received growingattention by marketing scholars. In many industries, pri-vate labels represent formidable competition to manufac-turer brands and manufacturers have a dilemma withselling to as well as competing with their retailers. Thisresearch may be the first study to investigate the reasonsmanufacturers engage in private labels based on twotheoretic views, market strategy and channel relationshipperspectives.

For manufacturers, this study shows key factorswhen manufacturers consider engaging in private labelproduction. From strategic management perspective,higher degrees of national brand reputation, pull market-ing investment, and product portfolio diversity have posi-

TABLE 2The Results of SEM for Test of Hypothesis

Dependent Std.H Predictor Variable Estimate S.E. C.R Results

H1 Marketing Capabilities PL Production -.401*** .094 -2.694 Supported

H2 Channel Bonding PL Production .987*** .283 3.985 Supported

H3 Firm’s Market Position Marketing Capabilities .024 .085 .443 Not Supported

H4 Pull Marketing Marketing Capabilities .443*** .120 6.040 SupportedInvestment

H5 NB Reputation Marketing Capabilities .245*** .100 4.071 Supported

H6 Product Portfolio Marketing Capabilities .176*** .109 2.620 Supported

H7 Manufacturer’s Asset- Channel Bonding .359*** .038 6.001 SupportedSpecificity

H8 Demand Volatility Channel Bonding -.157** .113 -2.003 Supported

H9 Market Diversity Channel Bonding .240 .219 1.603 Not Supported

H10 Trade Fairness Channel Bonding .280*** .100 4.651 Supported

Model Fit: χ2 = 932.551, degree of freedom = 600, χ2/df = 1.554, probability level =.000; RMSEA = 0.048,CFI = 0.938, NFI = 0.844, IFI = 0.938, TLI = 0.931 *** p < .01 ;** p < .05


tive relationship to firm’s consumer marketing capabili-ties. In addition, higher level of marketing capabilities hasgenerated negative relationship to private labeling en-gagement. In marketing field, dominant brands’ compa-nies which have higher level of marketing assets andcapabilities such as Coca-Cola, Heineken, Kellogg, Procter& Gamble, Gillette, and Nestle (in coffee) generally, donot engage in private label production (Kumar andSteenkamp 2007).

However, in spite of having higher brand position,manufacturers often make a decision to produce privatelabels due to the relationship with a retailer. According tochannel relationship perspective, higher level of asset-specificity and trade fairness is positively related to chan-nel bonding. Demand volatility has negative relationshipto channel bonding. In addition, higher level of channelbonding has positive relationship to private labeling en-gagement. Therefore, in spite of having higher marketingcapabilities, a manufacturer may engage in private labelproduction because maintaining a long-term relationshipwith a focal retailer is strategically more valuable andimportant to the manufacturer.

Then, which perspective is more convincing to theexplanation of private label production? In the results ofthis empirical research, channel bonding (ß = .987, p <.01) is more important factor than consumer marketingcapabilities (ß = -.401, p < .01) in consumer packagegoods industry in South Korea. The results show us thereason why manufacturers are engaged in private labelproduction. It is for the purpose of channel bonding withfocal retailers and less for consumer marketing capa-bilities.

Theoretical Implications

This is the first research to apply two different per-spectives to private labels production. We adopted firm’smarket strategy perspective including SCP paradigm andresources-based view and channel relationship perspec-

tive such as transaction cost analysis and relational con-tract theory, mainly using the variables/constructs of eachof theories, as underlying theoretical framework in thisstudy. These two perspectives are useful to explain a widevariety of business-to-business and relationship market-ing investigations. However, previous researches havelittle data to apply these theories to adoption of privatelabel production and are rather anecdotal. This paper alsoextends the scope of application of these two theories.

Theorists and researchers suggest that trade fairnessis important in developing effective marketing channelrelationships (Anderson and Weitz 1989; Dwyer et al.1987; Frazier 1983) even in a highly asymmetrical rela-tionships. If the vulnerable party is treated fairly by itsmore powerful partner, trust and commitment can bedeveloped. In this research, trade fairness (ß = .466, p <.01) is identified as the most important variable to buildingchannel bonding among other relationship factors.

Limitations and Future Research

This study also has some limitations. First, althoughwe considered various factors that influence private labelproduction, there are likely to be other precursors of thisconstruct. As such, the current model is underspecified, asin most research. Future research could examine differentvariables from different perspectives. Second, this studyis based upon limited field observation and sample size.Our sample is only composed of consumer package goodsindustry and the sample size is relatively small. Additionalresearch could attempt to collect a larger data set. Finally,this study was conducted only within the consumer pack-age goods industry, specifically, within the food andbeverage and consumer products. Whether the findingscan be generalized against other industry sectors or tocompletely different industries remains an empirical ques-tion. Therefore, subsequent research should utilize samplesof firms from other industries or other marketing channelsto discern whether or not our findings can be generalizedor if there are specific cross-industry differences.

ACKNOWLEDGEMENT

This research was supported by World Class Universityprogram funded by the Ministry of Education, Sci-ence and Technology through the National ResearchFoundation of Korea (R31–20002).

REFERENCES

Amit, R. and P.J.H. Schoemarker (1993), “Strategic Assetsand Organizational Rent,” Strategic ManagementJournal, 14 (1), 33–46.

Anderson. C.J. and W.D. Garbing (1988), “Structural

Equation Modeling in Practice: A Review and Rec-ommended Two Step Approach,” PsychologicalBulletin, 103 (3), 411–23.

Anderson, E. and B.A. Weitz (1989), “Determinants ofContinuity in Conventional Industrial ChannelDyads,” Marketing Science, 8 (4), 310–23.

Baltas, G. and P. Doyle (1998), “An Empirical Analysis ofPrivate Brand Demand Recognizing HeterogeneousPreferences and Choice Dynamics,” Journal of theOperational Research Society, 49 (8), 790–98.

Bordley, R. (2003), “Determining the Appropriate Depthand Breadth of a Firm’s Product Portfolio,” Journalof Marketing Research, 40 (1), 39–53.


Burton, S., D.R. Lichtenstein, R.G. Netemeyer, and J.A.Garretson (1998), “A Scale for Measuring AttitudeToward Private Label Products and an Examinationof Its Psychological Behavioral correlates,” Journalof the Academy of Marketing Science, 26 (4), 293–306.

Caloghirou, Y., A. Protogerou, Y. Spanos, and L. Para-giannakis (2004), “Industry versus Firm-SpecificEffects on Performance: Contrasting SMEs and Large-Sized Firms,” European Management Journal, 22(2), 231–43.

Churchill, Jr., G.A. (1979), “A Paradigm for DevelopingBetter Measures of Marketing Constructs,” Journalof Marketing Research, 16 (1), 64–73.

Collins, A. and S. Burt (2006), “Private Brands, Gover-nance, and Relational Exchange Within Retailer-Manufacturer Relationships: Evidence from IrishFood Manufacturers Supplying the Irish and BritishGrocery Markets,” Agribusiness, 22 (1), 1–20.

Day, G.S. (1994), “The Capabilities of Market-DrivenOrganizations,” Journal of Marketing, 58 (4), 37–52.

De Wulf, K., G. Odekerken-Schröder, F. Goedertier, andG. Van Ossel (2005), “Consumer Perception of StoreBrands versus National Brands,” Journal of Con-sumer Marketing, 22 (4), 223–32

Dhar, S. and S. Hoch (1997), “Why Store Brand Penetra-tion Varies by Retailer,” Marketing Science, 16 (3),208–27.

Dunne, D. and C. Narasimhan (1999), “The New Appealof Private Labels,” Harvard Business Review, 77 (3),41–52.

Dwyer, F.R., P.H. Schurr, and S.J. Oh (1987), “Develop-ing Buyer-Seller Relationships,” Journal of Market-ing, 51 (2), 11–27.

Euromonitor (2007), “Private Label-Potential in a Weak-ening Economy,” [http://www.portal.euromonitor.com/Portal/ResultsList.aspx].

Frazier, G.L. (1983), “On the Measurement of InterfirmPower in Channels of Distribution,” Journal of Mar-keting Research, 20 (2), 158–66.

____________, R.E. Spekman, and R. O’Neal (1988),“Just-In-Time Exchange Relationships in IndustrialMarkets,” Journal of Marketing, 52 (4), 52–67.

Ganesan, S. (1994), “Determinants of Long-Term Orien-tation in Buyer-Seller Relationship,” Journal of Mar-keting, 58 (2), 1–19.

Grant, R.M. (1991), “The Resource-based Theory ofCompetitive Advantage,” California ManagementReview, 33 (1), 114–35.

Hoch, S.J. and S. Banerji (1993), “When Do PrivateLabels Succeed?” Sloan Management Review, 34(4), 57–67.

____________ (1996), “How Should National BrandsThink about Private Labels,” Sloan ManagementReview, 37 (2), 89–102.

Hofer, C.W. and D. Schendel (1978), Strategy Formula-tion: Analytical Concepts. St. Paul: West Publication

Co.Hunt, S.D. and R.M. Morgan (1995), “The Comparative

Advantage Theory of Competition,” Journal of Mar-keting, 59 (2), 1–15.

Joskow, P.L. (1985), “Vertical Integration and Long-Term Contracts: The Case of Coal Burning ElectricGenerating Plants,” Journal of Law, Economics andOrganization, 1 (1), 33–80.

Kim, K.S. (2001), “On the Effects of Customer Condi-tions on Distributor Commitment and Supplier Com-mitment in Industrial Channel of Distribution,” Jour-nal of Business Research, 51 (2), 87–99.

Kim, S.K., R.G. McFarland, S.G. Kwon, S.G. Son, andD.A. Griffith (2011), “Understanding GovernanceDecisions in a Partially Integration Channel: A Con-tingent Alignment Framework,” Journal of Market-ing Research, 48 (3), 603–16.

Klein, S., G.L. Frazier, and V.J. Roth (1990), “A Transac-tion Cost Analysis Model of Channel Integration inInternational Markets,” Journal of MarketingResearch, 27 (2), 196–208.

Kumar, N., L. Scheer, and J.E.M. Steenkamp (1995),“The Effects of Supplier Fairness on VulnerableResellers,” Journal of Marketing Research, 32 (1),54–65.

____________ and J.E.M. Steenkamp (2007), PrivateLabel Strategy. Cambridge, MA: Harvard BusinessSchool Press.

MacNeil, I.R. (1980), The New Social Contract. NewHaven, CT: Yale University Press.

Marian, P. (2010), “Talking shop: UK Grocers Focus onPrivate-Label Innovation,” [http://www.justfood.com/analysis/uk-grocers-focus-on-private-label-innovationid112556.aspx].

Masten, S.E., J.W. Meehan, and E.A. Snyder (1991),“The Costs of Organization,” Journal of Law, Eco-nomics and Organization, 7 (1), 1–25.

McDermott, M.A. (2003), An Empirical Investigation ofCore Competence and Firm Performance, Ph D.Thesis, State University of New York, Albany.

Miller, D. (1987), “The Structural and EnvironmentalCorrelates of Business Strategy,” Strategic Manage-ment Journal, 8 (1), 55–76.

Nunnally, J.C. (1978), Psychometric Theory, 2nd ed. NewYork: McGraw-Hill Book Company.

Peteraf, M.A. (1993), “The Cornerstones of CompetitiveAdvantage: A Resource-Based View,” Strategic Man-agement Journal, 14 (3), 179–91.

Planet Retail (2007), “Private Label Trends Worldwide,2007,” (accessed to June 23, 2008), [http://www.planetretail.net/Onlineshop/RetailShopHome.aspx?Section=2&Page ID=6&ProfileTypeID=1&tocProductType=3&tocCatalogueID=45164].

Porter, M.E. (1980), Competitive Strategy. New York:Free Press.

Quelch, J.A. and D. Harding (1996), “Brand versus Pri-vate Labels: Fighting to Win,” Harvard Business


Review, 74 (1), 99–109.Richardson, P.S., A.K. Jain, and A. Dick (1996), “House-

hold Store Brand Proneness: A Framework,” Journalof Retailing, 72 (2), 159–85.

Rindfleisch, A. and J.B. Heide (1997), “Transaction CostAnalysis: Past, Present, and Future Applications,”Journal of Marketing, 61 (4), 30–54.

Scherer, F.M. and D. Ross (1990), “Industrial MarketStructure and Economic Performance,” University ofIllinois at Urbana-Champaign & apos;s Academyfor Entrepreneurial Leadership Historical ResearchReference in Entrepreneurship.

Spekman, R.E. (1988), “Strategic Supplier Selection:Understanding Long-Term Buyer Relationships,”Business Horizons, 31 (4), 75–81.

Teece, D.J., G. Pisano, and A. Shuen (1997), “DynamicCapabilities and Strategic Management,” Strategic

Management Journal, 18 (7), 509–33.Verhoef, P.C., E.J. Nijssen, and L.M. Sloot (2000), “Stra-

tegic Reactions of National Brand ManufacturersTowards Private Labels: An Empirical Study in theNetherlands,” European Journal of Marketing, 36(11/12), 1309–26.

Vorhies, D.W. and M. Harker (2000), “The Capabilitiesand Performance Advantages of Market-DrivenFirms: An Empirical Investigation,” Australian Jour-nal of Management, 25 (2), 145–71.

Williamson, O.E. (1985), The Economic Institutions ofCapitalism. New York: The Free Press.

____________ (1991), “Comparative Economic Organi-zation: The Analysis of Discrete Structural Alterna-tives,” Administrative Science Quarterly, 36, 269–96.

For further information contact:Ho-Taek Yi

Department of Global Service ManagementWCU Program

Sogang UniversitySinsudong #1, MapoguSeoul, Korea 121–742

Phone: 82.10.9148.0530Fax: 99.10.705.8519



QUANTIFYING THE SALES IMPACT OF LOCATION-TARGETEDMOBILE ADS

Xueming Luo, University of Texas at ArlingtonZheng Fang, Sichuan University, China

Megan E. Keith, University of Mississippi

SUMMARY

Recent developments in mobile communication andgeo-positioning technologies present marketers with aradical new media channel: location-targeted mobile ads(LTA). LTA involves the provision of ad messages tocellular subscribers based on their geographic locations.Many companies are beginning to incorporate mobilechannels into their marketing strategies. For example,P.F. Chang’s China Bistro successfully launched geo-targeted mobile marketing efforts by offering free appe-tizers to customers who checked in via location-basedapplications (apps) such as Foursquare, Yelp, andFacebook. It is popular in Asia and Europe for movietheaters to send specific movie ads and promotions tomobile users when these potential customers are close tothe theater geographic locations.

According to the 2011 Wharton Customer AnalyticsInitiative on mobile customer behavior, cellular devicespossess unique characteristics such as portability andpersonal nature unavailable in traditional and electronicmedia. LTA opens up an innovative conduit to deliver adsand coupons that are customized to an individual’s tastes,geographical location, and time of day. With LTA, adver-tisers could deliver ad messages via short message service(SMS) contextually through the media on a geographi-cally targeted basis and reach consumers when and wherethey are most likely to purchase. According to a survey byJiWire, 78 percent of U.S. consumers use location-basedapps on their phone and a growing 17 percent have alreadymade a purchase in response to a LTA. No wonder thatindustry analysts boldly predict that LTA messages are 20times more effective than online ads and create five to tentimes higher click-through rates compared to Internet admessages.

However, extant marketing literature lacks researchon the sales impact of LTA. In benchmarking whether ornot any marketing spending is paying off, one needs tounderstand the corresponding returns in sales and profits.Although LTA seemingly offers practitioners tremen-dous potential given the ubiquitous nature of mobiledevices, we have very little empirical evidence about itsimmediate and cumulative effectiveness. The goal of ourresearch is to fill this gap. In particular, we seek answers

to the following questions: How effective is LTA ingenerating mobile sales? What are the dynamics of theeffectiveness of LTA in terms of the short-term, long-term, and decay duration time patterns? Are there asym-metric sales impacts across various user groups? Howdifferent are the effects of LTA between the high-interestand low-interest segments?

This study makes several contributions to the litera-ture. First, it adds to the growing studies on mobilemarketing. Prior research has examined the personaliza-tion of mobile music via an adaptive system, the intensionto use mobile services such as SMS, payment, and gam-ing, and the social networking effects of using mobilemarketing in general. None of prior studies have investi-gated the sales impact of LTA, perhaps because of the lackof large-scale company data. We employ a real-worldcompany archival dataset with more than three millionwireless users to quantify the sales impact of LTA.

Second, our work advances the advertising account-ability literature (Luo and Donthu 2001, 2006). Withoutquantifying the sales impact of LTA, marketers wouldquestion the return on mobile marketing and location-based targeting tools. Media and brand managers arepressured to demonstrate advertising accountability andmobile ad value. This pressure is increasing as “theperceived lack of accountability has underminedmarketing’s credibility, threatened marketing’s standingin the firm, and even threatened marketing’s existence asa distinct capability within the firm” (Rust, Lemon, andZeithaml 2004, p. 76). Indeed, because most marketingexecutives today demand a “more scientific approach tohelp defend marketing strategies from CFOs” (Luo andDonthu 2006, p. 70), it is important to link new advertisingplatforms such as LTA to sales impact and mobile buying.We not only quantify the immediate short-term salesimpact of LTA but also track the cumulative effectivenessover time. This would allow managers to dynamicallymonitor returns to LTA and make adjustment decisionsover time.

Third, our research contributes to the informationsystems literature on mobile technology applications.Prior studies have focused on the adoption of mobile dataservices (Hong and Tam 2006), usage of mobile web


browsing, and the substitution or complement effectsbetween mobile voice service and SMS. While mostinformation systems literature depends on the technologyacceptance model (TAM), our study complements it withtargeting theory (Goldfarb and Tucker 2011) and hedonic

value of LTA. We uncover asymmetric effects of LTAacross different user segments for the same technology ofLTA. Thus, managers should practice different targetingstrategies to reflect the heterogeneity among technologyusers.


Eunice & James L. West Distinguished Professor of MarketingThe University of Texas at Arlington

701 S West StArlington, TX 76019Phone: 817.272.2279

E-Mail: [email protected]://www3.uta.edu/faculty/luoxm/


HOW ADVERTISING WORKS EMBEDDED IN NEW MEDIA:CONSUMER MEDIA EXPERIENCE MODEL

Mark Yi-Cheon Yim, Canisius College, BuffaloMinette E. Drumwright, The University of Texas at AustinVincent J. Cicchirillo, The University of Texas at Austin

SUMMARY

Scholars and advertisers have long been challengedto explain how advertising works. Reflecting the diffi-culty of reaching a conclusive answer, many attemptshave been made from diverse perspectives. Prior tradi-tional, integrative advertising models focused on themessage recipients’ characteristics and information pro-cesses (e.g., MacInnis and Jaworski 1989; Greenwald andLeavitt 1984), but prompted by the changing media envi-ronment, recent advertising models have tended to paymore attention to advertising effectiveness within thecontext of media. Each medium with its different func-tional features provides users with a different mediaexperience, and therefore, each medium may result indifferent advertising effects (Bronner and Neijens 2006).Thus, some recent models have included media character-istics in understanding how advertising works. However,the recent advertising models have some limitations inthat they have been somewhat narrowly applied to aspecific medium, the Internet. Furthermore, they haveemphasized the functional media feature of interactivity,generally suggesting that highly interactive media elicitmore positive consumer evaluations of advertising (e.g.,Cho 1999), and they have ignored other media featuressuch as vividness. Given the rapidly changing mediaenvironment that extends beyond the Internet, anothermodel with a focus on the general media experience,including both interactivity and vividness, is needed toassess advertising effectiveness more appropriately.

The current study examines several key componentsthat may significantly affect consumer media experienceswithin the context of emerging new media, and it proposesan integrative model explaining how diverse new mediaexperiences are associated with consumer responses. Themodel assumes that the consumer media experience isconstructed based on four primary factors: media novelty,attention, presence, and irritation. Specifically, our pro-posed model highlights the prominent roles of medianovelty and presence in enhancing advertising effective-ness in two innovative media – one that emphasizes vivid-ness (e.g., stereoscopic 3-D display) and another thatemphasizes interactivity (e.g., augmented reality). Thenovelty effect, created by the newness of the medium, hasthe power to attract viewers’ attention, and the increased

attention enhances their sense of presence, the experienceof being plunged into a new virtual world that advertisersconstructed. These sequential relationships result in posi-tive measures of advertising effectiveness, such as impro-ved product knowledge and increased enjoyment, andultimately more favorable attitudes toward the ad. Also,the proposed model considers how irritating factors ineach medium hinder ad viewers by reducing their atten-tion to the ad and their enjoyment of it.

The proposed advertising model was tested withinthe context of stereoscopic 3-D (Study 1) and augmentedreality (Study 2). Both studies consistently identifiedsignificant relationships among media novelty, attention,presence, enjoyment, informativeness, attitude towardadvertising, irritation, and attention. Specifically, the find-ings highlighted the prominent roles of media novelty andpresence in enhancing advertising effectiveness. The nov-elty effect, created by the newness of the medium, had thepower to attract viewers’ attention, and the increasedattention enhanced viewers’ sense of presence, which isthe experience of being plunged into a new, virtual worldthat advertisers constructed (Kim and Biocca 1997; Witmerand Singer 1998). These sequential relationships resultedin positive measures of advertising effectiveness, such asimproved product knowledge and increased enjoyment.In addition, these relationships resulted in more favorableattitudes toward the ad. Also, our findings partially sup-ported that an irritation effect in new media can hinder adviewers in processing advertising by reducing their atten-tion to the ad and their enjoyment of it. However, asignificant relationship between previous media experi-ence and irritation was not detected in either study. It isparticularly noteworthy that these relationships were foundin two very different, new media that rely on differentaspects of presence – stereoscopic 3-D, which emphasizesvividness, and augmented reality, which emphasizesinteractivity. In summary, the current study proposed andtested a new advertising model in two very different,innovative new media – stereoscopic 3-D and augmentedreality. It expanded our knowledge of how advertisingworks embedded in a new, innovative medium by exam-ining the impact of the medium, isolated from advertisingcontent or messages. As such, it enabled us to assess theoverall advertising effects more appropriately and effec-tively.


While the current study tried to provide a comprehen-sive model of assessing advertising embedded in a new,innovative medium, the exploratory nature of this studyhad some limitations. The proposed model is most suit-able for predicting advertising effectiveness within a new,innovative medium (e.g., stereoscopic 3-D, augmentedreality), rather than a traditional medium (e.g., newspaper,magazine). This is mainly because traditional media wouldnot be likely to generate enough media novelty to attracta high degree of attention from ad viewers. As such, as thenovelty of a medium wears off and viewers become boredwith it (Tellis 1997). One of the ways to avoid this limi-tation would be to incorporate into the model noveltyeffects caused not only by media but also by content. This

change would allow the model to explain how and whenadvertising within a traditional medium could be moreeffective than advertising in a new, innovative medium.Furthermore, the current findings are limited to the respon-ses of college students. Given that college students arerelatively more upscale and technologically savvy thanthe general population, the results might be different whenthe model is tested on respondents with a different demo-graphic profile. For example, older participants mightexperience a higher sense of media novelty than collegestudents. Therefore, future research should replicate themodel test on a more varied population. References areavailable upon request.

For further information contact:Mark Yi-Cheon Yim

Department of Marketing & Information SystemsThe Richard J. Wehle School of Business

Canisius College2001 Main Street

Buffalo, NY 14208Phone: 716.888.3264



VIRAL ADVERTISING AND ITS PLACE IN THEADVERTISING FRAMEWORK

Maria Petrescu, Barry UniversityPradeep Korgaonkar, Florida Atlantic UniversityTamara Mangleburg, Florida Atlantic University

Ann Root, Florida Atlantic University

SUMMARY

Despite the growing potential of online marketing,academic research regarding viral advertising and socialmedia has been sparse, especially compared with otherareas of Web 2.0 research (Allsop, Bassett, and Hoskins2007). There is still a limited understanding of the viralprocess (Graham and Havlena 2007) and previousresearch has noted the necessity to study the nature andcharacteristics of viral advertising as a means of multiply-ing a brand’s popularity (Chiu, Hsieh, Kao, and Lee2007).

The article integrates key advertising variables in aframework based on attitudes theory and their influenceon behavioral intentions, in order to assess the relation-ship between two attitudinal elements – attitude towardthe ad and attitude toward the brand – and two behavioralelements – viral intentions and purchase intentions.

We define viral advertising as unpaid electronic(e-mail, web, or social media) distribution of business oruser generated advertisements from consumer to con-sumer, based on ad content likeability, entertainment andcontroversial characteristics.

In the context of viral advertising, our model tests thekey classical hypotheses studied and confirmed in mul-tiple previous studies: that attitude toward the ad posi-tively affects attitude toward the brand, which has apositive effect on consumers’ intentions to purchase theadvertised product. We maintain the same relationbetween attitude toward the ad and attitude toward thebrand, and the relationship Ab – behavioral variables.

According to previous research, we estimate thatattitude toward the brand will positively influence thebehavioral variables in this model, viral intentions andpurchase intentions (Biehal, Stephens and Curlo 1992;MacKenzie et al. 1986). Given previous studies, wehypothesize a positive relation between consumers’ viralintentions and purchase intentions for the advertised prod-uct and brand. We include in the model key demograph-ics, such as age and gender.

The variables were operationalized through estab-lished scale, such as attitude toward the ad, attitude towardthe brand and purchase intention (MacKenzie et al. 1986;MacKenzie and Lutz 1989). We operationalized our keyvariable, viral intentions, by using previous used scales,adapting some behavioral intentions scales such as inten-tion to purchase and intention to recommend and evengenerating new items (Chiu et al. 2007; MacKenzie et al.1986).

In order to test our survey, we distributed the ques-tionnaire to a national consumer sample of 400 individu-als provided by Qualtrics. The national consumer sampleincludes consumers from every U.S. state, 53.4 percentmen, distributed over all age and income categories.

Principal component factor analysis using Varimaxrotation was performed in order to assess the fitness of ourmeasures. All items loaded as expected, with loadings ofover .6. The Cronbach alphas for the measurement scaleswere all above .8. We employed structural equationsmodeling (SEM), using Lisrel 8.8, to test the model.

As recommended by numerous researchers, we useda combination of goodness-of-fit indices (Hu and Bentler1999; Kline 2005) to assess the fitness of our model. TheCFI and NFI fit into the most restrictive guidelines pre-sented by Hu and Bentler (1999), who recommend valuesequal or higher than .95. RMSEA at .1 indicates only amarginally acceptable fit (Browne and Cudeck 1993);however, the SRMR at .06 fits into the most restrictiveguidelines presented in the literature (Hu and Bentler1999). Given these results, we considered the model fit.

We found that attitude toward the ad has a significantpositive effect on attitude toward the brand (MacKenzieet al. 1986, Mitchell and Olson 1981). The data analysisalso showed a positive relationship between attitudetoward the brand and purchase intentions, just as hypoth-esized and showed in previous studies (Biehal, Stephensand Curlo 1992).

The results found a positive relationship betweenattitude toward the brand and viral intentions. This estab-


lishes and places viral intentions as a significant variablein relation to classical advertising variables.

The key relationship tested in this model is the influ-ence of viral intentions on purchase intentions. Viraladvertising’s potential to increase consumer communica-tion and lead to increased sales and reduced marketingcosts has especially been debated in practitioner journals(Dobele et al. 2007; Fattah 2000). The results concludethat consumers’ intention to forward an ad positivelyinfluences their intention to buy the advertised product,ending the controversy related to the viral ads’ potential tolead to sales. This finding represents a significant contri-bution to the advertising literature, to the incipientresearch on viral ads, and a major confirmation for prac-titioners regarding the importance of viral advertising.

We also found support for a negative relationshipbetween education and viral intentions, conforming toresearch on market mavens, stating that they tend to beslightly less educated than other consumers are in general(Feick and Price 1987). Other demographic variablessuch as age and gender were not significant in our study.This might be due to the widespread use of the Internet andthe relative homogenization of Internet usage.

Overall, this analysis represents a significant contri-bution to the modern advertising research and practice,and a key step forward in the relatively sparse viraladvertising research. It establishes the role of viral inten-tions and ads in the classical advertising framework, andprovides a steady base for future research. References areavailable upon request.

For further information contact:Maria Petrescu

Barry University11300 NE 2nd Ave.

Miami Shores, FL 33161Phone: 954.628.2334



THE ROLE OF FACEBOOK FOR ADVERTISING: ADVERTISINGEFFECTIVENESS OF SOCIAL NETWORKS COMPARED TO

TRADITIONAL ONLINE ADVERTISING, INCLUDINGSYNERGIES AND TIME LAGS

Jens-Christian Reich, RWTH Aachen University, GermanyMalte Brettel, RWTH Aachen University, Germany

SUMMARY

Advertising effectiveness is of ongoing interest toboth researchers and practitioners (Rust et al. 2004).Many studies have analyzed different aspects of adver-tisements (ads) and have proven their effectiveness byshowing which ad works, when, for which customer, andunder what circumstances (e.g., Tellis et al. 2005). How-ever, advertising is subject to increasing cost pressures,reinforced by recent economic turbulences that have ledto calls for improved accountability. In reaction to suchcalls, advanced models have been developed that demon-strate that an ad effect can be carried-over to the next dayand beyond (time lags; cf., Naik and Raman 2003, whocalculate a carry-over of 93% for TV and 37% for printads). Advanced models have also shown that simulta-neous use of different ads can increase their effectiveness(synergies; cf., Naik and Peters 2009, who find synergiesfor TV, print and search ads).

With the advent of social media, new advertisingchannels have emerged which play an increasingly impor-tant role in today’s marketing mix (Forrester 2009). Dueto their highly interactive character, social networks openup unprecedented possibilities and challenges for manag-ers and researchers that can be used to market a product orservice (Trusov et al. 2009). Indeed, early studies basedon surveys and experiments in small communities demon-strate that a social network can foster customer retention(Bagozzi and Dholakia 2006) and facilitate customeracquisition (Trusov et al. 2009). However, recent studiesalso indicate comparably low ad effectiveness (Mabryand Porter 2010).

Nevertheless, little is known about how strong thesales impact of a big social network like Facebook as partof an integrated advertising campaign can be taking intoaccount both time lags and synergies. We thus addressfour main research questions: (1) How strong is the short-term sales impact of social network ads compared totraditional ads (Winer 2009)? (2) How long do these adeffects last (Osinga et al. 2010)? (3) What is the long-termsales effect of social network ads compared to traditionalads (Wang et al. 2009)? (4) Are there synergies betweensocial networks and other online ads (Rust et al. 2004)?

We use aggregate-level, daily field data obtainedfrom one of Germany’s top ten e-commerce retailers. Thisunique sample spans a period of 365 days from October2010 to September 2011 and includes search ads, targetedbanners, email newsletter and Facebook ads. We apply anintegrated direct aggregation approach to account for timelags (Srinivasan and Weir 1988; Herrington and Dempsey2005) and calculate interaction terms to model synergies(Green 1973; Völckner and Sattler 2006). Based on theestablished framework of Vakratsas and Ambler (1999),we derive our hypothesis on how advertising works:

Short-term, we expect positive effects that are com-parably strong for search ads (H1a); weak for targetedbanners (H1b); strong for email ads (H1c); and strong forFacebook ads (H1d).

We hypothesize carry-over effects that are compara-bly short for search ads (H2a); moderate for targetedbanners (H2b); long for email ads (H2c); and moderate tolong for Facebook ads (H2d).

Long-term, we expect positive effects that are com-parably weak for search ads (H3a); weak for targetedbanners (H3b); strong for email ads (H3c); strong forFacebook ads (H3d).

We infer positive synergies and their sales impact tobe comparably weak for search ads and targeted banners(H4a); strong for search and email ads (H4b); strong forsearch and Facebook ads (H4c); weak for targeted ban-ners and email ads (H4d); moderate for targeted bannersand Facebook ads (H4e); strong for email and Facebookads (H4f).

We find that all ad channels in our analysis positivelyaffect sales, with the strongest short-term impact forsearch and email ads, followed by targeted banners andFacebook ads (H1a-c fully, H1d partially supported). Ouranalyses also reveal a high carry-over for targeted bannersand email ads, followed by a moderate time lag forFacebook and an immediate sales impact for search ads(H2a and H2c-d fully, H2b partially supported). Theseresults lead to the long-term effect that is by far strongestfor email ads and is followed by a comparatively moderate


effect for targeted banners and a weak total sales impactfor search and Facebook ads (H3a and H3c fully, H3b andH3d partially confirmed). Out of our six hypotheses onsynergies, one is fully (H4b), three are partially supported(H4a and H4c-d). Search ads benefit most from synergiesas this is the only channel with significant synergies withall other channels in our analysis and shows the strongestsynergy overall with email ads. Unlike the other channels,Facebook does not build significant synergies except fora rather weak synergy with search ads.

For future research, our results underscore the impor-tance of incorporating different time lags for each adchannel as well as synergies when modeling advertising

effectiveness. Practitioners can use this approach to opti-mize their advertising. Generalizing our empirical results,a marketer should use search ads to generate an immediateeffect and targeted banners to stimulate long-term reac-tions. Email ads are highly effective in both the short runand the long run. Within an integrated campaign, alsoFacebook should play a significant role, but there is moreto social network advertising than creating group sitevisits which we analyze here. We encourage more re-search in this field to gain a solid understanding of howadvertising works and what the drivers are for successfuladvertising in a social network (e.g., clarify the role of anactive contribution). References are available uponrequest.

For further information contact:Jens-Christian Reich

RWTH Aachen UniversityKackertstr. 7, 52072Aachen, Germany

Phone: +49.241.80.99396Fax: +49.241.80.92371



BUYING THE FORTHCOMING: A PRELAUNCH INFORMATION ANDVALUE CONGRUENCE MODEL

Kyung-Ah Byun, Texas Tech University, LubbockJunghwan Kim, Texas Tech University, LubbockGavin L. Fox, Texas Tech University, Lubbock

SUMMARY

The market trend of rapid product replacement andobsolescence accelerates consumers’ behavioral patternsin terms of the timing of information seeking and ofpurchase intention. For example, AT&T sold out 200,000pre-order stocks for the iPhone 4S in the first 12 hours ofpreordering (Satariano 2011). The preorder behavior isintriguing not only because it commits consumers topurchasing and ownership well before taking physicalpossession, but also because it amplifies the level ofincomplete information when making purchasing deci-sions by removing tactility. Little marketing researchinvestigates prelaunch purchase behavior and most stud-ies about new-product purchase behavior focuses on theinitial post-launch period. Therefore, this study investi-gates the motivational processes wherein consumers’intrinsic motivations lead them through incomplete infor-mation-seeking, value-congruence, and purchase com-mitment.

In the prelaunch stage, consumers do not have theability to observe or experience the products. Consumerswho are interested in forthcoming products, therefore,likely are attempting to fill in as many informational gapsas possible to paint pictures of such products. The onlyvalidation available for such information is the prelaunchinformation itself, such that consumers must select whatto believe and how much weight to place on differentcomponents. With respect to information evaluation,Cognitive Evaluation Theory (CET) provides a theoreti-cal foundation as CET highlights that an external eventwith behavior-related information has an impact on indi-viduals’ intrinsic motivations by facilitating perceptionsof autonomy and competence (Deci 1975; Deci and Ryan1985; Ryan 1982). Drawing on CET, we suggest that theprelaunch product information plays a role as an externalevent in facilitating consumers’ intrinsic motivation forpurchase decision-making.

The Prelaunch Information and Value CongruenceModel (PIVCM) presented in this research explains howconsumers’ motivations induce engagement in informa-tion seeking, which in turn contributes to purchase com-mitment when the source of their decision-making islimited to prelaunch information.

In the prelaunch stage, a select set of motivatedconsumers are involved with searching information aboutforthcoming products. We consider two types of intrinsicmotivations as meaningful antecedents: innovativenessand playfulness. First, personal innovativeness refers tothe tendency of consumers to learn about and adoptinnovations or new products within a specific domain ofinterest (Goldsmith and Hofacker 1991). Consumers whoseek novelty necessarily are receptive to new ideas andtend to make independent decisions (Midgley and Dowling1978). They necessarily express a desire for novel expe-riences because such experiences are congruent withcertain self-concepts (Hirshman 1980; Cowart et al. 2008).As independent decision-makers and novelty seekers,innovative consumers become engaged in seeking infor-mation about forthcoming products.

As another motivation, playfulness is identified ascognitive spontaneity in information-seeking activitiesbased on Webster and Martocchio (1992). In playfulstates, individuals are more task-focused, aroused in termsof sensory or cognitive curiosity, and involved in activi-ties simply for pleasure and enjoyment rather than forextrinsic rewards (Moon and Kim 2001). Consequently,playfulness stimulates individuals in the prelaunch stageto seek additional information about forthcomingproducts.

Innovative and playful consumers are likely to seekinformation about new forthcoming products to fulfilltheir intrinsic motivations. When information createshigher-order beliefs and higher-order affect, purchasecommitment increases (Smith and Swinyard 1982). Inparticular, strong engagement intensifies attraction to avalue target (Higgins 2006). Therefore, consumers’ strongengagement in additional information search is likely toattract them to forthcoming products even without directproduct experience.

The attraction-repulsion mechanism operates basedon value congruence, which is defined as the belief thatthe value of a product will correspond to an individual’spersonal value. Value congruence has direct and positiveimpacts on affective commitment to brands and service(Zhang and Bloemer 2008, 2011). In the prelaunch stage,consumers will be more likely to commit to preorderingwhen value congruence from prelaunch information is


high, as they expect that the product will match the valuethey regard as important.

In sum, consumers’ purchase commitment to forth-coming products is influenced by the motivational pro-cess wherein innovativeness and playfulness facilitateinformation-searching engagement and value congru-ence occurs through the searching activities. In addition,source credibility is expected to increase the likelihood ofpurchase commitment by facilitating the relationshipbetween available information and value congruence.

This study contributes to the advertising and newproduct launch literatures by explicating how consumersreach states of purchase commitment for products beforethey are launched into the market, based on intrinsicmotivations such as innovativeness and playfulness.Although at present this research is limited to a conceptualframework, managerial implications can be found inestablishing prelaunch marketing strategies for forthcom-ing products. References are available upon request.

For further information contact:Kyung-Ah Byun

Area of MarketingRawls College of Business Administration

Texas Tech UniversityLubbock, TX 79409–2101

Phone: 806.834.4899Fax: 806.834.2199



THE ROLE OF FINANCIAL GIFT FORMATS ON THEPERSUASIVENESS OF THE ADVERTISING

MESSAGES

Qing Yao, University of Science and Technology Beijing, ChinaRong Chen, Tsinghua University, China

SUMMARY

Gift cards are claiming an ever increasing role inconsumers’ gift purchases and gift consumption. Previ-ous gift card studies have focused on the motivations ofgift givers to enable marketers to effectively market giftcards and maximize their sales (e.g., Waldfogel 1993;DuCasse 2009; Khouja et al. 2011). For example, thepopularity of gift cards stems from the purchasingconsumer’s ability to fulfill gifting obligations whileconveniently reducing the risk of poor gift selection(Waldfogel 1993). Despite the advances in our under-standing of the gift cards themselves, little is knownregarding whether the recipients of gift cards perceive andspend gift-card funds in the same way they would ofequivalent cash gifts.

This article proposes that gift formats (i.e., gift cardsvs. cash) can influence the persuasive impact of advertis-ing messages via different levels of construal. Our predic-tion is based on the model of goal-pursuit (e.g., Gollwitzer,Heckhausen, and Steller 1990) which suggests that thepreactional process of goal-pursuit entails two distinctphases: choosing between potential action goals and pro-moting the implementation of the chosen goal. We positthat if we view the purchase of a product as a consumer’sfinal action, gift-card users (vs. cash-gift buyers) wouldbe psychologically closer to the action because most ofthem already adopt a goal of redeeming all of the creditsfor hedonic aims (White 2008). That is, gift-card users aremore likely to focus on evaluating how a potential productcan satisfy their need to reward themselves (i.e., feasibil-ity), whereas cash-gift users still deliberate on whether orwhy they are willing to exert monetary resources to attaina goal (i.e., desirability).

Specifically, we posit that whereas consumers whoreceive a financial gift in the form of a gift card tend toconceptualize information at a low-level construal, thosewho receive a financial gift in the form of cash are moreinclined to construe information at a high-level construal.Furthermore, we hypothesize that when there is a corre-spondence between gift format and the level at whichmessages are construed, the evaluation of advertisedproducts is more favorable than when such correspon-dence is absent. Three studies test these hypotheses.

Study 1 primed the gift fund as either gift card or cashgift by asking participants to recall and describe in detailan occasion on which they received a gift card or cash asa gift. Then the participants classified objects that theywould take with them on a camping trip. The choice of aclassification task as the dependent variable was based onthe well-accepted premise that people who construe infor-mation at a high level will use fewer categories to classifyobjects in comparison to those who construe informationat a low level (e.g., Lee, Keller, and Sternthal 2010). Theresult that participants who wrote about gift cards usedmore categories (M = 6.02) than those who wrote aboutcash gifts (M = 4.87, p < .02) offered support for ourprediction that financial gift presented as gift cards aremore likely to activate low-level construal than those thatare presented as cash.

Study 2 tested the hypothesis that the fit with giftformat would enhance product evaluation and providedsupport for the underlying theory that the subjectiveexperience of fit induces more engagement in messageprocessing. Similar to Study 1, participants were ran-domly assigned to gift card or cash gift conditions. More-over, half of the participants were exposed to a high-levelconstrual advertisement for a trainer and the other halfwere exposed to a low-level construal advertisement.Then the participants reviewed the advertisement, evalu-ated the product and worked on an incidental anagramtask. As predicted, participants in the gift card conditionhad more favorable product attitudes (M = 4.96) andsolved more anagrams (M = 6.76) when the product wasdescribed at a low construal rather than a high construallevel (M

attitude = 3.92, p < .02; M

anagram = 3.92, p < .001). In

contrast, participants who recalled spending cash giftfunds evaluated the product more favorably (M = 4.88)and solved more anagrams (M = 6.96) when it wasdescribed at a high construal level rather than a lowconstrual level (M

attitude = 4.18, p < .08; M

anagram = 4.52, p <

.001).

Study 3 replicated the effect of fit on product evalu-ation by simulating a purchase scenario using either giftcard or cash gift fund and examined the mediating role ofengagement on the effect of fit with gift formats onproduct evaluations. Corroborating the findings of Study 2,participants who used gift cards had more favorable


product evaluation when the product was described at alow construal level (M = 5.02) than a high construal level(M = 4.23, p < .01), whereas those who used cash gifts hadmore favorable product attitudes when the product wasdescribed at a high construal level (M = 4.63) than a lowconstrual level (M = 4.02, p < .08). Furthermore, engage-ment was significantly increased in the fit conditions incomparison to the non-fit conditions (p < .001). Theproduct evaluation for the fit conditions was significantlyhigher than that for the non-fit conditions (p < .01).Finally, when we regressed product evaluation on bothfitness and engagement, the effect of fitness was no longersignificant, whereas the effect of engagement was signifi-

cant (p < .001). The Sobel test for mediation was alsosignificant (p < .01).

Together the studies further our understanding of thedifferent cognitive processes and consequences associ-ated with gift cards versus cash gifts. The research isimportant as such knowledge could help companies deter-mine what type of advertising messages would be mostsuitable for encouraging purchases by consumers witheither type of gift.

This work was supported by the National NaturalScience Foundation of China (Grant No. 70872057 and71172011).

For further information contact:Qing Yao

Dongling School of Economics and ManagementUniversity of Science and Technology Beijing

Beijing, China 100084Phone: 86.1381.032.2363



A FRAMEWORK TO UNDERSTAND CUSTOMER DATA QUALITY INCRM SYSTEMS FOR FINANCIAL SERVICES FIRMS

Debra Zahay-Blatz, Northern Illinois UniversityJames Peltier, University of Wisconsin – Whitewater

Anjala Krishen, University of Nevada, Las Vegas

SUMMARY

Recent changes in the digital environment related toinformation storage, collection, and dissemination haveresulted in firms developing and maintaining databaseswith large quantities of customer information. As a con-sequence, companies are increasingly being confrontedwith massive amounts of data contained in widely dispar-ate and often inconsistent databases. In fact, researchindicates that marketing professionals will use or willwork around data repositories without trust in the under-lying data quality. Thus, the identification of processeswhich can establish systems to not only capture customerdata and increase customer knowledge but also maintainthe quality and consistency of the data after the fact isbecoming a paramount business concern.

Still, few firms implement relational frameworks thatprovide a 360° view of their customers’ transactional,psycho-demographic, and behavioral profiles. This trendpersists in part because customer data quality is of utmostimportance for businesses as firms attempt to mergemultiple methods of communication together in inte-grated campaigns. Although companies appear to under-stand the need to generate customer information, theystruggle with the process of integrating that informationthroughout the organization, in other words, the creationof customer knowledge competence. Central to customerdata quality in CRM systems is the need for ongoing andaccurate customer information both to manage campaignsand determine customer value. We approach this area byproviding a model which explains how intra-organiza-

tional structures in firms can lead to increased customerdata quality in CRM systems.

For our study, after a pretest, a survey was mailed to525 executives in the financial services industry, specifi-cally the banking industry. The financial services industrywas selected because many firms in this area, includingthe Royal Bank of Canada and Charles Schwab, have beencited for superior quality customer information use. Nameswere obtained from Hoover’s database and the unit ofanalysis was the business unit. Self-reports of perfor-mance were used since performance data is not usuallyavailable on a business unit level. A total of 166 usablequestionnaires were analyzed using multiple regression.

Our findings illustrate a path to performance as aresult of collecting information about the customers in alearning context. In essence, this framework suggests thatcustomer information must be managed through effortswhich include knowledge identification, capture, selec-tion, storage, sharing, application, creation, and selling.Building from this research on knowledge orientation,data collection in this study is seen as a customer data-oriented learning activity that is a precursor to the qualityof the CRM system as measured by the quality of its data.Such data tells managers how to organize their prioritiesin terms of data learning activities to create the highestquality CRM data so that improved performance canensue. This study suggests that capturing the results ofpersonalized communications as well as detailed transac-tions of customer interactions will yield the highest resultsand therefore should be the organization’s strategic focus.References are available upon request.

For further information contact:Debra Zahay-Blatz

Northern Illinois UniversityBH 128I

DeKalb, IL 60115Phone: 815.753.6215



THE ROLE OF FIT AND SIMILARITY IN SOCIALSPONSORSHIP COMMUNICATIONS

Ravi Pappu, University of Queensland, AustraliaT. Bettina Cornwell, University of Oregon

SUMMARY

Corporations are becoming active in promoting so-cial causes (Szykman, Bloom, and Blazing 2004) andwith good reason. Such alliances, called social sponsor-ships (Madill and O’Reilly 2010), can lead to highersponsor recall, favorable consumer attitudes, reinforcesponsor brand positioning and enhance equity for thesponsors (Cornwell et al. 2006; Cornwell, Weeks, andRoy 2005; Simmons and Becker-Olsen 2006). Manycorporations engage in social sponsorships with the ob-jective of gaining goodwill and positive associations.

Sponsorship fit, the perceived match in the sponsor-ing relationship, is considered an important factor influ-encing consumer evaluations of social sponsorships andis a widely researched topic in advertising and consumerresearch (Olson and Thjomoe 2011). The findings forsocial sponsorships are, however, mixed. For example,research shows that high-fit social sponsorships favor-ably influence consumer attitudes and behavioral inten-tions toward the cause (Samu and Wyner 2009) as well asthe nonprofit’s brand image (Becker-Olsen and Hill 2006).On the contrary, some studies have shown that corporate-sponsored messages may not be as effective as thosecommunicated by nonprofits themselves (Szykman et al.2004) and that fit may not be as important in nonprofitevaluations (Menon and Kahn 2003).

Moreover, relationship fit is malleable. The conceptof fit found in a particular sponsorship association can becreated by making shared associations salient (Simmonsand Becker-Olsen 2006), and by explaining or articulat-ing a possible linking idea in the relationship (Cornwellet al. 2006). This seems particularly important when thereis a goal to create goodwill, as when an oil companysponsors environmental protection. In addition to thereactions of individuals to the sponsorship relationship, aperson may hold a basic, and somewhat independentperception of the similarity of the two partners. That is tosay, there is the response to fit of the partnership and aresponse to the similarity between the two entities. Both ofthese may be important to sponsorship success.

This potential difference between perceptions of re-lationship fit and entity similarity stems in part from thefact that sponsors and properties are multifaceted andbring unique characteristics (e.g., trustworthiness, famil-

iarity, category performance and reputation) as well aspotentially negative or simply mismatching characteris-tics, all of which have the potential to influence targetmarket perceptions (Bower and Grau 2009). In cause andnon-profit sponsoring, there are also heightened concernsof trust and possible exploitation as one partner is utilizedin the communications platform of the other. Although therole of trustworthiness has been widely researched in thesource credibility of a spokesperson (e.g., Priester andPetty 2003; Tormala and Petty 2004), the role of sponsortrustworthiness in social sponsorships has not been fullyexplored. Thus, the objective of this research is to exam-ine the role of fit, similarity and trustworthiness, in theevaluation of social sponsorship relationships.

Conceptual Framework

We explain the moderating role of sponsor-nonprofitsimilarity in the relationship between fit and social spon-sorship evaluations using categorization theory. In keep-ing with research in the area of sponsorship (Simmons andBecker-Olsen 2006), we predict that high-fit social en-dorsements will be viewed more favorably and generategreater clarity of brand positioning for the endorsedproperty; and that both consumer attitudes toward theendorsement and clarity of positioning influence attitudesand behavioral intentions toward sponsored nonprofit. Interms of moderation, we predict that high similaritybetween sponsor and nonprofit to be problematic whenthere is low fit of the sponsorship relationship. When theentities hold some similarity (alcohol distributor, alcoholabuse prevention) and the sponsorship relationship fit islow, it is argued that individuals become skeptical and thismay influence dependent variables of interest.

Method

Experimental designs for two charity product catego-ries (blood donations and cancer prevention) were exam-ined with structural equation modeling (Bagozzi 1977).Participants in each experiment were undergraduate stu-dents and were randomly assigned to one of the fourexperimental conditions in a 2 (social sponsorship: high-fit versus low-fit) x 2 (brand: Red Cross versus LeukemiaFoundation) between-subjects design. The procedureemployed was identical for both experiments. Stimuliwere developed based on pretesting. Fast-food chainSubway, which offers deli sandwiches was included as the


high-fit sponsor whereas Kentucky Fried Chicken (KFC)was included as the low-fit sponsor in Experiment 1 (N =189). In the second experiment, a regional supermarketchain was the high-fit sponsor and Coca Cola brand softdrink was the low-fit sponsor (N = 210). Treatmentconditions were manipulated in a fictitious newsletter.Established scales were used to measure constructs ofinterest (e.g., Ohanian 1990; Simmons and Becker-Olsen2006). Items were measured on seven-point scales.

Results

Results from the two experiments were consistentwith the study’s predictions. High-fit social partnershipsimproved clarity of positioning and generated more fav-

orable attitudes toward social sponsorships but highersimilarity blurred clarity of positioning, and generatedless favorable attitudes toward the focal sponsorships.Moreover, the impact of fit varied according to the degreeof similarity between the social partners. Higher similar-ity blurred clarity of positioning for low fit sponsorshiprelationships. Trustworthiness of the sponsor also en-hanced the clarity of positioning and generated morefavorable attitude toward the sponsorship. The findingssuggest that aspects of similarity between partners raisesuspicion regarding the motivations for the relationshipand this is especially pronounced with there is poor fit interms of the sponsorship relationship. References areavailable upon request.

For further information contact:Ravi Pappu

Business, Economics and LawUniversity of Queensland

Brisbane 4072 QLDAustralia

Phone: +61.7.3346.8089Fax: +61.7.3365.6988



SHOULD FIRMS NAME COMPETITORS IN THE ADS?

Chun-Kai Tommy Hsu, Old Dominion University, NorfolkLeona Tam, Old Dominion University, Norfolk

SUMMARY

Comparative advertising has been widely used in theUnited States. It is generally believed that comparativeadvertising is more effective than non-comparative adver-tising in terms of memory, claim acceptance, and Con-sumer Perceptions. With the growing popularity of com-parative advertising in recent years, we believe there is aneed to further our research to specifically study on theeffects of different types of comparative advertising.

One question has been asked repeatedly by manycompanies: Is naming a specific competitor in a compara-tive advertisement a good idea? In other words, what is theeffectiveness of direct comparative advertising? In theliterature, the discussion of comparative advertising hasfocused on direct comparative advertising alone (Miniardet al. 2006), and largely ignored the important comparisonof direct versus indirect comparative advertising (with theexception of Neese and Taylor 1994) However, comparedto direct comparative advertising, indirect comparativeadvertising’s potential for positioning the advertised prod-uct has received far less attention in the literature (Miniardet al. 2006). In this paper, we will address this research gapby investigating the effects of direct versus indirect com-parative advertising and the boundary conditions of direct/indirect comparative advertising effects in terms of adver-tising valence and attribute typicality.

Direct/Indirect Comparative Advertising

Based on Pechmann and Ratneshwar (1991), directcomparative advertising is an advertising strategy that theadvertiser specifically names its competitors in the adver-tisement to compare itself to the named competitors. Incontrast, in an indirect comparative advertisement, theadvertiser does not identify particular competing brands,but instead refers to unnamed competitors, such as “theleading brand,” “other brands,” or “all other brands”(Miniard et al. 2006). While both direct and indirectcomparative advertising encourage the creation of com-parative evaluations in viewers’ minds, the effectivenessof these two types of comparative advertising shoulddiffer based upon viewers’ reference points (Miniard et al.2006).

Moderating Role of Advertising Valence

Comparative advertisements can be classified bywhether they are positive or negative (Jain 1993). Positivecomparative advertising compares brands with selectedattributes to make the claim that the advertised brand issuperior to the compared brand while in a negative com-parative advertisement it focuses on negative aspectsassociated with the compared brand. When consumers areexposed to direct comparative advertisements, they tendto pay more attention and simultaneously comparebetween the advertised and compared brands. Since nega-tive comparative advertisements are more memorable(Sorescu and Gelb 2000), it can be more efficient to createcomparative evaluating process in consumers’ minds. Incontrast, since the advertiser compares itself to the leadingbrand or to all other brands in an indirect comparativeadvertisement, trying to attack all other competitors or animplicit brand can lead to have negative consumer atti-tudes or confusions. Thus, positive comparative adver-tisements are more effective in the situation which theindirect advertising is used.

Moderating Role of Attribute Typicality

Product or brand attributes can be classified on aspectrum ranging from typical to atypical. When a com-parative advertisement uses a typical attribute to compare,it is more likely for consumers to be involved in analyzingthe comparison thoughtfully and having a “piecemealreview” of product attributes (Pillai and Goldsmith 2008).Therefore, the evaluating processes will pose seriousthreats to consumers’ current attitudes toward both theadvertised and compared brands and then create counter-argumentation in their minds. On the other hand, when theattributes that are compared in the comparative advertise-ment are atypical, consumers are less likely to havecounter-argumentation as the information provided by thecomparative advertisement is less threatening to the com-pared brands in consumers’ minds (Pechmann andRatneshwar 1991).

Results and Discussions

Using two experimental studies, this paper aims atunderstanding the effects of direct versus indirect com-


parative advertising and investigating two moderatingvariables: advertising valence and attribute typicality.From study 1, we found that indirect comparative adver-tisements generated more positive attitude toward theadvertised brand when the advertisements were posi-tively-worded; while direct comparative advertisementswere more effective when the advertisements were nega-tively-worded. As hypothesized, we found in study 2 thatwhen the compared attribute was typical, direct compara-tive advertisements generated more positive attitude to-ward the advertised brand. However, contrary to ourexpectation, indirect comparative advertisements gener-ated more positive attitude toward the advertised brandwhen the compared attribute was atypical. We speculateif it is better for those firms which want to comparethemselves with more than one competitor to developtheir own “specialized attributes” rather than focus onsome typical attributes that have been well provided andserved by many other companies. It will be beneficial to

further investigate this in the future to understand theunderlying driving factors.

This paper also provides several managerial implica-tions and applications. First, the findings suggest that, ifthe company wants to attack one particular competitor,not only should it uses negatively-worded comparativeadvertisements but also it has to compare a typical attri-bute to effectively influence consumers’ attitudes towardthe advertised brand. In the cases that the company wantsto claim it is better than others in general, it needs to utilizepositive comparisons in its advertisements regardless ofwhether typical or atypical attributes are used. We believethis paper not only advances our current understanding onthe effectiveness of comparative advertising, but alsocreates a new research stream that specifically addressesdirect and indirect comparative advertising. Referencesare available upon request.

For further information contact:Chun-Kai Tommy Hsu

Old Dominion University5115 Hampton Boulevard

Norfolk, VA 23529Phone: 757.337.9444



REGULATORY FOCUS AND DAILY DEAL MESSAGE FRAMING:ARE WE SAVING OR GAINING WITH GROUPON?

Iryna Pentina, University of ToledoDavid G. Taylor, Sacred Heart University, Fairfield

SUMMARY

Growing consumer demand for coupons, discountsand deals, attributed to the recent recession and sloweconomic recovery, is a fertile ground for a multiplicity ofonline daily deal sites selling vouchers for steeply dis-counted local offers. Such relatively new websites asGroupon, LivingSocial, and others, have attracted numer-ous small and medium businesses as an alternative to localadvertising that is targeted, controllable, and accountable(Streitfeld 2011). They have also been popular with con-sumers who enjoy both the savings and the novelty of theofferings delivered straight to their email boxes and(increasingly) to their smart phones. Recently, however,the attractiveness of this business model to both sellersand consumers has been questioned: the merchants areincreasingly disappointed with lack of loyalty (and prof-its) from the new consumers they attract through deals,while consumers complain about being flooded with andenticed into impulsively buying offers that they ultimatelydo not need. While some customers are enthusiastic andexperience “Groupon anxiety” (eager anticipation of eachnew Groupon offer), others exhibit “Groupon fatigue”having exhausted their urges for “fish pedicure and indoorsky-diving” (Salmon 2011). These conflicting sentimentsand arguments emphasize complexity of the daily-dealsphenomenon and possible lack of understanding abouthow to best utilize this new tool to accomplish businessobjectives. It is possible that wide diversity of opinionsand assessments reflects diverse goals and motivations ofdaily-deal users, both on business and consumer sides.

This paper focuses on consumer perceptions andbehaviors regarding online daily deals. It addresses theissue of compatibility of the daily-deal marketing toolwith consumer personal characteristics, and its effect onthe perceived value of the deal. We utilize regulatory fittheory (Higgins et al. 2003) to suggest that the matchbetween consumer regulatory orientation (promotion vs.prevention) and daily-deal message framing (gain vs.non-loss) would affect the perceived value of the offer.We conduct an experiment to test our supposition and tounderstand whether the mechanism of the regulatory fitphenomenon operates in the daily deal context. We exam-ine the effect of daily-deal message framing (gain vs. non-loss) on the choice between two otherwise very similarGroupon offers for a three-day get-away in a Floridaresort by individuals with different primed regulatory

foci. We also assess how the perceived value assigned tothe resorts selected by participants in different regulatoryfocus conditions is moderated by its promotional framing(gain vs. non-loss).

Seventy-nine marketing students (20–55 years old,30% undergraduate, 60% female) participated in the studyfor a class credit. All parts of the experiment were con-ducted in the form of an online survey. We used a 2(regulatory focus: promotion vs. prevention) x 2 (offermessage frame: non-loss vs. gain) mixed-factorial designwith the message frame as a repeated measure. Theparticipants were randomly assigned to one of the tworegulatory focus groups. The manipulation of promotionand prevention foci followed Higgins et al. (1994), wherebyparticipants in the promotion (prevention) focus condi-tion were asked to write a short essay on their hopes(duties) as a consumer and things they wanted to acquireor experience. Following this task, they were asked tocarefully review two Groupon offers that appeared ontheir computer screens in random order and to answerquestions about each of them. The two offers were authen-tic Groupon deals for three-day Florida get-away vaca-tions in resorts of identical star ratings, each showing thepicture of a swimming pool on the blue ocean back-ground, with the text below detailing comparable ameni-ties and rates

Our findings confirmed the applicability of regula-tory compatibility framework to online daily deal offers,showing that matching message framing of daily deals toshoppers’ regulatory focus. Thus, the level of persuasive-ness of a promotion may be increased by emphasizingnon-loss for prevention-oriented individuals, and con-versely by emphasizing gain for promotion-oriented cus-tomers. It is possible that framing promotional messagesonline in a manner compatible with a website visitor’schronic regulatory focus can facilitate conversion of onlineshoppers to buyers. This study also supports the phenom-enon of transfer of value from perceived regulatory com-patibility to the value of the selected product, potentiallyincreasing reference price of the product in the consumer’smind. This higher reference price for the product itself, ifstored in memory, may increase the probability of itsfuture purchases at a non-discounted price, thus reducingnegative effects of online sales promotions on brandimage. References are available upon request.


For further information contact:Iryna Pentina

Department of Marketing and International BusinessUniversity of Toledo

MS 103Toledo, OH 43606

Phone: 419.530.2093Fax: 419.530.4610



AN INVESTIGATION OF THE ATTITUDE RESISTANCE PROCESS ONNEGATIVELY FRAMED COMPARATIVE ADS

Russell Laczniak, Iowa State University, AmesKristine Ehrich, University of San Diego

Darrel Muehling, Washington State University, PullmanAkshaya Vijayalakshmi, Iowa State University, Ames

SUMMARY

Since the Federal Trade Commission (FTC) endorsedthe use of comparative ads (defined as ads in which asponsor directly compares itself to a “leading brand” – thecomparative referent), researchers have noted that manyare antagonistic and negative, especially in their portrayalof the comparative referent or “compared-to brand” (cf.,Grewal, Kavanoor, Fern, Costley, and Barnes 1997).

The recent increase in such ad campaigns is likelybased on research findings (e.g., Sorescu and Gelb 2000)which suggest that negatively framed comparative ads(i.e., those which convey information on the inferiority ofcomparative referents, rather than promoting the superi-ority of the sponsoring brand) tend to be more effectivethan their positively framed counterparts. Yet, some ques-tion the degree to which negatively framed comparativeads are as effective as initially thought. For example, in amore recent study, Shiv, Britton, and Payne (2004) foundthat in certain situations, receivers perceived negativelyframed comparative ads as being unfair and, as a result,were less influenced by them (as opposed to positivelyframed ads). Thus, it seems that researchers are far frommaking definitive conclusions about the presumed advan-tage of negatively framed comparative ads.

While prior research investigating the effectivenessof positively versus negatively framed comparative adshas studied effects based on receivers’ perceptions of thesponsoring brands, no study to date has looked at theirimpact with respect to receivers’ views of the compared-to brand (i.e., comparative referent). Since one of the mainobjectives of comparative ads is to induce users of thecomparative referent to switch brands (Aaker and Myers1982), it is important to investigate these receivers’ viewsof their brand of usage after exposure. The present studyaims to investigate both users and non-users post-expo-sure views of the comparative referent after exposure tonegatively and positively framed comparative ads withimmediate and delayed (two week) measures. The presentstudy will aid researchers by determining the extent towhich persuasion resistance can be used as a theoreticalframework in the comparative advertising context.

Tormala and Petty (2002) describe the persuasionresistance process as one in which there is an absence ofattitude change (regarding both the advocated positionand the initial position held by the receiver of the persua-sive message) that occurs in response to counter-attitudi-nal information that is presented to receivers who are“committed” to an alternative position. The framework isbased on the premise that when committed receivers viewcounter-attitudinal information as being an attack on theirposition, they will cognitively reject it. Two importantadvertising implications are that after exposure to a counter-attitudinal ad message, receivers are not likely to changetheir attitudes toward the advocated brand and will becomemore confident in their attitudes toward their brand of useand since persuasion resistance is a motivated process(Tormala and Petty 2004) it follows that the expectedincrease in attitude confidence should be maintained overtime. One could question whether any comparative adver-tisement will be viewed by all consumers as an attack onthe comparative referent (compared-to brand). We con-tend that this will more likely be the case for users of thecomparative referent when they are exposed to negativelyframed comparative ads. Given that positively framedcomparative ads are less likely to be perceived as attacks(or as derogating the comparative referent), we believethat the persuasion-resistance processes will less likely beevoked by users exposed to such ads – the absence ofnegative information will be less motivating for users tocognitively support their position. We contend that usersof the comparative referent should respond less adverselyto positively framed comparative ads than to negativelyframed comparative ads. Further, since nonusers of thecomparative referent do not have a “stake” in the com-parative referent, they should also be less likely than usersto employ these processes outlined by the persuasionresistance framework. As a result, we suggest that users(compared to nonusers) of the comparative referent shouldbe more confident in their post-exposure brand attitudestoward the comparative referent after exposure to nega-tively (as opposed to positively) framed comparative ads.

H1: Users of the comparative referent, as compared tononusers, should have higher levels of post-exposureattitude confidence (for the comparative referent)


after exposure to negatively framed comparative ads.(Supported)

H2: Users of the comparative referent exposed to nega-tively framed comparative ads should have higherlevels of post-exposure attitude confidence (regard-ing the comparative referent) than users exposed topositively framed comparative ads. (Not supported)

H3: Users of the comparative referent, as compared tononusers, will generate more counterarguments tonegatively framed comparative ads. (Not Supported)

H4: Users of the comparative referent will generate morecounterarguments to negatively framed comparativeads as compared to positively framed comparativeads. (Supported)

H5: Users of the comparative referent, as compared tononusers, should have higher levels of delayed mea-sures of attitude confidence (for the comparativereferent) after exposure to negatively framed com-parative ads. (Supported)

H6: Users of the comparative referent exposed to nega-tively framed comparative ads should have higherlevels of delayed measures of attitude confidence(regarding the comparative referent) than usersexposed to positively framed comparative ads. (Notsupported)

The main purpose of the study was to determine theeffects of negatively versus positively framed compara-tive ads on the post-exposure self-reported attitude confi-dence of users versus nonusers of the comparative refer-ent. It appears that results support the hypothesized notionthat users exposed to a negatively framed comparative adwill have higher levels of post-exposure attitude confi-dence in the comparative referent than do nonusersexposed to the same ad. Moreover, it appears that theattitude frames were unsuccessful in differentially influ-encing post-exposure attitudes toward the sponsoringbrand leading to the belief that the use of attack compara-tive ads are not more effective than positive comparativeads.

For further information contact:Kristine Ehrich

University of San Diego5998 Alcala Park, Olin 235

San Diego, CA 92110Phone: 619.260.7629



THE EFFECT OF ANALYTIC AND HOLISTIC THINKING ONCONSUMERS’ ATTITUDES’ TOWARD HOLISTIC OR

ATTRIBUTE ADVERTISING

Beichen Liang, East Tennessee State University, Johnson City

SUMMARY

Advertising information has a strong impact on con-sumers’ responses to advertisements and brands(Abernethy and Franks 1996). Many scholars have com-pared the information content between Eastern and U.S.advertisements. However, almost all but one study (i.e.,Taylor, Miracle, and Wilson 1997) were based on contentanalysis. Therefore, we still know a little about howadvertisements with different information content influ-ence consumers’ attitudes. Moreover, cross-cultural adver-tising scholars have relied too much on Hofstede’s cul-tural dimensions (Okazaki and Mueller 2007) and fewscholars have investigated the effect of analytic versusholistic thinking on advertising, especially the informa-tion content of advertising. In this paper, I apply theanalytic and holsitic theroy to advertising. Moreover, thisstudy will also investigate whether cognitive elaborationor quantity of information cues moderate the effect ofculture on consumers’ responses toward ads with differ-ent information emphasis.

For East Asians, it is not the part but the whole thatexists. Thus, East Asians look at everything “in its totality,not in parts” (Moore 1968, p. 3). Thus, when exposed toads with product attribute information only (attribute ad),East Asians tend to think holistically to consider addi-tional factors not provided in such ads, such as price andavailability, because failing to consider a factor – even aminor factor – may result in a bad judgment or purchasingdecision. For East Asians, although product attributeinformation is still important, in isolation it is far fromenough to support evaluation, judgment, and decisionmaking.

Since Westerners tend to think analytically (Nisbettet al. 2001), they pay more attention to product attributeinformation. For them, product attribute information isthe predominant factor influencing their evaluation, judg-ment, and purchase decision. Thinking about attributes inisolation is quite normal because they tend to focus on theanalytic parts, not the holistic totality. This discussionleads to the following hypotheses.

H1a: Individuals from East Asian cultures will generatemore favorable attitudes toward holistic ads thanindividuals from Western cultures.

H1b: Individuals from East Asian cultures will generatemore favorable attitudes toward holistic ads thantoward attribute ads.

H2a: Individuals from Western cultures will generatemore favorable attitude toward attribute ads thanindividuals from East Asian cultures.

H2b: Individuals from Western cultures will generatemore favorable attitude toward attribute ads thantoward holistic ads.

The purpose of Study 1 is to test how culture affectsconsumers’ attitudes toward different ads. To assess thestated hypotheses, the study employed a 2 (culture: Eastvs. Western) x 2 (ad: attribute vs. holistic ad) between-subjects factorial design. A digital camera was used as thetarget product. Ad stimuli were adapted from a real ad ina magazine to increase the external validity. Sixty Cauca-sian American students from a public university in thesoutheastern U.S. and sixty Chinese students from apublic university in southwestern China participated inthe study. As expected, Chinese participants generatedmore favorable attitudes toward holistic ads than towardattribute ads while Americans generated more favorableattitudes than the Chinese when exposed to the attributead. However, in contrast to identified expectations, Ameri-cans generated more favorable attitudes toward the holis-tic ad than the attribute ad. Moreover, both Americans andthe Chinese generated similar attitudes toward the holisticad.

The surprising finding in Study 1 may be the result ofeffortful cognitive elaboration. Therefore, Study 2 wasdesigned to examine whether people’s attitudes towardattribute and holistic ads were moderated by effortfuldeliberation. Seventy-nine Chinese students from a pub-lic university in southwestern China and eighty-fourAmerican students from a public university in the south-eastern U.S. participated in this study. The study utilizeda 2 culture (China vs. U.S.) x 2 ad (holistic vs. attribute ad)x 2 time pressure (high vs. low) between subject factorialdesign. However, the results show that cultural differencein attitudes was not significantly moderated in the lowtime pressure condition.

Study 1 still has one limitation: the holistic ad hadmore information cues than the attribute ad. Unlike the


previous studies, Study 3 contained four pieces of infor-mation for both attribute and holistic ads. Forty Chinesestudents at a large university in southwestern China andforty American students at a university in the southeasternU.S. participated in this study. The findings repeatedthose from study 1. However, reducing attribute cues inthe holistic ad slightly (not significantly) moderated theattitude differences between the holistic and attribute ad.

Although culture may not explain why consumersfrom both Eastern and Western cultures prefer holistic adsto attribute ads, culture can partially explain consumers’attitudes toward the ad with different information empha-sis. Yet why do both Americans and the Chinese preferholistic ads to attribute ads? A possible explanation isconsumers’ need for holistic information. Consumersneed information to make a better purchase decision eventhough they are not planning to make a purchase decisionimmediately or in the near future (Schmidt and Spreng1996). Second, consumers also need more information tojustify their decisions to others (Simonson 1989) or evento themselves (Schmidt and Spreng 1996). Third, con-sumers need information to make an optimal decisionrather than an acceptable decision (Swan 1969).

Individuals may have two knowledge systems: cul-tural and individual knowledge (Briley and Aaker 2006).Although consumers learn consumption knowledge (e.g.,how to make a wise decision; which brand is trustworthy)much later than they learn their culture knowledge, con-sumption knowledge may be woven into their culturalknowledge and used to correct shortcomings of theircultural knowledge since consumers always want to makea wise decision, increase the utility of their purchases, andreduce risks.

This study also has some managerial implications.The current study’s findings suggest that advertisers inboth East Asian and Western cultures should use adver-tisements with holistic information to reduce perceivedrisks and help consumers make better decisions. Thefindings further suggest that holistic ads should be usedwhen consumers are under high time-pressure situations(e.g., when consumers are reading a roadside billboard ona highway). The findings in this study may not be limitedto the advertising only and may be applied to other areas,such as website design or direct marketing communica-tion. References are available upon request.

For further information contact:Beichen Liang

Department of Management and MarketingEast Tennessee State University

Johnson City, TN 37604Phone: 312.636.0646



USING SELF- VERSUS OTHER-BENEFIT MESSAGES IN ADS FOR GREENPRODUCTS: THE MODERATING ROLE OF PERCEIVED CONSUMER

EFFECTIVENESS AND CONSUMER GUILT

Ceren Ekebas, Old Dominion University, NorfolkKiran Karande, Old Dominion University, Norfolk

SUMMARY

Green marketing efforts are being increasingly usedby companies to enhance their corporate image. In spite ofits increased importance, there is limited research on theefficacy of green advertising strategies.

The benefit of green marketing efforts can be framedby highlighting the benefit to the self and to others. Thenotion of influencing consumers by framing the messageto highlight the benefit to the self and benefit to others hasbeen investigated in other contexts such as donor behav-ior. To our knowledge, the effect of message framing (byusing self or other benefit green marketing advertisingappeals) on the consumer has not been investigated. Theperception of consumers about the societal responsibilityof the firm has become crucial as consumers pay more andmore attention to the green activities undertaken by com-panies. Therefore, our study investigates the effect ofusing self and other-benefit messages on consumer per-ception about the societal responsibility of the company,which in turn influences consumer response (attitudetoward the brand, purchase intention, and word of mouthintention). Further, we investigate whether the effect ofmessage framing on perceived societal responsibility ismoderated by the consumers’ perceived control overenvironmental problems and consumer guilt.

We define a self-benefit message frame as “the claimof green product which is beneficial for the consumerhimself or the very close family” and an other-benefitmessage as “the claim of the green product which isbeneficial for others such as society and future genera-tions.” We propose that self-benefit appeals enhance theselfish and egoistic motives of consumers and other-benefit appeals enhance the altruistic and empatheticmotives of consumers. When either motive is enhanced, itleads to a favorable perception about the company’ssocietal responsibility. We also propose that a high per-ceived societal responsibility results in a better attitudetoward the brand, purchase intention, and word of mouth.The strength of the effects during the two stages dependsupon consumer characteristics such as perceived con-sumer effectiveness (PCE) and consumer guilt.

Moderating Role of Perceived Consumer Effective-ness

Perceived consumer effectiveness (PCE) is the extentto which consumers believe they individually can influ-ence and solve environmental problems (Ellen et al.1991). People who believe their individual effort canchange the conditions have high PCE. For consumerswith high PCE, we propose that other-benefit messagescreate a better perception of societal responsibility for thecompany. Individuals who think their effort would notmake any difference to change the conditions have lowPCE. For low PCE consumers, we hypothesize that self-benefit messages create a better perception of societalresponsibility for the company.

Moderating Role of Consumer Guilt

Consumer guilt is the extent to which consumers feelnegative emotions for not doing enough for benefit of theenvironment (Berger and Corbin 1992). Using other-benefit messages help alleviate the guilt felt by a con-sumer, and can therefore lead to a perception that thecompany is responsible to the society. Thus, we hypoth-esize that other-benefit messages will work better than aself-benefit oriented message when the guilt felt by theconsumer is high. On the other hand, when the consumerfeels less guilty, the self-centered tendencies of the indi-vidual will be addressed by the self-benefit message,which will be more effective in improving the company’sperceived societal responsibility.

Results

We test the hypotheses using a 2 (Message frame) X2 (PCE) X 2 (consumer guilt) between subjects experi-mental design where message frame was manipulated andPCE and consumer guilt were measured. The self-benefitmessage was manipulated in the ad by emphasizing ben-efits to the consumer and their immediate family, and theother-benefit was manipulated by emphasizing benefits tosociety and future generations. Findings indicate thatwhen consumers’ PCE is high, the company is perceivedas more socially responsible with other-benefit messages


than with self-benefit messages. On the other hand, whenconsumers’ PCE is low, there is no difference in thecompany’s perceived societal responsibility between self-benefit and other-benefit appeals. It may be that under lowPCE conditions, consumers might believe that as indi-viduals cannot influence environmental problems, thecompany cannot influence them as well. With regard toconsumer guilt, when consumer guilt is high, consumersperceive the company to be more socially responsiblewith other-benefit messages than with self-benefit mes-sages. However, when consumer guilt is low, we find nodifference in the perceived societal responsibility be-tween two different message frames. This may be becauseless guilty individuals have low involvement and are lessconcerned about environmental issues. Finally, when theperceived societal responsibility is high, the attitude to-

ward the brand, purchase intention, as well as word ofmouth is better.

Our study contributes to the literature on green adver-tising by investigating the effect of using different mes-sage frames (self- and other-benefit messages) on per-ceived societal responsibility of the company, and throughit on consumer’s attitude toward the brand, purchaseintention and word-of-mouth intention. In terms of mana-gerial implications, advertisers can motivate consumerswith different message frames and influence their percep-tion about the company’s societal responsibility. A betterperceived responsibility will enhance attitude toward thebrand, purchase and word of mouth intentions. Referencesare available upon request.

For further information contact:Ceren Ekebas

Old Dominion UniversityCollege of Business and Public Administration 2160 Constant Hall

Norfolk, VA 23529Phone: 216.785.1128



THE IMPACT OF VIOLENT HUMOR ON ADVERTISING SUCCESS:A GENDER PERSPECTIVE

Kunal Swani, University of Massachusetts, AmherstMarc Weinberger, University of Massachusetts, Amherst

SUMMARY

Empirical evidence suggests that aggressive humor isnow used in a significant number of television ads (Sharreret al. 2006). In an analysis of over 4,000 broadcasttelevision ads, Scharrer et al. (2006) found some form ofaggression in 12.3 percent of the ads. They further foundthat 53.5 percent of the ads featuring aggression alsocontained elements of humor. This represents a signifi-cant departure from the predominant styles of non-ten-dentious humor that had been common in advertising.Gulas, McKeague, and Weinberger (2010) found in 1989,just 13.6 percent of ads using humor involved aggressiveor disparaging humor, none of which used physical vio-lence. In 1999, aggression was used in almost a third of theads using humor, and physical humor mainly aimed atmen became more common. In 2009, over 70 percent ofthe humorous ads used some form of aggression. Morenotable is that 13 out of 34 of these ads use males as thevictims of physical violence, in contrast to only one ad inwhich women were the victims. The advertisers usingphysical violence with male victims in the 2009 sampleinclude Coke Zero, Bud Light, Audi, Frito-Lay, Doritos,Castrol, SoBe, and Pepsi Max. The pattern illustrates aclear increase in the use of disparagement, particularlyphysical aggression, targeting males over the 20-yearcross section of ads. Examples they highlight are ofwaiters being tripped and falling through plate glasswindows, men being hit in the groin by flying objects, andmen being tossed from meetings out second floor win-dows for forgetting the beer. There are numerous otherillustrations of this genre of violent advertising cloaked inhumor.

Brown and his colleagues (2010) examined the impactof violence in an on-line environment and found that moreextreme violence contributed to greater ad likability,recall and the likelihood that the ad would be passed alongto others in an on-line environment with the implicationthat this form of disparaging humor may benefit advertis-ers. The growing use of violence with humor in advertis-ing begs the question whether the level of violence mat-ters, whether it is accepted by men and women and

whether it works better than more traditional advertisingexecutions. The current study extends Brown’s investiga-tion by looking at the intensity of humor violence in thecontext of gender and in comparison to other non-violentadvertising.

We examine the effect of this emergent aggressivephysical humor in an experiment where male and femalesubjects were exposed four times to either a high or lowintensity violent humor ad and as well as a non-humorousad and an incongruity humorous ad.

Hypothesis H1 is: Viewed by men, violent ads withhumor will be perceived as more humorous, liked more(Aad) and have stronger brand attitudes (Ab) than thesame ads viewed by women. Furthermore, we expect agender by violence intensity humor ad type interaction. Inhypothesis H2 we pose an over-arching question byasking whether violent humor works better or worse thanmore playful humor or no humor at all. We test this issuein the context of gender preferences. The expectation isfor a gender by ad type interaction in H2.

The results support a gender interaction for H1 andH2. The gender differences play out as expected fromprior humor appreciation literature. Men do like ads withhigh intensity violent humor more than women thoughthere is no gender difference for low intensity humorviolence. For the non-violent humor ads, the femalesubjects do like the ads with playful humor more than themen. It is noteworthy that even men liked the ad with theplayful humor more than the ad in the Hi intensity violenthumor condition. The comparison with the non-humor adgives additional caution to using the violent humor formatsince violent humor did not outperform either the non-humor or incongruity humor ads on Aad or Ab. The resultsprovide a tale of caution for advertisers who use violence.Though Brown et al. (2010) suggest that such violencemay be passed along more in an on-line environment andmay be recalled better, the downside risk in comparison toless tendentious humor or no humor appears to be inpossible sacrifice to Aad and Ab, when using high inten-sity violent humor to reach female audiences


For further information contact:Kunal Swani

Isenberg School of ManagementUniversity of Massachusetts

121 Presidents DriveAmherst, MA 01003Phone: 216.212.1851



AD STRATEGY FOR MULTI-ETHNIC MARKETS: THE INFLUENCEOF COSMOPOLITANISM

Enrique P. Becerra, Texas State University – San MarcosSindy Chapa, Texas State University – San Marcos

Delonia O. Cooley, Texas Southern University, Houston

SUMMARY

Multi-ethnic markets present marketers with an adver-tising strategy conundrum. Should they use ethnic spe-cific ads or standardized ads? Ethnic specific ads usespecific cues related to the targeted ethnicity. Standard-ized or integrated ads are not tailored to any specific ethnicgroup. Conflicting information on using ethnic specificads versus standardized ads warrants further scrutiny todetermine the optimal advertising strategy for a multi-ethnic market. Besides, the extant scrutiny of ethnicspecific ads has mostly compared the ad effectiveness ondifferent ethnic groups using ads with ethnic specificspokespersons, instead of comparing the ad effectivenesson different ethnic groups using ads with ethnic specificactors versus using ads with a mix of ethnic actors, whichbetter reflects a multi-ethnic market and can be standard-ized across ethnic groups. Furthermore, in multi-ethnicmarkets, such as in the US, young adults are becomingmore cosmopolitan due to increased inter-ethnic mar-riages and ethnic diversity in schools, neighborhoods andthe work place. Therefore, this study compares, acrossthree different young adult US ethnic groups, the effectsof using integrated ethnic ads, that is, ads using a mix ofethnic spokespersons against the effects of using ads withethnic specific spokespersons on attitudes toward the ad,attitudes toward the brand, and brand purchase intentions.Specifically, the study tests the effects of using an ad withWhite, African American, or Hispanic American actorsversus using an ad with a mix of actors of these threeethnic groups of young adults on an ethnic group’s atti-tude toward the ad, attitudes toward the brand, and brandpurchase intentions. The study accounts for the influenceof cosmopolitanism and prior ethnic contact on ad effects.

Cosmopolitanism engenders greater preference forproducts outside one’s culture and/or community sug-gesting a desire for diversity. Because cosmopolitanismleads individuals to experience products outside their owncommunity and/or culture, cosmopolitanism may alsolead them to expect ads with ethnic and/or culture diver-sity. In other words, cosmopolitanism increases consum-ers’ desire for diversity including ads depicting diversity,such as ads with ethnically diverse spokespersons. Conse-quently, the increased exposure to ethnic groups in theU.S., particularly for young adults, leads to greater cos-mopolitanism and reduces the need for ethnic specific ads

but increases the need for ads reflecting diversity, whichcan be standardized.

The study and the pretests were conducted in severaluniversities in the southern region of the United States; themain study was conducted during a period of 10 months.Each respondent saw an ad out of the four, randomlyassigned regardless of ethnic makeup. So, a White respon-dent saw one of the fours ads for a 4 (ads) x 3 (ethnicgroups) study. Previously validated and published scaleswere used to measure the variables of interest. Ethnicitywas self-reported as customary in this type of studyfollowing the six options (e.g., White American non-Hispanic, African American non-Hispanic, and Hispanic)provided by the census bureau. However, participantswere allowed to choose only one option, including “other”to describe their ethnicity. The data was analyzed usingmulti-group Structural Equations Modeling (SEM) inLISREL 8.72.

The results uphold the premise that cosmopolitanisminfluences ad effects across three different types of ads.Findings indicate that the difference in effects across allthree types of ads is due to the direct and indirect influenceof cosmopolitanism on intentions toward the brand adver-tised. In other words, if cosmopolitanism is not taken intoaccount, advertisers may select ethnic specific ads as thebest option to advertise to young adult ethnic groups in theUS. However, when cosmopolitanism is taken intoaccount, ads displaying diversity exert a greater influenceon intentions toward the brand than ethnic specific ads.Results also indicate that the affinity toward one’s ownethnic group is quite strong and influences the direct effectof attitude toward the brand on intentions toward thebrand advertised. The strong desire for seeing one’sethnicity reflected in the ad may be confusing advertisersif they don’t include cosmopolitanism in their targetmarket analysis. Our findings indicate that as cosmopoli-tanism increases so thus its influence on intentions towardthe brand and this relationship is the strongest for diversityads. Therefore, advertisers could still appeal to the strongdesire for seeing one’s ethnicity reflected in the ad as longas the ad includes diversity features or cues.

The study advances our knowledge on advertising toethnic groups, specifically young adults, who make up agrowing target market in the U.S. However, it needs


further corroboration. Future studies can provide morecredence to our findings by using a sample of adults thatincludes more than just college students. While usingcollege students for studies of this type is valid, college

students only represent a portion of the population ofyoung adults. In addition, college students may have morecontact with diversity than non-college students. Refer-ences are available upon request.

For further information contact:Enrique P. Becerra

Texas State University – San Marcos601 University Drive

San Marcos, TX 78666Phone: 512.245.1277



STRONGER ENVIRONMENTAL NORMS INCREASE “GREEN” BUYINGINTENTIONS BUT NOT BEHAVIOR: IMPLICATIONS

FOR ADVERTISERS

Melissa M. Bishop, University of New Hampshire, DurhamNelson A. Barber, University of New Hampshire, Durham

SUMMARY

Growth in the market for “green” or environmentallyfriendly (EF) offerings has been slower than expected, asgreen attitudes and intentions have not been a goodpredictor of purchase behavior. The most common expla-nations for this gap between attitudes and behavior con-cern the perceived quality of EF produced goods and thehigher price that these products usually entail (Bazoche,Deola, and Soler 2008; Loureiro 2003). Another reasonfor this gap could be due to the way the “green” movementhas been marketed.

Environmental messages and slogans are often pessi-mistic and threatening, attempting to guilt or scare peopleinto doing the “right” thing. By doing so, message sendersare trying to establish, strengthen and appeal to anindividual’s norms – informal rules requiring that oneshould act or should not act in a particular way in a givensituation (Nyborg 2003). For example, some of the mostpopular green slogans from ThinkSlogans.com are asfollows: “Protect our earth today for our children’s tomor-row,” and “Don’t let the water run in the sink, our life’s onthe brink!”

Beliefs based in norms – or normative beliefs (NBs) –are important determinants of intentions to act for a widerange of behaviors, including intentions to engage in pro-environmental behaviors. The main normative beliefspromoted by companies producing EF products espousethat people should simply buy products that are good forthe environment – and avoid those that are not – even if itcosts them more money. However, NBs are individual-specific, such that some “buy into” the movement morethan others, and have thus adopted stronger NBs. Othershave not, and the convictions to purchase EF offerings andbe willing to pay more for them are not held as strongly bythese individuals. Therefore, we propose:

H1: Individuals who indicate higher environmental NBswill state (a) higher purchase intentions, and (b) awillingness to pay (WTP) more for EF offerings thanindividuals with lower NBs.

Norm-Intention-Behavior Linkage

There has been ample evidence across a wide varietyof behaviors to support the linkage between norms and

behavioral intentions. However, there is much less evi-dence and far fewer studies that attempt to validate thenorm-intentions-behavior relationship (Rivis 2009), par-ticularly in the environmental domain, as it is much moredifficult to document actual behavior compared to inten-tions. It is essential to know if intentions derived from NBsresult in a positive relationship with the behavior ofinterest. If NBs only drive intentions but not the corre-sponding behavior, it would be useless to try to appeal toindividuals’ norms to impact behavior. Therefore, beyondpurchase intentions, this study will also examine theimpact of NBs on ultimate purchase behavior.

Study 1

A total of 120 individuals (who met the appropriatecriteria) in Connecticut participated first in a survey, inwhich they answered several questions regarding pur-chase intentions (PIs) and expressed WTP for a bottle ofenvironmentally friendly wine. They also answered spe-cific questions about NBs regarding the purchase of EFofferings. Approximately one month later, participantstook part in a Vickrey auction. The purpose of the auctionwas to measure actual WTP of an EF wine, as measuredby participants’ maximum bid prices.

Survey Results

Using a mean split, individuals were categorized ashigh NBs (n = 76) and low NBs (n = 44), depending on thepressure they reported regarding purchasing EF offerings.High NBs reported significantly greater PIs (M = 5.34) ofan EF wine than low NBs (M = 3.03) [t(118) = 6.884, p <.05]). This supports H1a. Further, individuals expressedhow much in dollars they would be willing to pay for EFwine (M = $21.49). Those reporting high NBs expressedhigher WTP (M = $24.55), while low NB individualsindicated they would pay significantly less (M = $16.38)[t(118) = 7.822, p < .05]. This supports hypothesis 1b.

Auction Results

Comparing stated WTP (survey) to actual WTP (auc-tion), high NBs significantly overstated their actual WTP,with a stated WTP of $24.55 and an actual WTP of $17.65[t(76) = 7.21, p < .05]. On the other hand, low NBs did notoverstate their actual WTP. The $16.75 auction bid was


not statistically different from the price they expressed topay during the survey ($16.38) [t(44) = .667, p > .05].Comparing the actual WTP amounts between the twogroups, high NBs were willing to pay $17.65 while thosewith low NBs were willing to pay $16.75, though thedifference was not significant [t(118) = 1.10, p >.05].

Study 2

A second study was conducted to essentially repli-cate the findings from Study 1 and to extend them tolocally produced wine. Very similar results were foundamong the 98 new participants, such that those withhigher NBs (concerning EF or locally produced offerings)did report higher PIs and greater stated WTP. However,high NBs were found to overstate their actual WTP again,while low NBs did not for both EF and locally producedwine. Further, no significant differences were foundbetween groups in terms of their actual WTP.

Discussion

Intentions based on NBs that are founded in guilt and/or fear may not translate to behavior, and this may be thereason for such a large gap between what people say andwhat they actually do. The purpose of this research is toshow that those who do evidence stronger normativebeliefs to buy EF and locally produced offerings mayindicate greater intentions, but this does not actuallytranslate to behavior; thus, a large disparity occurs. Alter-natively, those less influenced by normative pressure (lowNBs) did not overstate in a survey what they wouldactually pay in practice. Further, there was no significantdifference between the NB groups in their actual WTP foran EF or locally produced product. These findings poseproblems for the norm-intention-behavior link. Hence,the primary implication for advertisers (and organiza-tions, etc.) is that those who spend their time and moneytrying to establish, create, or increase normative pressurethrough the use of guilt or fear appeals may be wastingtheir resources. References are available upon request.

For further information contact:Melissa M. Bishop

University of New Hampshire15 Academic Way

Durham, NH 03824Phone: 603.862.3370



#LOWESHATESMUSLIMS: CONSUMER ETHICAL DECISION MAKINGRELATED TO CORPORATE ADVERTISING CHOICES

R. Nicholas Gerlich, West Texas A&M University, CanyonKristina Drumheller, West Texas A&M University, Canyon

Emily Kinsky, West Texas A&M University, CanyonMeagan Brock, West Texas A&M University, CanyonMarc Sollosy, West Texas A&M University, Canyon

ABSTRACT

When Lowe’s Home Improvement bowed to pres-sure from Florida Family Association and pulled its adsfrom TLC’s All-American Muslim, #loweshatesmuslimsquickly hit Twitter feeds, forcing Lowe’s to respond.Using the theory of planned behavior, SEM showed thatsubjective norms affected whether consumers planned toavoid purchasing from Lowe’s.

INTRODUCTION

When Lowe’s Home Improvement (Lowe’s) decidedto pull its advertising from TLC’s All-American Muslim,a firestorm swelled as #loweshatesmuslims soon appearedon Twitter. In an effort to manage its image, Lowe’sapologized on its Facebook page, resulting in thousandsof vitriolic responses – so many that Lowe’s pulled themfrom the page and replaced its original statement with asecond apology. It is imperative that organizations usecare in managing their brands, but social media presentadditional challenges when organizations respond to ex-ternal stakeholders. Pulling advertising from TLC’s All-American Muslim created an organizational crisis forLowe’s. Drawing on the work of Shaw, Shiu, Hassan,Bekin, and Hogg (2007) to study ethical decision makingrelated to planned behavior, this study seeks to understandconsumer decision making in relation to the advertisingand social media choices of organizations, in this case,Lowe’s. Consumers are more likely to purchase fromorganizations that are viewed as socially responsible(Wigley 2008; Seeger 1997), which likely includes ethi-cal organizational responses in brand management efforts.

Branding is the process an organization engages in tocreate an identity that is unique among competitors (Aaker1997; Keller 1998; de Chernatony 1999; Martin, Johnson,and French 2010). Brand management involves maintain-ing one’s corporate identity with varied and multiplestakeholders (Cheney 1991; de Chernatony 1999), likelyby aligning the corporation’s goals and vision with thosevalued by the larger social system (Seeger, Sellnow, andUlmer 1998). However, organizational decision-makingis then naturally limited by the values espoused. As thenotion of corporate identity has evolved, many corpora-

tions understand that they now must be responsible andresponsive to the public. Communication processes func-tion to lay the foundation for value claims while creatinga unique sense of self, or identity for the organization(Cheney and Christenson 2001). “One of the externalchallenges is engendering confidence amongst multiplestakeholders about their brand’s ability to consistentlydeliver valued outcomes” (de Chernatony 1999, p. 158).As corporations interact with external and internal stake-holders, organizational boundaries become blurred, espe-cially with the rise of social media.

With the advent of social networks such as Facebookand Twitter, organizations have realized the importanceof communicating with their stakeholders and managingtheir brands via social media. Facebook, which recentlycelebrated its eighth anniversary, has as its mission: “togive people the power to share and make the world moreopen and connected” (Facebook Info 2012). Facebooknow has more than 800 million users (Velaigam 2012).Twitter, another popular social networking site, waslaunched in 2006. By March 2008, Twitter had 1.3 millionregistered users, and by September 2011, that number hadblossomed to 200 million with 50 million active users ona daily basis (Buck 2011). Besides individual users, manyorganizations have their own Facebook pages and Twitterstreams. These social networking sites, among others,have become a new channel for relationship buildingbetween organizations and their stakeholders. Thus, reach-ing this socially networked world of consumers is essen-tial for any brand, but it has moved beyond simple promo-tion. Benjamin (2011) states, “Gone are the days whensocial media was all about marketing through Twitter andFacebook. Brands are increasingly using the medium tobe responsive and contactable and to resolve customerissues in real time” (p. 33). More companies are develop-ing, enhancing, and monitoring their brands via socialmedia (Magee 2012). The effectiveness of such socialmedia use depends on how organizations communicatetheir brand in a variety of situations.

Two companies that were recently named top socialmedia users by PRWeek include Carnival and BP, whichlike Lowe’s, have both used social networks related tocrises. Magee (2012) said BP has focused much attention


on its social media channels since the 2010 DeepwaterHorizon crisis. With regard to Carnival, after the recentCosta Concordia accident off the coast of Italy, “Facebookhas been the most active channel for discussion of theaccident” (p. 26). While Costa and Carnival’s socialmedia channels are active, “Carnival’s corporate site isnot playing a significant role” (Magee 2012, p. 26). Thissuggests a giant shift in corporate communication forbrand management. Lowe’s also utilized Facebook tointeract with its stakeholders, specifically to respond toconsumer reaction to its advertising actions, the outcomeof which was controversial.

Lowe’s Under Fire

TLC, a subsidiary of the privately held DiscoveryCommunications, Inc., described All-American Muslimas a program that “follows the daily lives of five AmericanMuslim families in Dearborn, Michigan, one of the mostestablished and largest concentrations of American Mus-lims in the country and home to the largest mosque in theUnited States.” Eight All-American Muslim episodes werecreated, and were aired between November 13, 2011 andJanuary 8, 2012. According to Allen (2011), All-American Muslim averaged 1.2 million weekly view-ers and had a high of 1.7 million viewers for the premierepisode. The December 11 episode, the day after Lowe’spulled its support, dipped to under one million (Levin2011).

The program faced criticism from those concernedthat the program might convert others to follow Islam, butalso from Muslims who thought that it did not showtraditional Muslim lifestyles or the variety of Muslimbeliefs and backgrounds (Allen 2011). The criticism,coupled with an economic downturn (Norris 2012), likelyreinforced the need to be concerned about advertisingsupport for a controversial program. Shortly after theprogram’s airing, a Florida-based religious organizationknown as the Florida Family Association (FFA) spoke outagainst the program, stating that it “riskily hides theIslamic agenda’s clear and present danger to Americanliberties and traditional values” (FFA 2012). Reacting topressure from the FFA, Lowe’s decided to pull its adver-tising support from the program as of December 10, 2011.Although FFA contacted more than 100 companiesencouraging them to pull advertising from the program,and many did, Lowe’s seemed to take the brunt of theattack from users of social media. Although social mediarelationships are an important part of brand management,interactions on social media sites can impact consumerplanned behavior, particularly when stakeholders ques-tion the ethical dimension of corporate decisions.

Theory of Planned Behavior and Consumer Ethics

Central to the theory of planned behavior (TPB) isthat individual intentions to engage in a particular behav-

ior indicate the likelihood of actually engaging in thatbehavior in the future (Ajzen 1985, 1991). The theory ofplanned behavior extends the theory of reasoned action(Ajzen and Fishbein 1980), by adding the concept ofbehavioral intentions. Behavioral intentions are a func-tion of three distinct constructs: attitudes, subjective norms,and perceived behavioral control. Work on TPB by Ajzenand colleagues found that the more favorable the attitudesand subjective norms, and the greater the perceived behav-ioral control toward a particular behavior, the stronger theintention for performing that behavior will be. Whenindividuals have a stronger motivation or intention toengage in a behavior, the TPB proposes that individualsare more likely to follow through with that behavior.Building on the work of Ajzen and others, Peruigini andBagozzi (2001) contend that desire to engage in a focalbehavior is the antecedent of intent to engage in saidbehavior. Desire is defined as “a state of mind whereby anagent has a personal motivation to perform an action or toachieve a goal” (Peruigini and Bagozzi 2001, p. 71).

Shaw and Shiu (2003) extended the theory of plannedbehavior to develop a model for understanding ethicalconsumer choices. “Ethical concerns are often ongoingand irresolvable” (Shaw and Shiu 2003, p. 1486), requir-ing more effort in decision-making. Most models ignoresocietal-centered concerns in favor of hedonic ones. Inorder to address the complexity of ethical considerationsrelated to intent to behave in a particular manner, Shaw etal. (2007) proposed the need to examine attitude (ATT),subjective norm (SN), and perceived behavioral control(PBC), as well as ethical obligation and social identity asantecedents of planned behavior. These antecedents werethen tested in the context of purchasing goods manufac-tured by sweatshop labor (Shaw et al. 2007).

Critics of TPB argue that motivations and volitionunderlying behavioral intent have received little atten-tion; Shaw et al. (2007) thus view the model from theperspective of plan and desire being distinct from inten-tion. In particular, their model looks specifically at desire(DES), intention (INTENT), and plan (PLAN) to avoidpurchasing sweatshop clothing. In their study, “desirewas found to be pertinent in fully mediating the effect ofattitude and partially mediating the effect of subjectivenorm on intention” (p. 2). Moreover, desire was found tobe distinct from intention. In fact, it was found to be arequirement, as attitude alone did not impact intention.Specifically, consumers have to desire the “positive atti-tude” associated with avoiding sweatshop clothing beforeintent kicks in. Furthermore, important others “can serveto impact personal motivation to act in terms of desire bypositively supporting personal motivation or through nega-tively influencing desire to avoid sweatshop apparel”(p. 2). Discourse related to corporate decisions and iden-tity is facilitated by social media.


Social media have the potential to affect TPB becauseof the introduction of multiple voices exerting pressure onuser attitudes. Group norms and social identity, whenadded to TPB related to online communities, were foundto significantly predict behavioral intentions (Bagozzi,Dholakia, and Mookerjee 2006). This suggests that onlinediscussions could crystallize norms and identity, whichcan influence ethical consumer decision-making. Inter-estingly, it has been found that subjective norms failed tosignificantly predict intentions related to online commu-nities (Bagozzi et al. 2007), and were not sufficient forcapturing the means by which consumer self and socialidentity influence consumer intentions (Thorbjørnsen,Pedersen, and Nysveen 2007). Extending this researchrelated to the potential influence of social media discourseon TPB, and consonant with Shaw et al. (2007), it ishypothesized:

H1: (a) ATT, (b) perceived SN, and (c) PBC will have astrong positive relationship with DES to avoid pur-chasing at Lowe’s.

H2: (a) ATT, (b) perceived SN, and (c) PBC will have astrong positive relationship with INTENT to avoidpurchasing at Lowe’s.

H3: The relationship between (a) ATT, (b) perceived SN,(c) PBC and INTENT is mediated by DES.

H4: The relationship between (a) ATT, (b) perceived SN,(c) PBC and PLAN is mediated by DES andINTENT.

METHOD

Procedure and Sample

Individuals at least 18 years of age were solicited viasocial media sites such as Facebook and Twitter, as wellas other electronic media to complete an electronicQualtrics survey. The survey included an assessment ofparticipant demographics, Facebook usage, TV viewing,and Lowe’s shopping behavior and opinions. In addition,participants were asked to read an embedded article aboutthe Lowe’s decision to pull its advertising money from theTV show All-American Muslim, as well as the text of theFacebook apology Lowe’s posted shortly afterward. Next,participants were asked to complete a variety of scalesadapted from the theory of planned behavior, as well asdemographic items and other measures assessing theirview of the ethicality of Lowe’s decision.

A total of 446 surveys were received, although 67were eliminated because of being incomplete. The finaloperational sample (N = 379) was evenly balanced ongender (48% male and 52% female), and the majority

(85.4%) were Caucasian (with 4.3% Asian American, 4%Latino, 3.2% African American, 1.1% Native American,and 2.1% other). Participants’ ages ranged from 18 to 76,with a mean age of 35. Participants self-identified asmarried (46.7%), not married (47.2%) or in a domesticpartnership (6.1%). In addition, they identified as Demo-crats (34.9%), Republicans (28%), Independents (32.3%),or Libertarians (4.8%). Participant education ranged froma doctorate (5.9%), through master’s (14.4%), bachelor’s(35.1%), some college (37%), and high school (7.7%).Finally, all 50 states and Washington, D.C. were repre-sented in the sample with the most common location beingTexas (29.4%).

Measures

Demographics. Items were collected on eachparticipant’s age, gender, ethnicity, marital status, educa-tion, political affiliation, and state of residence.

Planned Behavior (TPB). Scale items in the TPBsurvey were adapted to fit the Lowe’s scenario, with theprimary goal of assessing respondents’ planned avoid-ance or non-avoidance of Lowe’s. Wording of these itemswas as consistent as possible to that set forth in Shaw et al.(2007), whose application regarding purchasing productsmanufactured by sweatshop labor was modified to reflectattitudes and intent toward purchasing from Lowe’s fol-lowing the company’s controversial decision to stop adver-tising on All-American Muslim. The Cronbach’s alpha forthe full scale was .76.

Consistent with prior studies (e.g., Ajzen and Fishbein1980; Ajzen and Madden 1986; Shaw et al. 2007), overallattitudes toward Lowe’s were measured by askingrespondents to evaluate the behavior using five 7-pointsemantic differential scales. Participants used the follow-ing adjectives to appraise Lowe’s: (a) foolish – wise; (b)harmful – beneficial; (c) boring – exciting; (d) enjoyable –unenjoyable (reverse coded); and, (e) good – bad (reversecoded). Two additional items were added to the attitudeconstruct, in order to reflect nuances of the apology issuedon Facebook. These two items, though, were ultimatelyeliminated after Exploratory Factor Analysis (using SPSS)and Confirmatory Factor Analysis (using AMOS) wereperformed on the data. The attitude subscale of the TPBhad a Cronbach’s alpha of .91.

Subjective norms (SN) (e.g., Ajzen and Fishbein1980; Ajzen and Madden 1986; Shaw et al. 2007) wereassessed by asking whether respondents believed thatthose who are important to them think they should shopwith Lowe’s and whether those referent others wouldapprove of this behavior. Responses to these two itemsranged from 1 should to 7 should not. The subjectivenorms subscale of the TPB had a Cronbach’s alpha of .81.


Perceived behavioral control (PBC) (e.g., Ajzen andFishbein 1980; Ajzen and Madden 1986; Shaw et al.2007) was measured using three items. Consistent to pastresearch using the TPB (e.g., Ajzen and Madden 1986),PBC assessed the participants’ perception of control overshopping at Lowe’s. Responses were made on a 7-pointscale (1 strongly disagree to 7 strongly agree). The per-ceived behavioral control subscale of the TPB had aCronbach’s alpha of .81.

Desire to avoid purchasing at Lowe’s was measuredusing two-items. Responses were made on a 7-point scale(1 strongly disagree to 7 strongly agree). The desiresubscale of the TPB had a Cronbach’s alpha of .97.

Behavioral intentions (BI) (e.g., Ajzen and Fishbein1980; Ajzen and Madden 1986; Shaw et al. 2007) weremeasured with two items asking participants their likeli-hood of avoiding making purchases at Lowe’s and intentof avoiding purchasing at Lowe’s. Respondents used a 7-point scale ranging from 1 very unlikely to 7 very likely.The behavioral intentions subscale of the TPB had aCronbach’s alpha of .88.

Finally, planned behavior was measured using two-items. Consistent with past research, the plans partici-

pants have made to avoid shopping at Lowe’s wereassessed. Responses were made on a 7-point scale (1strongly disagree to 7 strongly agree). The desire subscaleof the TPB had a Cronbach’s alpha of .96.

RESULTS

Confirmatory factor analysis (CFA) and structuralequation modeling (SEM) analyses were conducted usingAMOS 4.0 software (Wothke and Arbuckle 1996) withmaximum-likelihood estimation. Structural EquationModels were calculated for a Base Model (ATT, SN andPBC effect on PLAN), Partially Mediated Model 1 (ATT,SN, and PBC relationship to PLAN as mediated by DES),Partially Mediated Model 2 (ATT, SN, and PBC relation-ship to PLAN as mediated by INTENT), and Fully Medi-ated Model (ATT, SN and PBC relationship to PLAN asmediated by DES and INTENT). An examination of thedifferences between the measured model (CFA) and theproposed models (SEM) revealed reasonably good over-all fits. Table 1 compares the goodness of fit between themodels, along with other relevant measures. A compari-son of the four proposed models and the measured model(CFA) supported the fully mediated model as providingthe most acceptable path relative to the measured model.This was determined by comparing the X2s and the normed

TABLE 1Model Comparison

Partially- Partially- Fully-Base Meadiated Mediated Mediated Measured

Model Model 1 Model 2 Model Model

SEM CFA

X2 82.147 132.559 109.809 167.630 158.539

Degrees of Freedom 38 58 58 79 75

Probability .000 .000 .000 .000 .000

GFI .961 .949 .957 .944 .948

RMSEA .055 .058 .049 .054 .054

CI of RMSEA .039-.072 .045-.071 .034-.062 .043-.066 .042-.066

Normed X2 2.162 2.285 1.893 2.122 2.113

Incremental Fit Measures

NFI .974 .971 .974 .971 .973

CFI .986 .983 .987 .984 .985

RFI .962 .960 .965 .962 .962

Parsimony Measure

AGFI .933 .920 .932 .915 .916

PNFI .673 .722 .724 .731 .695


X2s of the models. The fully mediated model presented nosignificant change of the X2 (9.091) nor a significantincrease in the normed X2, with only an increase of .009from the measured X2 of 2.113 to 2.122 for the fullymediated model. The model also returned a CFI of .984(CFI; Bentler 1990) and a RMSEA of .054 (RMSEA;Steiger and Lind 1980) both indicative of acceptable fit(Bentler and Bonett 1980; Browne and Cudeck 1992; Huand Bentler 1999). This suggested no other alternativepath from that of the proposed model was needed.

Table 2 reports the regression weights for the pathmodel, with Figure 1 depicting the relationships. Resultsshow that ATT had a weak yet significant negative rela-tionship to DES, while DES mediated INTENT. The roleof SN is significant as it affected DES, and subsequentlyINTENT and PLAN. Based on these results, support wasthus found for all hypotheses except H1a, H2a, and H2c(see Table 2).

DISCUSSION

The negative relationship between ATT and DESindicates that respondents’ views toward Lowe’s apologyare of little or no consequence in their decision to avoid/not avoid purchasing at Lowe’s. Instead, it is peer pressure

as reflected in SN that is wielding the most influence. Inother words, the strong positive relationship between SNand DES, and subsequently INTENT and PLAN, indicatethat participants are more concerned with the perceptionof how their peer group will view them, rather than theirown attitudes toward the store.

Furthermore, the weak relationship between PBCand DES indicates that being able to avoid shopping atLowe’s is not the issue. It may have been different ifeverything the respondent needs is at Lowe’s, but this isnot likely the case. Instead, the results show that, for thissample, it is the perception of what others might think thatis driving shoppers to declare their avoidance of Lowe’s.

The findings point to the impact social media canhave on people’s planned behavior. The majority ofinformation disseminated about Lowe’s decision wasdelivered via social media by both the company andthousands of users which, attesting to the legitimacy ofthese media as well as the inherent risks. The public forumof a corporate Facebook page, as well as the completelyunregulated and un-moderated Twitter feed, allow any-one with an opinion to voice it easily. It is entirely possiblethat the results reported herein are a result of just thisphenomenon.

FIGURE 1Fully Mediated Path Model

ATT

SN

PBC

PLAN

INTENT

DES-0.15

1.07

0.15

-.02

0.12

0.72

1.06


It is also possible that the effects of SN are magnifiedin the social media era because the notion of one’s peergroup has been redefined. No longer are individualslimited to face-to-face interaction, nor temporal nor geo-graphic exchanges. Instead, social media allow for theinteraction between a person and his/her friends acrossmultiple layers of one’s life, and, in the case of corporateand organization pages, with people completely unknownto the user. These layers are connected easily by socialmedia, and can span decades and continents.

It is evident that brand perception (in this case, that ofLowe’s) is affected by SN via social media. Whereas 10–20 years ago individuals would be exposed to subjectivenorms only through their temporal spheres of face-to-facecontact, they are now exposed to the perceived views ofmany by virtue of computer mediated contact. The impactin this instance is a plan to avoid shopping at Lowe’s.Furthermore, that SN would outweigh an individual’sown attitudes is a significant indicator of how plannedbehaviors (and by extension, brand perception) can beaffected via social media.

This study is limited in that it is but a small samplelimited to self-reported Facebook users. It is possible thatmany other shoppers became aware of the controversywithout benefit of having seen the dialogue on Facebook

or other social media. There may also be geographicdifference, given that Lowe’s is a more dominant com-petitor in the U.S. South.

The controversy occurring immediately before thewinter holidays may also have reduced the impact onpersons than had it occurred when there were fewer socialdistractions. Shoppers may have been more attuned togift-buying than participating in boycotts. Still, the sampleis diverse geographically, with good representation acrossage groups and gender, and there is little reason to suspectthat any other sample would respond differently.

Finally, it should be considered that PLAN is notnecessarily ACTION. In other words, while respondentsmay indicate they will avoid purchasing at Lowe’s, onlya longitudinal study would allow for examination ofactual purchase (or non-purchase) behavior resulting fromthe controversy. It is possible that shoppers might avoidshopping at Lowe’s in the short-run, only to return therein the long-run; it is also possible that, while shoppersexhibit the effect of peer pressure when in the context ofplanned behavior, there may be a disconnect when itcomes to actually refraining from their stated plan. Afuture examination of Lowe’s sales data at both theaggregated and same-store level would be able to reveallong-term effects.

TABLE 2Fully Mediated Model

Hypothesis Construct Construct Estimate S. E. C.R. p Conclusion

Η1a DES ΑΤΤ −0.15 0.06 − 2.60 .009 not supported

H1b DES SN 1.07 0.08 14.22 *** supported

H1c DES PBC 0.15 0.08 2.01 .044 supported

H2a INTENT ATT -0.09 0.03 -3.31 *** not supported

H2b INTENT SN 0.12 0.05 2.51 .012 supported

H2c INTENT PBC -0.02 0.04 -0.66 .509 not supported

H3 INTENT DES 0.72 0.04 16.47 *** supported

H4 PLAN INTENT 1.06 0.05 22.61 *** supported


REFERENCES

Aaker, Jennifer L. (1997), “Dimensions of Brand Person-ality,” Journal of Marketing Research, 34 (August),347–56.

All-American Muslim. TLC. Retrieved Feb. 3, 2012, from[http://blogs.discovery.com/all_american_muslim/].

Allen, K. (2011, December 19), “TLC Show BringsMuslims in America out in the Open,” USA Today[online], p. A1. Retrieved February 6, 2012, from[http://www.usatoday.com/news/nation/story/2011-12-18/muslims-america-series/52057894/1].

Ajzen, I. and M. Fishbein (1980), Understanding Atti-tudes and Predicting Social Behavior. EnglewoodCliffs, NJ: Prentice-Hall.

____________ (1985), “From Intentions to Actions: ATheory of Planned Behavior,” in Action-Control:From Cognition to Behavior, J. Kuhl and J. Beckman,eds. Heidelberg: Springer, 11–39.

____________ and T.J. Madden (1986), “Prediction ofGoal-Directed Behavior: Attitudes, Intentions, andPerceived Behavioral Control,” Journal of Experi-mental Social Psychology, 22 (5), 453–74.

____________ (1991), “The Theory of Planned Behav-ior,” Organizational Behavior and Human DecisionProcesses, 50, 179–211.

Bagozzi, R.P., U.M. Dholakia, and A. Mookerjee (2006),“Individual and Group Bases of Social luence inOnline Environments,” Media Psychology, 8, 95–126.

Benjamin, K. (2011), “Social Media – Battle of theBrands,” Revolution, (November 30), 33. RetrievedFeb. 10, 2012, from LexisNexis.

Bentler, P.M. and D.G. Bonett (1980), “SignificanceTests and Goodness of Fit in the Analysis of Covari-ance Structures,” Psychological Bulletin, 88, 588–606.

____________ (1990), “Comparative Fit Indexes in Struc-tural Models,” Psychological Bulletin, 107 (2), 238–46.

Browne, M.W. and R. Cudeck (1992), “Alternative Waysof Assessing Model Fit,” Sociological MethodsResearch, 21 (2), 230–58.

Buck, S. (2011), “A Visual History of Twitter,” (Septem-ber 30). Retrieved February 14, 2012, from [http://mashable.com/2011/09/30/twit ter-history-infographic/].

Cheney, G. (1991), Rhetoric in an Organizational Soci-ety: Managing Multiple Identities. Columbia, SC:University of South Carolina Press.

____________ and L.T. Christensen (2001), “Organiza-tional Identity: Linkages Between Internal and Exter-nal Communication,” in The New Handbook ofOrganizational Communication: Advances in Theory,Research, and Methods, L.L. Putnam and F.M. Jablin,eds. Thousand Oaks, CA: Sage, 231–69.

de Chernatony, L. (1999), “Brand Management ThroughNarrowing the Gap Between Brand Identity andBrand Reputation,” Journal of Marketing Manage-ment, 15 (1/3), 157–79, doi: 10.1362/026725799784870432.

Facebook Info. (2012), Retrieved February 14, 2012,from [http://www.facebook.com/facebook#!/facebook?sk=info.].

Florida Family Association (2012, January 12), “Emailsto All-american Muslim Advertisers Made a Differ-ence: 101 out 112 Companies Did Not Return to theShow,” Retrieved February 9, 2012, from [http://floridafamily.org/full_article.php?article_no=108].

Hu, L. and P.M. Bentler (1999), “Cutoff Criteria for FitIndexes in Covariance Structure Analysis: Conven-tional Criteria Versus New Alternatives,” StructuralEquation Modeling: A Multidisciplinary Journal, 6(1), 1–55.

Keller, Kevin Lane (1998), “Branding Perspectives onSocial Marketing,” in Advances in ConsumerResearch, Joseph W. Alba and J. Wesley Hutchinson,eds. Provo, UT: Association for Consumer Research,25, 299–302.

Levin, G. (2011), “TLC’s All-American Muslim CouldProfit from Protests,” USA Today, Retrieved Febru-ary 10, 2012, from [http://www.usatoday.com/life/television/news/story/2011-12-13/All-American-Muslim-boycott/51885502/1], 3D.

Magee, K. (2012), “Social Media – The FTSE 100 SocialMedia Index,” PRWeek, Retrieved February 10, 2012,from LexisNexis, 26.

Martin, K.D., J.L. Johnson, and J.J. French (2011), “Insti-tutional Pressures and Marketing Ethics Initiatives:The Focal Role of Organizational Identity,” Journalof the Academy of Marketing Science, 39, 574–91.doi: 10.1007/s11747-010-0246-x.

Norris, F. (2012, February 4), “A Bleak Outlook forLong-Term Growth,” The New York Times, RetrievedFebruary 10, 2012, from LexisNexis.

Perugini, M. and R.P. Bagozzi (2001), “The Role ofDesires and Anticipated Emotions in Goal-DirectedBehaviors: Broadening and Deepening the Theory ofPlanned Behavior,” British Journal of Social Psy-chology, 40, 79–98. doi: 10.1348/014466601164704.

Seeger, M.W. (1997), Ethics and Organizational Com-munication. Cresskill, NJ: Hampton Press.

____________, T.L. Sellnow, and R.R. Ulmer (1998),“Communication, Organization, and Crisis,” Com-munication Yearbook, 21, 231–75.

Shaw, D. and E. Shiu (2003), “Ethics in Consumer Choice:A Multivariate Modeling,” European Journal of Mar-keting, 37 (10), 1485–98. doi: 10.1108/03090560310487202.

____________, ____________, L. Hassan, C. Bekin, andG. Hogg (2007), “Intending to Be Ethical: An Exami-nation of Consumer Choice in Sweatshop Avoid-ance,” Advances in Consumer Research (ACR),Orlando, FL.


Steiger, J. H. and J.C. Lind (1980), “Statistically-BasedTests for the Number of Common Factors,” Paperpresented at the annual Spring (May) meeting of thePsychometric Society, Iowa City, IA.

Thorbjørnsen, H., P.E. Pedersen, and H. Nysveen (2007),“‘This Is Who I Am’: Identity Expressiveness and theTheory of Planned Behavior,” Psychology and Mar-keting, 24, 763–85. doi: 10.1002/mar.20183.

Velaigam, M. (2012, February 2), Facebook and theSocial Media Boom. Investors Chronicle. Retrieved

February 10, 2012, from LexisNexis.Wigley, S. (2008), “Gauging Consumers’ Responses to

CSR Activities: Does Increased Awareness MakeCents?” Public Relations Review, 34 (3), 306–08.

Wothke, W. and J.L. Arbuckle (1996), “Full-InformationMissing Data Analysis with Amos,” in AdvancedStructural Equation Modeling, Techniques: Issuesand Techniques,G.A. Marcoulides and R.E.Schumakcer, eds.Mahwah, NJ: LEA.

For further information contact:R. Nicholas Gerlich

West Texas A&M UniversityWT Box 60809

Canyon TX 79016Phone: 806.651.2492



BRAND HERITAGE IN THE LUXURY INDUSTRY: CREATING ANDDELIVERING CONTINUOUS VALUE TO CONSUMERS

Nadine Hennigs, Leibniz University Hannover, GermanySteffen Schmidt, Leibniz University Hannover, Germany

Thomas Wuestefeld, Leibniz University Hannover, GermanyKlaus-Peter Wiedmann, Leibniz University Hannover, Germany

SUMMARY

Heritage brands stand for longevity and authenticity,as proof that the core values and performance of the givenproducts are reliable. Focusing on the market for luxurygoods where the origin and heritage of a brand is some-thing that consumers are increasingly aware of, the aim ofthe present study is to analyze the value drivers andoutcomes of brand heritage and luxury brand perception.The empirical results show significant effects of brandheritage and luxury brand perception on customer per-ceived value related to the given brand.

In the luxury industry, most brands are based oncraftsmanship in family business with well-known found-ing fathers; their history is deep-rooted and authentic whatcreates the heart of brand heritage. The heritage aspect isa crucial part of a luxury brand as it has to appear bothperfectly modern to the society of the day and at the sametime laden with history. Starting from an integral cus-tomer perceived value concept, to investigate the relation-ships between brand heritage, brand luxury perceptionand customer perceived value, in our questionnaire, weused already existing and tested measures for brand heri-tage, brand luxury perception, and customer perceivedvalue. In detail, the measures for brand heritage and brandluxury are by definition and based on the nature of theircontent conceptualized as formative drivers causing thelatent constructs, whereas the reflective indicators ofcustomer perceived value are defined as being caused bythe underlying construct. All items were rated on five-point Likert scales and specified to the brand CHANEL,one of the world’s leading luxury brands with a strongheritage and a broad product range encompassing bothhaute couture fashion as well as accessible goods such asaccessories, makeup, jewelry, fragrances, skincare etc.For data collection, an Internet survey with a snowballsampling method was employed. The interviewees wereinvited to actively participate in the survey via personal-ized emails and links on selective web pages (e.g., Facebookprofile pages). A total of 333 valid questionnaires inGermany were received in winter 2011. All respondentshave bought or used the brand CHANEL at least once toensure sufficient brand usage knowledge.

To examine the drivers and outcomes of brand heri-tage as well as brand luxury and to empirically test ourhypotheses, we used Partial Least Squares (PLS) struc-tural modeling. Due to the fact that our conceptual modelincludes formative and reflective constructs, PLS wasconsidered as the appropriate method. Our PLS pathmodel for antecedents/drivers and consequences/reflec-tors of brand heritage and brand luxury perception deter-mines the indicators that create brand heritage as well asbrand luxury perception using formative measurementand the four types of consumer perceived value usingreflective measurement.

With reference to the evaluation of the formative andreflective measurement models, the results revealed satis-factory values. To test the significance of the relationsbetween the latent variables, we applied a nonparametricbootstrapping procedure. The results of the test of ourhypotheses revealed the following insights: In H1, apositive effect of brand heritage on brand luxury waspostulated. These relation is positive and significant (p <.01), providing full support for H1. Moreover, the postu-lated effects of brand heritage as well as brand luxuryperception on all four dimensions of customer perceivedvalue are positive and significant (p < .01). Consequently,these empirical findings are in line with H2a to H2d andH3a to H3d.

The results indicate that brand heritage has a strongimpact on brand luxury. Besides, both constructs have astrong effect on all examined dimensions of customerperceived value. Moreover, the results give evidence thatthe respective formative indicators of brand heritage aswell as brand luxury differ in their impact upon thecorresponding construct. Brand managers should concen-trate their marketing activities on indicators with thegreatest importance to build and sustain a meaningfulbrand.

In a global economy, where competitive products orcounterfeits are easily available, brand managers shouldidentify and concentrate on the specific value dimensionthat is regarded as the most important driver of consump-tion for their brand. Based on this, the market communi-


cation should stress the perceived values and emphasizethe benefits of the given brand over competing brands orfake products. Even if low-cost counterfeit luxuries allowtheir buyers to be in tune with fashion without spending anexorbitant amount of money, a counterfeit product willnever be able to provide the same pleasure or satisfy theindividual need for sensory gratification. Consumers whoplace importance on the heritage aspect or hedonistic andmaterialistic product features might have a negative atti-tude toward a counterfeit purchase because they are awareof the self-deceiving aspect of this behavior. In order to be

successful and to obtain a high perceived value in theircustomers’ eye, luxury brand managers will have toaddress all relevant value dimensions: To be consideredas a luxury brand in the eyes of the customers, it’s aboutunderstanding the customers’ evaluation and accentuat-ing the brand appropriately to appeal to both their cogni-tive needs and affective desires. In sum, successful luxurybrands balance the timelessness of brand heritage withinnovative market communication and brand positioningto address contemporary consumers’ needs and valueperception. References are available upon request.

For further information contact:Nadine Hennigs

Leibniz University HannoverKoenigsworther Platz 1 Hannover

Lower Saxony, 30163Germany

Phone: 49(0)511.762.4862Fax: 49(0)511.762.3142



LOSS OF BRAND EQUITY IN CRISES: THE IMPACT OF EMOTIONSAND ATTRIBUTIONS IN PRODUCT AND NON-PRODUCT

NEGATIVE PUBLICITY

Haodong Gu, The University of New South Wales, AustraliaAshish Sinha, The University of New South Wales, Australia

SUMMARY

Companies expend a substantial amount of resourceto become distinctive in the mind of consumers throughbranding. Brand equity is a fragile strategic asset thattakes many years to build, the value of which could be lostovernight due to negative publicity and the over-reachingpower of mass media. Due to its strategic importance,restoring brand equity is essential for companies mired innegative publicity.

Extant literature examined the link between negativepublicity and brand equity (Dawar and Pillutla 2000; Jing,Dawar, and Lemmink 2008; Klein and Dawar 2004;Laczniak, DeCarlo, and Ramaswami 2001). However,several research questions remain unanswered. First, theimpact of consumer emotions on brand equity is unex-plored. Modern consumers no longer simply buy productsand services. Instead they purchase the emotional experi-ence around what is being sold (Fournier 1998; Morrisonand Crane 2007). While the term emotional branding hasreceived wide-spread attention from marketing practitio-ners for years, to date it is still a construct seldom re-searched by academic researchers. By linking consumers’post-scandal emotions and brand equity, we highlight apreviously veiled example of emotional branding. Sec-ond, extant research usually looks at instances of product-harm crises. However, the effect of negative publicity isby no means uniform. Different types of negative public-ity may influence brand equity in different ways, andrequires tailored responses from the company (Pullig,Netemeyer, and Biswas 2006; Tybout and Roehm 2009).It is still unclear whether consumers adopt similar pro-cessing strategies when facing different scandals, espe-cially product and non-product scandals. Product scandalis defined as the wide-spread negative information thatindicates the defect or danger embedded in the products orservices, whereas non-product scandal is defined as thereported egregious social performance of the company orthe brand. For example, product-harm crisis is a type ofproduct negative publicity, and excessive carbon emis-sion is a type of non-product negative publicity. In theliterature, the former is better understood and more inten-sively researched than the latter. The comparison betweenthe two, however, is rare. Third, when investigating theemotions triggered by product and non-product scandals,it should be noted that the product-based negative public-

ity activates emotions that are mostly self-focused, wherethe value of the brand in the eyes of its customers isdamaged by the scandal. On the other hand, the emotionsarise from non-product scandals are mostly other-focused(moral emotions), where the value of other stakeholdersor the society is impaired. The impact of moral emotionson brand equity has not been examined before.

We propose two types of negative publicity: productand non-product, as well as two reaction processes: emo-tions and attributions, and test the influence of theirinterplay on various accounts of brand equity, includingbrand association, quality perception, and brand loyalty.After negative publicity happens, both attributions andemotions can be triggered in consumers’ minds (Hartel,McColl-Kennedy, and McDonald 1998). Attributions andemotions can influence brand association and qualityperception through a learning process (Keller 1993; vanOsselaer and Janiszewski 2001). Brand loyalty, as thedependent variable, is influenced by brand associationand quality perception (Aaker 1991; Keller 2003). Withthis conceptual framework, this paper aims to test therelative importance of attributions and emotions in prod-uct and non-product negative publicity. We postulate thatthe decision process in product related scandals is fea-tured by higher involvement than that of non-productscandals. As a result, attributions (emotions) should forma more (less) informative learning cues in product thannon-product negative publicity. On the other hand, infor-mation from non-product scandals should be less infor-mative to make judgement on perceived product quality,so both attributions and emotions of non-product scandalsshould be less influential on brand equity compared to thereactions to product negative publicity.

A time series Structural Equation Model (SEM) wasconstructed to reflect the change in brand equity duringthe scandal exposure, and the impact of consumer emo-tions and attributions on brand equity. The structuralequations approach may be applied to repeated data overtime (Congdon 2003). We included the pre-scandal brandequity in the model to solve the problem of individualheterogeneity in assessing the brand equity. A hypotheti-cal winery brand was used to test the hypotheses. Theproduct negative publicity described a whistle blowerexposing that the winery company had switched a lower-quality wine for its medal-winning Decanter range, which


was being widely sold on the market. In the non-productscandal scenario, the winery company was involved in afatal workplace injury, in which a 20-year-old part-timeemployee was killed by a snake bite while working in thevineyard, because she had not been provided propersafety equipment. There were no co-workers or supervi-sors around to help her when the tragedy happened. Thesurvey data obtained from the responses of 525 Austra-lians indicates that: (1) both emotions and attributions caninfluence brand equity in product negative publicity; (2)attributions influence all the dimensions of brand equitywhereas emotions only influence brand association in anon-product scandal; (3) emotional impact is similarbetween the two types of scandals, but attributional im-pact is stronger in the product scandal except for the effecton the perception of product quality; (4) the impact ofnegative publicity on consumers’ willingness to pay andbrand loyalty is fully mediated by brand association andquality perception.

Our study recommends managers to recognize thatthere are different types of scandals. Therefore, response

strategies should be tailored to be consistent with the typeof scandals. For example, in a product-harm crisis, con-sumer attributions are of more importance than a non-product scandal. Thus, a defensive reaction might bejustifiable to decrease unnecessary suspicious on corpo-rate integrity. In contrast, in a social scandal like environ-mental pollution, actions such as affective repair mightappease the public emotions by offering possibilities foremotion or problem-focused coping. Several areas are leftas avenues for future research. First, we eliminate ratherthan study the consumer heterogeneity in our model.Future studies can investigate the impact of age, gender,and personality on the links between negative publicityand brand equity. Second, there might be other informa-tion-processing routes in addition to emotions and attribu-tions. The identification of such routes can improve theunderstanding of corporate scandals. Third, we did notdiscuss the relationship between attributions and emo-tions in our paper due to the complexity and potentialreciprocity. Future studies can adopt longitudinal designsto disentangle the connection between the two. Refer-ences are available on request.

For further information contact:Haodong Gu

School of MarketingThe University of New South WalesASB, UNSW, Sydney, NSW 2052

AustraliaPhone: +61.2.9385.1612



INVESTIGATING HOW WORD-OF-MOUTH DRIVES INFORMATIONSEARCH FOR A BRAND

Andrew M. Baker, San Diego State UniversityNaveen Donthu, Georgia State University, Atlanta

V. Kumar, Georgia State University, Atlanta

SUMMARY

In this study we investigate how word-of-mouthabout a brand (WOM) influences subsequent informationsearch activity by the WOM recipient. We assess theimpact of both WOM-level and brand-level characteris-tics in a series of hierarchical models using data from a (1)longitudinal, nationally representative panel that tracksconsumer brand-related WOM of more than 1,000 brandsand (2) brand-level measures brand equity (Equitrend).Key findings from this study include the following: (1)positive WOM has an asymmetrically larger absoluteeffect on information search than negative WOM, (2)online WOM tends to drive more information search thanWOM received offline, although the positive impact ofthe online channel is attenuated somewhat in the case ofnegative WOM, (3) WOM received from strong socialties tends to motivate information search more than WOMreceived from weak social ties, but this difference isattenuated if the conversations are generally negative orpositive about a brand, (4) information search is a morelikely consequence of both positive and negative WOM ifbrand has higher overall brand equity, and (5) informationsearch is a more likely consequence from a WOM episodewith a weak social tie if the brand has higher overall brandequity.

Based on attraction/avoidance theories and informa-tion search behavior in the face of skepticism, we hypoth-esize the following:

H1&2

: Positive WOM has a positive influence on inten-tions to seek additional information, and negativeWOM has a negative effect. Positive WOM has agreater absolute effect.

Electronic marketplaces have been characterizedas platforms that substantially lower search costs(Bakos 1997) and thus increase consumer informa-tion search relative to offline channels (Lynch, Jr.and Ariely 2000).

H3: WOM occurring offline tends to decrease inten-

tions to seek additional information compared withonline WOM.

Furthermore, it is likely that the arousal/avoidanceeffect of message valence on consumer information search(H

1) will be moderated by the cost of searching out

additional information about the brand.

H4: The positive effect of positive WOM on intentions to

seek additional information is attenuated when theconversation occurs offline, whereas the negativeeffect of negative WOM on intentions to seek addi-tional information is accentuated when the conversa-tion occurs offline. Researchers have also linked howWOM disperses across consumer groups to market-ing performance. Systematic differences in social tiecommunication styles between weak and strong tiessuggest that how the brand-related valence in a WOMconversation affects intentions to seek additionalinformation will be moderated by the social tiestrength.

H5: The positive effect of positive WOM on intentions to

seek additional information is attenuated when theconversation occurs between strong social ties,whereas the negative effect of negative WOM onintentions to seek additional information is accentu-ated when the conversation occurs between strongsocial ties.

Brands with characteristically high levels of brandequity are expected to be relatively more well-known andhave more overall positive associations about them (Keller2001). Consequently, when a WOM recipient receivesnegative sentiment about a well-known and generallyliked brand, there should be a greater motivation to seekadditional information about the brand to seek confirma-tion or disconfirmation of the brand sentiment.

H6: Compared with brands with lower levels of overall

brand equity, brands with higher levels of overallbrand equity will have higher levels of intentions toseek additional information when a consumer re-ceives negative WOM about the brand.

For brands with high levels of brand equity, thesearch behaviors of the WOM recipient should be lessimpacted by who transmitted the WOM.


H7: Compared with brands with lower levels of overall

brand equity, brands with higher levels of overallbrand equity have higher levels of intentions to seekadditional information when a consumer receivesWOM from a weak social tie.

The WOM conversation data analyzed in this studycomes from a proprietary database developed and main-tained by a marketing consulting firm specializing inproviding clients with highly detailed information aboutU.S. consumers’ WOM activity both offline and online.

The valence of the WOM conversations about abrand is reported by the panel respondent as being eitherpositive, negative, neutral, or mixed. We coded the re-ported communication medium as either offline or online.BrandChat participants also identify their relationshipwith the person (coded as strong or weak tie). The 2007and 2008 BrandWeek Superbrands annual special reportswere used to compile brand equity measures. The depen-dent variable, seeking additional brand-related informa-

tion, is an 11-point scale using the BrandChat panelparticipants’ self-report of their propensity to seek addi-tional information about the brand as a result of the WOMconversation. Six hundred and fifty-four brands (level 2)were selected for analysis from the BrandChat databasecomprising a total of 188,510 conversations (level 1).This full sample was used to evaluate H

1 – H

5. Not all

brands in the BrandChat database were included in theEquitrend database. A subsample of brands was used toevaluate H

6 and H

7. A total of 226 brands were represented

in this sub analysis comprising 125,460 WOM conversa-tions.

A random coefficient regression model was used totest the hypotheses. The population regression intercept isallowed to vary across brands. Likewise, the main effectsof WOM valence, channel, and social tie strength allowedto randomly deviate across brands. H1, H2, H3, and H5were supported. H4 was not supported. H6 was partlysupported and H7 was fully supported. References areavailable upon request.

For further information contact:Andrew M. Baker

College of Business AdministrationSan Diego State University

5500 Campanille DriveSan Diego, CA 92182Phone: 619.594.4306

Fax: 619.594.3272


EMPLOYEES AS INTERNAL BRAND CONSUMERS: THE NATUREOF RELATIONSHIPS

Lucy Gill-Simmen, Imperial College London, United KingdomAndreas B. Eisingerich, Imperial College London, United Kingdom

SUMMARY

In the past few decades, as the role of the brand hasevolved, so too has the role of the employee in branddelivery. In particular, the employee is considered a brandenabler and co-creator of brand value (Merz, Yi, andVargo 2009). If the free press is any indication, companiessuch as Starbucks, Ritz-Carlton, Zappos, and SouthwestAirlines appear to have achieved the goal of turningemployees into brand champions, i.e., those who consis-tently deliver on the brand promise, and in so doing haveset themselves apart from the rest within highly competi-tive market situations. Thus the employee as an advocatefor his/her own brand has achieved heightened attentionof late, such that the employee may be viewed as a brandconsumer in his/her own right.

However, despite a wealth of research into consumerrelationships with brands (e.g., Aaker 1997; Fournier1998), with some notable exceptions (see Hughes andAhearne 2010; Morhart, Herzog, and Tomczak 2009;Wieseke et al. 2009) how employees relate to their ownbrands and subsequently exhibit specific brand-behaviorsis poorly understood. Particularly, little attention has beendirected toward the concept of the employee as an internalbrand consumer (for exceptions see Gilly and Wolfinbarger1998; Berry 1995; Schneider and Bowen 1985; Merz, Yi,and Vargo 2009).

The aforementioned brand success stories raise thefollowing questions:

1. What is the nature of the employee-brand relation-ship?

2. What mechanisms prompt employees to connectwith and to identify with their brands?

3. How may companies enable brand-specific behav-ior?

The aim of our study is to address these questions byinvestigating the nature of the relationship between thebrand and the internal consumer and more specifically toadvance a conceptual framework and propositions thatexplain the role of brand benefits, employee self-brandconnections and employee brand identification in elicit-ing brand-specific behaviors.

Given the relative paucity of literature on employeesand brands we adopt a grounded theory approach andconduct a series of depth interviews to access employees’insights and experiences with the brand. The themes thatemerged were used to construct a conceptual framework.

Our findings propose that, in the same ways asconsumers, (a) employees seek symbolic and emotionalvalue beyond functional value from the brand they workfor, (b) employees draw on brand associations in the formof benefits to assess their self-congruence with the brand,(c) employees form employee self brand connections andthey identify with the brand, and (d) employees exhibitbrand specific behaviors dependent on levels of employeeself brand connections and employee brand identifica-tion.

Our study makes a number of key contributions.Firstly, it expands on existing consumer brand relation-ship theory (Fournier 1998; Aggarwal 2004) to provide abroad view of the relationship components between brandsand internal consumers, i.e., employees. Our findingscontribute to self-congruence theorization in consumers(Aaker 1999; Sirgy 1982) in suggesting employees drawon brand associations in this case brand benefits whichthey “use” to form connections based on congruence withtheir self-concept which in turn lead to brand identifica-tion. Our research is distinct in bridging the dividebetween consumers, i.e., those consuming brands outsideof the company and those inside the company. Secondly,by positing that self-brand connections and brand identi-fication are the constructs which lie at the heart of therelationship employees appear to have with their brand,the study goes some way to addressing scholars’ calls toidentify the antecedent conditions for brand specific behav-iors (Morhart, Herzog, and Tomczak 2009). In addition toexpanding the consumer brand relationship researchdomain, our research contributes to the fairly recentscholarship on how and why employees interact with theirown brands and how this interaction affects their brand-specific behaviors (Hughes and Ahearne 2010; Morhart,Herzog, and Tomczak 2009; Wieseke et al. 2009). Finally, besides new theoretical implications for aca-demic research, the research has important managerialimplications, to maximize the positive effects of emplo-yee and brand relationships, managers must realize theirown capabilities in enabling brand specific behaviors. Ifmanagers can maximize the brand value proposition for


employees the connection and identification with thebrand that follows will have a positive effect on theirbehaviors toward the brand. Furthermore, managers mustadopt the view that brand-building is not only an externalconsumer-oriented strategy but indeed an internal con-

sumer strategy. A limitation of this study is that it exam-ines employees of particularly well-known brands. Futureefforts should examine employees of lesser-known brands.References are available upon request.

For further information contact:Lucy Gill-Simmen

Imperial College Business SchoolImperial College London

LondonSW7 2AZ

United KingdomPhone: +44(0)7876350671



TO BE FAMILIAR OR TO BE THERE? BRAND FAMILIARITY, SOCIALPRESENCE, AND ONLINE RETAILING

Xiaoping Fan, Zhejiang University, ChinaFeng Liu, California State University – Sacramento

Jia Zhang, Zhejiang University, China

SUMMARY

While there is little doubt about the importance ofweb store image on online retailing, which factors andhow such factors may influence the establishment of astrong web store image, consequently the formation ofconsumers’ purchase intention, however, are still not wellunderstood. In this study, by employing a theoreticalframework based on the stimulus-organism-response(SOR) paradigm, we investigate the roles of both brandfamiliarity and social presence regarding web store emo-tional image improvement and, sequentially, consumers’online purchase intention.

The stimulus-organism-response (SOR) paradigm,developed by Mehrabian and Russell (References includ-ing this are available on request), fits nicely with theonline marketing context, as online consumers are con-stantly facing various website features and other factorsthat are critical in shaping consumers’ vendor evaluationsand their subsequent purchase intentions. This study con-siders brand familiarity (the number of brand-relateddirect or indirect experiences that have been accrued bythe consumer) and social presence (interactivity and viv-idness) as the stimuli that activate online consumers’affective and cognitive processes, because these twofactors are widely documented in the literature as relatedto web store image improvement (the nature of the rela-tionships, however, has never been depicted conclusively).The cognitive states investigated in this study are webstore images perceived by consumers. In alignment withthe SOR paradigm, we focus on the emotional store imageand its three major dimensions: usefulness, entertainmentand trustworthiness. Finally, online purchase intentionserves as the response construct in our application of SORparadigm. The sequence of effects in our theoreticalmodel is that brand familiarity and social presence (S)influence consumers’ perceptions of web store emotionalimage (O), which in turn influences customers’ onlinepurchase intentions (R). Based on this model, we developthe following hypotheses,

H1: Increased brand familiarity will result in greater webstore emotional image.

H2: Increased social presence will result in greater webstore emotional image.

H3: Increased brand familiarity will result in greater pur-chase intention in a web store.

H4: Increased social presence will result in greater pur-chase intention in a web store.

H5: Web store emotional image will mediate the relation-ship between brand familiarity/ social presence andconsumers’ online purchase intention.

We develop a 2X2 between-subjects factorial designexperiment on Business-to-Consumer apparel retailingsites, conducted with 160 subjects, all undergraduatestudents in a large public university in China. Two levelsof both brand familiarity (low versus high) and socialpresence (low versus high) were manipulated as indepen-dent variables, generating four experimental conditions.All the constructs included in the study are measured ona seven-point Likert scale. Scale reliabilities are evaluatedusing Cronbach’s Alpha (α). We develop nine web storeemotional image items and conduct a factor analysis withVarimax rotation (the measurements are deemed appro-priate by checking KMO and the Barlett’s test of spheric-ity) to identify entertainment, trustworthiness, and useful-ness. Furthermore, we conduct manipulation checks on allthe three constructs.

Our analysis reveals that social presence affects bothweb store emotional image and purchase intention, whereasthe influences of brand familiarity on these constructs arelimited. We also find that among the three dimensions ofweb store emotional image, only the entertaining dimen-sion plays a mediating role in the relationship betweensocial presence and purchase intention. The other aspectsof store image, although important in determining pur-chase intention, do not intercede or change the way howsocial presence affects purchase intention. These findingshave significant managerial implications. We also discusslimitations and some interesting avenues for futureresearch.


For further information contact:Feng “Oliver” Liu

College of Business AdministrationCalifornia State University – Sacramento

6000 J StreetSacramento, CA 95819–6088

Phone 916.278.7106E-Mail: [email protected]


DO PERSONALITY TRAITS INFLUENCE BRAND RELATEDACTIVITIES IN SOCIAL NETWORKS? AN

EMPIRICAL STUDY

Philipp A. Rauschnabel, University of Bamberg, GermanyBjörn S. Ivens, University of Bamberg, Germany

Gunnar Mau, University of Siegen, Germany

SUMMARY

Facebook is getting more and more important all overthe world. More than 700,000 registered users use thisonline social network to build up and maintain interper-sonal relationships and to spend their freetime. Compa-nies can use Facebook as a new channel to interact withtheir stakeholders, especially with customers. Most of thecompanies use so called “fanpages” (synonym: brandpage). Registered users can connect their personal pro-files with those fanpages and become a “fan” of a brand byclicking the “I Like”-Button. Some brands, for exampleRed Bull, have more than 30 million fans on Facebook.

Marketers can use those fanpages as a channel forpublishing contents to their fans and for interacting withthem. For example, they can publish pictures of newproducts and ask their fans about their opinion. However,only a part of the registered users is fan of brands onFacebook or even actively engaging on those brand pages.But who are those active fans? In this empirical study, weexamine the relationship between the “Big Five” person-ality traits and brand related behavior on Facebook. Resultsare tested on a sample of 865 German “brand fans” onFacebook (mostly students) using moderated regressionanalyses.

Prior research, mostly from computer science, com-munications, or psychology investigated the relationshipbetween the user personality and private usage behavioron the internet, especially in social media. Summarizingthe findings from those studies, especially extraversionplays an important role in how people use social media. Tothe best of our knowledge, however, no research on theeffects of personality traits on brand related behavior insocial media is available. Hence, the scholarly call forsuch research remains still unanswered.

We measure the independent variables, the “BigFive,” with a reliable German scale with five items foreach dimension. Based on qualitative interviews we iden-tified possible brand related activities on fanpages. Thoseactivities were aggregated to indices based on the three-dimensional structure of “consumer online brand relatedactivities (“COBRAs”) from Muntinga et al. (2011). Con-

suming consists of passive activities, such as watchingbrand related photos or reading reviews from other fans.Contributing consists of activities with a medium level ofinteraction, for example to vote in polls or take part indiscussions with other fans. Creating is the strongest formof COBRAs, which includes activities like publishingown brand related pictures or writing reviews.

Results from regression analyses reveal that open-ness to experience and conscientiousness both have asignificant effect on all three dimensions of COBRAs.However, the predictive powers of the analyses are low(all R² < 5%).

In a second step, we included moderators into ourmodel. The results are surprising and lead to a significantincrease of the predictive power: The effects are, with oneexception, only significant within the female respondents.Additionally, Facebook-involvement was identified as amoderator: The effects of the personality traits on theCOBRAs are stronger when users show a high involve-ment with Facebook.

The results have important theoretical and manage-rial implications: To the best of our knowledge, this is thefirst study examining the relationships between personal-ity traits and brand related activities in social networks.Additionally, the study gives suggestions why the explan-atory power of prior studies was very low: most of theother studies did not include moderators in their analyses.The findings also underline the importance of the samplein personality research. Samples with a high share of malerespondents might lead to weak or even insignificanteffects. Additionally, our results are also contrary to theprivate usage of social networks: in the field of brandrelated activities, extraversion plays a less important role.Marketers can use these findings to optimize their com-munication.

Future studies should adopt brand related character-istics (e.g., emotional brand attachment or brand love) andmotivations. As studies from clinical psychologistsreveal, Facebook usage has an impact on several con-structs related to social well-being, as Facebook helpsusers to maintain their personal networks. It would beinteresting to analyze if brand related activities on fanpages


also have an impact on those constructs. This seems to beplausible, as users can build up relationships with userswho have similar (brand) interests, or, in a more abstractway, with a brand. Furthermore, fans have the opportunity

to fulfill themselves by creating contents and to getpraised by other fans. Finally, replications in other coun-tries and with representative samples should be con-ducted. References are available upon request.

For further information contact:Philipp A. Rauschnabel

Otto-Friedrich-University BambergFeldkirchenstrasse 21

96052 BambergGermany

Phone: +49(0)951.863.3864Fax: +49(0)951.863.5566



IDENTITY PRESENTATION AND CONFLICT IN A SOCIAL NETWORK:IMPLICATIONS FOR BRANDING IN SOCIAL GAMES

Ginger Killian, University of Georgia, AthensJohn Hulland, University of Georgia, Athens

SUMMARY

Reasons for consumer participation in social net-works vary widely; these motivations can influence anindividual’s online identity as well as his or her willing-ness to interact with brands in a social network setting.McCann (2010) suggests that while some users may viewsocial networks as a tool to learn about new products,others view them as a way to build relationships and hangout in a virtual setting. As social networks become moreintegrated into consumers’ daily lives, they increasinglybring together more diverse groups of individuals. Thepresent research seeks to understand how consumersmanage the individual communities within which theyparticipate as well as the varying digital identities theyreveal to other network members. As a user’s socialnetwork size increases to accommodate new friends andinterests, she may feel conflict in selecting a social identitythat simultaneously satisfies multiple communities. Thiscan create a tension between presenting a single, consis-tent identity to all users versus managing distinct identi-ties aimed at different community groups. Such a tensionoften arises in the context of social gaming, a context weuse to investigate the questions posed above.

Methodology

In order to provide a rich description of the socialgaming community culture, a combination of netnographyand interviews was used (Kozinets 1998). Analysis ofinterview findings followed procedures outlined by Corbinand Strauss. Participant interviews were open-ended andexplored a variety of topics related to self-presentation,coping mechanisms, and impressions of brand relatedadvertising. Interview participants were selected from adiverse group of gamers, and interviews were conducteduntil saturation occurred and additional interviews gener-ated no additional insights. A total of 18 social networkgamers agreed to take part in this research.

Sources of Conflict

Social networking sites often bring vastly differentgroups of individuals together, which can create internalconflict in attempting to satisfy the conflicting expecta-tions of various groups simultaneously (Biddle 1986).Users reported role conflict in managing three areas oftheir self presentation. The most prominent conflict that

users manage is the one that exists between the profes-sional and personal presentation of self. Second, Facebookalgorithms suggest potential Friends based on a user’sexisting network of Friends, often resulting in users receiv-ing Friend requests from colleagues, friends, acquaintan-ces, and even strangers. Conflict arises in assessing whetheror not to allow these individuals to be a part of one’snetwork. On the one hand, users do not wish to offendsomeone who requests ‘Friend’ status. On the other hand,they may also feel that access should be limited to thosewho are known in real life. Finally, gamers suggest thatself -presentation issues are exacerbated by the differ-ences between friends and family. Interactions with fam-ily members are likely to be quite different than those withfriends, and expectations between the two groups alsovary considerably.

Coping Mechanisms

Our research identified three types of coping mecha-nisms used to address the tensions described above. First,users cope with various tensions by determining whichindividuals will be allowed into their social network.Once Friends have been granted access, tension can stillarise in determining how to present oneself to others. Thesecond type of coping involves censoring oneself to herentire network by carefully choosing objects for display.Several informants noted that they try to be more moder-ate on Facebook than they are in face-to-face, recognizingthe diverse audience to which they communicate online.The final type of coping mechanism is to segregate indi-viduals into distinct groups and to then present different,targeted information to each group. Some participantsutilize the group function as a courtesy to others who mayfind some posts irrelevant or annoying.

What Does This Mean for Brands and MarketingManagers?

Social network game-related advertising campaignscan be grouped into four categories that fall along twodimensions. The first dimension refers to the action con-text. Advertising campaigns can require gamers to takeaction online or offline. When the action is online, thegamer experiences minimal disruption in her activity flowand is easily able to maneuver between the activitiesrequested by the ad and the game itself. Offline actionsdisrupt the gamer’s behavior and may cause internal


conflict as the gamer attempts to keep her game-relatedworld separate from the rest of her life. The seconddimension is related to the game context. An advertiser’scampaign may take place solely within the game or mayrequire gamers to exit the game (but remain online) inorder to complete a task. Campaigns housed solely withina social network game allow gamers to keep game relatedbehaviors private, whereas out-of-game campaigns maycause conflict in requiring gamers to publicly show theyare playing games. The two dimensions create a typologyof brand related advertising that can be used to generatediffering levels of benefits for firms. We label the fourcategories created by these dimensions as Awareness, Co-branding, Fans, and Sales.

The first category (“Awareness”) involves an adver-tising campaign designed to take place strictly within thegame. Campaigns in this category build awareness byrequiring gamers to complete brand-related in-game tasks,having similar (but stronger) effects to those observed forproduct placements in movies and video games. Further-more, unlike traditional product placements, the directeffects of these campaigns can be assessed. This categoryis likely to minimize the tension associated with self-

presentation to distinct social network communities. Thesecond category (“Co-branding”) is designed to generateoffline traffic related to the game. This cell representsretailers looking to gain exposure to social network gamerswithout creating their own extensive online presence. Thebenefits of this cell are primarily to Zynga and the retailercarrying the card. As was the case for category one, thisstrategy is not likely to create increased cross-communityself-presentation conflict for the gamer. The third cat-egory (“Fans”) is focused on generating an increase insocial network traffic for the brand. Gamers are requiredto leave the game and complete a separate online task –such as becoming a brand Fan—in order to receive anin-game benefit. Unlike categories one and two, however,the “Fan” strategy publicly identifies the gamer as a brandFan to the members of her social network, thereby increas-ing her need to manage the potential for self-presentationconflict. The final category (“Sales”) is designed to gen-erate brand sales offline. With gamers already willing tospend cash online within a game, social network gameadvertising can be a profitable venture for brands. How-ever, this category also has the potential to induce self-presentation conflict for gamers. References are availableupon request.

For further information contact:Ginger Killian

University of GeorgiaAthens, GA 30605

Phone: 404.545.4512Fax: 706.542.3738



RISK REDUCING PORTFOLIO EFFECTS: UNCERTAINTY AVOIDANCEIN BRAND EXTENSIONS

Matthew A. Hawkins, ESADE, Ramon Llull University, SpainJatinder Jit Singh, ESADE, Ramon Llull University, Spain

ABSTRACT

An individual’s level of uncertainty avoidanceimpacts brand extension evaluations but uncertainty avoid-ance is composed of two sub-dimensions: risk aversionand ambiguity tolerance. This study finds high uncer-tainty avoidance individuals prefer extensions from broadbreadth brands over narrow brands which conflicts withprior studies and this finding holds across sub-dimen-sions.

INTRODUCTION

Introducing a brand extension is a strategic but riskyendeavor. Extensions build upon the parent brand’sequity and can benefit from positive attributes transfer-ring to the extension (Boush and Loken 1991; Keller1998; Milberg, Park, and McCarthy 1997; Nijssen andAgustin 2005). Thus, the expected promotional costs arelower and trial rates are higher for brand extensions thanwhen introducing the product under a new brand (Keller1998; Keller and Aaker 1998; Klink and Athaide 2010).However, to obtain these benefits the parent brand’simage and future sales are offered as collateral (Dacin andSmith 1994). Brand image can be unintentionally adjus-ted due to a change in the portfolio composition or worsethe image can be negatively impacted if the extension fails(Keller 1998; Milberg, Park, and McCarthy 1997). Priorresearch has found that brand extension attitudes fluctuateacross countries (Bottomley and Holden 2001; Buil, deChernatony, and Hem 2009; Martínez, Montaner, andPina 2009). Additionally, openness to innovation has alsobeen found to impact extension evaluations (Klink andAthaide 2010); and, innovation acceptance and uncer-tainty avoidance are both cultural and personality traits(de Mooij 2004; Xie 2008). Therefore, understandinghow uncertainty avoidance tendencies vary across con-sumer markets is a vital concern for brand managers.

Current research has established that consumers’uncertainty avoidance tendencies impact brand extensionevaluations; but, uncertainty avoidance is composed oftwo sub-dimensions: risk aversion and ambiguity toler-ance (Sharma 2010). Despite repeated calls for in-depthinvestigations into culture’s role on new product intro-ductions (Bottomly and Holden 2001; Buil, de Chernatony,and Hem 2009; Laroche 2007; Xie 2008) the authors areunaware of any study that investigate the role these sub-

dimensions have on extension similarity, portfolio breadthand attitude ratings. In particular, Henseler and colleagues(2010) found that consumers’ level of uncertainty avoid-ance impacts extension evaluations. However, they pro-vided no insight into which component of uncertaintyavoidance leads to evaluation variations.

Additionally, prior research has found that consum-ers prefer similar extensions from brands with a narrowportfolio and dissimilar extensions from wide breadthbrands (Boush and Loken 1991). However, there is lim-ited research on the impact uncertainty avoidance and itssub-dimensions have on brand extensions from parentbrands with varying portfolio breadth. Thus, this papercontinues to advance the understanding of brand exten-sion acceptance at both the higher uncertainty avoidancelevel and at the sub-dimension level and on parent brandswith varying product portfolios breadths.

Accordingly, a review of the current understandingof brand extensions is presented. Then, the methodologyand data analysis sections are detailed. The paper con-cludes by summarizing the major findings and offeringfuture research ideas stemming from potential study limi-tations.

EXTENSIONS AND UNCERTAINTYAVOIDANCE

Uncertainty Avoidance

Categorization theory, the underlying assumptionbehind brand extension research, argues that the positiveassociations of the parent brand will transfer to the new,extended product (e.g., Boush and Locken 1991; Klinkand Smith 2001; McCarthy, Heath, and Milberg 2001;Völckner and Sattler 2006). The importance of categori-zation theory is supported by the consistent finding thatthe more similar the extension is to the parent brand thehigher extension favorability (e.g., Aaker and Keller1990; Keller and Aaker 1998; Völckner and Sattler 2007).While it has been argued that consumers might view thefirm as lacking the manufacturing skills necessary toproduce a quality product if the extension is too dissimilarfrom the parent brand (Aaker and Keller 1990), morerecent research indicates that the lack of similarity limitsthe formation of attribute transferring schema between theparent and extension (Lau and Phau 2007), as categoriza-


tion theory argues. Thus, the more distant the extensionthe more likely consumers form extension valuationsconsidering a limited amount of parent brand attributescreating purchase risk thereby elevating the importance ofuncertainty avoidance.

A respondents’ level of uncertainty avoidance (UNA)reflects their willingness to try new products and exten-sions. In particular, consumers from cultures with lowUNA tend to be more accepting of brand extensions,regardless of fit, over consumers from a culture with highUNA (de Mooij 2004; Henseler et al. 2010; Klink andAthaide 2010; Klink and Smith 2001; Xie 2008). de Mooij(2004) and Klink and Smith (2001) argue that consumerswho are open to innovation and accept uncertainty are lessconcerned with a lack of information prior to making adecision. Current research indicates that consumers moreaccepting of innovation provide more positive attituderatings for both similar and dissimilar extensions. Addi-tionally, it should be noted that this study measuresextension attitude with behavioral-based indicatorsrelated to product purchase which should favor risk takingconsumers. Accordingly, the following dual-hypothesisis proposed:

H1: Low UNA respondents will provide higher extensionattitude ratings than high UNA respondents whenrating similar (dissimilar) extensions.

Uncertainty Avoidance Sub-Dimensions

Uncertainty avoidance can be decomposed into twosub-dimensions: risk aversion (RSK) and ambiguity tol-erance (AMB) (Sharma 2010). Prior research has sug-gested that the acceptance of risk is the primary driver forconsumers’ acceptance of distant extensions (Henseleret al. 2010; Klink and Athaide 2010) however no studyhas directly investigated the impact of these two posi-tively correlated uncertainty avoidance sub-dimensions.Thus, hypothesis one investigates these sub-dimensionsat the aggregated level and the following hypothesisaddress the sub-dimensions individually.

It is commonly accepted that trying a new product isrisky. Further, it is argued that early adopters tend to bemore open to risk (Yorkston, Nunes, and Matta 2010).This study conceptualizes risk aversion as “the extent towhich people are reluctant to take risk or make riskydecisions” (Sharma 2010, p. 791). Under thisconceptualization, extensions lacking a perceived fit couldbe risky to purchase because there is less informationtransference to base a decision. Then, this lack of informa-tion generates purchase risk impacting extension evalua-tions. For similar extensions, less RSK respondents willprovide higher favorability ratings because they are com-fortable with the risky decision. Further, for dissimilarextensions the difference between low and high RSK

respondents should be more pronounced; again, becausethese consumers are not as impacted by the lack ofinformation transference. Accordingly, the followinghypothesis is proposed:

H2: Low RKS respondents will provide higher extensionattitude ratings than high RSK respondents whenrating similar (dissimilar) extensions.

Prior research (e.g., Henseler et al. 2010; Yorkston,Nunes, and Matta 2010) has not distinguished betweenacceptance of risk and acceptance of ambiguity but ratheruse the concepts interchangeably. However, Henseler andcolleagues (2010) and de Mooij (2004) indicate thatavoidance of ambiguity leads to lower rates of newproduct acceptance. Sharma (2010) operationalizes ambi-guity as “the degree to which people can tolerate ambigu-ity and uncertain situations” (p. 791). Thus, trying a newextension creates a situation in which the performanceand quality levels are ambiguous. Further, as the per-ceived similarity between a parent brand and extensiondecreases, ambiguity related to perceived product perfor-mance should increase due do the reduced amount ofparent brand attributes transferring to the extension. There-fore, the following hypothesis is proposed:

H3: Low AMB respondents will provide higher exten-sion attitude ratings than high AMB respondentswhen rating similar (dissimilar) extensions.

Portfolio Breadth

Prior research has found that portfolio breadth alsoimpacts extension favorability ratings. In particular, Boushand Loken (1991) found that similar extension fromnarrow portfolio brands receive higher favorability thanfrom broad breadth brands, with breadth measured by sizeof product portfolio not brand image. Conversely, theyfound that dissimilar extensions received higherfavorability ratings when introduced by broad breadthbrands over narrow brands. The enhanced extend-abilityof broad brands can be attributed to consumers interpret-ing past success as an indicator of ability to producequality products (Aaker 1996a) and developing moreattribute transferring schema, collectively across numer-ous products. Thus, narrow and broad breadth brandsbenefit from their size, albeit under different circum-stance. However, prior cultural studies typically don’tdistinguish between portfolio breadths (e.g., Henseleret al. 2010; Martínez, Polo and de Chernatony 2008) and,thus, have neglected investigating the effect portfoliobreadth and uncertainty avoidance have on extensionevaluations. Maintaining consistency with prior findingsthe following hypotheses are proposed:

H4a: When rating similar extensions, respondents willprovide higher extension attitude ratings for nar-


row breadth brands than broad brands, regardlessof uncertainty tendencies.

H4b: When rating dissimilar extensions, respondentswill provide higher extension attitude ratings forbroad breadth brands than narrow brands, regard-less of uncertainty tendencies.

METHODS AND DATA COLLECTION

Pretests

To determine the global brands and extensions forthose brands, a series of pretests were conducted based onMartínez, Polo, and de Chernatony (2008). In pretest one,international graduate students (n = 33), who were spe-cifically sought out to obtain a wide range of globalbrands, were asked to write down two global brands, one

narrow and one broad portfolio brand. Next, the respon-dents listed two similar and dissimilar extensions for eachglobal brand. The pretest resulted in two narrow and threebroad portfolio brands to be tested in pretest two. Inpretest two, another group of international graduate mar-keting students (n = 44) rated one narrow and one broadbreadth brand and the fit of their respective extensions.Extension similarity was measured using one global simi-larity indicator. The respondents also rated the perceivedportfolio breadth in terms of size: two indicators to mea-sure portfolio breadth (To your knowledge, (parent brand)sells products in how many different product categories?;Do you consider (parent brand) to have a related orunrelated product portfolio?) were original to this study.The third indicator for portfolio breadth was based onMartínez, Montaner, and Pina (2009) and Boush andLoken (1991) (How many products does (parent brand)sell?). Nestlé was selected to be the broad brand with a

TABLE 1Main Construct Indicators

Construct Source Indicator Scale

Monga & John 2007 How consistent do you feel (extension) with the 7pt Likert: veryother products offered by (parent brand)? inconsistent-very

consistent

Keller & Aaker How logical is selling (extension) for (parent 7pt Likert: not at allPerceived 1992; brand)? logical-very logicalFit Diamantopoulos, How good/bad is the fit between (parent 7pt Likert: very

Smith, & Grime brand) and (extension)? bad-very good2005

Diamantopoulos, Overall, how would you rate the similarity of 7pt Likert: verySmith, & Grime (extension) to the products currently offered dissimilar-very2005* by (parent brand)? similar

Aaker 1996b How likely are you to recommend (brand 7pt Likert: not veryextension) make by (parent brand) to others? likely-very likely

Aaker & Keller How likely would you be to try a (extension) 7pt Likert: not1990; Klink & sold by (parent brand) assuming that you are very likely-very

Extension Smith 2001; planning to purchase a product in this category? likelyAttitudes Martínez, Polo, & Indicate how favorable you are towards 7pt Likert: not at

de Chernatony (extension) made by (parent brand)? all favorable-very2008 favorable

McCarthy, Heath, How likely would you be to buy a (brand 7pt Likert: not& Milberg 2001 extension) made by (parent brand), if you very likely-very

were on vacation, and you needed to buy a likely(extension)?

*Global perceived fit indicator used in pretest 2.


similar extension of nutritional snacks and a dissimilarextension of watches. Coca-Cola was selected to be thenarrow brand with juice and mobile phones as the similarand dissimilar extensions respectively. Real global brandswere used because they improve the validity of the exten-sion evaluations (Dacin and Smith 1994; Völckner andSattler 2006). The complete questionnaire was tested witha small group of international MBA graduate studentsbefore being fully implemented.

Survey Design

The complete survey packet was primarily composedof existing scales; only the two indicators for portfoliobreadth are original to this study. The survey packetcontained additional constructs for use in another study,such as measuring parent brand image. The main con-structs used in the study and their sources are included inTable 1.

In the first part of the survey respondents answeredquestions related to the parent brand. Then, they werepresented with one potential extension and answeredquestions related to the extension. Lastly, the respondentsfilled out Sharma’s (2010) cultural measurement scale,followed by a few demographic questions (gender, age,nationality). Sharma’s scale consists of 40-items. For thisstudy four indicators were used to measure RSK and fourindicators for AMB, the two sub-dimensions of uncer-tainty avoidance; thus, UNA was measured by combingthese eight indicators.

Data Collection

The sample consists of 146 international studentsenrolled in an MBA program in Barcelona, Spain. Threerespondents did not complete the individual cultural scales;therefore, the final sample size was 143 subjects fromforty different countries and five continents. Internationalstudents were preferred since a wide variance in uncer-tainty avoidance dimensions could be collected and toreduce any home-country bias towards the global brand.

DATA ANALYSIS

Manipulation Tests

The data was first analyzed to determine if the port-folio manipulation was effective. The data confirmed thatthe respondents viewed the narrow brand has having lessproducts than the broad brand. The other manipulation tobe checked was the extension similarity ratings. The fourfit indicators were averaged into one composite score,with a Cronbach’s Alpha of .953. The means were com-pared using an independent samples test. The results

indicate the perceived fit was successfully manipulated,(t(141) = 17, p > .00) with reported means of M

sim = 4.38,

Mdis

= 1.61.

Extant Confirmation

Existing research indicates similar extensions re-ceive higher attitudinal ratings (e.g., Völckner and Sattler2007). The four extension attitude indicators were aver-aged to form a composite attitude score, with a Cronbach’sAlpha of .898. The data confirms prior findings, (t(141) =12, p > .00), with favorability means of M

sim = 4.22, M

dis =

2.09. Overall, the survey successfully manipulated theportfolio and extension similarity variables as well asproviding initial results consistent with established litera-ture.

Uncertainty Avoidance

Respondents’ self-reported uncertainty avoidancescores were averaged and coded into dummy variables. Aconservative median split analyses (Monga and John2007) to classify respondents as either high or low on allthree studied cultural dimensions was used: UNA with aCronbach’s Alpha of .826, RSK with a Cronbach’s Alphaof .745, and AMB with a Cronbach’s Alpha of .795.Additionally, a PCA on uncertainty avoidance resulted ina two-factor solution explaining 60% of the variance,further supporting the two dimensionality of UNA. To testour first hypothesis we performed two separate indepen-dent samples tests. The first tested low (n = 42) and high(n = 22) UNA respondents’ attitude ratings for similarextensions. As expected, the results indicated a differ-ence, albeit insignificant, in means between raters, (t(62) =-1.08, p = .29), with favorability means of M

lowUNA-Sim =

4.44, MhighUNA-Sim

= 4.11. A similar analysis was performedfor dissimilar extensions, the difference betweenfavorability ratings between low (n = 27) and high (n = 52)UNA respondents was minimal and insignificant, (t(77) =-.50, p = .62), with attitude means of M

lowRSK-Dis = 2.02,

MhighUNA-Dis

= 2.13.

Risk Aversion

The above test was repeated, by classifying respon-dents as either low or high RSK, a sub-dimension of UNA.As with UNA, respondents with low (n = 29) RSK ratedsimilar extensions higher than respondents with highRSK (n = 35), (t(62) = -1.46, p = .15), M

lowRSK-Sim = 4.46,

MhighRSK-Sim

= 4.03. When rating dissimilar extensions, lowRSK (n = 34) indicated being less favorable toward theextension than high RSK (n = 45), (t(77) = -1, p = .33).This statistically insignificant result is contradictory thanexpected, with means of M

lowRSK-Dis = 1.98, M

highRSK-Dis =

2.19.


Ambiguity Avoidance

The data was analyzed to determine if respondents’AMB impact extension evaluations. For similar exten-sions, respondents who have low AMB (n = 42) providedhigher favorability ratings than high AMB (n = 22).However, the results were not significant at (t(62) = .31,p = .76), with means of M

lowAMB-Sim = 4.23, M

highAMB-Sim =

4.16. For dissimilar extensions, both low (n = 26) and high(n = 53) AMB provided similar favorability ratings,(t(77) = .03, p = .98), M

lowAMB-Dis = 2.10, M

highAMB-Dis = 2.09.

Portfolio Effects

To test the final hypotheses, a three-way ANOVA(2 x 2 x 2) was conducted to test the effects of portfolio(narrow, broad), extension similarity (similar, dissimilar)and either UNA, RSK, or AMB (low, high) have onextension favorability. Three separate analyses were ran,one for UNA and one for each sub-dimension; thus UNAcan be considered a composite score of the eight indica-tors for RSK and AMB.

The between-subjects effects for the UNA model,showed, as expected, a main effect for extension, (F(1,135) = 152.41, p < .000) and a portfolio main effect, (F(1,135) = 2.06, p = .11). Further, there was an interactioneffect between portfolio breadth and extension distance,(F(2, 135) = 5.80, p = .02). The predicted three-wayinteraction effect among uncertainty avoidance, portfolioand extension was not significant (F(1, 135) = .08, p =.79).

Similar results were found with the sub-dimension,RSK. A main effect was found for extension distance(F(1, 135) = 153.52, p > .000) and a relatively significantportfolio main effect, (F(1, 135) = 2.81, p = .10). Further,the interaction effect between portfolio and extension wasmaintained, (F(1, 135) = 6.13, p > .02). Again, the pre-dicted interaction three-way interaction was not found tobe significant, (F(1, 135) = .04, p = .85). However, aninteraction effect between aversion to risk and extensiondistance was slightly significant, (F(1, 135) = 3.32, p =.07).

The AMB model repeated the main effect of exten-sion, (F(1, 135) = 149.45, p > .000) and portfolio maineffects, (F(1, 135) = 2.65, p = .11). Unlike risk aversionthere was no interaction effect between ambiguity toler-ance and extension. However, the portfolio and extensioninteraction effect remained, (F(1, 135) = 5.49, p = .02).And, the predicted three-way effect was insignificant,(F(1, 135) = .01, p = .94).

Due to a slightly significant interaction effect wasfound with extension distance and risk aversion this issue

was further explored. In particular, an independent samplestest was performed on RSK, extension distance and port-folio breadth on extension favorability ratings. Only therelationship between high risk adverse respondents, rat-ing similar extensions varied by portfolio breadth wassignificant, (t(33) = -2.90, p = .007). High risk adverserespondents appear to prefer similar extensions frombroad brands, M

highRSK-Sim-Narrow = 3.54, M

highRSK-Sim-Broad =

4.54. This finding encouraged us to see if the relationshipbetween high RSK preferring similar extensions frombroad brands is evident in the other uncertainty dimen-sions.

Additional tests confirmed that high AMB respon-dents prefer similar extensions from broad brands overnarrow brands, (t(40) = -2.71, p = .01), M

highAMB-Sim-Narrow =

3.85, MhighAMB-Sim-Broad

= 4.70. This relationship held at thehigher UNA level as well (t(40) = -3.06, p = .004) withmeans of M

highUNA-Sim-Narrow = 3.66, M

highRSK-Sim-Broad = 4.60

and, respectively.

FINDINGS

The first three hypotheses looked at the impact uncer-tainty related tendencies have on similar and dissimilarextension evaluations. Based on prior new product andextension research, it was hypothesized that respondentslow on uncertainty related tendencies would be moreaccepting of extensions than respondents high on uncer-tainty related tendencies. The data indicates this is par-tially true. First, while not highly statistically significant,low UNA, RSK, and AMB respondents rated similarextensions higher across all three uncertainty measures.However, while not significant, high UNA and RSKrespondents provided higher ratings for dissimilar exten-sion contradictory to prior conceptualization of risk tak-ing behavior. This could indicate that individuals with lowrisk aversion are actually better at evaluating or judgingrisk and are not free-spirited risk takes. Or, perhapsextensions from broad brands are not seen as risky sincelarge brands project stability and the ability to manufac-turer quality products. Thus, it appears that under catego-rization theory it is not the amount of information trans-ferred by it is the value placed on the information trans-ferred that matters. In this case, the quality and stability oflarge portfolio brands overcomes the lack of specificproduct related information.

When portfolio breadth is introduced into the analy-sis, the data indicates that respondents with high uncer-tainty related tendencies prefer similar extensions frombrands with a broad portfolio breadth. This finding isparticularly interesting and deserves further attentionsince prior research has found narrow brands receivehigher favorability ratings for similar extensions (Boushand Loken 1991). These results suggest that high uncer-


tainty avoiders can overcome their aversion to risk byplacing trust in larger, established brands, essentiallyplacing greater weight on the past success of largerbrands. Again, this finding tentatively suggests that undercategorization theory it is the value placed on the informa-tion transferred is more important than quantity of infor-mation transfer. The other relationships tested proved tobe statistically insignificant.

LIMITATIONS AND CONCLUSION

Future Brand Image and Size Effect

This study sought to identify if risk aversion orambiguity tolerance, both sub-dimensions of Hofestede’s(2001) uncertainty avoidance cultural dimensions, is themain driver behind brand extension acceptance. Themajor finding of this study is high uncertainty avoidanceindividuals prefer similar extensions from broad breadthbrands over narrow brands. This finding conflicts withprior studies; however, Boush and Loken (1991) mostlikely used American students, who tend to be lower inuncertainty avoidance. This does indicate that much ofour current knowledge on brand extensions is heavilybiased towards American cultural and personality values.Including a moderate brand extension (Barone, Miniard,and Romeo 2000; Boush and Loken 1991) or even a

moderate portfolio breadth brand could help in this endea-vor. Perhaps, replicating this study but with brand sizemeasured by perceived brand image over product portfo-lio breadth will add additional insights. Further, the studysuggests that future studies should investigate the valueconsumers place on the attributes transferred to exten-sions over quantity of parent brand attributes transferredunder a categorization theory framework. For marketers,selecting countries to introduce extensions it is recom-mend to consider the views toward company size held bythe citizenry.

Clearly sample sizes and lack of significance can bemajor limitations for empirical studies; however for anexploration into the impact the sub-dimensions of uncer-tainty avoidance have on brand extensions exercisingcaution can conserve resources and facilitate the develop-ment of more pointed follow-up studies. Additionally, theuse of an American brand and European brand could haveimpacted response but this critique is limited by the use ofinternational subject from forty different countries andfive continents. In conclusion, this study directs the wayfor future research projects to go more in-depth into howconsumers’ uncertainty avoidance tendencies impact ex-tension evaluations, as there appears to be interestinginteraction effects waiting to be revealed.

REFERENCES

Aaker, David A. (1996a), Building Strong Brands. NewYork: Free Press.

____________ (1996b), “Measuring Brand Equity AcrossProducts and Markets,” California Management Re-view, 38 (Spring), 102–20.

Aaker, David A. and Kevin L. Keller (1990), “ConsumerEvaluations of Brand Extensions,” Journal of Mar-keting, 54 (1), 27–41.

Barone, Michael J., Paul W. Miniard, and Jean B. Romeo(2000), “The Influence of Positive Mood on BrandExtension Evaluations,” Journal of Consumer Re-search, 26 (03), 386–400.

Bottomley, Paul A. and Stephen J.S. Holden (2001), “DoWe Really Know How Consumers Evaluate BrandExtensions? Empirical Generalizations Based onSecondary Analysis of Eight Studies,” Journal ofMarketing Research, 38 (11), 494–500.

Boush, David M. and Barbara Loken (1991), “A Process-Tracing Study of Brand Extension Evaluation,” Jour-nal of Marketing Research, 28 (2), 16–28.

Buil, Isabel, Leslie de Chernatony, and Leif E. Hem(2009), “Brand Extension Strategies: Perceived Fit,Brand Type, and Culture Influences,” European Jour-nal of Marketing, 43 (12), 1300–24.

Dacin, Peter A. and Daniel C. Smith (1994), “The Effectof Brand Portfolio Characteristics on ConsumerEvaluations of Brand Extensions,” Journal of Mar-keting Research, 31 (05), 229–42.

de Mooij, Marieke (2004), Consumer Behavior and Cul-ture: Consequences for Global Marketing and Ad-vertising. Thousand Oaks, CA: Sage Publications.

Diamantopoulos, Adamantios, Gareth Smith, and IanGrime (2005), “The Impact of Brand Extensions onBrand Personality: Experimental Evidence,” Euro-pean Journal of Marketing, 39 (1/2), 129–48.

Henseler, Jö, Casilla Horváth, Marko Sarstedt, and LorenzZimmermann. (2010), “A Cross-Cultural Compari-son of Brand Extension Success Factors: A Meta-Study,” Journal of Brand Management, 18 (9), 5–20.

Hofstede, Geert (2001), Culture’s Consequences: Com-paring Values, Behaviors, Institutions, and Organi-zations Across Nations. Thousand Oaks, CA: SagePublications.

Keller, Kevin L. and David A. Aaker (1992), “The Effectsof Sequential Introduction of Brand Extensions,”Journal of Marketing Research, 29 (2), 35–60.

____________ and ____________ (1998), “The Impactof Corporate Marketing on a Company’s Brand Exten-sions,” Corporate Reputation Review, 1 (Summer),356–78.


____________ (1998), Strategic Brand Management:Building, Measuring and Managing Brand Equity.Upper Saddle River, NJ: Prentice Hall.

Klink, Richard R. and Gerard A. Athaide (2010), “Con-sumer Innovativeness and the use of New VersusExtended Brand Names for New Products,” Journalof Product Innovation Management, 27 (1), 23–32.

____________ and Daniel C. Smith (2001), “Threats tothe External Validity of Brand Extension Research,”Journal of Marketing Research, 38 (8), 326–35.

Laroche, Michel (2007), “Introduction to the SpecialIssue on the Impact of Culture on Marketing Strat-egy,” Journal of Business Research, 60 (3), 177–80.

Lau, Kong C. and Ian Phau (2007), “Extending SymbolicBrands using their Personality: Examining Anteced-ents and Implications Towards Brand Image Fit andBrand Dilution,” Psychology & Marketing, 24 (5),421–44.

Martínez, Eva, Teresa Montaner, and José M. Pina (2009),“Brand Extension Feedback: The Role of Advertis-ing,” Journal of Business Research, 62 (3), 305–13.

____________, Yolanda Polo, and Leslie de Chernatony(2008), “Effect of Brand Extension Strategies onBrand Image,” International Marketing Review, 25(1), 107–37.

McCarthy, Michael S., Timothy B. Heath, and Sandra J.Milberg (2001), “New Brands Versus Brand Exten-sions, Attitudes Versus Choice: Experimental Evi-dence for Theory and Practice,” Marketing Letters,12 (Feb), 75–90.

Milberg, Sandra J., C. Whan Park, and Michael S.McCarthy (1997), “Managing Negative FeedbackEffects Associated with Brand Extensions: TheImpact of Alternative Branding Strategies,” Journalof Consumer Psychology, 6 (3), 119–40.

Monga, Alokparna B. and Deborah R. John (2007), “Cul-tural Differences in Brand Extension Evaluation: TheInfluence of Analytic Versus Holistic Thinking,”Journal of Consumer Research, 33 (3), 529–36.

Nijssen, Edwin J. and Clara Agustin (2005), “BrandExtensions: A Manager’s Perspective,” Journal ofBrand Management, 13 (10), 33–49.

Sharma, P. (2010), “Measuring Personal Cultural Orien-tations: Scale Development and Validation,” Journalof the Academy of Marketing Science, 38, 787–806.

Völckner, Franziska and Henrik Sattler (2006), “Driversof Brand Extension Success,” Journal of Marketing,70 (4), 18–34.

____________ and ____________ (2007), “EmpiricalGeneralizability of Consumer Evaluations of BrandExtensions,” International Journal of Research inMarketing, 24 (06), 149–62.

Xie, Yu H. (2008), “Consumer Innovativeness and Con-sumer Acceptance of Brand Extensions,” Journal ofProduct & Brand Management, 17 (8), 235–43.

Yorkston, Eric A., Joseph C. Nunes, and Shashi Matta(2010), “The Malleable Brand: The Role of ImplicitTheories in Evaluating Brand Extensions,” Journalof Marketing, 74 (1), 80–93.

For further information contact:Matthew A. Hawkins

ESADE, Ramon Llull UniversityAv. de la Torre Blanca, 59

Sant Cugat, SpainPhone: 34.93.554.35.11

Fax: 34.93.204.81.05E-Mail: [email protected]


SUB-BRANDING AFFECT TRANSFER: THE ROLE OF PRODUCTCATEGORY CROWDEDNESS AND BRAND LOYALTY

Yi He, California State University, East BayQimei Chen, University of Hawaii

Ruby P. Lee, Florida State UniversityLeona Tam, Old Dominion University

SUMMARY

Today, it is widely accepted among managers andscholars that brands are valuable intangible assets of afirm that can significantly contribute to its performanceand financial value (Bahadir, Bharadwaj, and Srivastava2008; Morgan and Lego 2009). To capitalize on the valueof existing brands, the burgeoning brand extensionresearch has provided valuable managerial insights intohow an existing brand can be extended to different prod-uct categories such as Heinz cheese cracker (Keller andAaker 1992) or Frito Lay’s partially baked pizza (Oakleyet al. 2007). However, brand extensions have been con-sidered as a double-edged sword. While successful brandextensions provide new sources of revenue and enhancebrand equity, failed extensions damage family brands,squandering millions of dollars in building the familybrands’ equity (Keller and Sood 2003). To take advantageof a positive spillover from a family brand and at the sametime to avoid the family brand being diluted from apossible failed extension, some firms opt for a sub-branding strategy, which is a combination of the familybrand name and a new brand name such as Courtyard byMarriott (Kirmani, Sood, and Bridges 1999). Despite thenoted importance of sub-branding, there is little empiricalevidence to guide managers’ sub-branding decisions.Even though marketing literature is replete with findingssupporting the transfer of affect from a brand to itsextensions (Keller and Aaker 1992, 1993), the directempirical evidence on the occurrence of affect transferfrom a family brand to its sub-brand has rarely beendocumented (see Milberg et al. 1997 for a rare exception).

Against this backdrop, we aim to examine the occur-rence and nature of affect transfer from a family brand toits sub-brand. Further, extant research has yet to identifythe boundary conditions of sub-branding affect transfereffect. Building on prior research, we propose three mod-erators, product category crowdedness and competing/family brand loyalty, on sub-branding affect transfer.Although product category crowdedness has receivedconsiderable attention in the new product introductionliterature that supports pioneer advantage and order ofentry effects (Carson, Jewell, and Joiner 2007), empiricalevidence on product crowdedness in a sub-brand intro-duction context is scarce despite its clear linkage with

critical sub-branding decisions such as entry timing andcustomer loyalty (Morgan and Rego 2009). Brand loyaltyhas been described as the ultimate corporate objective inthe new millennium due to the widely-accepted linkbetween loyalty and firm performance (Aaker 1996;Chaud-hurl and Holbrook 2001; Dubin 1998; Harris andGoode 2004); however, systematic research on its impli-cations to branding development decisions (e.g., sub-branding) is lacking.

To build an overall nomological network, we callupon the associative network theory (Anderson 1983) indelineating the nature of sub-branding affect transfer. Instudying family brand attitude activation and sub-brandattitude formation (Herr et al. 1996; Zimmerman, Redker,and Gibson 2011), we conducted four experiments to test(1) whether there is brand affect transfer from a familybrand to its sub-brand (Experiment 1); (2) whether thebrand affect transfer effect will be moderated by productcategory crowdedness (Experiment 2); and (3) whetherthe moderated brand affect transfer effect in Experiment 2will further interact with consumers’ loyalty to the familybrand (Experiment 3) as well as a competing brand(Experiment 4).

Experiment 1 demonstrates that there exists a posi-tive relationship between family brand attitude and itssub-brand attitude (B = .51, t = 3.10, p < .01). Experiment2 replicates the results found in Experiment 1, confirmingthe occurrence of family brand affect transfer to its sub-brand in a less crowded product category setting (B = .56,t = 3.80, p < .001). Experiment 2 demonstrates thatproduct category crowdedness moderates affect transfer,that is, the affect transfer from the family brand to the sub-brand attitude disappears in more crowded product cat-egory setting (ts < .98, ps > .32). The findings fromExperiment 3 reveals that the suppressed affect transfer inmore crowded category setting from Experiment 2 can berestored for consumers who are highly loyal to the familybrand (B = .63, t = 2.92, p < .01). The findings fromExperiment 4 show that loyalty to the competing brandcan offset the desired affect transfer in the less crowdedcategory setting (B = -.05, t = -.25, p > .08).

To conclude, our research advances marketing theoryby detailing affect transfer from a family brand to its sub-


brand. Our results further demonstrate major differencesin affect transfer due to the competitive environment (i.e.,product category crowdedness) and brand loyalty. These

findings provide important insights for researchers andmarketers attempting to effectively manage a company’sbrand portfolio.

For further information contact:Yi He

College of Business and EconomicsCalifornia State University, East Bay

25800 Carlos Bee BoulevardHayward, CA 94542Phone: 510.885.3534



EXPLICIT AND IMPLICIT MEASURES OF BRAND INFORMATIONPROCESSING AND ITS IMPACT ON BRAND PERCEPTION AND

BRAND-RELATED BEHAVIOR

Sascha Langner, Leibniz University Hannover, GermanyNadine Hennigs, Leibniz University Hannover, GermanySteffen Schmidt, Leibniz University Hannover, Germany

Klaus-Peter Wiedmann, Leibniz University Hannover, Germany

SUMMARY

In our paper we introduce a formative construct ofbrand information processing (BIP) combining both im-plicit (EEG, SC-IAT) and explicit measures (brand be-lievability, brand attitude) to one integrated framework.To validate BIP we present a structural model linkingbrand information processing, brand perception (BP) andbrand behavior (BB). Hypotheses on the formative con-struct as well as on the interrelations of BIP, BP and BBare proposed and the model is empirically tested usingcausal analysis. The results show that the combination ofexplicit and implicit measures can reliably explain evensmall changes in brand perception and behavior.

In an attempt to combine explicit and implicit mea-sures of brand perception and behavioral responses to-ward the brand, our study relies on the two-dimensionalframework of Camerer et al. (2005). They distinguishbetween explicit and implicit processes, and betweencognition and affect. The combination of these two di-mensions defines information processing. The four quad-rants form the basis for the development of our hypoth-eses.

Our questionnaire includes already existing and testedmeasures for brand believability and brand attitude as ourdefined core elements of BP as well as for price premiumand buying intention as essential constituents of BB. Allitems were rated on a five-point Likert scale (1 = stronglydisagree to 5 = strongly agree). With regard to the multi-dimensional BIP construct, the explicit-cognitive dimen-sion was assessed using a believability measure, while theexplicit-cognitive dimension was captured employing anattitude measure. Both measures were rated on seven-point semantic differentials. For measuring the implicit-cognitive dimension, a single category IAT (SC-IAT) wasprogrammed. Via an EEG, the implicit-affective dimen-sion was measured with regard to record mediation acti-vation. Partial Least Squares (PLS) structural modelingwas used to examine the drivers and outcomes of BIP andto empirically test our additional hypotheses.

With reference to BIP all explicit and implicit mea-sures reach satisfactory values. Especially for implicitmeasures, the achievement of a substantial internal con-sistency is a continuous challenge. Moreover, this is thekernel of a bright acceptance as a reliable measure ap-proach within a traditionally explicit measure dominatedresearch stream as it is in marketing research. Both im-plicit measures in this study, the reaction time based singlecategory implicit association test (SC-IAT) as well as theEEG-based mediation measure display a satisfying reli-ability. The evaluation of the formative BIP constructwith the four calculated variables believability, attitude,SC-IAT score and mediation led to satisfactory results aswell. The same holds for the reflective measurement of BPand BB. Moreover, the results reveal a high predictiverelevance of the PLS structural equation model. Withregard to the test of our hypotheses the results giveevidence that BIP is indeed constituted by a combinationof explicit and implicit measures, providing a holisticpicture of its significant impact upon behavioral outcomessuch as BP and BB.

Although in the last 30 years many authors havetested various technologies measuring implicit reactionsin order to better understand the fundamentals of informa-tion processing regarding brand related messages, most ofthe research concentrated on very specific research prob-lems like memory processes, attention tracking or emo-tional reactions. Particularly the number of studies thatfollow a holistic approach that combines both implicit andexplicit measures are rather scarce. In this light our studyoffers a promising idea of uniting conscious and subcon-scious reactions in one integrated framework. With ourconcept of BIP, we present a new way to better determinenot only how, when and in which combination implicitand explicit perceived stimuli are processed, but also offeran alternative approach to more precisely measure brandstimuli and their effects on BP and BB. References areavailable upon request.


For further information contact:Sascha Langner

Leibniz University Hannover, GermanyKoenigsworther Platz 1

Hannover, Lower Saxony, 30163Germany

Phone: ++49(0)511.762.3057 or -4862Fax ++49(0)511.762.3142



DIMINISHING EFFECTS OF PERCEIVED FIT ONVERTICAL EXTENSIONS

Nicolas Pontes, Monash University, AustraliaMauricio Palmeira, Monash University, Australia

Colin Jevons, Monash University, Australia

SUMMARY

A common finding in brand extension literature isthat extension’s favorability is a function of the perceivedfit between the parent brand and its extension (Aaker andKeller 1990; Park, Milberg, and Lawson 1991; Volcknerand Sattler 2006) that is partially mediated by perceptionsof risk (Milberg, Sinn, and Goodstein 2010; Smith andAndrews 1995). In other words, as fit between the parentbrand and its extension increases, parent brand beliefsbecome more readily available, thus increasing consumercertainty and confidence about the new extension, whichresults in more positive evaluations. On the other hand, asperceived fit decreases, consumer certainty about theparent brand’s ability to introduce the extension is re-duced, leading to more negative evaluations. Building onthe notion that perceived fit of vertical line extensions isa function of the price/quality distance between parentbrand and its extension (Lei, de Ruyter, and Wetzels2008), traditional brand extension knowledge predicts adirectionally consistent impact of perceived fit on evalu-ations of vertical extensions. Hence, vertical (upscale ordownscale) extensions that are placed closer to the parentbrand in the price/quality spectrum should lead to higherfavorability ratings compared to more distant ones.

In contrast to existing literature, this research showsthat the extension’s direction is a key moderator of theperceived fit effect on vertical line extension evaluations.Specifically, we argue that consumers rely on perceivedfit to diminish risk perceptions and increase favorabilityratings for upscale but not for downscale extensions. Toaccount for such asymmetric effect we argue that per-ceived brand expertise plays an important role. Becauseintroducing an upscale extension implies a change inbrand expertise, perceived fit is evoked to diminish risinglevels of uncertainty. Conversely, because there is nochange in brand expertise for downscale extensions, con-sumers do not need to search for a second risk reductionmechanism which in turn leads to a diminish effect ofperceived risk. We examine this proposition in two em-pirical studies that offer converging evidence of the mod-erating role of the extension’s direction in the perceivedfit – extension evaluation relationship.

Study 1 was an online 2 (distance: close vs. far) x 2(extension direction: up vs. down) between-subjects fac-

torial design experiment to test the hypothesis that theeffect of perceived fit on vertical extension evaluations ismoderated by the extension’s direction. In particular, wetest the hypothesis that the greater the perceived fitbetween an upscale extension and its parent brand, thegreater the perceived risk, which in turn results in morefavorable evaluations of the extension. We show that thisis actually an asymmetrical effect that does not arise indownscale scenarios. Consumers will not perceive a moredistant downscale extension to be of higher risk comparedto a closer extension.

Two-way ANOVA results showed no significantmain effect for perceived fit on extension evaluations(F(1, 131) = 1.54, p > .10), but a significant main effect fordirection (F(1, 131) = 12.39, p < .05) on the extensionevaluations. More importantly there was a significantinteraction effect between perceived fit and direction onextension evaluations (F(1, 131) = 4.54, p < .05). Thisinteraction effect was in the hypothesised direction. Inparticular, one-way ANOVA results show that upscaleextension evaluations are significantly lower in the farcompared to the close condition (M

fu = 4.76 vs. M

cu = 5.39;

F(1, 64) = 4.84, p < .05). In contrast, no significantdifferences between far versus close downscale exten-sions were found in the extension evaluation (M

fd = 5.82

vs. Mcd

= 5.66; F(1, 67) = .47, p > .10). We conducted anadditional analysis to test the proposition that the effect ofperceived fit on upscale extension evaluations is mediatedby perceptions of risk. The bias-corrected bootstrap (basedon 5,000 bootstraps samples) reveal that the mean medi-ating effect is positive (a x b = .12) and significant with a90% confidence interval excluding zero (.004 to .408). Inthe indirect path, a unit increase in perceived fit reducesrisk by a = - .69 units and a unit increase in risk reducesextension evaluation by b = - .17. The effect of the directeffect c (.51) is only marginally significant (p = .086).According to Zhao et al. (2010), because a x b x c (.06) ispositive, this is a complementary mediation.

Study 2 was also an online 2 (distance: close vs. far)x 2 (direction: up vs. down) between-subjects factorialdesign experiment that sought to explain why there is adiminishing effect of fit on the extension’s evaluation indownscale scenarios. Weexamined the effect of brandexpertise on the extension evaluation in the upscale anddownscale scenarios and tested whether perceived risk


mediates this relationship in both directions (up vs. down).Our results replicated the findings of study 1 in a differentmanipulation setting; the extension price was kept con-stant while the parent brand price was manipulated. Thebias-corrected bootstrap analysis (based on 5,000 boot-straps samples) reveal that the mean mediating effect ofperceived risk on the fit-evaluation relationship is positive(a x b = .2423) and significant with a 95% confidenceinterval excluding zero (.0522 to .4806) when brandexpertise was used as a covariate for upscale extensions.In contrast, the bias-corrected bootstrapping (using 5,000sample) show that mean mediating effect of perceived riskon the brand expertise-extension evaluation is positive(a x b = .2216) and significant with a 95% confidenceinterval excluding zero (.0457 to .4687) when perceivedfit was used as a covariate. In both cases, the covariate wasnot significant (p > .20).

Taken together, the two studies provide strong evi-dence that consumers systematically use perceived fit asa risk reduction mechanism for upscale extensions whilebrand expertise plays that role for downscale extensions.First, we show that the effect of perceived fit on verticalextension evaluations is moderated by the extension’sdirection. Results show a positive effect of perceived fiton upscale evaluations but not for downscale extensions.Study 2 replicated this finding and further showed thatthat risk perceptions are lower for downscale extensionscompared to upscale extensions because it is the effect ofbrand expertise on the extension evaluation that is medi-ated by perceived risk in the downscale setting while theeffect of perceived fit on the extension’s evaluation ismediated by perceived risk in upscale scenarios. Refer-ences are available upon request.

For further information contact:Nicolas Pontes

Department of MarketingMonash UniversityLevel 7, Building S

Sir John Monash DriveCaulfield, VIC 3145

AustraliaPhone: +61(03)9903.1884



MANAGING BRAND ARCHITECTURE: STATE-OF-THE-ART ANDDECISION-MAKING

Christopher Kanitz, University of Bremen, GermanyMichael Schade, University of Bremen, Germany

Christoph Burmann, University of Bremen, Germany

SUMMARY

The complexity of corporate brand portfolios hasincreased significantly within the last years (Aaker 2004;Muzellec and Lambkin 2009). In most corporations alarge variety of brands are managed simultaneously(Schuiling and Kapferer 2004). Hence, the managementis faced with major challenges. First, brand portfolioshave to be coordinated with the corporate strategy. In thiscontext, the brand architecture research focuses on thebrand portfolio management in multiple brand corpora-tions (Strebinger 2004).

The current state of brand architecture research leavesout important questions unanswered (Muzellec andLambkin 2009). For example, how do companies makebrand architecture decisions? For this purpose, practicallyapplicable decision-making criteria are glaringly missing(Strebinger 2004). The other major criticism is the mea-surement of the perceived brand architecture (Rao,Agarwal, and Dahlhoff 2004). Current methodologiesfade out behavioral relevant brand associations and focusonly on the assessment of brand logos’ perception. Thepresent paper is set to develop a conceptual framework forthe management of brand architectures with the aim ofclosing major research gaps.

The first important result is the identity-based brandarchitecture process (iBAP). It illustrates the theoreticalframework of managing brand architecture and is dividedin four steps: (1) Hierarchizing brand portfolios, (2)Corporate-strategic development of brand architectureconsisting of two steps: (a) Identifying courses of action,and (b) Evaluating and deciding courses of action, (3)Translating brand architecture in brand strategies, and (4)Monitoring. The implementation-oriented perspectivetranslates the brand architecture in individual strategiesfor every single brand within the portfolio. The finalmonitoring perspective includes the internal acceptanceof the brand architecture by the employees and the exter-nal perception of the brand architecture by the customers.

As a second major result besides the vertical dimen-sion of brand architecture, as already known in the litera-

ture (e.g., Aaker 2004; Strebinger 2004), a horizontal anda distribution-oriented dimension will be integrated intothe conceptual brand architecture model. The verticaldimension refers to the intensity of the connection be-tween brands on different hierarchy levels within thebrand portfolio (branded house vs. house of brands). Thehorizontal dimension states, whether a market segment istreated with either one or multiple brands in parallel. Thedistribution-oriented dimension defines whether a com-pany exclusively offers products as a classic brandedmanufacturer or further provides other manufacturerswith ready-made products using their product brands inthe market or even white label brands of major retailers.

The third result is the derivation of internal andexternal evaluation criteria for brand architecture deci-sion-making. These criteria will be derived for the verticaldimension of the brand architecture model as it is themajor driver of brand architecture decision-making. Forinternal criteria the internal acceptance by the employees,the risk balance within the brand portfolio and the require-ments of resources are divided. For external criteria theacceptance of external stakeholders, the skimming ofmarket potential and the strategic flexibility of the brandarchitecture will be differentiated. With the help of theseevaluation criteria companies can make profound brandarchitecture decisions. The last important result is theoperationalization of the mentioned evaluation criteria. Inthis context, especially a new approach of measuring theperceived brand architecture is developed, implementinga profound brand image measurement instead of focusingon a brand logos’ perception only.

It is shown, that the increasing complexity of brandportfolios in many corporations requests a systematicbrand portfolio management. Therefore, the identity-basedbrand architecture process (iBAP) and the evaluationcriteria for strategic brand architecture decision-makingoffer a solid approach for multiple brand corporations toface this challenge and improve their brand architecturemanagement. References are available upon request.


For further information contact:Christopher Kanitz

Innovative Brand Management (LiM®)Business Administration

University of BremenHochschulring 4WiWi Building28359 Bremen

GermanyPhone: +49.421.218.66579



BRANDING WITH POLITICAL THEORIES:AN INTERDISCIPLINARY APPROACH

Tony Yan, New Mexico State University, Las Cruces

ABSTRACT

This paper tries to illustrate the way that politicaltheories and models may be useful for the study ofbranding. For example, branding focuses on the construc-tion of specific brand and corporate values, meanings,cultures, and myths, while political framing is used tomobilize political support and loyalty. Political theoriesthus may provide cognitive, organizational, and strategicassistances to the study of branding.

INTRODUCTION

Branding is an important strategy for firms to main-tain or build up their images in competing with otherfirms. To establish, maintain, and extend the meaningsand values of brands, which are also through branding, arevital for companies’ operation and development. Thispaper tries to show that political theories may offer sometheoretical assistance to the study of branding. The studyfirst discusses brandings and theories of branding. Sec-ond, the study defines some useful political theories andhow they can be employed to study branding. The paperthen illustrates the application of political theories to thestudy of branding by examining two cases: branding andpolitical framing; and the institutionalization of consumerloyalty and identity via brand community. The paperfinally concludes with the suggestion that some politicalmodels may offer interdisciplinary assistance to the studyof branding.

BRANDING AND THEORIES OF BRANDING

According to David Aaker, branding mainly meansto build a strong brand for a corporation, a product, aperson, or a symbol (Aaker 1991). Douglas Holt arguesthat branding is constructive; that is, “a brand emerges asvarious ‘authors’ tell stories that involve the brand” (Holt2004, p. 3). These authors include: companies, cultureindustries, social media, and consumers. Thus, we per-ceive branding as a specific type of storytelling withmultiple-authors (Fog et al. 2005).

Branding is important because it may generate somesignificant advantages for the company or consumers.These advantages include: product identification, shop-ping efficiency, risk reduction, product acceptance, enhan-ced self-image, and enhanced product loyalty (Ferrell andHartline 2008, pp. 197–199).

Indeed, conventional studies on branding focus onthe product dimension. These studies explore the wayfirms design brand, via logo or verbal signs, to expand theimage of their products and services (Johansson and Holm2006). Branding is necessary because it helps firms differ-entiate their products or identities from those of otherfirms. Indeed, branding is crucial in delineating the bound-ary between “them’ and “us” in consumers’ minds, whichhelps firms maintain or increase their market shares.

Recent research focus on the dynamic and construc-tive process of building corporate identities that incorpo-rates culture of the corporation (Schultz and Chernatony2002). Basically, the study of brand focuses on threeimportant fields: brand positioning, brand resonance, andbrand value chain (Ferrell and Hartline 2009). Brandpositioning describes how to maximize firms’ competi-tive advantages. It is an art of showing the difference ofmeanings and highlighting the personality of the firm’sproduct or service (Tybout and Sternthal 2005, pp. 18–20). Brand resonance describes how to create and consoli-date positive relationship with consumers. This perspec-tive defines dynamic process of branding; that is, thereactions from consumers. Brand value chain describeshow to explore the process of creating values to under-stand the financial influences of marketing expendituresand investments.

Some studies stress building and enhancing brandculture, involving profound meaning-construction in multi-elements context. These studies show that brands canreflect a myth, history, or value that the corporationadheres to. More important, brand culture can exist beyond the products, services, or even the corporationbecause it supports some social values that people desire(Holt 2004). Once the brand culture is established, it mayattract supporters via a contagious-like transmissionmechanism, assuming that consumer’s motivations, per-ceptions, and feelings can be transmitted among consum-ers like virus (Hatfield et al. 1994). Brand culture hencehelps to consolidate brand identification, inducing con-sumers to support the product, service, brand, or the firm.In other words, brand culture becomes an ID for a groupof people who belongs to specific groups or communities.

Moreover, conventional theories argue that brandculture or identity is mainly created by firms, Holt andother authors argue that consumers may also create brandand brand culture (Holt 2004, pp. 3, 28–29; Schultz andChernatony 2002).


Siding with Holt, this study believes that the power ofconsumers to create brand culture to some extent balancesthe power of firms, which is beneficial to consolidateconsumer democracy and reduce the feeling of “alien-ation” among consumers (Gobé 2001, xv-xviii; Schultzand Chernatony 2002; Tucker 1970, pp. 133–135). Theprocess is similar to what Robert Dahl once suggested thatthe extensive participation of labors in management willincrease economic equity, efficiency, and democracy(Dahl 1989).

POLITICAL THEORIES AND THE STUDY OFBRANDING

In political arena, political parties also compete forselling their ideas or policies to constituencies. Thus,political theories also study the relationships betweenparties and voters. Political theories may offer theoreticalor intuitive assistance to the study of branding at leastfrom three perspectives.

First, cognitive models in political science argue thatperceptions or misperceptions, depending on availabilityof information and the extent of rationality of actors,significantly determine actors’ activities (Zaller 1992).This dimension provides a constructive analysis to studysymbols, values, meanings, and changes of individual andorganizational behaviors. For instance, in politicaleconomy model, symbols may be deemed as a “signalingmechanism” to sift candidates or choose policies frompolicy pools (Spence 1974); that is, specific symbolsindicate, if not equal to, particular values of policies orcompetency of individuals and organizations, providingfundamental evaluating parameters to decision makers.Brand can also act as a “signaling mechanism” for con-sumers to choose products or show their corporate align-ment because specific brands may embody guaranteedquality or preferred values to consumers and thereforedecrease consumers’ cost of selection.

In political theory, the building of culture or norms isoften labeled as the “internalization” of values, meaningthat people voluntarily accept, advocate, and spread someparticular values without coercion from political authori-ties. According to this model, more outsiders will soonjoin the dominant group or community due to the“bandwagoning effect,” or they will be marginalized bythe mainstream ideology (Key 1960; Shattschneider 1960;North 1990). This phenomenon is also depicted by theItalian communist Antonio Gramsci in the 1920s that theruling ideas of an epoch are the ideas of its ruling class(Gramsci 1971). In political economy, the mechanism iscalled “network effect,” exemplified by the fact that themore users of a type of computer software, the morevalues of the software in the future (Pindyck and Rubinfeld2004). As have mentioned, brand culture also helps toadvocate or legitimize specific brand communities

(McAlexander et al. 2002). Thus, once a brand culture isbuilt, it will have longer longevity than other brands.

Political theories also evidence that people or organi-zations may misjudge something due to insufficient infor-mation or “bounded rationality,” and political organiza-tions or leaders may also manipulate people’s perceptionsto serve some particular interests (Lindblom 1959).

By contrast, branding may engender perceptions ormisperceptions among consumers and it can also manipu-late information or identities to achieve some specificgoals. The extension of brand may also dilute the originalbrand image and bring negative impression of the brand orthe firm. In addition, as Jurgen Habermas finds that theloss of awareness and belief of the government or politicalparty may cause the government’s or the party’s “legiti-mation crisis,” Albert M. Muniz and Thomas C. O’Guinnalso argue that brand affects legitimacy of brandingcommunity, because brand shows the essence of the firmand differentiates it from others (Habermas 1975; Munizand O’Guinn 2001). Perceptions and misperceptions arethus not only important topics in political science, they arealso essential for studying branding.

Second, organizational models have been employedto explain collective behaviors of political parties, politi-cal movements, and social networks. These models andtheories define the controlling, maintenance, and evolve-ment of political organizations and movements and exam-ine transaction scope, cost, and efficiency of certainpolicies and behaviors. In organizational dimension, poli-tical organizations have to build strong organizational orpolicy identity in the society, which is like companies tocreate strong corporate or product brand.

In political theory, especially in voting models, poli-tical parties or candidates often use advertisement toaffect people’s party identification and hence their parti-san alignment or realignment (Key 1960). Using politicalbranding or labeling of specific stories, images, slogans,or policies, political parties or candidates can mobilize orenhance political support. In this regard, politicians orpolitical parties must build “the power to persuade”(Neustadt 1960). Political parties or candidates may alsouse negative advertisements to derogate their rivals’images. Even for a governing party, political brandingalso affects problem definition and hence the policy-agenda building. With financial or managerial constraints,firms may also have to choose from a variety of potentialbrands to build brands that will most likely benefit firms,indicating the most highlighted brands are more likely tobe chosen. Such “brand portfolio strategy” is oftenweighted carefully by brand entrepreneurs to extend thecurrent brand and to build the firm’s future growth withinternal and external constraints. In other words, brandentrepreneurs must design mix of brands to maximizefirm interests (Aaker 2004 a).


In political models, legislators’ reputation will affectwhich kind of bills to be transmitted to or proposed bythem. For instance, controversial bills will more likely tobe proposed by legislators who have the “maverick”reputation, while “tough bills” will more likely to be firsttransmitted to legislators who are branded as easygoing(Tullock 1960). All these models help us understand howfirms use advertisements to enhance their own brands orto use “negative advertisements” to depreciate their ri-vals’ brands.

Third, political theories study games and strategiestaken by various actors in given contexts. This is the studyof interactive and dynamic relations among actors. Accor-ding to political theories and economic models, partici-pants in any transactions will act or react accordingly totheir counterparts’ strategies or activities. Game resultsare decided by the power balance of all participants. Suchpower balance is further determined by strategies taken byparticipants, the asymmetry information each participantowns, and the timing of implementing strategies. Thegame can be one-shot or sequential, depending on trans-action rules or transaction cost. If the game is a one-shotgame, then the first mover will more likely to gain at theexpense of other players. If it is a sequential game, theneach player must carefully calculate the stance and thefollowing steps that others will take. Such reciprocityhelps to build stable relationship among game players(Axerold 1984). More important, political and economictheories assume that everyone has the freedom to makedecision based on given information and knowledge.They may show their preferences in the form of loyalty,voice, or just exit (Hirschman 1978). These preferences,however, are not always stable and they will shift whenexternal or internal cognition of the issue changes (Downs1972).

To build strong consumer loyalty, firms must con-sider reactions from consumers, employees, and rivalcompanies to their branding strategies. The efficiency andeffectiveness of branding is more likely to be determinedby interactive or reciprocal mechanisms among variouscomponents (Fournier 1998; McAlexander et al. 2002;Csaba, Faurhold, and Bengtson 2006; Dacin and Brown2006). Consumers do affect cultural branding. Accordingto Holt, unlike emotional branding, viral branding, andmind-share branding, the delineation of “authors” and“audiences” in cultural branding is not necessarily clearand the reciprocity between firms and consumers affectthe building of brand values and brand identities (Holt2004, pp. 13–16). For example, the successful branding ofCorona beer was mainly driven by consumers’ perceptionof the beer as producing “party myth” (Holt 2004, pp. 16–20).

In competing with other products or firm, a goodunderstanding of game theories may also help firms

design brands that enhance the firms’ competitiveness inthe market. Hence, the brand the firm wants to create mustbe creative and difficult to be copied. The essence is thatthe brand must effectively help the firm or product bedifferentiated from their peers, meaning that the firmsmust try to take steps to maintain or enhance brand valuesby preventing potential competitors from constructing thesame or much similar brands (Muzellec 2006).

In addition, social-contract model argues that eachsocial member should act according to the social contractshe has agreed upon. Again, those who refused to acceptsocial contract terms have the freedom to exit (Rousseau2003). By contrast, we can also study branding as aprocess of building contractual relationship among firms,consumers, and other related actors such as the society(McMurrian and Washburn 2008, 5–22). This normativestudy explores branding ethics as well as mutual trustbetween consumers and firms. Due to the “voting” role ofconsumers and the reciprocity between firms and con-sumers, dishonest branding often brings negativeresponses to firms or brands, so branding must try to fulfillits promises to consumers.

Indeed, political theories can be applied to the studyof branding, which is detailed in the following two com-parative cases.

CASE 1 – BRANDING AND FRAMING: A TALEOF TWO CONSTRUCTS

One important strategy that is often used in politicalprocess is framing, which is very similar to branding inmarketing. According to political theories, parties orpolitical individuals differentiate themselves from theircounterparts generally through framing, which means tocreate or even manipulate specific values, meanings, orsymbols to achieve political goals. According to standardpolitical model, framing is “not simply an expression ofpreexisting group claims, but is an active, creative, con-structive process (McAdam et al. 2001, pp. 16–18, pp.132–137).

To build a good image of products or firms, brandingnot only has to focus on physical features of products orservices; it also has to explore deep meanings and valuesof firms or the society. Similarly, to political organization,framing not only aims at building specific policies orideas, it also centers on advocating political values, cul-tures, and long-term goals of political parties or politicalorganizations. Verbal and non-verbal indicators, there-fore, may be used in political framing. Political entrepre-neurs, for example, may use mass media to proposepolitical opinions and build or destroy images of policies,political leaders and parties (Paletyz 2002). Similarly,branding also need the help of mass media in creating,maintaining, and developing identities and values.


As framing is often used by political parties or politi-cal entrepreneurs to mobilize support to partisan doctrinesor goals, branding is often taken by firms to build ordevelop consumer support. The basic similarities of brand-ing and framing are that both can be used to construct ormanipulate meanings in given contexts and they areessential to maintain or increase the capacity of organiza-tions and momentum of a movement. In other words, bothbranding and framing can be deemed as specific tools torealize strategic goals of individuals, firms, and politicalorganizations. Just like framing, branding may also usesymbols and mass media to help build or destroy images,or to advocate particular values that company or consum-ers prefer (Elliott and Davies 2006).

Moreover, Laurent Muzellec points out that brandingmust consider resonance from all corporate brand audi-ences, meaning any “radical” or eccentric brand will harmthe firm (Muzellec 2006). By contrast, political theoriesintroduce the “median voter” model, which argues thatpolitical candidate must try to attract more voters as hecan. Therefore, the candidate must try to ensure that hispolicy platform (or political brand) is not radical andconforms to majority’s preference (Condorcet 1795; Riker1962).

To compete with rival firms or political organiza-tions, branding and framing both are intending to build“discursive hegemony” that helps firms or political par-ties survive and develop. If we say concrete quality ofproducts or services are hard power of firms, then such“discursive hegemony” is the “soft power” that firms orpolitical organizations often eagerly in pursuit of. Origi-nated from Plato and Aristotle, rhetoric is important inpersuading others to accept your point, which constitutesdiscursive power (Plato 1998). In an extreme sense, wemay label persuasion as “brainwashing,” the effect andthe process of which has been vividly illustrated byAlexander Solzenitzen and George Orwell (Ericson andMahoney 2009; Orwell 1946). Orwell, for example, saidthat political prose was issued “to make lies sound truthfuland murder respectable, and to give an appearance ofsolidity to pure wind.” Orwell hence complained thatpolitical writing, adied by vague or meaningless lan-guages, was used to hide the truth rather than express it. Inaddition, this ambiguous prose was a “contagion” tospread even to those who had no intent to hide the truth,and it concealed a writer’s thoughts from himself andothers (Orwell 1946). Thus, Orwell’s semantic criticismagainst lunguage manipulation supports LudwigWittgenstein’s point that words may not necessarilyexpress what they have intended to (Wittgenstein 1953).

Branding, however, is more often appeared in differ-ent forms. This feature means branding may design effec-tive symbols and verbal or non-verbal signs to “persuade”or convince consumers to accept their services, products,

or particular values and cultural modes (Johansson andHolm 2006). Branding also helps to integrate organiza-tional structures of firms by building employees’ identifi-cation of the firm or increasing employees’ morale (Aaker1991, 1996). Thus, effective branding not only intends tosecure external understanding and support, it also tries tomobilize internal support or identification of specificvalues or culture (Chong 2007; Ind 2001). The importantquestion is: to what extent can we stabilize and elongateinternal and external advocacy? To build and strengthenbrand community may be an effective answer to thisquestion.

CASE 2 – INSTITUTIONALIZING BRANDIDENTITY AND LOYALTY VIA

BRAND COMMUNITY

Branding is used to engender or enhance consumerawareness of and loyalty to the products, services, and theculture of the firm. To stabilize or institutionalize suchloyalty and awareness, it is vital to build a brand commu-nity. Muniz and O’Guinn define brand community as “aspecialized, non-geographically bound community, basedon a structured set of social relationships among users ofa brand”, and “Like other communities, it is marked by ashared consciousness, rituals and traditions, and a sense ofmoral responsibility” (Muniz and O’Guinn 2001). Thisdefinition stresses that brand community is non-geographically bounded and is driven by consumers’share of particular values of the brand.

Another definition of brand community is proposedby McAlexander et al. highlighting three dimensions ofthis construct: geographic concentration, social context,and temporality (McAlexander et al. 2002). McAlexanderet al. find that brand community establishes and strength-ens four relationships, including the consumer-product,consumer-brand, consumer-company, and consumer-consumer relationships. These interwoven and cumula-tive relationships help to build long term consumer iden-tification and loyalty. Members belong to the communityhave the similar consuming experience, value perception,and interactive personal relationship that function as exitbarriers. Aaker also finds that by leveraging brands,companies can build or increase consumer awareness andloyalty (Aaker 2004 b). The crucial point of these relation-ships is the creation and negotiation of meaning or culturethat strengthens consumers’ understanding and identifi-cation of the product, the brand, and the company (Balmer2006; McAlexander et al. 2002). The example is theconsumers of Jeep. They not only consume the vehicle,but also build strong identity and multi-actor relationshipsamong themselves with the Jeep brand as locus of all theserelationships. In this regard, brand is not entirely objec-tive, it is also the subjective image perceived or created byothers.


Similarly, framing in political process is also used topropagate political ideas, goals, and images of politicalorganizations, political leaders, or policies via positive ornegative languages (McAdam et al. 2001). In addition,framing is used to institutionalize support or emotionaladvocacy from different geographic areas, social strata,and political backgrounds, which constructs exit barriers.Regarding this, “information is not neutral”, and “infor-mation is strategically manipulated to serve different aimsfor different elements in the policy process.” The manipu-lated information is aimed to “provide meaning, clarifica-tion, and identity” (Zahariadis 2003, pp. 21–22) and isprimarily used to help political entrepreneurs “sell” theirpet solutions to the government or decision makers; or toinstitutionalize people’s support to the party, the leader, orthe policy. Just like consumers use Jeep as a symbol tobuild common identity, the Coke continuously reinterpretCoca Cola in the past decades, or the Morris Philipsinterpreted Marlboro reflected the “frontier spirit” toentrench the selling of the product to blue-collar workers(McAlexander et al. 2002; Holt 2004, pp. 22–28).

Indeed, to increase consumers’ or constituents’ aware-ness of, and support to, the specific goal and to acquire andmaintain consumer loyalty, firms or political parties mustintegrate their goals via branding or framing. Just likeforming consumer loyalty to firms, framing helps to buildup political loyalty to, or namely legitimacy of, politicalorganizations. If constituencies no longer believe in ide-ologies or values proposed by this political organization,they may turn to advocate other organizations. To secureconstituent advocacy, political entrepreneurs generallyhave to resort to framing or even manipulation to retainpeople’s support (Riker 1962, 1976).

When people do not believe any more, politicalparties may take another strategy to retain support; that is,coercion (Machiavelli 1989). However, in democraticregimes, the cost of such Machiavellian tactic can beprohibitive. Hence, to build a stable belief system amongthe people via peaceful means is essential for politicalparties’ survival and development. One strategy is toestablish informal institutions, which are “internallyenforced codes that modify behavior” (North 1990, pp.41–42). Informal institutions help to solve coordinationproblems in collective actions where the “free rider”problem is frequent (North 1990, pp. 40–41; Olson 1960).Informal institutions also help political activists mobilize

political movement by building advocacy networks amongparticipants across geographic borders (Keck and Sikkink1998).

As for branding, consumers or the firm can buildsome informal institutions to guide their behaviors. Accor-ding to Holt, Aaker, Muniz, and O’Guinn, Gobé, Schultz,and Chernatony, and other authors, successful brandingmust build a strong brand identity, brand myth, and, ifpossible, brand culture to coordinate multiple-level rela-tionships among firms, consumers, and shareholders acrossgeographic regions.

CONCLUDING REMARKS

This paper studies related theories in political scienceand marketing and suggests that interdisciplinary meth-ods may help to extent research scale and scope ofbranding theory. The paper finds that successful brandingand political framing depend on effective creation ofmeanings, values, cultures, and myths.

First, theories from political science may providesome useful analytical models or intuitions to the study ofbranding because they all involves the creation , maintain-ing, and development of values, ideas, cultures, andmeanings that affect the comparative or competitive ad-vantages of related players. Branding and political fram-ing both have the constructive implications of sellingideas or products.

Second, this paper also indicates that practicallyinterdisciplinary theories may provide assistance to pro-fessionals in the other field. For instance, knowing some-thing about framing may help businessmen enhance theirbusiness interests by igniting consumers’ imagination ofparticular values of proposed brands. There are ampleevidences to show how Chinese businessmen in the 1930sframed nationalism to increase their share in the tobacco,kerosene, alkaline, and textile market by labeling thepurchases as patriotic (Wang 1995).

In sum, branding focuses on the construction ofspecific brand and corporate values, meanings, cultures,and myths, while political framing is used to mobilizepolitical support and loyalty. Political theories thus mayprovide cognitive, organizational, and strategic assistancesto the study of branding.

REFERENCES

Aaker, David A. (1996), Building Strong Brands, NewYork: The Free Press.

____________ (2004 a), Brand Portfolio Strategy. FreePress.

____________ (2004 b), “Leveraging the CorporateBrand,” California Management Review, 46 (3), 6–18.

Axelrod, Robert (1984), The Evolution of Cooperation.Basic Books.

Balmer, John M. (2006), “Corporate Brand Cultures,” in


Brand Culture, Jonathan E. Schroeder and MiriamSalzer-Mörling, eds. New York: Routledge, 34–49.

Borgerson, Janet L., Martin Escudero Magnusson, andFrank Magnusson (2006), “Branding Ethics: Negoti-ating Benetton’s Identity and Image,” in Jonathan E.Schroeder and Miriam Salzer-Mörling, eds. BrandCulture. New York: Routledge, 171–85.

Brown, Stephen (2006), “Ambi-brand Culture: On a Wingand Swear with Ryanair,” in Brand Culture, JonathanE. Schroeder and Miriam Salzer-Mörling, eds. NewYork: Routledge, 50–66.

Chong, Mark (2007), “The Role of Internal Communica-tion and Training in Infusing Corporate Values andDelivering Brand Promise: Singapore Airlines’ Exper-ience,” Corporate Reputation Review, 10, 201–12.

Condorcet, Marquis de (1785), Essay on the Applicationof Analysis to the Probability of Majority Decisions,at [http://gallica.bnf.fr/ark:/12148/bpt6k417181].

Csaba, Fabian Faurhold and Anders Bengtsson (2006),“Rethinking Identity in Brand Management,” in BrandCulture, Jonathan E. Schroeder and Miriam Salzer-Mörling, eds. New York: Routledge, 118–35.

Dacin, Peter A. and Tom J. Brown (2006), “CorporateBranding, Identity, and Consumer Response, Jour-nal of the Academy of Marketing Science, 34 (Spring),95–98.

Dahl, Robert (1989), A Preface to Democratic Theory.California University Press.

Downs, Athony (1957), An Economic Theory of Democ-racy. New York: Harper and Row.

____________ (1972), “Up and Down with Ecology: The‘Issue-Attention Cycle’,” The Public Interests, 28,28–50.

Elliott, Richard and Andrea Davies (2006), “SymbolicBrands and Authenticity of Identity Performance,” inBrand Culture, Jonathan E. Schroeder and MiriamSalzer-Mörling, eds. New York: Routledge, 155–70.

Ericson, Edward E. Jr. and Daniel J. Mahoney (2009), TheSolzhenitsyn Reader: New and Essential Writings,1947–2005. Wilmington, DE: ISI Books.

Ferrell, O.C. and Michael D. Hartline (2008), MarketingStrategy, 4th ed. Mason, OH: Thomson-South West-ern.

Fog, Klaus, Christian Budtz, and Baris Yakaboylu (2005),Storytelling: Branding in Practice. Berlin: Springer.

Fournier, Susan (1998), “Consumers and Their Brands:Developing Relationship Theory in ConsumerResearch,” Journal of Consumer Research, 24(March) 343–73.

Gobé, Marc (2002), Emotional Branding. New York:Allworth Press.

Gramsci, Antonio (1971), Selections from the PrisonNotebooks. Lawrence and Wishart.

Habermas, Jurgen (1975), Legitimation Crisis. Boston,MA: Beacon Press.

Hatfield, Elaine, John T. Cacioppo, and Richard L. Rapson(1994), Emotional Contagion. Cambridge, UK: Cam-

bridge University Press.Hirschman, Albert O. (1970), Exit, Voice, and Loyalty:

Responses to Decline in Firms, Organizations, andStates. Boston, MA: Harvard University Press.

Holt, Douglas B. (2004), How Brands Become Icons: ThePrinciples of Cultural Branding. Cambridge, MA:Harvard Business School Press.

Ind, N. (2001), Living the Brand. London: Kogan Page.Johansson, Ulla and Lisbeth Svengren Holm (2006),

“Brand Management and Design Management,” inBrand Culture, Jonathan E. Schroeder and MiriamSalzer-Mörling, eds. New York: Routledge, 136–52.

Kapferer, Jean-Noël (2006), “The Two Business Culturesof Luxury Brands,” in Brand Culture, Jonathan E.Schroeder and Miriam Salzer-Mörling, eds. NewYork: Routledge, 67–76.

Keck, Margaret E. and Kathryn Sikkink (1998), ActivistsBeyond Borders: Advocacy Networks in Interna-tional Politics. Ithaca, NY: Cornell University Press.

Key, V.O. (1960), “Interrelations of Opinion” in PublicOpinion and American Democracy, V.O.Key, ed.New York: Alfred A. Knopf, Inc.

Lindblom, Charles (1959), “The Science of MuddlingThrough,” Public Administration Review, 19, 79–88.

McAdam, Doug, Sidney Tarrow, and Charles Tilly (2001),Dynamics of Contention. Cambridge: Cambridge Uni-versity Press.

McAlexander, James H., John W. Schouten, and HaroldF. Koenig (2002), “Building Brand Community,”Journal of Marketing, 66 (January), 38–54.

McMurrian, Robert and Judith H. Washburn (2008),“Branding: A Social Contract Between a Businessand Its Consumer,” in Contemporary Thoughts onCorporate Branding and Corporate Identity Man-agement, Melewar T.C. and Elif Karaosmanoglu eds.New York: Palgrave Macmillan, 5–22.

Machiavelli, Niccolò (1989), The Prince. Cambridge:Cambridge University Press.

Muniz, Albert and Thomas O’Guinn (1996), “BrandCommunity,” Journal of Consumer Research, 27(March), 412–32.

Muzellec, Laurnet (2006), “What is in a Name Change?Re-Joycing Corporate Names to Create CorporateBrands,” Corporate Reputation Review, 8 (Winter),305–21.

Neustadt, Richard (1960), “The Power to Persuade” inPresidential Power. New York: John Wiley & Sons,Inc.

North, Douglass (1990), Institutions, Institutional Changeand Economic Performance. Cambridge: CambridgeUniversity Press.

Olson, Mancur (1960), Logic of Collective Action. Cam-bridge, MA: Harvard University Press.

Orwell, George (1946), “Politics and the English Lan-guage,” [http://www.calvinvanhoek.com/articles/2007/04/politics-english-language/].

Paletyz, David, L. (2002), The Media in American Poli-


tics. Longman Publishers.Pindyck, Robert S. and Daniel L. Rubinfeld (2004),

Microeconomics, 7th ed. Prentice Hall.Plato (1998), Gorgias. Oxford: Oxford University Press.Riker, William H. (1962), The Theory of Political Coali-

tions. New Haven, CT: Yale University Press.____________ (1976), The Art of Political Manipulation.

New Haven: Yale University Press.Rousseau, Jean Jacques (2003), On the Social Contract.

New York: Dover Publications.Schattschneider, E.E. (1960), The Semisovereign People:

A Realist’s View of Democracy in America. NewYork: Rinehart and Winston.

Schroeder, Jonathan E. and Miriam Salzer-Mörling (2006),Brand Culture. Routledge.

Schultz, Majken and L. de Chernatony (2002), “SpecialIssue on Corporate Branding,” Corporate Reputa-tion Review, 5.

____________ and Mary Jo Hatch (2006), “A CulturalPerspective on Corporate Branding: The Case of

LEGO Group, in Brand Culture, Jonathan E.Schroeder and Miriam Salzer-Mörling, eds. NewYork: Routledge, 15–33.

Simon, Herbert A. (1957), Models of Man. New York:Wiley.

Tucker, Robert (1970), The Marx-Engels Reader. W.W.Norton & Company.

Tullock, Gordon (1960), Private Wants, Public Means:An Economic Analysis of the Desirable Scope ofGovernment. New York: Basic Books, 1–29.

Wang, Huilin (1995), Modern Chinese History. BeijingNormal University Press.

Wittgenstein, Ludwig (1953/2001), Philosophical Inves-tigations. Blackwell Publishing.

Zahariadis, Nikolaos (2003), Ambiguity and Choice inPublic Policy: Political Decision Making in ModernDemocracies. Washington D.C.: Georgetown Uni-versity Press.

Zaller, John (1992), The Nature and Origins of MassOpinion. Cambridge: Cambridge University Press.

For further information contact:Tony Yan

Marketing DepartmentNew Mexico State University

Las Cruces, NM 88003Phone: 575.646.6344



CONSTRUCTING BAYESIAN NETWORK AND NOMOLOGICALNETWORK OF PERFORMANCE-BASED USABILITY

OF MOBILE DEVICES

Taewon Suh, Texas State University, San MarcosJohn Ford, Old Dominion University, Norfolk

Jang-Ho Park, Pohang University of Science and Technology, PohangKyungdoc Kim, Pohang University of Science and Technology, PohangChi-Hyuck Jun, Pohang University of Science and Technology, Pohang

SUMMARY

This article builds up on the previous research by Suhet al. (2008). It begins with a definition of mobile devicesand a review of the usability literature. A basic definitionfor Usability of Mobile Devices (PUMD) is then pre-sented and followed by a discussion of the scale develop-ment process. Continuing Suh et al.’s (2008)’s basic scaledevelopment, two subsequent studies that were under-taken and added. Our first study, utilizing a set of 200subjects, employs a complementary tool, Bayesian Net-work (BN), to revalidate and expand the results from theprevious validation of PUMD. Second study, utilizing aseparate sample of 190 students, attempts to assess nomo-logical validity with a comparison of the final scale withother relevant marketing constructs.

In the previous research, PUMD were developed andvalidated through confirmatory factor analysis and ItemResponse Theory using two different research samples.The four-factor, 24 item scale was now deemed appropri-ate for further analysis (Suh et al. 2008). Our first studycontinues to describe how a Bayesian Network (BN) canbe used to make probabilistic statement about the multi-dimensional marketing construct. Similar to SEMs, BNsgraphically display the nature of relationships amongvariables and can be interpreted causally (Anderson andVastag 2004; Lauritzen 2000). Graphical analysis anddescriptive analysis were conducted based on the resultedBN. First, graphical analysis was based on the currentfour-factor structure of PUMD. Our initial analysis foundsix clusters in the BN, and, after few iterations, we couldget the expected number of clusters. Therefore, the graphi-cal analysis produced a similar result to our previousfactor structure. While the traditional factor analysis onlyreveals the factor structure among the measure items, BNportrays the probabilistic statements between measureitems.

To properly test for nomological validity, it is neces-sary to build a strong theoretical rationale for the nature ofthe relationships of the construct in question (PUMD) andother constructs. Thus, we build a nomological network to

examine the relationship of the measure with other impor-tant theoretical constructs to assess nomological validity.This study also allows for the exploration of the differentinfluences of the four sub-dimensions of PUMD in thenomological set. We conclude that among the four usabil-ity factors, Interface Usefulness (IU) is the most criticalfactor in the long-term relationship between a brand andits customers. It was interesting to note that the relation-ship for Ease of Learning and Use (ELU) was significantbut negative on all dimensions of brand relationshipsexcept for Sense of Community. As a result, being easy tolearn or use may be detrimental to the user’s relationshipwith the brand in certain situations; however, this conclu-sion needs further study. As a follow-up to the previousfinding, two path models were tested to shed further lighton the sets of relationships involved using both the advan-ced and basic user samples. In this case consumer satisfac-tion was measured using readily used and recognizedmeasures. IU directly influences the four brand relation-ship constructs and indirectly influences them throughsatisfaction’s mediation. In contrast, IU has no significantdirect impact on the relationship-based constructs in thebasic user model. Its role is totally mediated by satisfac-tion in the basic user model.

This research is both scientific and pragmatic innature. Following a strong argument for the relevance ofthe design mode in business research (Romme 2003), thisresearch is focused on the science-design interface tobridge the important gap between theory and practice.Built on the work of Herbert Simon (1996), the design-mode camp in the research debate argues the main ques-tion, “Will it work?” rather than “Is it valid or true?” Thisallows the researchers to pursue the true synergiesbetween science and design. Thus, in our attempt to gainthe complementarities which the design camp posits, it isimportant to stretch beyond the traditional approach ofdeveloping a methodological instrument and representa-tional knowledge in the subject matter and to seek to bemore pragmatic in the development of knowledge in theservice of action, which is normative and synthetic innature (Simon 1996).


This type of design-science interface concerns bothtesting in practical contexts and grounding in empiricalfindings (Romme 2003). Hopefully with this kind of

effort, the study of marketing may become increasingpluralistic in nature as it evolves (e.g., Achrol and Kotler2012). References are available upon request.

For further information contact:Taewon Suh

Texas State University601 University Drive

San Marcos, TX 78666Phone: 512.245.3239



INVESTIGATING THE EFFECTS OF VAGUENESS IN ADVERTISINGSLOGANS ON BRAND PREFERENCE

Widyarso Roswinanto, University of North Texas, Denton

SUMMARY

Vagueness embedded in advertising slogans play animportant role in determining consumers’ experiencesand attitudes upon receiving the message. Vagueness wasgenerally defined as violation of semantic standard ofinformativeness, explicitness and clarity (Kaufner 1983);incomplete definition (Bosch et al. 1983); and the oppo-site of precision (Cook 2007). This paper investigated theeffects of vagueness on cognitive response by measuringthe magnitude of evoked thought and the effects of bothrelated constructs on consumer’s attitude and purchaseintention. Preliminary study of content analysis applyingMeta Model coding (Bandler and Grinder 1976) was usedto set the foundation as to how various vagueness levelsexisted in the slogan, how vagueness was recognized,coded and analyzed. The result of the content analysis alsoshowed how vagueness is pervasively found in advertis-ing slogan and in what kind of patterns it usually appears.

Extant literature in advertising was extensive with theinvestigation of vivid (vs. abstract) concept in the adver-tising message in its effect on recall and attitude. How-ever, such concept of “vividness” was problematic and themeasures of vividness such as Vividness of Visual Imag-ery Questionnaire/VVIQ (Marks et al. 1995) did notcorrelate highly with many other variables (Benjafield1997). The influence of vividness on consumer’s recalland persuasiveness is inconsistent across literature(Simpson and Borgida 1991; Taylor and Thompson 1982;Collines et al. 1988; Frey and Eagly 1993). This paperfocused on the vagueness (vs. clarity) quality of advertis-ing messages with the attempt to further scrutinize one ofthe building blocks of the abstractness (vs. vividness)construct.

Drawing on the extant literature, seven hypotheseswere developed. Perceived vagueness was hypothesizedas having negative association with the (1) level of evokedthought and (2) attitude toward brand. The level of evokedthought would be positive association with (3) the attitudetoward brand and (4) purchase intention. The level of (5)evoked thought and attitude toward brand would mediatethe relationship between perceived vagueness of the slo-gan and purchase intention. Two moderators; (6) need forcognition / NFC and (7) purchase involvement; werehypothesized as amplifying the effect of perceived vague-ness on the level of evoked thought and on the attitudetoward brand.

Constructs were measured using 7-point Likert-stylescales for attitude toward brand (Spears and Singh 2004),purchase intention (modified from Bower and Landreth2001), perceived vagueness (developmental), evokedthought (modified from Unnava and Burnkrant 1991),cognitive dimensions of brand experience (Brakus et al.2009), the need for cognition (Cacioppo et al. 1984), andpurchase involvement (Mittal 1989). The sample wascollected from a Southwestern public university in theUnited States. In filling out the questionnaires, partici-pants were asked to recall ad slogans they noticed in thepast two weeks. There were 433 out of 512 responses usedin the final study. Early and late survey respondents werecompared and the result showed lack of non-responsebias.

The used measures were validated using item-to-totalcorrelations and confirmatory factor analysis usingLISREL 8.8. (Jöreskog and Sörbom 2006). The global fitindices were reported. Individual composite reliabilitiesand the average variance extracted (AVEs) were calcu-lated using procedures suggested by Fornell and Larcker(1981). All of the factor loadings were significant, andcomposite reliabilities range from 0.75 to 0.81. The AVEsranged from 0.93 to 0.96, above the recommended 0.50level (Bagozzi and Yi 1988; Fornell and Larcker 1981;Hair et al. 2006). Discriminant validity was examined bycomparing the correlations among constructs and theAVE.

The structural model was analyzed and showed accep-table overall fit of the model with the χ2 (183, N = 403) =624.34, χ²/df = 3.41, NFI = .97, NNFI = .97, CFI = .98,SRMR = .061, and RMSEA = .078. The path coefficientswere examined and all of the hypothesized associationswere significant. As expected, perceived vagueness nega-tively predicts both evoked thought (γ = -.16) and brandattitude (γ = -.32), and evoked thought was positivelyassociated with brand attitude (β = .26), while the indirecteffect of perceived vagueness to brand attitude (β = -.04).Evoked thought has a positive impact on purchase inten-tion (β = .18) and similarly but stronger positive effect ofbrand attitude on purchase intention (β = .58), while theindirect effect of evoked thought on purchase intention(β = .15) and indirect effect of perceived vagueness onpurchase intention (β = .24). This model was better thanother two alternative models that were analyzed.


Multi-group analysis was used to test the moderatingeffects of need for cognition and purchase involvementusing LISREL (Schumacker and Lomax 2010). Purchaseinvolvement was significant as a moderator Δχ²(23) =35.43, p-value = 0.047, while the NFC was not significantΔχ²(23) = 19.8, p-value = 0.065.

The study contributes to the advertising literatures insupporting the existence and the patterns of vagueness inadvertising slogans which at the same time sets a founda-tion for studying vagueness in persuasion message in thefuture, the influence of vagueness in advertising slogan onsome critical variables in consumer preference, the nega-

tive relationships of vagueness as one of the dimensionsof vividness construct on consumer’s attitude and behav-ior, and the mediated relationship between perceivedvagueness on brand attitude and purchase intention byevoked thought. One of the managerial implications isthat slogans need to be designed in ways that effectivelyaim the specific targeted segments. This can be done byentertaining them with clarity and precision reflected inthe wording of the slogans, which at the same timeproviding enough vagueness to be relevant to the widerscope of audiences. References are available uponrequest.

For further information contact:Widyarso Roswinanto

Department of Marketing and LogisticsCollege of Business

University of North Texas1155 Union Circle #311160

Denton, TX 76203–5017Phone: 940.565.3338

E-Mail: [email protected] / [email protected]


CONSUMING ETHICALLY: THE ROLE OF EMOTIONS

Diana Gregory-Smith, Durham University Business School, United KingdomAndrew Smith, Nottingham University Business School, United Kingdom

Heidi Winklhofer, Nottingham University Business School, United Kingdom

SUMMARY

Previous studies about emotions have enhancedunderstanding of their influence in generic consumptiondecisions but a detailed examination of emotions in thecontext of ethical consumption has been lacking. Previousresearch in psychology has established that some emo-tions (e.g., guilt, shame) play a pivotal role in choices witha moral/ethical dimension (Haidt 2003; Tangney, Stuewig,and Mashek 2007). The term ‘ethical consumption’implies an ethical/moral dimension, where the hedonicfunction and/or product utility are, to some extent, subor-dinated by concerns about right and wrong and conse-quences of consumption acts (Starr 2009) in areas such ashuman welfare, animal welfare and environmental wel-fare (Low and Davenport 2007).

In order to examine the manifestation and role ofemotions in the context of ethical consumption, in-depthinterviews with consumers with different degrees of ethi-cal orientation were undertaken. A theoretical thematicanalysis was employed in order to systematically code andanalyze the data (Thompson 1997; Braun and Clarke2006).

Overall, the findings indicated that self-consciousbehavior that is perceived to be ethical is context depen-dent more than person dependent, with the majority ofconsumers displaying patterns of complex behavior invol-ving both ethical and unethical choices. It was also foundthat consumers compensate their unethical choices withethical ones in many categories and stages of consump-tion and disposal, with emotions playing a key role in thiscompensatory process. Positive emotions (e.g., feelinggood, pleased, happy, proud) were found to encouragefuture ethical choices while negative emotions (e.g., feel-

ing uncomfortable, guilty, ashamed, regretful, disap-pointed) delayed or restrained unethical consumption.Emotions were experienced at different stages within theconsumption cycle (i.e., before, during and after pur-chase) and emerged in different areas of concern. Thefindings also identified different types of strategiesemployed by consumers to manage their guilt: expediency-oriented actions, introspection, diminishing net impacts,and the use of positive emotions.

This study has both theoretical and practical implica-tions. Firstly, this research adds further evidence about themanifestation of emotions in consumers’ decision mak-ing. Secondly, it contributes to the literature on ethicalconsumption by moving the debate further from cogni-tion-related variables and by highlighting the role playedby emotions both in relation to ethical and unethicalchoices. Within this context, emotions can explain theattitude-behavior gap for consumers with various degreesof ethical orientation. Thirdly, knowledge about howconsumers manage their guilt is of use for marketers sincecounteracting strategies could be employed in marketingcommunications, i.e., aimed at neutralizing the tech-niques that allow consumers to justify and sustain theirless ethical behavior.

Future research needs to be undertaken to validatesome of the present results. A longitudinal study couldobserve consumers’ choice and behavior over time as thefindings presented here rely on respondents’ memory.Future research could also investigate the efficacy ofmarketing communications employing positive or nega-tive emotions in determining attitude change and encour-aging ethical behavior. References are available uponrequest.

For further information contact:Diana Gregory-Smith

Durham University Business SchoolQueen’s Campus, Stockton University BoulevardStockton on Tees, United Kingdom, TS17 6BH

Phone: 44(0)191.33.40112Fax: +44(0)191.3340010



DOES IN-STORE MARKETING REDUCE THE ATTITUDE-BEHAVIORGAP FOR ETHICAL PRODUCTS?

Gulberk Keysan, PriceWaterhouseCoopers, LondonVishal Talwar, London School of Economics, London

Amitav Chakravarti, London School of Economics, London

SUMMARY

In-Store Marketing and “Ethical Gap”

The enigma around why positive consumer attitudestoward ethical products do not seem to translate intoactual purchase behavior has led many researchers toexplore the “ethical attitude – behavior gap.” However,till date, the focus of research has been on the underlyingreasons behind this ethical gap, whereas the effectivenessof the strategies eliminating these reasons has not beentested empirically. In this paper, the effectiveness of twoin-store marketing variables – fair trade labeling as well astemporary price reductions on consumers’ purchase behav-iors are explored. In order to evaluate their effects com-pared to demographical, attitudinal or behavioral vari-ables, this study analyses these two factors’ interactionwith other situational context elements – mainly shoppinggoals and shopping trip type.

A discrete choice experiment (DCE) was designed todetermine the significance of certain in-store marketingvariables in customers’ purchase decisions for Fair Tradechocolate compared to personal attributes. The most con-crete variables – Fair Trade labeling as well as temporaryprice reductions – were chosen as the attributes to beassessed efficiently in a paper-based survey overcomingsocial desirability bias at the same time. The variables FairTrade labeling and temporary price reductions were mani-pulated whereas the variables shopping trip type andshopping goals were measured. In terms of shoppinggoals, participants were divided into two categories –planned vs. impulsive shoppers (using median split basedon how much they plan ahead before grocery shopping).Likewise, for the shopping trip type, participants weresplit into two categories – fill-in vs. major shoppers (usingmedian split based on the time they spend on groceryshopping in general). In order to measure the effect ofsituational variables on the purchase of Fair Trade prod-ucts, we chose to focus on one product category – choco-late. Being a hedonic product, chocolate is considered tobe bought mostly as part of unplanned purchases whichcan be triggered effectively by in-store stimuli. However,in order to isolate the effect of situational dimensions frombrand loyalty, which is commonly cited as another in-

tervening factor, anonymous chocolate brands were intro-duced in this study.

Findings reveal that the presence of the Fair Tradelogo increases consumers’ purchase decisions signifi-cantly whereas temporary price reductions ironically haveno significant impact. Furthermore, it is noted that theeffect of Fair Trade labeling on purchase decisions islarger than the effect of ethical attitudes, highlighting theimportance of situational context in ethical decision-making. Besides, the interaction between different situ-ational elements such as shopping trip type or shoppinggoals is found to influence the effectiveness of in-storemarketing initiatives in different ways.

It is found that the existence of a Fair Trade logoincreases the probability of purchasing a chocolate by 25percent, confirming our hypothesis about the positiveeffect of Fair Trade labeling on purchase decisions. FairTrade labeling is assumed to overcome consumer skepti-cism about a product’s origins and its ethical credentialsby providing clear and credible information. The signifi-cant impact of Fair Trade labeling supported the con-sumer arguments for the lack of credible informationbeing an obstacle to purchasing Fair Trade products.Likewise, the positive effect of Fair Trade labeling can bejustified due to its role in mainstreaming Fair Tradeproducts – differentiating them in a mass market byimproving the connection between producers and con-sumers. Furthermore, Fair Trade labeling is considered tobe instrumental for building a brand image. The impor-tance of the Fair Trade logo in consumers’ purchasedecisions indicates that it creates brand awareness; form-ing a brand image subsequently when accompanied withthe relevant message in a market setting.

On the other hand, it is found that the price differencevariable does not have a statistically significant impact,disproving our hypothesis about the positive effect oftemporary price reductions on purchase decisions. Theinsignificant impact of price cuts actually provides evi-dence for the “ethical premium” concept. Although it isargued that consumers consider high prices of Fair Tradeproducts to be a key impeding factor, they have beenincreasing their support for Fair Trade and becomingmore likely to pay a premium for it.


In terms of ethical decision-making, shopping triptype and planning ahead are found to have no significantimpact on individuals’ purchase decisions. However,when ethical decision making of different shopper groupsis compared, it is found that the effect of certain in-storevariables or attitudinal variables differed across shoppergroups. Firstly, planned shoppers are found to be moreinfluenced by the Fair Trade logo as well as temporaryprice reductions when compared to impulsive shoppers.This is in contrast to previous studies which claim thatconcreteness of shopping goals reduces the influence ofin-store actions. Interestingly, this finding may supportanother perspective about planned shoppers, i.e., plannedshoppers are more likely to consider product image as thekey factor in their purchase decisions. When the effect ofFair Trade labeling in the formation of brand image istaken into account, it may be concluded that the Fair Tradelogo has an indirect impact on the purchase decisionthrough the product image which may be more importantfor planned shoppers.

Furthermore, it is noted that once consumers planahead for chocolate purchase, they seem more likely topurchase Fair Trade brands. This finding actually sup-ports the assertion that following the accumulation ofinformation about Fair Trade, consumers have started toform Fair Trade loyalty. In the long run, this may developthe Fair Trade market much further.

In terms of shopping trip type, this study finds thatfill-in shoppers who go grocery shopping more frequentlyto meet urgent needs are more likely to be influenced byin-store activities. On the other hand, past purchase fre-quency for Fair Trade products seems to be a betterindicator for future purchase decisions of major shoppersrather than the existence of in-store stimuli.

Understanding the effect of in-store marketing onpurchase decisions can help Fair Trade brands to positiontheir products better and establish long-term customerrelationships. The significant impact of the Fair Tradelogo suggests that more sophisticated Fair Trade labelingincluding vivid visuals/images from the lives of producersas well as wider use of the logo in retail promotions mightform an emotional bond with consumers. This would helpsuch brands build an identity and create customer loyaltyin the long-run. The consumers’ insensitivity to differentprice levels suggests that instead of considering tempo-rary price reductions, marketers should focus on thepremium positioning of Fair Trade products. Once high-quality perceptions are established in customers’ minds,another underlying factor behind “ethical gap” – theskepticism about product quality – may be eliminated.

Our findings which relate to distinctions between fill-in and major shoppers in terms of their susceptibility toin-store stimuli may be useful for retailers in several ways.Retailers can plan their in-store activities so that theywould attract fill-in shoppers. Likewise, our study showsthat consumers who purchase Fair Trade products morefrequently are more likely to buy Fair Trade chocolate.This indicates that cross-selling opportunities exist withindifferent Fair Trade product ranges. Thus, a more exten-sive Fair Trade consumer segmentation should incorpo-rate not only attitudes or demographics but also pastpurchase behaviors.

Given the relevant implications of this study, it wouldbe useful to extend the product range in a future study oreven compare the effect of in-store marketing variables ondifferent product categories in an experimental setting.References are available upon request.

For further information contact:Gulberk Keysan

PriceWaterhouseCoopers LLP7 More London Riverside

London, United Kingdom SE1 2RTPhone: +44(0)7586.757.061



UNDERSTANDING TEENAGE POKER GAMBLING: POLICY ANDCONSUMER BEHAVIOR IMPLICATIONS

Sudhir H. Kale, Bond University, AustraliaNatalina Zlatevska, Bond University, Australia

SUMMARY

Poker has become a popular pastime for many teen-agers all over the world. Once characterized as a gameplayed in dingy basements reeking of stale cigar smokewith tattered cards and grimy chips, poker gambling isnow perceived as the “sport of choice” among mediacelebrities, catwalk models, Hollywood actors, and any-one aspiring to quick fame and fortune. Stories in popularmedia seem to suggest that the popularity of poker isluring increasingly large number of young people togamble, creating a new cadre of youth gamblers andproblem gamblers.

Despite the vast media attention poker gambling hasthus far received, there are very few academic studiesdevoted to poker, especially to young people’s gamblingon poker. Consequently, it is very hard for policy makers,regulators, educators, and parents to respond to youthpoker gambling in an informed and effective manner. Ata crucial time when several states and provinces aredeciding whether to legalize internet poker gambling, weneed solid data to assess the prevalence of poker gam-bling, especially among our youth, and to better compre-hend the antecedents and consequences of this phenom-enon. In a recent debate on legalizing internet poker, Rep.Frank Wolf characterized internet poker as “the crackcocaine of gambling.”

Our research sought to focus on teenage poker withthree key objectives: (1) establish general gambling par-ticipation and poker gambling prevalence among teenag-ers in Australia; (2) understand how poker gamblers differfrom other gamblers and non-gamblers on the trait ofimpulsive sensation seeking; and (3) to understand thedegree of gambling related distortions among poker play-ers relative to those teenagers who gamble on other thingsbesides poker.

Based on an extensive literature search and on focusgroups with 25 Australian teenagers, we formulated threeresearch hypotheses:

H1: Teenagers who have gambled will score higher on the

impulsive sensation seeking scale than those whohave never gambled.

H2: Impulsive sensation seeking, cognitive distortions,

and problem gambling tendencies will be positivelyintercorrelated.

H3: Poker players will score higher than other gamblers

on impulsive sensation seeking, cognitive distor-tions, and problem gambling.

Data were collected online from 2,000 teenagers,aged 15–19, from the Australian state of Victoria. Underan arrangement with the commercial firm that provided usaccess to the sample, prospective respondents signed up toparticipate in online surveys in exchange for points, witharound three dollars worth of points per survey earned byeach respondent. The points earned can then be cashed forpurchases.

The online questionnaire took 20–25 minutes tocomplete. It comprised of questions relating to respon-dents’ level of impulsive sensation seeking, their gam-bling behaviors and gambling-related cognitions, theirpoker gambling activity, and their proclivity for problemgambling. Other information gathered but not addressedin this summary included information on respondents’consumption of tobacco, recreational drugs, and alcohol,parental awareness of teenagers’ gambling, and informa-tion on help-seeking behavior from teenagers who hadfaced issues relating to relationships, academic problems,alcohol-related problems, and gambling-related problems.

The sample comprised of 1003 females and 997males. Just over 88 percent of the participants went toschool or university full-time, 3.8 percent worked full-time, and 46.5 percent worked part-time. Just over half ofthe sample (54.7 percent) had engaged in some form ofgambling at least once in their lifetime. Around two-thirds(66.3 percent) of the teenagers had played poker formoney at some time in their life. Four hundred and fiftyteenagers (22.5 percent) had played poker for money atleast once in their life, and 308 respondents (15.4 percent)had done so in the preceding twelve months. Of the 450teenagers who had gambled on poker, 245 were males and205 were females.

Participant scores on impulsivity and sensation seek-ing were computed by summing responses to the 19-item


Impulsive Sensation Seeking Scale (ImpSS) adapted fromthe Zuckerman-Kuhlman Personality Questionnaire(Zuckerman and Kuhlman 2000). Participants were askedto indicate whether statements such as “I like doing thingsfor the thrill of it” were true (coded as one) or false (codedas zero). The possible scores on ImpSS ranged from zeroto nineteen. The mean ImpSS score was 9.49 with astandard deviation of 3.95. The scale showed acceptablereliability (Cronbach’s alpha = 0.76). Our analysisrevealed a significant difference on ImpSS between gam-blers (M = 9.92, SD = 3.88) and non-gamblers (M = 8.97,SD = 3.98), t

(1,1998) = 5.40, p < 0.001. Hypothesis H

1 is thus

supported.

To test the prevalence of irrational beliefs amonggamblers, the Gambling Related Cognitions Scale (GRCS,Raylu and Oei 2004) was used. This scale comprises of 23items with five underlying dimensions: interpretive con-trol or bias, illusion of control, predictive control, gam-bling-related expectancies, and a perceived inability tostop gambling. This scale exhibited excellent internalreliability (Cronbach’s alpha = 0.96).

The proclivity for problem gambling was measuredusing the nine-item Problem Gambling Severity Index(Ferris and Wynne 2001). Four items assess problemgambling behaviors (e.g., How often have you bet morethan you could afford to lose?) and five items seekinformation on the adverse consequences of gambling(e.g., How often have you felt you might have a problemwith gambling?). The nine questions were answered on a0–3 scale (0 = Not in the last year or never; 1 = Sometimes;2 = Most of the time; and 3 = Almost always). Cronbach’salpha for this scale was 0.93.

Analysis of data revealed a weak but positive corre-lation of teenagers’ scores on the ImpSS and PGSI scales,r = 0.13, p < 0.01, and between ImpSS and GRCS, r = 0.18,p < 0.001. The PGSI and GRCS scores were highlycorrelated with r = .66, p < 0.001, thus supporting H

2. The

GRCS scores for poker players vis-à-vis other gamblersreveal that teenagers who have gambled on poker evenonce scored considerably higher (M = 45.42, SD = 17.41)compared to those gamblers who have never bet on poker(M = 37.11, SD = 14.76, F

(1,1093)=72.29, p < 0.001). The

mean ImpSS rating for poker players was 10.43 (SD =3.85), while that for non-poker gamblers was 9.56 (SD =3.86). This difference in means is statistically significant(F

(1,1093) = 13.34, p < 0.001). Finally, the mean PGSI score

for poker players was 3.21 (SD = 5.08), while that forother gamblers was 1.18 (SD = 3.02, F

(1,1093) = 68.36, p <

.001). These results are supportive of H3. Closer inspec-

tion of the results on the PGSI scores of poker gamblersvis-à-vis other gamblers reveals that poker players arealmost three and a half times more likely to be problemgamblers compared to other teenagers who have gambled,but never on poker (18.4% vs. 5.4%). Furthermore, while70 percent of those who have gambled on poker at leastonce in their life fall into the “Non-problem” or “low-risk”category, almost 30 percent face moderate to high gam-bling risks.

Our cumulative analysis suggests that while pokergambling is a popular pastime among Australian teenag-ers, it is nowhere as rampant as some media stories wouldlead one to believe. While more than half the 15–19 yearolds have gambled at least once in their life, slightly overone in five have ever gambled on poker. What we do findconcerning is that poker players exhibit a far highertendency to become problem gamblers. Even havinggambled on poker just once dramatically increases ateenager’s risk of becoming a problem gambler.

This is the first reported large-scale study to assesspoker playing among Australian teenagers. Our findingscorroborate previous studies suggesting that poker play-ers tend to have more distorted cognitions than othergamblers (Wohl et al. 2005). Poker players also tend to bemore impulsive and sensation seeking oriented comparedto other gamblers. If other specific individual and contex-tual factors influencing problem gambling in poker play-ers can be identified, advances toward possible interven-tions for prevention and treatment of problem gamblingamong poker players can be made (cf., Dickerson 1993;Mitrovic and Brown 2009). Given the high scores ofpoker gamblers on GRCS, cognitive behavioral therapyshould prove particularly effective with poker playerswho face high problem-gambling risks. References areavailable upon request.

For further information contact:Sudhir Kale

Bond UniversityGold Coast, QLD

AustraliaPhone: +61.755.952214



“IT’S NOT MY FAULT” AND “MONEY CAN BUY ANYTHING”:RECOVERING FROM ETHICAL FAILURES

Sekar Raju, Iowa State University, AmesPriyali Rajagopal, Southern Methodist University, Dallas

SUMMARY

Research has found that denying ethical failureselicits more favorable attitudes than accepting them (Ferrinet al. 2007) due to the differential inferences people drawfrom the failure (Reeder and Brewer 1979). An ethicalfailure is more diagnostic in judging moral character andhence likely to have greater informational value andimpact on judgment. This suggests that the effect ofaccepting ethical failures is difficult to overcome. How-ever, the literature on moral disengagement (White,Bandura, and Bero 2009) provides insight into recoverystrategies from ethical transgressions. We focus on twosuch alternatives that are often observed in a marketingcontext – external attribution and monetary compensa-tion.

Attributing a failure to outside sources is commonlyadopted by firms (e.g., Ford/Firestone). External attribu-tion shifts the culpability of the unethical act away fromthe firm. By attributing the failure to an outside party,firms can restrict the negative association of having com-mitted an unethical act and consequently reduce the futurelikelihood of repeating the act. In other words, the diag-nostic value of the ethical failure for consumers is re-duced. Thus, when the failure is attributed externally,consumers will have more favorable attitudes and inten-tions toward the firm, greater trust in the firm and be lesslikely to believe that the firm will make similar errors inthe future, as compared to when the firm accepts thefailure. Finally, there will be no differences in attitudes,intentions, future likelihood estimates or trust in the firmwhen the firm externally attributes the failure as comparedto when the firm denies the failure.

Monetary compensation, however, attempts to shiftthe focus away from the failure and its consequences toresolution of the failure. Therefore, we expected to findthat this response would improve consumer attitudes butwould not improve intentions or reduce the future failurelikelihood.

In real life, consumers often hold varying levels oftrust in a company. We therefore also hypothesize thatunder conditions of low prior trust, there will be nodifferences in consumer responses toward the firmbetween offering monetary compensation, accepting,externally attributing or denying the failure.

Study 1 (N = 154; Table 1, Appendix)

Design: 2 (prior trust: high versus low) x 3 (recovery:accept, deny, external attribution) between subjects study.The dependent measures included attitudes toward thefirm, trust in the firm, likelihood of future transgressionsby the firm and behavioral intentions toward the firm (7-point scales).

We found a significant interaction between prior trustand company responses on attitudes (F (1, 147) = 7.62,p < .01), intentions (F (1, 147) =10.3, p < .01, post-failuretrust (F (1, 147) = 4.89, p < .01) and future likelihood offailure (F (1, 147) = 2.73, p = .06). Planned contrastsrevealed that when trust was high, externally attributingthe failure led to results that were no different fromdenying the failure (p’s > .1). However, when trust waslow, externally attributing the failure did not improve anyof the dependent measures as compared to accepting thefailure.

In addition, a mediation analysis revealed that theresponse significantly predicted post-failure trust in thecompany (F (1, 71) = 7.14, p < .01) and likelihood offuture failures (F (1, 71) = 6.97), p < .01); the likelihoodof future failures significantly predicted post-recoverytrust in the company (β = .68, p < .05), and when both thetype of response and likelihood of future failures wereincluded as predictors of post-recovery trust, only likeli-hood of future failures remained a significant predictor (F(1, 70) = 19.85, p < .01) and the type of response was non-significant (F (1, 70) = 2.33, p > .1). Thus, externalattribution reduces the diagnosticity of the ethical failureand thereby improves consumer perceptions of the firm.

Study 2 (N = 147; Table 2)

In study 2, we varied the failure context (environ-mental damage) and conducted a 2 (prior trust: highversus low) x 3 (response: accept, deny, external attribu-tion) between subjects study. We replicated the findingsfrom Study 1 and found a significant interaction betweenprior trust and type of response on consumer attitudes (F(2, 140) = 3.57, p < .05) and post-recovery trust (F (2,140) = 3.49, p < .05). Under high trust, external attributionelicited responses that were no different as compared todenying the failure (p’s > .1) but were significantly higherthan when the failure was accepted (p’s < .05) but under


low trust, external attribution elicited results that were nodifferent from accepting the failure.

Study 3 (N = 200; Table 3)

Design: 2 (prior trust: low versus high) x 4 (responsesto failure: accept, deny, monetary compensation, externalattribution) between subjects study. We used the samescenario and dependent measures used in study 1.

We found the expected interaction between priortrust and company responses on attitudes toward the firm(F (3, 192) = 3.0, p < .05), intentions toward the firm (F (3,192) = 4.39, p < .05) and post-failure trust in the firm (F(3, 192) = 2.29, p = .07). We replicated the pattern ofresults for external attribution that was found in study 1.Thus, external attribution improved attitudes and inten-tions and lowered future likelihood of failure when trustwas high.

When prior trust was high, monetary compensationimproved attitudes, but not intentions, trust or future

failure likelihood, as compared to accepting the failure.Thus, externally attributing the failure was better thanoffering monetary compensation in terms of future inten-tions and diagnosticity. Also as predicted, there were nodifferences between any of the four conditions on atti-tudes, intentions, post-failure trust in the firm or thelikelihood of future failures when prior trust in the firmwas low (all p’s > .1). Thus, any response to an ethicalfailure appears to require a modicum of trust in the firm inorder to be successful.

While prior research has shown that denying ethicalfailures is the only option to retain favorable consumerattitudes and intentions, we identify exceptions to thisfinding. We show that external attribution reduces theperceived diagnosticity of ethical failures, leading tomore favorable attitudes/intentions toward the firm whilemonetary compensation can improve attitudes, but notfuture intentions. We also identify an important modera-tor to these effects – prior trust in the company. Referencesare available upon request.

For further information contact:Sekar Raju

Iowa State University3212 Gerdin Business Bldg.

Ames, IA 50011Phone: 515.294.8463



WORD-OF-MOUTH BEHAVIOR, ONLINE ACTIVITY AND COMPANY’SCOMMUNICATION STRATEGY

Danny P. Claro, Insper Education and Research Institute, BrazilSilvio A. Laban Neto, Insper Education and Research Institute, BrazilPriscila B.O. Claro, Insper Education and Research Institute, Brazil

SUMMARY

Word-of-mouth (WOM) communication has beenrecognized for quite some time as a powerful means ofdisseminating information about products and services(Brooks 1957). For example, Schmitt, Skiera, and Vanden Bulte (2011) found that referral programs generatemore customers in both the short- and long-term; Cheemaand Kaikati (2010) found that high-uniqueness consum-ers (consumers who prefer to differentiate themselvesfrom members of their respective reference groups) weremore likely to recommend products they themselves con-sumed; and Court, Gordon, and Perrey (2010) proposedusing “word-of-mouth equity” as an index of a brand’spower to generate messages that influence consumers’purchasing decisions. The importance of the WOM phe-nomenon has been shown in studies that describe theeffects of WOM on purchasing decisions (e.g., Chevalierand Mayzlin 2006). However, the vast majority of theseminal work and observations about informal, unsolic-ited WOM communication have been based on a market-ing world untouched by the internet (Brown, Broderick,and Lee 2007; Dellarocas 2003; Godes et al. 2005; Hennig-Thurau et al. 2004).

Online communities have increased in size, number,and character, and companies have begun to recognize thegrowing importance of WOM in that context. Firms’investment in online technology is increasing, and signifi-cant resources are allocated in digital and online market-ing venues. The widespread, but misguided, belief amongbusiness managers is that companies’ websites must pro-vide community-generated content if they are to buildbrand loyalty and achieve high sales figures. This belief isposing an enormous challenge for such companies. Theopportunities for managers to monitor and influence WOMcommunication lie in the increasing accessibility andreach of the internet. Companies then need to be vigilantto WOM behavior and online activity that allow for abetter understanding of what makes WOM effective andhow to measure its impact. While WOM Behavior refersto the motivations and beliefs that drive individuals topass information, Online Activity refers to the exercise ofinformation dissemination by using online means.

We therefore aim to study WOM behavior and ana-lyze the impact of companies’ on- and offline communi-

cation and consumers’ online activities on consumption.For example, online customer reviews are an importantsource of information about product quality for interestedconsumers, complementing, and very often substitutingfor, other aspects of a company’s communication strat-egy. By collecting information about consumers’ per-spectives on companies’ communication strategies andanalyzing it together with information about their ownbehavior and activities, we offer insights into how con-sumers react to and influence the online environment.This study is an attempt to contribute to the WOM litera-ture and provide a model with four hypotheses as well asevidence of a survey data drawn from a population ofinternet heavy users. Our sample yielded a response rateof 9.2 percent (313); after excluding incomplete question-naires, we had 248 usable responses.

Using the antecedent variables, mediator variablesand outcome variable specified in our research model, weanalyzed three sets of ordinary-least-square regressions.The first set of equations took the two dimensions ofWOM behavior as dependent variables and the company’scommunication strategies as the explanatory variables. Inthe second set of equations, the four dimensions of onlineactivity were considered the dependent variables and thedimensions of WOM behavior were considered the explan-atory variables. In the third set of equations, the WOMoutcome was the dependent variable and the communica-tion strategies and WOM activity were the explanatoryvariables. A total of seven equations were modeled; allincluded the control variables.

Our results provide evidence to address other schol-ars’ calls for research that leads to a better understandingof WOM behavior and that identifies those who influenceonline activity and consumption (Brown and Reingen1987; Schmitt, Skiera, and Van den Bulte 2011). It hasalso been suggested that studying companies’ communi-cation strategies might provide indications as to whichstrategies are more effective in encouraging online con-sumption. This paper presents evidence that such commu-nication influences consumption through WOM behaviorand online activity.

The direct impact of some types of communicationson consumption is, interestingly, negative. A communica-tion strategy employing both on- and offline media was


shown to have negative impact on online consumption.Consumers may look with suspicion on any kind of directmanipulation of the press or of customer evaluations. Webelieve that the extremely high volume of informationavailable today has altered the balance of power betweencompanies and consumers. Consumers have become over-loaded and are skeptical about traditional company-drivencommunications. However, company-driven communi-cation can impact online consumption through WOMbehavior and online activity. Companies cannot controlwhat consumers tell other consumers. But marketers canuse insights from research on WOM to shift from conse-quential to intentional marketing. WOM communicationcan be effective at encouraging a consumer to consider aproduct in a positive light in a way advertising simply

cannot, no matter how much is spent on it. The mosteffective communication messages echo within interestedsocial networks and expand their influence, affectingproduct perceptions.

We found that respondents’ engagement in activeWOM did not have an impact on their online consump-tion. One may argue that consumers will not spend timeproviding product reviews for which they are not directlycompensated. A company’s competitors may benefit fromits investment in online recommendation systems, andperhaps most importantly, companies do not have muchcontrol over community-generated content. Referencesare available upon request.

For further information contact:Danny P. Claro

MarketingInsper Education and Research Institute

R.Quatá. 300São Paulo 04547–002

BrazilPhone: (11)4504.2419



THE ROLE OF RESISTANCE AND POSITIVE COPING INCONSUMPTION OF MOBILE APPS

Anubha Mishra, University of Idaho, MoscowMary Ann Eastlick, University of Arizona, Tucson

SUMMARY

Although mobile app is one of the fastest growingmarkets, data from smartphone owners suggest that not allusers actively download and use apps (Wayne 2012;Dowell 2010). However, little research investigates whyconsumers elude technology after ownership (Kleijnen,Lee, and Wetzels 2009) and the strategies they employthat ultimately lead to their decision to continue/discon-tinue using a technology. The proposed researchemployed cognitive appraisal theory in a process-based model to explore consumer resistance as a partof their decision-making process during consumption.The research model proposed appraisal of perceivedfacilitators and inhibitors in using mobile apps to influ-ence consumers’ coping strategies in the form of positivecoping and resistance toward using apps. The copingstrategies then influence consumers’ intentions of contin-ued use of mobile apps.

Data were collected via a self-administered, web-based survey from 375 smartphone users who had down-loaded app(s) on their smartphone since the time ofpurchase. SPSS and Lisrel 8.8 were used to analyze thedata. Consumers’ appraisal of mobile apps was exploredvia the diametric anchors of paradoxes of technology asproposed by Mick and Fournier (1998). Exploratory fac-tor analysis suggested three distinct factors underlying theparadoxes. The first factor, perceived benefits, highlightsthe extent to which using mobile apps provide control,freedom, newness, fulfillment of needs and assimilationto the users. Compared to the concept of “optimism” fromthe Technology Readiness Index (Parasuraman 2000),perceived benefits provide a richer dimension by incorpo-rating consumers’ perceptions of connectedness in thescale. The second factor, perceived apprehension, addres-ses consumers’ perceptions of disruption in ones’ life,dependence, fear of obsolescence, isolation, and addi-tional needs as a result of using mobile app(s). Perceivedapprehension is more detailed as compared to the conceptof “discomfort” (Parasuraman 2000) that mostly explainsanxiety toward the use of technology itself and the fear ofmalfunctioning. Lastly, consumers’ perceptions of in-competence and inefficiency in using mobile app(s) encom-passed a third factor, termed as perceived obscurity.Unlike perceived obscurity, scale based on technologyparadoxes, e.g., TRI, does not propose such a factor thatexplains the confusion and ambiguity within individuals

as a result of using a technology. Items intended tomeasure perceived competence and efficiency in usingmobile app(s) did not load significantly on any factor.

All indicators intended to measure rejection loadedon a single factor comprised of behavioral tendenciessuch as, distancing, abandoning, and neglecting. Positivecoping strategies such as partnering and mastering alsoloaded on a single factor. Continued usage was measuredby three items. CFA supported the factor structure. Theoverall fit for the measurement model was χ2 = 2040.62(df = 865; p < .001); CFI = .98; IFI = .98; NFI = .97;RMSEA = .061; χ2 /df = 2.36 indicating a good fit. Thereliabilities ranged from .80 to .96. The AVE for eachconstruct was greater than 50 percent except for positivecoping (44.39%) (Fornell and Larcker 1981). Discrimi-nant validity was successfully assessed by examining theconfidence interval for the cross-construct correlations(Bagozzi, Yi, and Phillips 1991).

Structural modeling results supported most of thehypotheses (χ2 = 2068.54 (df = 869; p < .001); CFI = .98;IFI = .98; NFI = .97; RMSEA = .061; χ2 /df = 2.38). Thepaths from perceived benefits (β = .66, p < .001) andperceived apprehension (β = .31, p < .001) to positivecoping, and in turn, on continued usage (β = .39, p < .001)highlights consumers’ use of proactive coping (Aspinwalland Taylor 1997). Individuals use proactive coping tocultivate their capabilities in an effort to build generalresources that either facilitate overcoming challenginggoals or enhance personal growth. In the present context,consumers who might see the use of mobile apps asbeneficial or challenging may engage in positive copingto gather more resources and continue their use or out-weigh their difficulties in using apps.

Interestingly, perceived obscurity had no influenceon positive coping but a strong positive influence onrejection (β = .40, p < .001). These relationships suggestthat if a smartphone owner had already tried using appsand still doubts his or her ability to use apps then she/hewas likely to strictly limit the use of apps or reject usingthem all together. These respondents showed no inten-tions to try and master the art of using apps. This reasoningcan also explain the positive effect of perceived appre-hension on rejection (β = .53, p < .001) which, in turn,negatively influence continued usage (β = -.49, p < .001).The negative path from perceived benefits to rejection


(β = -.07, p < .01), and in turn, to continued usagesupported Mick and Fournier’s (1998) findings. In that,consumers’ perceptions of low benefits derived from theactual use of mobile apps were found to result in limitedor discontinued use of mobile apps, or being ignoranttoward using apps. The above mentioned paths canexplain the plausible reason for the recent market researchreport advocating that some consumers do not use app(s)after initial downloading (Realwire 2010).

The results of the present study give us valuableinformation regarding consumers’ decision-making pro-cess during consumption of mobile apps. The findingsand implications are timely in the current market given the

exponential growth of the apps market, and the use ofmobile technologies in the retail environment, e.g., QRcodes which permit consumers, via an app, to see greaterdepth of product information via videos on their smartphones, apps for self-scanning, apps for coupon redemp-tions, etc. (Greene 2011). In addition, interactive tech-nologies are penetrating many different product lines andservices such as, the recent announcement that Toyota isgoing to include interactive windshields in some of itsautomobiles; some stores are installing interactive win-dows as display windows, etc. (Technology 2011). Hence,there are many new technologies on the near horizon forwhich the study results may have implications. Refer-ences are available upon request.

For further information contact:Anubha Mishra

College of Business & EconomicsUniversity of Idaho875 Campus DriveP.O. Box 443161

Moscow, ID 83844–3161Phone: 208.885.6090



STRATEGIES FOR INCREASING ONLINE SURVEY PARTICIPATION

Mary Foster, Ryerson University, TorontoAnne Warner, Ryerson University, Toronto

Adam Froman, Delivina, Toronto

SUMMARY

Given the increasing use of online technology togather information and the importance of consumer opin-ions and experiences in driving decisions about the typeand range of products offered, marketers are interested inunderstanding how to engage consumers in sharing theiropinions and increasing the quantity and quality of par-ticipation. The purpose of this paper is to identify strate-gies for encouraging online survey participation.

Researchers have identified a number of motiva-tional constructs related to knowledge sharing that mayhave application to online survey participation. Theseinclude trust and shared norms, self-efficacy, exchangeand reciprocity, altruism, social interaction and privacyconcerns. One issue that has not been addressed byprevious research is whether these motivational con-structs are differentially important within online usergroups. This project examines motivation within foursocial media user group segments to gain insight into whyknowledge is shared online and to test the appeal ofvarious strategies for increasing online participation.

We chose the sample of 1501 from the 160,000member “Asking Canadians” panel to be representative ofthe Canadian population in terms of regional distribution,age and gender according to the latest information fromStatistics Canada. The instrument was developed basedon questions from the Foster, Francescucci, and West(2011) study to identify social media user group seg-ments, and the existing literature on motivational con-structs.

Using SPSS 19 and AMOS 19, we constructed amodel to describe online behavior that considered bothtypes of behavior and frequency. Three distinct types ofbehavior emerged: creating, socializing, and info seeking.Using K-means cluster analysis, we segmented the sampleinto four user groups: (a) Social Media Technology Ma-vens, representing 7 percent of the sample and scoringhigh on need for information and need for interaction insocial networks; (b) Info Seekers, representing 18 percentof the sample and scoring high on the need for informationand low on the need for interaction; (c) Socializers,representing 26 percent of the sample and scoring high onthe need for interaction and low on the need for informa-tion; and (d) Minimally Involved, representing 49 percent

of the sample scoring low on both the need for informationand the need for interaction.

Analysis of variance and Chi2 were used to under-stand the strength of the various motivational constructsfor each of the user groups and to identify strategies withthe most appeal. Three motivational constructs are impor-tant to all user group segments: (a) trust in sponsor, whenthe survey is associated with a familiar sponsor and onethat is revealed to them in advance of participating; (b)reciprocity sponsor, when respondents receive feedbackfrom the sponsor about the impact of the informationprovided, and/or when there is a tangible incentive toparticipate; and (c) privacy, when respondents have noconcern about the confidentiality or security of theirresponses. These top motivators are within the sponsors’control. They are not dependent on peer response orintrinsic motivators within an individual.

The findings suggest that the most effective strategiesfor motivating participation in online surveys are linked tothe most important motivations. Participants are moti-vated by “trust in sponsor” and this relates to their reportof an increased likelihood to share opinions if sponsorsprominently display how they protect respondents’ per-sonal information. “Reciprocity sponsor” motivation islinked to the reported positive impact of the opportunityfor respondents to earn points for the quality of theircontributions, which can later be used for rewards. Fi-nally, identifying privacy issues as a motivator is consis-tent with the positive response to enforcing an online codeof conduct.

In terms of differences within user groups, SMTMavens appear intrinsically motivated to participate online,and respond positively to any number of online strategiesdesigned to increase participation. Info Seekers may bethe most promising segment to target because they aretechnologically engaged, are motivated by a variety ofconstructs to share opinions online and respond positivelyto a number of proposed community formats and featuresfor increasing participation. Socializers appear limited intheir scope of expertise and comfort in that their onlineactivities seem to consist mainly of Facebook. Few haveprogressed beyond this entry level technology. The Mini-mally Involveds are technological laggards and it is un-likely that any targeted strategy is going to result in anincrease in participation.


While this study was conducted with a representativedemographic sample, it only included those who arecurrently part of an online panel. A broader sample mayhave yielded different results. References are vailableupon request.

ENDNOTE

1 The research on which this paper is based was fundedby the Applied Research and CommercializationInitiative of the Federal Economic DevelopmentAgency for Southern Ontario (Fed Dev Ontario).

For further information contact:Mary Foster


Toronto, ON M5B2K3Phone: 416.979.5000, Ext. 6734



TOO MUCH FACEBOOK: AN EXPLORATORY EXAMINATION OFSOCIAL MEDIA FATIGUE

Stacy Landreth-Grau, Texas Christian University, Fort WorthLaura F. Bright, Texas Christian University, Fort Worth

Susan Bardi Kleiser, Texas Christian University, Fort Worth

SUMMARY

Social networking sites like Facebook and micro-blogging sites like Twitter have become some of theprimary sources of communication for people to find outabout people in their network and brands with which theyidentify. Facebook has become the world’s largest socialnetworking site with more than 800 million active usersand with more than 50 percent of those active userslogging onto Facebook in any given day (Facebook 2012).Similarly, Twitter has 100 million active user accountsand one billion tweets sent every seven days (Sullivan2011). Overall media usage has become fractured withpeople spending 24 percent of their time online (secondonly to television at 40%). But while Facebook has grownin popularity over the years, it appears it may have reachedits peak in terms of usage. There appears to be a usage-related phenomenon setting in among users of Facebookand other social media websites – users are visiting socialmedia sites less. This phenomenon is called “social mediafatigue” (Goasduff and Pettey 2011).

Social media fatigue is defined as “social mediausers’ tendency to back away from social media usagewhen they become overwhelmed with too many sites, toomany pieces of content, too many friends and contacts andtoo much time spent keeping up with these connections”(Technopedia 2011). Social media fatigue is rooted in theidea that too much information stemming from socialmedia can lead to feelings of being overwhelmed. TheLimited Capacity Model (Lang 2000) suggests that peoplehave a limited amount of mental resources to processinformation. As such, LCM is being utilized in thisresearch to determine whether information overload playsa conceptual role in social media fatigue. Therefore, thepurpose of this study is to identify the concept of socialmedia fatigue and understand its antecedents.

There are several antecedents that we want to exam-ine for social media fatigue. Social media confidencedescribes a person’s perception of their ability to usesocial media in an effective manner. For the purposes ofthis research, confidence refers to the level of certitude orassurance consumers have for dealing with content onsocial media websites. Self-efficacy is belief in one’sability to organize and execute a particular course ofaction – in this case, engagement and participation in

social media (Bandura 1986). As such, social media userswho become more self-efficacious will increase theirexpectations of obtaining specific outcomes from theirsocial media use. In turn, the more self-efficacious, themore likely the user will continue engaging and participat-ing in social media. As social media grows, the issue ofprivacy becomes ever more important. Facebook contin-ues to have a liberal viewpoint on privacy and Googlerecently announced major changes to privacy for all of itsproperties. We would expect those with higher privacyconcerns to also have social media fatigue due to the“threshold beyond which social contact becomes irritatingfor all parties” (Schwartz 1968, p. 741) meaning thatpeople might experience being too accessible or receivingtoo much information from too many people. Social mediahelpfulness refers to the extent to which users gainresources and helpful information from their explorationof social media sites. Research has explored why peopleparticipate in social media and social networks (Foster,Francescurri, and West 2010). It is hypothesized that allfour of these antecedents will have an impact on consum-ers’ level of social media fatigue in positive (privacy) andnegative (self-efficacy, helpfulness, and confidence) ways.

A 210-item questionnaire was administered to an opt-in subject pool recruited for web-based research (i.e.,online panel). A representative sample of social mediausers between the ages of 18–49 was recruited for thesurvey. Data was gathered from a total of 750 participantsover a seven-day period to insure an even distribution ofrespondents on each day of the week (i.e., weekdays andweekend days). The final sample consisted of 747 currentsocial media users with a Facebook account. Amongst thissample, 47.5 percent (N = 355) were male and 52.5 percent(N = 392) were female. Respondents ranged in age from18 to 49 (M = 32.52, SD = 9.1). Respondents belonged toa variety of social media websites including Facebook,Twitter, Google+, Pinterest, and LinkedIn.

To test the hypothesized conceptual framework forsocial media fatigue, a step-wise regression was under-taken using IBM SPSS version 19.0 statistical software.The results of this investigation indicate that all fourantecedents, social media privacy, self-efficacy, confi-dence and helpfulness, all had significant relationshipswith social media fatigue. However, two antecedents,self-efficacy and helpfulness, had positive and significant


predictive effects for social media fatigue which was theinverse of the hypothesized model. Privacy concerns andsocial media confidence were significant in their ability topredict social media fatigue in line with the hypothesizedmodel.

Results showed that respondents with greater levelsof confidence about their social media usage were lesslikely to experience social media fatigue. However, wefound a positive relationship between social media self-efficacy and social media fatigue that was not the hypoth-esized relationship. This relationship may be the result ofburnout experienced when using social media. We found

that there was a relationship between privacy concernsand social media fatigue whereas those respondents withhigher levels of privacy concerns experienced socialmedia fatigue. Last, we found a positive relationship ofsocial media helpfulness and greater levels of socialmedia fatigue. However, contrary to our hypothesis, whenrespondents experience high levels of social media help-fulness they also experience high levels of social mediafatigue. Even as exploratory research, this has theoreticaland managerial implications. Future research needs toexplore the social media fatigue construct further as therecould be additional antecedents, moderators, mediatorsand outcomes. References are available upon request.

For further information contact:Laura F. Bright

Texas Christian UniversityTCU Box 298060

Fort Worth, TX 76129Phone: 817.257.5251



ON MATERIALISM, COPING AND CONSUMER BEHAVIORS UNDERTRAUMATIC STRESS

Ayalla A. Ruvio, Temple University, PhiladelphiaEli Somer, University of Haifa, Israel

SUMMARY

Focusing on existential stress and guided by theTerror Management Theory (TMT), scholars have theo-rized that materialistic behaviors may serve as a copingmechanism with stress (Arndt, Solomon, Kasser, andSheldon 2004; Mandel and Smeesters 2008). Neverthe-less, very little empirical support has been provided forthese theoretical arguments and in most studies the meth-odological procedures entailed the induction of mortalitysalience among participants in well-controlled lab set-tings (e.g., Mandel and Smeesters 2008; Rindfleisch,Burroughs, and Wong 2009). Thus, although this pioneer-ing body of literature highlights the potential contributionof materialism to the study of coping, it has also left someunresolved theoretical and methodological issues.

To address the lacunae in the traumatic stress andconsumer behavior literature, we explored how exposureto traumatic stress, posttraumatic reactions and material-istic values impact coping and consumption behaviors inreal life traumatic conditions. We collected data in twoIsraeli towns during a prolonged period of hostilitiesalong the Gaza-Israel border. Our respondents weresampled in Sderot, an Israeli community that had beensubjected to the launching of thousands of mortar shellsand rockets during this conflict, and in a comparison city.

In this study we hypothesize that compared to a groupexposed to low levels of stress, respondents residing in ahigh-level stress environment will report higher levels ofposttraumatic distress (PTSD), coping and maladaptiveconsumption behaviors (shopping escapism, hedonic shop-ping and impulsive consumption). We also predicted thatmaterialism would be a significant risk factor for thestudy’s dependent variables.

Method

Participants and Procedures. Our sample consistedof two groups of civilians: 139 participants who lived inSderot, a town in the southern part of Israel, located closeto the Gaza Strip. Data were collected when this area wassubjected to extensive mortar shell and rocket attacksfrom the Gaza Strip. Participants in the low-stress groupwere 179 residents of a large city that was not threatenedby rocket attacks.

Measures

Materialism. We used Richins’ (2004) 9-item Mate-rial Values Scale (MVS) to measure materialism (α =0.84). Posttraumatic distress, measured using 17-itemself-reported PTSD Symptom Scale (PSS-SR; Foa, Riggs,Dancu, and Rothbaum 1993). Reliability of the scale inthis study was α = 0.95. Coping. The Brief COPE (Carver1997), 17-items scale was utilized to measure coping withthe following sub-dimensions: Direct Active coping (5-items); Interpersonal Expressive coping (5-items); Avoid-ance and Distancing (3-items); Positive Restructuring (4-items). Coefficient alpha for the scale was α = 0.85.Shopping escapism. We used Mathwick et al. (2001) 3-item scale (α = 0.92). Compulsive buying. We used Babin,Garden and Griffin’s (1994) 5-item scale (α = 0.81).Hedonic Shopping. We used Arnold and Reynolds (2003)3-itmes “gratification shopping, factor of their hedonicshopping motivation scale (α = 0.90).

Results and Discussion

To shed further light on the interconnectedness ofexposure to stress, coping, consumer behavior and post-traumatic distress we investigated the role of materialisticvalues in this matrix. A main effect for materialism wasfound for all the distress, coping and consumer behaviorsunder study, with the exception of Interpersonal Expres-sive coping. Highly materialistic individuals under duresstended not to intensify Interpersonal Expressive coping.These findings are aligned with the main premise ofmaterialism that places a high premium on possessionsrather than social relationships (Richins and Dawson1992). It seems that even in times of upheaval highlymaterialistic individuals tend to seek solace in materialpossessions rather than social relationships.

Our data also show a significant interaction betweenmaterialism and exposure to traumatic stress with regardto PTSD, Direct Active coping, Interpersonal Expressivecoping, Avoidance coping and compulsive buying. Whenexposed to severe psychological stress, highly materialis-tic individuals reported the highest level of PTSD. Thesefindings support previous literature that demonstrated thedetrimental effect of materialism on an individual’s well-being (e.g., Burroughs and Rindfleisch 2002; Kasser andKanner 2005). Nevertheless, the mechanism causing


materialistic individuals to be more vulnerable to PTSDrequires further investigation. Values have been docu-mented to be relevant to the experience of stress(Bouckenooghe et al. 2005). Kishon-Barash and her asso-ciates (1999) demonstrated that individuals reportinghigh levels of other-directedness were less likely to sufferfrom PTSD symptoms. The attraction of highly material-istic persons objects could be incompatible with theirability to benefit from social support, a known moderatorof stress in general (e.g., Field and Schuldenberg 2011),and PTSD in particular (e.g., Haden et al. 2007; Scarpaet al. 2006). In line with this view is our data showing thatin the high-stress group Interpersonal Expressive coping,reflecting an inclination to utilize social support, washighest among mildly materialistic individuals.

Under normal conditions, people who had been upsetby adverse life events were more likely to endorse mate-rialistic values and more inclined to react with maladap-tive consumer behaviors. However, under extreme andprolonged duress, PTSD seemed to have been unrelated toboth hedonic shopping and compulsive buying, but ap-peared to have been significantly associated with shop-ping escapism. Thus, to cope with their distress, civilians

living under continuous shelling shopped to escape fromthe hardships of their threatening milieu, but were notmore likely to engage in spendthrift purchases or to reactwith indulgent, extravagant shopping. Escapist shoppingwas probably motivated by the need to capture attentionresources and offer a distraction from the bleak environ-ment, rather than to satisfy materialistic values and theneed for ownership of objects. Again, these findingsimply that materialistic individuals who are exposed totraumatic stress, experience shopping as a temporarymeans to alleviate distress when higher-level psychologi-cal coping skills are unavailable.

Our data indicates a significant relationship betweenmaterialism and the consumer behaviors we measured,regardless of how stressful the environment had been.However, in the high-stress group materialism, as well asthe interaction between PTSD and materialism, weresignificant predictors of the three measured consumptionbehavior constructs. That is to say, under extremely stress-ful conditions, PTSD tended to exacerbate the effect ofmaterialism on elevating maladaptive shopping patterns.References are available upon request.

For further information contact:Ayalla A. Ruvio

Fox School of BusinessTemple University

1801 Liacouras WalkPhiladelphia, PA 19122Phone: 215.204.4224



THE GREEN SIDE OF MATERIALISM

Pia Furchheim, Chemnitz University of Technology, GermanySteffen Jahn, Chemnitz University of Technology, Germany

Cornelia Zanger, Chemnitz University of Technology, Germany

SUMMARY

Individuals use material possessions to judge them-selves and others (Richins 1994, 1999). In serving assymbols of identity and self-expression, possessionsextend the owner’s self (Belk 1988). Closely related toinsecurity (Rindfleisch and Wong 2009), compulsiveconsumption (Rindfleisch, Burroughs, and Denton 1997)or traits like possessiveness, self-centeredness, greed andgeneral life-dissatisfaction (Belk 1983, 1985) material-ism is predominately viewed as a dark side variable ofconsumer behavior (Mick 1996).

Rising concerns about environmental or ethical issues(Shaw et al. 2005; Shaw and Shiu 2003), however, havemade many individuals rethink their consumption behav-ior. In the last years, ecological and economicalsustainability or ethical aspects have become a main-stream issue (Gordon, Carrigan, and Hastings 2011;Prothero, McDonagh, and Dobscha 2010; Shaw et al.2005; Shaw and Newholm 2002; Shaw and Shiu 2003)and are regarded important by companies, consumers, andgovernments alike (Sheth, Sethia, and Srinivas 2011).

Prior research has agreed on a mutually incompatibil-ity of materialistic and sustainable consumption (e.g.,Banerjee and McKeage 1994; Doran 2009; Kilbourne andPickett 2008; Muncy and Eastman 1998; Prothero,McDonagh, and Dobscha 2010; Shaw et al. 2005).Burroughs and Rindfleisch (2002) even propose that theconflict between material values and collective-orientedvalues creates psychological tension that lowers subjec-tive well-being.

By contrast, the present paper delineates howsustainability and materialism can be aligned by present-ing a conceptual framework that links both phenomenafrom three perspectives. In a first step, we show thatmaterialism and sustainability need not be at odds. Goingbeyond the normative view of green consumption, theconcept of consumer value (Holbrook 1994, 2006) helpsto explain why green consumption may entail statuselements. Holbrook (2006) identifies four key types ofconsumer value (economic, hedonic, social, and altruis-tic). Importantly, the four types of value potentially coex-ist in any consumption experience (Holbrook 1994). Thisnotion of compresence of consumer value types offers anew perspective on both altruistic and materialistic con-sumer behavior. Given that altruistic value includes

aspects that are most often related to green consumerbehavior and that social value refers to the case in whichconsumption experiences serve as a means to attain status-relevant responses from others, there need not be a con-flict between green and materialistic consumption.

Second, we enrich existing materialism research bydiscussing the nature of material possessions in moredetail. Researchers tend to agree that material possessionshave to be more expensive than other products and haveto be socially visible (Richins 1994). Accordingly, luxurygoods, cars, homes, money, or new clothes that symbolizestatus are typical objects of investigation in most materi-alism research (e.g., Burroughs and Rindfleisch 2002;Chaplin and John 2007; Richins 1994; Rindfleisch,Burroughs, and Wong 2009; Wong and Ahuvia 1998).Granted that those types of products are suitable to dis-cuss materialistic consumer behavior one has to bear inmind that it is not a particular feature or group of featureswhich guarantees the desired consumption outcomes.Considering the importance of possessions for identitypurposes as well as their influence on the respective socialenvironment (Solomon 1983), the reduction of materialpossessions to costly goods only is subject to debate.Rather, the actual value of possessions depends on thesocial environment’s (envious) response (Solomon 1983).Suitable possessions need to be rare, valued by the respec-tive social environment, and socially visible. This abstractclassification of material possessions allows inclusion ofproducts whose consumption is closely tied to scarceproperties, such as unique knowledge, intelligence, self-sacrifice, or farsightedness. Notably, knowledge, intelli-gence, self-sacrifice, and farsightedness are closely con-nected with green consumption.

In a third step we conceptualize why green productattributes can be attractive for materialists. Materialiststry to create, express, and shape themselves throughpossessions. While the use of possessions for identitysignaling purposes addresses an outward perspective ofmaterialism, the desire to enhance self-esteem focuses onan inward perception. Although correlated it seems fea-sible to draw a distinction between the two. Referring tooutward-signaling processes the theory of competitivealtruism provides further insight how green consumptioncan be utilized for status-signaling purposes. An altruisticact that is performed in public, e.g., buying green prod-ucts, serves to communicate both one’s willingness and


the ability to incur the cost of self-sacrifice for publicwealth (Griskevicius, Tybur, and van den Bergh 2010;van Vugt, Roberts, and Hardy 2007) and therefore can beassociated with signaling status. Green consumption sub-sumes the deliberate and conscious decision due to moraland ethical beliefs (Crane and Matten 2003) and conse-quently demonstrates an individual’s farsightedness,humanity, and intellectual capacity. As such, materialistsmay use these rather subtle signals to communicate one’sown identity to the social environment and thereforesignal affiliation to a desired group while drawing a cleardistinction from dissociative groups. In addition, consum-ers use products for self-definitional reasons (Belk 1988;Hunt, Kernan, and Mitchell 1996; Richins 1994). It islikely that individuals may consider attributes of greenconsumption (e.g., strength of character, humanity) to benoble values. Hence, green consumption may serve self-

signaling purposes which closely corresponds with typi-cal materialistic behavior (Richins 1999). In this vein, ouranalysis suggests that even green products can serve asenviable possessions. This does not imply that all greenproducts are used for materialistic reasons. If green prod-ucts are suitable to satisfy the conditions of outward andinward signaling, there is reason to assume that they canbe regarded as legitimate material possessions as areluxury goods.

The consideration of a green side of materialismcontributes to current discussions in transformative con-sumer research (Mick et al. 2012). This broader perspec-tive not only overcomes the often mentioned dark side ofmaterialism but also shows that green materialism canentail positive consequences for nature and environment.References are available upon request.

For further information contact:Pia Furchheim

Department of MarketingChemnitz University of Technology

Thueringer Weg 709126 Chemnitz

GermanyPhone: +49.371.531.36811Fax: +49.371.531.836811



SUBJECTIVE WELL-BEING AND ITS INFLUENCE ON CONSUMERPURCHASE BEHAVIOR OF LUXURY ITEMS DO MATERIALISM

AND AFFECTIVE AUTONOMY HAVE AN IMPACT?

La Toya M. Russell, Cleveland State University

ABSTRACT

In previous literature, researchers have studied sub-jective well-being; however, there appears to be limitedresearch on the relationship between subjective well-being and consumer purchase behavior of luxury items. Inthis paper, the author explores the relationship betweensubjective well-being, materialism and affective autonomyand consumer purchase behavior of luxury items.

(Keywords: subjective well-being, consumer behav-ior, marketing, affective autonomy, materialism.)

INTRODUCTION

Over the last few decades, psychologists and researc-hers in the field of psychology have studied subjectivewell-being as it relates to happiness and life satisfaction.However, there appears to be limited, if any, research onthe effect that subjective well-being has on consumerpurchase behavior of luxury items. Subjective well-beingmay need to be researched further in the field of marketingin order to determine how important the construct is in thefield of marketing. Marketers may want to know if happyconsumers or consumers who are satisfied with their liveshave different shopping behaviors than unhappy consum-ers or consumers who are dissatisfied with their lives.Consumers are promised greater well-being in the promo-tion of goods and services (La Barbera and Gurhan 1997).Marketing theory and practice have evolved from thefocus on competition and performance to customer satis-faction and consumer well-being (Sara and Surendra2010).

This paper explores subjective well-being and itsimpact on consumer purchasing behavior of luxury items.The purchase of luxury items has increased in variousmarkets over the last few years. For example, there hasbeen growth of the middle class in Asian economies,specifically the Chinese middle class. From 2005 to 2007,the Chinese middle class grew from 65.5 million to 80million and it is expected to grow to 700 million by 2020(People’s Daily Online 2010). Consumers in Chinesecities such as Zhuhai are estimated to account for 60percent of the growth of China’s luxury buyers in 2008and 2009 (Lannes and Han 2010). Luxury companies,such as Louis Vuitton have recognized the growth of theChinese middle class and have opened up stores in China(Adams and Elliot 2010).

Two variables, materialism and affective autonomy,will also be discussed in regard to how they affect con-sumer purchase behavior of luxury items. Materialismreflects the importance a consumer attaches his or herselfto worldly possession (Belk 1984). These possessions areassumed to be a central place in this person’s life andprovide sources of satisfaction or dissatisfaction in life(Belk 1984). Affective autonomy is the cultural emphasison the desirability of individuals independently pursuingaffectively their positive experiences (Schwartz 1999).

The first section consists of a literature review ofsubjective well-being, its antecedents – life satisfactionand happiness as well as materialism and affective auton-omy. The second section consists of the development ofthe propositions, which are based on the literature review.The final section consists of the discussion and futureresearch.

LITERATURE REVIEW

Subjective Well-Being

Subjective well-being refers to how people evaluatetheir lives. It involves how individuals view the meaningof their lives, their purpose, their potential for growth, andtheir experiences of emotions (Gaia 2005). People tend toexperience large amounts of subjective well-being whenthey feel many pleasant and few unpleasant emotions,when they are engaged in interesting activities, when theyexperience many pleasures and few pains, and when theyare satisfied with their lives (Diener 2000). Subjectivewell-being is a scientific concept that contains affectivecomponents (moods and emotions like joy, happiness,etc.) as well as cognitive evaluations of life satisfaction(i.e., global judgments of one’s life (Diener 2000; Freyand Stutzer 2000). As people fulfill more basic materialneeds, it is likely that subjective well-being may becomean even more valued goal (Diener 2000).

The two main antecedents discussed in this paper arehappiness and life satisfaction. At the individual level,good health, adequate education, fit between personalityand culture, personal growth, and purpose of life areamong the factors that can increase subjective well-being(Gaia 2005). Though some researchers use happiness andlife satisfaction interchangeably (Easterlin 2005) and theconstructs are correlated to some degree, they are two


distinctive constructs and should be treated as such (Tsouand Liu 2001). Happiness is sensitive to sudden moodchanges while life satisfaction is related to a cognitive orjudgmental state (Tsou and Liu 2001).

Happiness

Happiness is defined as the degree to which anindividual evaluates the overall quality of his or her life aswhole on a positive level (Veenhoven 1994). Many indi-viduals aspire to be happy throughout their lives. Previousresearch has shown that happiness is difficult to measureand has to be compared interpersonally (Ng 2002). Hap-piness is also an important indicator in identifying people’slife satisfaction (Susniene and Jurkauskas 2009). Researc-hers have found that happiness, as a construct, hasremained constant and factors such as prosperity or mis-fortune have a permanent effect on happiness (Inglehartet al. 2008).

Research has also shown that multiple variables areused to determine happiness. For example, Brereton et al.(2008) find environmental climate has a significant influ-ence on an individual’s happiness and that geography andthe environment have a larger influence on happiness. Thevariables used in this paper are income level and socialstatus to determine happiness.

Happiness and Income

Income may be evaluated relative to others (socialcomparison) or to oneself in the past (habituation) (Clarket al. 2008). Ball and Chernova (2008) discuss anindividual’s happiness as it relates to income in absoluteand relative terms. The author’s findings show happinessis affected by both relative and absolute income. Bartram(2011) finds that the relationship between income andhappiness is stronger between immigrants in the UnitedStates than with those individuals who were born in theUnited States.

On the other hand, other research has shown thatmoney may not buy happiness. Subjective well-being(and its antecedent) is positively associated with incomebut only to a certain point (Gaia, 2005). For example,dramatic increases in income, say from below poverty tofinancial sustainability, are considered strongly related tosubjective well-being. However, if an individual is abillionaire, acquiring $100,000 in a business transactionmay not have an impact on the individual’s subjectivewell-being or happiness.

Happiness and Social Status

Social psychologists remind us that we are socialanimals (Myers 2000). Scholars have studied the effectsocial status has on happiness. Studies have shown that

most people are happier when attached than when unat-tached (Myers 2000). Studies also show that marriedpeople tend to be happier than single or divorced people.For example, Selim (2008) finds that married people havethe highest degree of happiness in Turkey. Individualswho are married, have children or participate actively insocial and religious terms, are significantly happier thanthose who are not in such relationships (Haller and Hadler2006). In comparison with people who never marry, it hasbeen reported that married people are happier and moresatisfied with life (Myers 2000).

Life Satisfaction

Life satisfaction has been examined in fields such associology, psychology, and marketing (Bhardwaj et al.2011). It is a cognitive evaluation that is particularlydependent on social comparisons with other importantreference groups as well as individual’s desires, expecta-tions, and hopes (Selim 2008). Life satisfaction reflects anindividual’s life conditions, improved or demographicand physical conditions such as employment and health(Bhardwaj et al. 2011). Some research has associatedmaterialism with life satisfaction. For example, Belk(1984, as cited by Parker et al. 2010) states, “At the highestlevels of materialism, such possessions assume a centralplace in a person’s life and are believed to provide thegreatest sources of satisfaction and dissatisfaction inlife. . . .”

Several studies have associated life satisfaction withthe acquisition of goods (Bhardwaj et al. 2011). Forexample, researchers find that material possessions in anindividual’s life are related to happiness and satisfactionwith life (Andrews and Withey 1976). Some studies findthat life satisfaction has a negative on consumer behaviorand consumption. Yujie and Talpade (2009) find thatmaterialism will be stronger for older adults who are lesssatisfied with their lives.

Materialism

Materialism is defined as a “set of centrally heldbeliefs about the importance of possessions in one’s life”(Richins and Dawson 1992). Materialism is also said to bea “value that guides people’s choices and conduct in avariety of situations, including, but not limited to, con-sumption arenas” (Richins and Dawson 1992). Richins(1994) states materialism sometimes influences whatconsumer’s desire from their possessions and which prod-ucts they believe can fulfill their desires during consump-tion. Consumers who are highly materialistic tend to usepossessions and materialism to signal success and pres-tige (Wang and Wallendorf 2006).

When discussing consumer purchase behavior, spe-cifically consumer purchase behavior of luxury items,


materialism predicts consumer purchase frequency ofluxury products such as perfume, jewelry, antique furni-ture, etc. (Cleveland et al. 2009). Based on the research, itcan be inferred that materialism does affect a consumer’sconsumption patterns. Highly materialistic consumersmay purchase more items in order to gain or retain theappearance of having a high social or economic status.Highly materialistic individuals are usually status ori-ented and prefer a higher perceived social and economicstatus (Demirbag et al. 2010).

Affective Autonomy

When using the individual as the unit of analysis, setsof value priorities reveal the trade-offs an individualmakes in order to pursue a particular value (Siew Imm etal. 2007). Schwartz (1999) summarizes three culturaldimensions (Conservatism versus Intellectual and Affec-tive Autonomy; Hierarchy versus Egalitarianism; Mas-tery versus Harmony) and seven culture level values(conservatism, intellectual autonomy, hierarchy, mas-tery, egalitarian commitment, harmony) based onresearch done using 49 different countries.

In this paper, affective autonomy is used as a moder-ating variable for subjective well-being. Siew Imm et al.(2007) define affective autonomy as a society that recog-nizes individuals as autonomous entities who are entitledto pursue their stimulation and hedonism interests and

desires. Affective autonomy is the cultural value that isrelated to individualism and has an emphasis on thedesirability of individuals independently pursuingaffectively their positive experience (i.e., pleasure, excit-ing life, etc.) (Schwartz 1999). In societies emphasizingaffective autonomy, people are encouraged to pursueaffectively pleasant experiences and are expected toexpress their own ideas and find meaning in their ownuniqueness (Fischer and Boer 2011).

It is important to note that Hofstede’s four dimen-sions (masculinity, uncertainty avoidance, individualism,and power distance) were researched for this body ofwork. Though Hofstede’s work has been used in countlessamounts of research, Hofstede’s dimensions measuresculture at the national level (Earley 2006) and this paperlooks at the consumer at the individual level. Based onthis, it has been determined that since this paper is lookingat the individual consumer, Schwartz’s dimensions aremore appropriate.

PROPOSITION DEVELOPMENT


Based on research presented in this paper, Figure 1(below) shows the relationship between subjective well-being, materialism and affective autonomy and consumerpurchase behavior of luxury goods.

FIGURE 1Conceptual Framework


Previous research states happiness and life satisfac-tion are the antecedents of subjective well-being (Selim2008). A person that is happy and satisfied with his or herlife may have large amounts of subjective well-being. Asstated earlier, life satisfaction and happiness, thoughcorrelated are two distinct constructs that are used todefine subjective well-being. Happiness is sensitive tosudden mood changes and life satisfaction is related to acognitive or judgmental state (Tsou and Liu 2001).

Evans (1989) findings indicate that individuals whoprefer products that facilitate achievements in life aremore likely to buy well-known brands. One may considercertain luxury items as well known brands. However,research has also shown that though increased happinessand life satisfaction means that an individual’s subjectivewell-being will increase, it does not mean that the indi-vidual will be inclined to purchase more goods becausethey are happy with their lives. Based on this reasoning, Ipropose Proposition 1:

Proposition 1: Increased subjective well-being willhave a negative impact on consumer purchase behav-ior of luxury items.

Research supports the argument stating that in therelationship between subjective well-being and material-ism, materialistic people generally are found to beunhappy (La Barbera and Gurhan 1997). A number oftheories, including the telic and judgment theories, predictthat materialism has a negative correlation with subjectivewell-being (La Barbera and Gurhan 1997). Belk (1984)finds that the three aspects of materialistic orientation(possessiveness, nongenerosity, and envy) are negativelycorrelated with satisfaction. Materialistic individuals whohave a desire for better possessions and longing for thingsthey cannot afford are less satisfied with their possessionsin a certain category (Wang and Wallendorf 2006). Incomparison to materialistic people with higher levels ofeducation or income level, materialistic people with lowerlevels of education or income will feel frustrated with theirability to satisfy their materialistic desires (La Barberaand Gurhan 1997). Therefore, one can infer that peoplewho are highly materialistic may have lower levels ofsubjective well-being. It can also be inferred that materi-alism will have a positive impact on consumer purchasebehavior of luxury items. Based on this argument, Ipropose proposition 2:

Proposition 2: Materialism positively impacts con-sumer purchase behavior of luxury items.

Affective autonomy is used as a moderating variablein the relationship of subjective well-being and consumerpurchase behavior of luxury items. Consumers in anaffective autonomy society are encouraged to pursueaffectively pleasant experiences and are expected to ex-

press their own ideas and find meaning in their ownuniqueness (Fischer and Boer 2011). Sun et al. (2004)state that consumers with individualistic values are mostsatisfied with their lives, more satisfied financially, opti-mistic, more fashion conscious, and more brand-savvy.Based on this argument, I propose Proposition 3:

Proposition 3: Affective autonomy positively moder-ates the relationship being subjective well-being andconsumer purchase behavior of luxury items.

DISCUSSION AND FUTURE RESEARCH

The research in this paper is the first step in exploringthe relationship between subjective well-being and con-sumer purchase behavior of luxury items. Research hasshown that happiness and life satisfaction are the anteced-ents of subjective well-being (Selim 2008). Happiness hasbeen associated with variables such as income level andsocial status. Diener (2000) states that people tend toexperience a large amount of subjective well-being whenthe feel many pleasant and few unpleasant emotions,when they are engaged in interesting activities, when theyexperience many pleasures and few pains, and when theyare satisfied with their lives. Based on this, research tellsus that increased happiness and increased life satisfactionleads to an increased amount of subjective well being.

Materialism is used to show its direct relationshipwith consumer purchase behavior of luxury items.Research has shown that highly materialistic people tendto be unhappy with their lives (La Barbera and Gurhan1997) and tend to use possessions and materialism tosignal success and prestige (Wang and Wallendorf 2006).Affective autonomy is also used as a moderating variablein the relationship of subjective well-being and consumerbehavior toward luxury items. As discussed earlier, con-sumers in an affective autonomy society are encouragedto pursue affectively pleasant experiences and are expec-ted to express their own ideas and find meaning in theirown uniqueness (Fischer and Boer 2011).

Future Research

The next step for this body of research is to developtestable hypotheses in order to test the theories presented.A questionnaire may be developed and administered toconsumers in multiple markets (ex. the U.S. and China).An attempt will be made to measure the level of subjectwell-being the respondents possess.

Before the questionnaire can be administered to theparticipants, the questionnaire will go through cognitiveand field pretesting procedures. The cognitive procedurewill be done to ensure that the wording of the questionsand answers in the questionnaire are precise and under-standable. The pretest participants will be queried imme-


diately after completing the questionnaire in order todetermine what changes need to be made and what ques-tions need to be reworded, possibly eliminated or added.This procedure will also help in determining if the ques-tionnaire is too time consuming to complete. The fieldpretesting procedure will be used as a “dress rehearsal” ofthe questionnaire and will be used to remove any addi-tional issues that may appear during the administering ofthe questionnaire.

I propose that, once fully tested, this body of researchmay be used to help marketers explored which marketsthey should target for sales of luxury items. An investiga-tion on how happy consumers are could aid the marketerin the development of the best marketing plan or the bestmarketing strategy for a particular luxury item. In additionto this, a market that has a majority of materialisticconsumers could mean a change to the marketing plan orstrategy.

REFERENCES

Adams, Susan and Hannah Elliot (2010), “Master of theBrand: Bernard Arnault,” Forbes.com. Retrieved(2011, March), from [http://www.forbes.com/glo-bal/2010/1220/companies-bernard-arnault-lvmh-luxury-dior-master-of-brand.html].

Andrews, Frank M. and Stephen Bassett Withey (1976),Social Indicators of Well-Being: Americans’ Percep-tions of Life Quality. New York: Plenum.

Ball, Richard and Kateryna Chernova (2008), “AbsoluteIncome, Relative Income, and Happiness,” SocialIndicators Research, 88 (3), 497–529.

Bartram, David (2011), “Economic Migration and Happi-ness: Comparing Immigrants’ and Natives’ Happi-ness Gains From Income,” Social IndicatorsResearch, 103 (1), 57–76.

Belk, Russell W. (1984), “Three Scales to Measure Con-structs Related to Materialism: Reliability, Validity,and Relationships to Measures of Happiness,”Advances in Consumer Research, 11 (1), 291–97.

Bhardwaj, Vertika, Hyejune Park, and Youn-Kyung Kim(2011), “The Effect of Indian Consumers’ Life Satis-faction on Brand Behavior Toward a U.S. GlobalBrand,” Journal of International Consumer Market-ing, 23 (2), 105–16.

Brereton, Finbarr, Peter J. Clinch, and Susana Ferreira(2008), “Happiness, Geography and the Environ-ment,” Ecological Economics, 65 (2), 386–96.

Clark, Andrew, Paul Frijters, and Michael Shields (2008),“Relative Income, Happiness, and Utility: An Expla-nation for the Easterlin Paradox and Other Puzzles,”Journal of Economic Literature, 6 (1), 95–144.

Cleveland, Mark, Michel Laroche, and NicolasPapadopoulos (2009), “Cosmopolitanism, ConsumerEthnocentrism, and Materialism: An Eight-CountryStudy of Antecedents and Outcomes,” Journal ofInternational Marketing, 17 (1), 116–46.

Demirbag, Mehmet, Sunil Sahadev, and Kamel Mellahi(2010), “Country Image and Consumer Preferencefor Emerging Economy Products: The ModeratingRole of Consumer Materialism,” International Mar-keting Review, 27 (2), 141–63.

Diener, Ed (2000), “Subjective Well-Being,” AmericanPsychologist, 55 (1), 34.

Earley, Christopher P. (2006), “Leading CulturalResearch in the Future: A Matter of Paradigms andTaste,” Journal of International Business Studies, 37(6), 922–31.

Easterlin, Richard (2005), “Feeding the Illusion of Growthand Happiness: A Reply to Hagerty and Veenhoven,”Social Indicators Research, 74 (3), 429–43.

Evans, Marting (1989), “Consumer Behavior TowardFashion,” European Journal of Marketing, 23 (7), 7–16.

Fischer, Ronald and Diana Boer (2011), “What Is MoreImportant for National Well-Being: Money or Auton-omy? A Meta-Analysis of Well-Being, Burnout, andAnxiety Across 63 Societies,” Journal of Personality& Social Psychology, 101 (1), 164–84.

Frey, Bruno S. and Alois Stutzer (2000), “Happiness,Economy and Institutions,” Economic Journal, 110(466).

Gaia, A. Celeste (2005), “Culture and Subjective Well-Being,” International Social Science Review, 80 (1/2), 67–69.

Haller, Max and Markus Hadler (2006), “How SocialRelations and Structures can Produce Happiness andUnhappiness: An International Comparative Analy-sis,” Social Indicators Research, 75 (2), 169–216.

Inglehart, Ronald, Roberto Foa, Christopher Peterson,and Christian Welzel (2008), “Development, Free-dom, and Rising Happiness,” Perspectives on Psy-chological Science, 4 (4), 264–85.

La Barbera, Priscilla and Zeynep Gürhan (1997), “TheRole of Materialism, Religiosity, and Demographicsin Subjective Well-Being,” Psychology & Market-ing, 14 (1), 71–97.

Lannes, Bruno and Weiwen Han (2010), “Selling Luxuryto the Chinese Masses,” The Wall Street Journal.Retrieved (2011, March), from [http://online.wsj.com/article/SB10001424052748703866704575225303288815256.html].

Myers, David (2000), “The Funds, Friends, and Faith ofHappy People,” American Psychologist, 55 (1), 56.

Ng, Yew Kwang (2002), “East-Asian Happiness Gap,”Pacific Economic Review, 7 (1), 51.


Parker, R. Stephen, Diana Haytko, and Charles Hermans(2010), “The Perception of Materialism in a GlobalMarket: A Comparison of Younger Chinese andUnited States Consumers,” Journal of InternationalBusiness & Cultural Studies, 31–13.

China’s Middle Class to Reach 48% in 10 Years (2010),People’s Daily Online. Retrieved on April 30, 2011from [http://english.peopledaily.com.cn/90001/90778/90862/7071235.html ].

Richins, Marsha and Scott Dawson (1992), “A ConsumerValues Orientation for Materialism and Its Measure-ment: Scale Development and Validation,” Journalof Consumer Research, 19 (3), 303–16.

____________ (1994), “Special Possessions and theExpression of Material Values,” Journal of Con-sumer Research, 21(3), 522–33.

Sara, Hanson Dionne and Arjoon Surendra (2010), “Mar-keting for Happiness,” Advances in Management, 3(10), 7–14.

Schwartz, Shalom (1999), “A Theory of Cultural Valuesand Some Implications for Work,” Applied Psychol-ogy: An International Review, 48 (1), 23–47.

Selim, Sibel (2008), “Life Satisfaction and Happiness inTurkey,” Social Indicators Research, 88 (3), 531–62.

Siew Imm, Ng, Julie Anne Lee, and Geoffrey Soutar(2007), “Are Hofstede’s and Schwartz’s Value Frame-

works Congruent?” International Marketing Review,24 (2), 164–80.

Susniene, Dalia and Algirdas Jurkauskas (2009), “TheConcepts of Quality of Life and Happiness – Corre-lation and Differences,” Engineering Economics,63(3), 58–66.

Sun, Tao, Marty Horn, and Dennis Merritt (2004), “Val-ues and Lifestyles of Individualists and Collectivists:a Study on Chinese, Japanese, British and U.S. Con-sumers,” Journal of Consumer Marketing, 21 (5),318–31.

Tsou, Meng-Wen and Jin-Tan Liu (2001), “Happinessand Domain Satisfaction in Taiwan,” Journal ofHappiness Studies, 2, 269–88.

Veenhoven, Ruut (1994), “Is Happiness a Trait? Tests ofthe Theory That a Better Society Does Not MakePeople Any Happier,” Social Indicators Research,32 (2), 10.

Wang, Jeff and Melanie Wallendorf (2006), “Material-ism, Status Signaling, and Product Satisfaction,”Journal of The Academy of Marketing Science, 34(4), 494–505.

Yujie, Wei and Salil Talpade (2009), “Materialism ofMature Consumers in China and USA: A Cross-Cultural Study,” Journal of International Business &Cultural Studies, 21–14.

For further information contact:La Toya M. Russell

Department of MarketingMonte Ahuja College of Business Administration

Cleveland State University2121 Euclid Avenue

Cleveland, OH 44115Phone: 216.849.4210



THE IMPACT OF ESOTERIC PRODUCT SPECIFICATIONS ON CHOICE

Chelsea Wise, University of Technology, Sydney, AustraliaPaul F. Burke, University of Technology, Sydney, Australia

Sandra J. Burke, University of Technology, Sydney, Australia

SUMMARY

It is critical for firms to understand how demand fortheir products is impacted by the way in which productinformation is communicated. Despite research devotedto understanding how the provision of informationimpacts product evaluation and choice, research on thecommunication of product specifications is limited. Firmsoften use esoteric specifications to describe attributes. Incontrast to conventional specifications that are familiarand relatively easy to evaluate, esoteric specifications aredefined as descriptions of product attributes that areunfamiliar, novel, and difficult to evaluate even whencomparable alternatives are available to act as referencepoints. For example, the ‘zoom’ attribute of a digitalcamera may be described in a simple, easy-to-evaluatespecification (e.g., 7x), or equivalently described in anesoteric, difficult-to-evaluate way (e.g., 37–260mm).

Although the use of esoteric specifications is wide-spread, little is known about how esoteric specificationsimpact upon choice. Previous research suggests conflict-ing impacts of the use of esoteric specifications in influ-encing product evaluations. For instance, ambiguousattributes have been demonstrated to positively influence

product evaluations, but, paradoxically, consumers oftenignore product information that is difficult to evaluate,instead utilizing specifications that can be easily com-pared across alternatives. Given these conflicting views,the objective of this research is to understand how eso-teric specifications impact reliance on both esotericallyand conventionally described attributes in choice con-texts.

We demonstrate that attribute reliance is lower whendescribed by esoteric specifications than when describedby conventional specifications and, as a result, relianceshifts to those other attributes conventionally described.We propose that the shift in reliance to other convention-ally specified attributes depends on brand credibility.Finally, we demonstrate the moderating role of process-ing effort in attenuating the discounting of esotericallydescribed attributes in choice. Practically, the findingssuggest that firms should be mindful of the language usedto describe key attributes and, in particular, whetherconsumers can comprehend and compare the values ofstated specifications. Failure of firms to do so may lead toconsumers using attribute information in choice contextsin unanticipated ways. References are available uponrequest.

For further information contact:Chelsea Wise

University of Technology, SydneyP.O. Box 123, Broadway

NSW, 2007Australia

Phone: +61.2.9514.9818Fax: +61.2.9514.3535



CONFORMITY, UNIQUENESS, AND SOCIAL CLASS MOBILITY:CONSUMER SELF IDENTITY IN THE YARD

Elizabeth C. Hirschman, Rutgers University, New Brunswick and University of Virginia – WiseDavid L. Kendall, University of Virginia – Wise

ABSTRACT

The present study examines the conceptual and empi-rical links between two psychological traits – social con-formity and the need for uniqueness – and the sociologicalconstructs of social class and social mobility. Using depthinterviews with consumers drawn from the working,lower middle, middle, upper middle and lower uppersocial classes about their yards, we learned that the needfor uniqueness is not necessarily limited to persons havinghigher social status and that the motive of social confor-mity extends across the social class spectrum. Theoreticaldiscussion invokes Veblen’s historical notions of con-spicuous consumption and status display to link the psy-chological and sociological aspects of symbolic con-sumption

INTRODUCTION

One of the most fervently investigated aspects ofconsumer behavior is product symbolism. As early as1959, Levy noted that people purchased products not onlyfor what they do, but also for what they mean. In contem-porary American culture, the products and services onechooses to consume form the basis of both private self-identity and one’s public image (Aaker, Martina, andGaroloa 2001; Belk 1988; Dennis 2002; Hirschman 1985;McCracken 1988; Schor 1999). As a result, researchershave subjected a remarkably wide array of products andservices to scrutiny, attempting to determine the meaningsthey hold for consumers. Among these are apparel (Jolson,Anderson, and Leber 1981; Solomon and Anand 1985;Newell, Claiborne, and Sirgy 1991; Phaum and Prendergast2000), automobiles (Munson and Spivey 1981), motor-cycles (Schouten and McAlexander 1996), home furnish-ings (McCracken 1988; Ulver-Sniestrup 2008), artworkand musical preferences (Halle 1993; Holbrook and Huber1979), hairstyles (McAlexander and Schouten 1989),make-up usage (Miller and Cox 1982), and foods (Levy1959).

In virtually all of these consumption areas, bothpsychological and social structural characteristics arefound to guide consumer choices and usage patterns (Belk1988; Frank 1999; McCracken 1986; Simmel 1957). Forinstance, psychological traits such as conformity, appro-val seeking, materialism, innovativeness, desire for unique-

ness, and risk aversion have all been linked to symbolicproduct preferences (Aaker et al. 2001; Simonson andNowlis 2000; Nunes 2009; Bagwell 1996; Lynn andHarris 1997; Tian, Bearden, and Hunter 2001; Frank1999; Richins 1995). Concurrently, social structural vari-ables such as social class, upward mobility, status striv-ing, and social, economic and cultural capital have alsobeen determined to underlie possession meaning andacquisition (Frank 1999; Holt 1998; Veblen 1899).

Our present purpose is to discuss the relationshipbetween two of these psychological traits – conformityand the desire for uniqueness – and the sociologicalconstructs of social class and social mobility. As will bediscussed in the theoretical section below, there is someevidence that consumer conformity may be more com-mon among the working and lower middle classes; while,conversely, the desire for uniqueness has been identifiedas a trait found more often among those in the upper socialclasses who possess high levels of cultural capital (see,e.g., DiMaggio and Useem 1978; Fussell, Peterson, andDiMaggio 1975; Wilensky 1964).

To examine this potential relationship, we conducteddepth interviews with a cross-section of Americanhomeowners about the meanings of their yards. We choseyards as our focal consumption area, because they seemeda priori to be a site where social class membership, publicself-consciousness and the desire to differentiate oneselffrom others could all be displayed. Remarkably, theyard – a very socially conspicuous aspect of symbolicconsumption – remains unexplored by consumer behav-ior researchers (except as a peripheral cue, see, e.g., Holt1998), despite the fact that maintaining a lawn requiresvast amounts of time, effort and money on the part ofconsumers. Lawn care is a $30 billion annual business inthe United States (www.yardsandlandscapes.com), withhomeowners expending 150 hours per year, on average,mowing, trimming, weeding, fertilizing, and seeding theiryards. There are presently 24,000,000 acres of residentiallawn in the U.S. – an area the size of the state of Indiana(www.lawnsandlandscapes.com). Approximately sevenbillion gallons of fresh water are used every year toirrigate American lawns; and lawn mowing machinesconsume 580,000,000 gallons of gasoline annually. Obvi-ously, the household yard represents a substantial finan-cial investment for the American family. Our study delves


into the nature of consumers’ psychological and socio-logical investments, as well.

PSYCHOLOGICAL AND SOCIOLOGICALPERSPECTIVES ON SYMBOLISM

From their inception in Europe, especially Franceand Britain, during the eighteenth century, formal gardensand lawns were used to announce the elevated socialstatus and cultivated tastes of their owners. They areprominent among the examples of conspicuous consump-tion cited by Veblen (1899, p. 133) who concluded that theturn-of-the-century Victorian yard was a public display“requiring constant tending by gardeners employed by theaffluent in England and the Eastern Seaboard of the U.S.in order to announce their rank in society.” In other words,yards were, and are, examples of symbolic consumption.

Symbolic meanings have been invested in objects ona personal and social level within human populationssince at least the foundation of city-states around 5,000BCE; yet the systematic investigation of these meaningscan probably be dated to Thorstein Veblen’s (1899)sociological observations of the American upper classduring the Gilded Age. Veblen’s, and later Bourdeiu’s(1984), theories of consumption symbolism are premisedupon the linked social motives of status competition andimitation of the successful. More recently, evolutionarypsychologists, such as Richerson and Boyd (2005), haveproposed the same paired set of motivations as underlyinghuman social behaviors.

Analogously, Fussell (1992), a social class theorist,notes that “one of the unique anxieties [of the Americanconsumer] is the constant struggle for individual self-respect based on social approval . . . Despite our publicembrace of political . . . equality, . . . we arrange thingsvertically and insist on crucial differences in personal andsocial value” (www.pbs.org/peoplelikeus/resources/essays6.html, p. 4, accessed on 8/18/2011). Other observ-ers have proposed that the upper class Victorian confor-mity noted by Veblen has now given way to a search forways to publicly display individuality. For example,Peterson (1997, p. 1), writing in Poetics, argues that“highbrow snobbery” as a status marker has now beenreplaced by a “system of cosmopolitan omnivorousness.”Under this new system, one’s exposure to the exotic, therare, and the unusual represents the apex of social status.This would seem consistent with Holt’s (1998) proposalthat persons having high levels of cultural capital willexpend greater effort on self-expression and creativepursuits in order to differentiate themselves from others.

Within the consumer behavior literature, social strati-fication has been consistently linked to the symbolicmeanings attributed to products and services, with per-sons located along the social class continuum varyingsystematically in taste preferences. For instance, using

depth interviews with respondents possessing high andlow levels of cultural capital (see, e.g., Bourdieu 1984).Holt (1998, p. 3) reports that “consumption continues toserve as a consequential site for class reproduction in theUnited States.”

Conformity and Uniqueness

Consumer psychologists have investigated two per-sonality characteristics that are relevant to the presentstudy: the need for approval that may undergird socialconformity behaviors and the need for uniqueness thatmay motivate efforts at personal deviation from socialnorms (Tian, Bearden, and Hunter 2001). Thus far,research in this area has focused on personality traits suchas susceptibility to interpersonal influence (Bearden,Netemeyer, and Teel 1989) and attention to social com-parison information (Bearden and Rose 1990). Findingsfrom these studies suggest that both the need for socialconformity and the need for uniqueness may play impor-tant roles in consumer use of product symbolism. Forexample, the kinds of clothing or makeup worn (Jolson,Anderson, and Leber 1981; Solomon and Schapler 1982),reference group judgments as to product and style appro-priateness (Miniard and Cohen 1983), the potential forsocial approval or disapproval by relevant reference groups(Allen 1965), and expectations about how significantothers might respond to particular consumer behaviors(Calder and Burnkrant 1977) have all been found linkedto motivations for conformity or uniqueness. Thus, weanticipate that consumers’ lawn care practices will beinfluenced by concerns about the opinions of others,especially neighbors, who are publicly able to view thelawn.

Conversely, others’ expectations and social normsmay also serve as a point of differentiation for thoseconsumers who desire to create a distinct image forthemselves. Tian, Bearden, and Hunter (2001, p. 1) reportthat “consumers acquire and display material possessionsfor the purpose of feeling differentiated from otherpeople. . . . Consumers’ need for uniqueness is defined asan individual’s pursuit of differentiation relative to oth-ers . . . achieved through the acquisition, utilization anddisposition of consumer goods [in order to] develop andenhance one’s personal and social identity.”

Because household yards are publicly displayed pos-sessions, we anticipate that some homeowners will bemotivated to use their yards to differentiate themselvesfrom others in the neighborhood, while others will usetheir yards to display conformity to community standards.Following Holt’s (1998) propositions, we further antici-pate that persons lower in social class status will displaya higher level of conformity, while those higher in statuswill exhibit greater efforts to differentiate themselvesfrom their neighbors.


METHOD

Depth interviews were conducted with 26homeowners in two states in the northeastern U.S. Thesocioeconomic status of the homeowners ranged fromworking class to lower upper class using standard catego-rization procedures (see, e.g., Coleman 1983). Interviewswere conducted on-site by trained graduate students. Eachhomeowner consented to the interview and also to photo-graphs being taken of his/her yard. A semi-structured setof questions was used to guide the dialogue [Exhibit One],but interviewers were instructed to permit the respondentsto diverge from the format to enable flexibility in re-sponses.

Exhibit One: Yard Questionnaire

1. How long have you lived at this house?

2. What did the yard look like when you arrived here?

3. What changes have you made?

4. Are there things you would still like to do?

5. What is your favorite part of the yard?

6. What part of the yard do you like least?

7. If you could change your yard in any way, whatwould you do?

8. How are the seasons different in your yard?

9. What do you do during each season?

Interviews were either audio-taped or recorded infield notes. The transcripts were then examined by the twoprimary researchers, independently. After initial contentcategories had been identified by each researcher, a jointdiscussion followed during which further interpretationwas used to compare the data set with pre-existing theo-rization on consumption symbolism, generally, and socialclass, conformity and desire for uniqueness, specifically.

FINDINGS

The interview transcripts provide a much more intri-cate and nuanced view of social class status, conformityand uniqueness than is presently theorized. Indeed, thethree phenomena appear to be intimately connected inmany ways. First, we learned that social conformity andstatus competitiveness are often behavioral correlates.Frequently, homeowners would carefully observe whatothers in the area were doing in terms of landscaping theiryards, and then mimic the accepted “look,” while at the

same time trying to “outdo” the neighbors by adding moreexpensive plants and trees and installing architecturaldetails such as stone walkways, elaborate decks, swim-ming pools, ponds, waterfalls and gazebos. Significantly,this conformity-and-competition pattern was presentthroughout the social class spectrum.

Working to Lower Middle Class

We first discuss how this conformity/competitive-ness pattern was enacted among working to lower middleclass consumers. Ben, for example, is a Filipino registerednurse who lives in a lower-middle class housing develop-ment with his wife and three children. Because their houseis governed by association rules regarding yard appear-ance, Ben feels pressure to conform to the neighborhoodnorms, which include having tidy, mulched flower bedsand mulched trees surrounded by stone or brick borders.Ben states that although they live in a “cookie cutter”neighborhood, he strives to have “the best lawn” on hisstreet.

The family is upwardly mobile, having recently movedto their present residence from an older, working classneighborhood. Ben has recently acquired a set of “good toexcellent quality” lawn tools that he stores in a new shed.In order to make the shed “visually attractive” and con-form to association rules, Ben planted evergreen shrubsaround it. The family has not yet reached an income levelthat permits it to hire a professional lawn care service, soBen and his wife mow and trim the yard themselves.

Upon arriving at their new home, Ben and his wife,Jess, purchased a “trophy” tree, a weeping Alaskan cedar,they had desired to own for several years. Both Ben andJess also exhibit some signs of ‘status anxiety’ about theappearance of their yard. Ben is especially bothered by a“sink hole that does not drain properly” and he feels is“unattractive.” He tried planting a tree near the sink hole,but that effort was unsuccessful – the tree died. Jess statesthat she “feels bad” about a rock wall and area of redgravel at the front of their house which she believes are“ugly” and she is “embarrassed” for neighbors to “have tolook at.” In these instances, we see concern about inad-equate conformity being present, especially in the couple’sconcerns that their yard is not “up to par” with the rest ofthe neighborhood, because of some unattractive features.The concern is strong enough to cause them to feelpressure to remedy the “unsightly” aspects of the yard.Yet at the same time, they have prominently planted a rareand expensive tree in the front yard as a statement of theiraesthetic taste and likely a gesture of cultural superiority.Notably, this action on their part is counter to Holt’s(1998) proposition that such gestures of individuation anduniqueness would be confined to the upper classes and/orthose with high levels of cultural capital.


Another example is provided by Bhuban and herfamily, who immigrated to the U.S. from India, and nowlive in a lower middle class development which they alsodescribe as “cookie cutter,” because “all the houses lookso much the same.” They reported experiencing confor-mity pressure regarding the appearance of their lawn:“there are many expectations to uphold. What you do toyour yard affects your neighbors who have adjoiningyards. If you let weeds infect your lawn, then they canspread to your neighbor’s lawn, no matter how hard theymight try to prevent it.”

Both a sense of community belongingness andcompetition were present in Bhuban’s transcript. Shereported that she felt “a part of” the community andenjoyed that. But she also felt pressure to conform to herneighbors’ expectations. “What other lawns look like canset a standard that you strive to reach for your own lawn.”

She also noted that “competition can arise between differ-ent families and households.”

Bhuban notes that she is “confident in the appear-ance” of her yard, however, because she “is the only onewith a rose garden; often neighbors will knock on my doorasking for a rose or two.” She does, however, feel someunhappiness and insecurity about the tree in their frontyard. “Our tree was planted really late . . . in the wintertime. So while the other trees on our street bloom a lot andare very voluminous, our tree is little and wimpy bycomparison.” In this example, the tree represents aninadequacy in their efforts to conform to communitystandards, while the rose garden serves as both a point ofuniqueness and a marker of superiority. Importantly, thisconsumer is at the lower end of the socio-economiccontinuum, a place where such strivings were previouslybelieved to be absent.

PHOTOS OF FRONT AND BACK YARDS OF INTERVIEWEES’ HOUSES


Middle to Upper Middle Class

Consumers in the middle to upper middle class exhib-ited the same intertwining of social class status, confor-mity, and competition as we found with the classes belowthem, but gave much stronger voice to aspirational pos-sessions. Anne-Marie is a middle class homeowner whoseeks to move upward to the upper middle class. Sheseems to have studied the landscaping norms prevailing inthe upper middle class and desires to add them to her yard.When asked “How would you change your yard?” sheresponded, “I would tear up the entire deck and replace itwith the new [non degradable] material. I would put in ahuge hot tub and completely new backyard furniture . . .I would add an outdoor kitchen against the back wall. . . .That would be so convenient when I’m serving peopleoutside . . . I’ve always wanted one of those outdoor

fireplaces with couches, as well, and definitely a hugebar . . . perhaps connected to a built-in barbecue overthere.” In this example, Anne-Marie sees one route tosignaling her upward mobility to be the adding of highercultural capital assets to her yard. But in naming theaccoutrements she wishes to acquire, she is also display-ing social conformity – that is, she is conforming to thenorms she believes are present in the class above herpresent one.

A second middle class homeowner, Jeff, is alsoyoung and aspiring to become a member of the uppermiddle class. His list of desired yard acquisitions is inmany respects similar to that of Anne-Marie and includes“a fancy brick barbecue area that is contoured and terracedwith Italian block, . . . an in-ground pool, cabana, tenniscourts and a stone-paved area. . . .” In both these cases, the

FRONT AND BACK YARDS OF INTERVIEWEES’ HOUSES


upwardly mobile middle class consumers seem to have arelatively consistent set of yard acquisitions they feel arenecessary to announce upper middle class status. Wewould argue that this, also, is a sign of implicit socialconformity, even though it may be seen by the consumeras displaying individuality and uniqueness.

Thus, the interviews suggest that social class confor-mity norms may apply to one’s aspirational social class,as well as present social class. Those who see themselvesas “on the way up” have a fairly consistent “shopping list”of the things they want to purchase for their yards whenthey arrive.

Upper-Middle to Lower-Upper Class

The interviewed homeowners who were in the uppermiddle to lower upper class exhibited another aspect ofconformity not found in the interviews with those in theclasses below them. Specifically, they consistently descri-bed the use of professional lawn care and landscapingservices. These services were usually initially hired when

they moved into the neighborhood to “update” the yardwith the “plantings” then in style. In no instance did weencounter an upper middle class or lower upper classhomeowner who performed all of his/her yard work,especially the routine chores of mowing the grass orfertilizing the lawn. These repetitive and labor intensivetasks consistently were “hired out” to “a service.”

The homeowners at the highest end of the social classcontinuum were also much more active in restructuringthe appearance of their yards. Instead of purchasingsapling trees, which required several years to grow to anappreciable size, medium and full size trees would beplaced in the yard. Often the yard, itself, would be re-graded to smooth out overly sharp slopes or add curvesand rises to a “too flat” yard. Robert, for example, “hireda professional landscaper within the first year of movinginto set-up his yard in an aesthetically pleasing manner.”

As evidenced above, within this social stratum adifferent vocabulary was used to describe the yard and itsmaintenance. Grass, flowers and shrubs are no longer

FRONT AND BACK YARDS OF INTERVIEWEES’ HOUSES


referred to as “trimmed” or “cut,” but rather as “mani-cured” or “designed.” Homeowners in these social classesalso seem to be keenly aware of shifts in the fashionable-ness of various plant species and décor elements such asswimming pool shapes and materials, lawn furniturebrands, and the desirability of various types of stones usedfor walkways and patios. One homeowner had created a“courtyard area [which] mimics one in Greenwich Villagein New York.” This and other aesthetic references reflec-ted a cosmopolitan sensibility among these homeowners,much as Holt (1998) proposes. The reference groups usedwere not only the neighbors, but what the homeownersconsidered to be “tasteful” or “elegant” in other cultures,for example, Provencal France, the English Cotswolds, oreven more exotic locales, such as Thailand.

What Happened to Uniqueness?

As we have documented across all social strata stud-ied, there are examples of social conformity, and concur-rently, reports of efforts at being unique, of consumersdesiring to differentiate themselves from others throughtheir yards. Several families reported having a “special”tree, flower bed or item of décor that they felt ‘set themapart’ from others in the neighborhood. But as we pro-pose, usually these were implicitly intended as efforts toset oneself “above” others; that is, they were intended associally competitive gestures, not as personal statementsof distinctiveness, per se. Most consistently, the themeconveyed in the interviews was an implicit, or occasion-ally explicit, acknowledgment that there was a social codegoverning what yards in “their neighborhood” shouldlook like and a corresponding behavioral effort on the partof the homeowner to adhere to – or surpass – the basicrequirements of that code. Indeed, many respondentsexpressed a fear of negative feedback from neighbors, ifthey failed to maintain the minimum standards of thecode.

DISCUSSION

We now turn to an effort to merge the perspectives onconsumption symbolism in psychology and sociologyusing the findings from the present study. A foundationalpremise of this effort is that the construction of selfidentity is inherently and inescapably a social project. Webecome individuals through a lifelong process of interact-ing with others and receiving feedback from those withwhom we have relationships (Solomon 1983). Thus,individual lives may be singularly lived, but they aresocially enacted.

In considering the material from the present study, itbecomes clear that consumers’ perceptions of their yardshave simultaneously personal and social elements. Theconsumer may view his/her home and yard as being partof a “cookie cutter community” and him/herself as

immersed in a conformity demanding culture. It is hard todifferentiate the self in such a place. So, a special, rare treemay be purchased or a rose garden constructed in order toannounce an identity separate from one’s neighbors.

Yet the consumer’s judgment of what is appropriateto set oneself “apart” is intimately connected to the sur-rounding social context; how can I establish the appropri-ate level of differentiation (Snyder and Fromkin 1980),one not so distinct as to appear to be a radical or misfit, yetnot so similar as to go unnoticed by my neighbors? Thecountervailing motives of social conformity and indi-vidual uniqueness are calculated and satisfied by carefullygauging the possible responses of one’s social peers: whatwill they think of me? How do I appear to them? How willmy behaviors be interpreted (Fromkin and Snyder 1985)?The psychology of self-identity is never free from thesocial context in which it is enacted.

In our view, it would seem very fruitful to considerwhether status competition and the need for uniquenessmay, in practice, be the same phenomenon. This providesa clue to a key interpretive conundrum of the presentstudy: how are we to conceptually classify the behavioraldescriptions of those respondents who used aspects oftheir yards to differentiate themselves from neighbors inways that they believed established their yards as “better”or “superior?”

From Holts’ (1998) sociological perspective, suchactions would be classified as evidence of possessing highcultural capital. As he states (p. 3), “Cultural capitalconsists of a set of socially rare and distinctive tastes,skills, knowledges, and practices. . . .” High culturalcapital is also consistently associated with membership inthe upper classes, because only members of these classeshave the necessary prerequisites of a challenging educa-tion, childhood exposure to a variety of intellectuallystimulating experiences, an occupation requiring noveland analytical thought, and the financial capital enablingthe pursuit of specific interests and creative ventures (Holt1998; Bourdieu 1984).

This same proposal is made earlier by Kohn (1969;Kohn and Schooler 1983). Kohn’s (1969) study, forexample, found that, “Men of higher class position, whohave the opportunity to be self-directed in their work,want to be self-directed off the job, too, and come to thinkself direction [is] possible. Men of lower class position,who do not have the opportunity for self direction in theirwork, come to regard it as a matter of necessity to conformto authority, both on and off the job. (p. 12). In their laterstudy, Kohn and Schooler (1983, p. 167) found that “jobconditions [related] not only to values and orientations,but also with cognitive functioning as evidenced in . . . theintellectuality of leisure time activities.” Consistent withthis reasoning, Holt (1998, p. 4) states that “cultural


capital is expressed through consuming via aestheticinteractional styles that fit with cultural elite sensibilitiesand that are socially scarce.”

Within this conceptualization, the individual con-sumer expresses his/her personal, distinctive taste to thefullest only within the upper social classes where eco-nomic capital coincides with social and cultural capital.For example, Holt (1998, p. 8) described his High CulturalCapital homeowner respondents as “[viewing] their homesas canvases upon which they express their aesthetic sen-sibilities. . . . Decorating is a highly personalized activity.”The same group is described as “[emphasizing] cosmo-politanism, individuality and self-actualization” (p. 12).

However, Holt (1998, p. 14, 16) then continues hisdescription by stating, “They seek to avoid consumerbehaviors that are perceived to be common, homogeneousor widely-accepted. . . . HCC consumers explicitly dis-avow following a style that is widely adhered to and,instead, talk about how they mix and match to create theirown personal look.” By contrast, low cultural capitalconsumers are said to possess a “subjectivity [that]depends upon community acknowledgment of particulartaste and practices (Holt 1998, p. 16).”

Yet the data collected for the present study suggest adifferent interpretation. What we found was a consciousawareness of social norms operating across the classstructure ranging from working class to lower upper class.Though the specific requirements of these norms variedfrom class to class, they were still consciously invoked byconsumers at all levels in deciding how to decorate theiryards. Further, a sense of competitiveness was also foundat every social class level such that some individualhomeowners would seek to ‘outperform’ their neighborsby not only meeting, but surpassing, the prevailing deco-rative code within their community. In short, we foundevidence supporting the simultaneous presence of confor-mity and differentiation across all social strata studied.These strata ranged from working class to lower upperclass, with only the lower class (whose members usuallyare not homeowners) and upper-upper stratum beingomitted from our sample.

Finally, Holt’s (1998, p. 18) high cultural capitalinformants draw a distinction between the routine main-tenance of the yard and the opportunity be creative anddifferentiate themselves from others: “I like working inthe yard. I find it creative, . . . [but] I don’t cut the grass.We have some guys that come and do that” However, inour view, this practice is also an exhibition of conformity,because all of the higher social class consumers weinterviewed reported the same pattern; that is, routinemaintenance was assigned to hired workers, while deco-rative choices and activities were performed by the home-owner. However, a comparison of photos of the upper

class yard landscapes suggests a striking level of consis-tency across these yards, just as we see conformity withinthe other classes [see photos above]. Thus, even thoughthe upper classes may claim a high level of individualcreativity in yard décor, the results obtained look remark-ably alike. We suggest, therefore, that implicit conformitylevels are as high among upper status consumers as theyare among the rest of the social strata, although theseconsumers do seem to make a higher number of explicitclaims of uniqueness and differentiation.

Writing in 1899, Veblen observed, “Wherever theinstitution of private property is found . . . the economicprocess bears the character of a struggle between men forthe possession of goods (p. 24). . . . The motive that lies atthe root of ownership is emulation . . . [and] the possessionof wealth conveys honor; it is an invidious distinction (pp.25–26).” As noted at the outset to this study, notions ofpossession symbolism grounded in social status pervadevirtually every category of good yet examined. Cogently,Veblen also detected the intertwining of the pressures ofsocial conformity and social striving, which he identifiedas the source of the conspicuous consumption codesexisting among the social strata of his day.

As he notes (p. 30), even among the working classessocial standing requires at least a minimum display ofpublic conformity: “A certain standard of wealth . . . is anecessary condition of reputability. Those members of thecommunity who fall short of this . . . suffer in the esteemof their fellow men, and consequently they suffer also intheir own self esteem, since the usual basis for self respectis the respect accorded by one’s neighbors.” This wewitnessed among our working class homeowners whowere overtly concerned with maintaining at least theminimum standard of conformity to neighborhood normsin the landscaping of their yards.

As one ascends the social strata, Veblen proposes thatthe requirements of ‘maintaining the code’ become morecomplex and demanding, “In our time there is [expectedto be] knowledge of the dead languages, . . . of syntax andprosody; of the various forms of domestic music and otherhousehold art; of the latest proprieties of dress, furnitureand equipage; of games, sports and fancy-bred animals,such as dogs and race horses (p. 45).” This is whatBourdieu (1984) and Holt (1998) refer to as High CulturalCapital. But as we argue, consistent with Veblen’s inter-pretation, it is also a potent form of conformity pressure.Consumers in the upper middle class and the lower upperclass must display competence in their mastery of thiscode in order to gain the approval of their neighbors andalso to achieve an internal sense of self esteem. Competi-tiveness for social acclaim (and self-worth) in this realmrequires becoming the most distinctive in ways that areviewed as enviable by one’s peers.


Because of this code and competitiveness, the land-scaping we found in these homes is not “randomly”distinctive; rather it is conformitively distinctive. Thereare shared rules of appearance for the yard in upper middleclass and lower upper class neighborhoods which areadhered to with great consistency. As Veblen (p. 84)states: “The members of each stratum accept as their idealof decency the scheme in vogue. . . . They must conformto the accepted code.” Thus, as Snyder and Fromkin(1980) and others (see, e.g., Lynn and Harris 1997a,b)propose, the desire for uniqueness, for distinctiveness, fordifferentiation, even among those with high cultural capi-tal, is socially bounded. Indeed, efforts at differentiationwould seem to be part and parcel of the conformity codeof the upper social classes. One is expected to displayindividuality – but always within reason.

CONCLUSIONS

The present study is intended to stimulate a greaterdegree of rapprochement between sociological and psy-chological theories relevant to consumer behavior andmarketing. Perhaps too often research is cast within theconfines of a specific personality trait, social psychologyconstruct or sociological perspective. While the findingsresulting from such studies can be – and often have been– very useful, they also may not provide the field with thelarger context in which phenomena are embedded. One ofthe benefits of studying marketing and consumer behav-ior is that the objects of research within these fields areinherently personal and social, taking place in the mind ofthe consumer and the social structure simultaneously.This provides us with a great opportunity for buildingtheories that reside at multiple levels of abstraction andincorporate varied foci of interest.

REFERENCES

Aaker, Jennifer, Veronica Benet-Martinez, and JordiGarolera (2001), “Consumption Symbols as Carriersof Culture,” Journal of Personality and Social Psy-chology, 81 (3), 492–508.

Bearden, William O., Richard G. Netemeyer, and JesseTeel (1989), “Measurement of Consumer Suscepti-bility to Interpersonal Influence,” Journal of Con-sumer Research, 15 (March), 473–81.

____________ and Randall L. Rose (1990), “Attention toSocial Comparison Information: An Individual Dif-ference Factor Affecting Consumer Conformity,”Journal of Consumer Research, 15 (4), 461–71.

Belk, Russell W. (1988), “Possessions and the ExtendedSelf,” Journal of Consumer Research, 15 (Septem-ber), 139–68.

Bourdieu, Pierre (1984), Distinction: A Social Critique ofthe Judgment of Taste. Cambridge, MA: HarvardUniversity Press.

DiMaggio, Paul and Michael Useem (1978), “SocialClass and Arts Consumption,” Theory and Society, 5(March), 141–61.

Brewer, M.B. and C.l. Pickett (1999), “DistinctivenessMotives as a Source for the Social Self,” in ThePsychology of the Social Self, T.R. Tyler and R.M.Kramer, eds. Mahwah, NJ: Erlbaum, 71–87.

Cheema, Amar and Andrew M. Kaikati (2010), “TheEffect of Need for Uniqueness on Word of Mouth,”Journal of Marketing Research, 47 (3), 553–63.

Coleman, Richard P. (1983), “The Continuing Signifi-cance of Social Class to Marketing,” Journal ofConsumer Research, 10 (December), 265–80.

Durham, William (1991), Coevolution: Genes, Cultureand Human Diversity. Stanford, CA: Stanford Uni-versity Press.

Frank, Robert (1999), Luxury Fever: Money and Happi-

ness in an Era of Excess. Princeton: Princeton Uni-versity Press.

Fussell, Paul (1992), Class: A Guide Through the Ameri-can Status System. Philadelphia: Touchstone.

Gilbert, Dennis (2002) The American Class Structure.New York: Wadsworth.

Halle, David (1993), Inside Culture: Art and Culture inthe American Home. Chicago: University of ChicagoPress.

Hirschman, Elizabeth C. (1985), “The Creation of Prod-uct Symbolism,” Advances in Consumer Research,Richard Lutz, ed. Provo, UT: Association for Con-sumer Research, 13, 327–31.

Holbrook, Morris B. and Joel Huber (1979), “The Deter-minants of Aesthetic Value and Growth,” Advancesin Consumer Research, Jerry Olson, ed. Associationfor Consumer Research, Provo, UT: 7, 134–39.

Holt, Douglas B. (1998), “Does Social Class StructureAmerican Consumption?” Journal of ConsumerResearch, 25 (June), 1–25.

Hyatt, Eva (1992), “Consumer Stereotyping,” Advancesin Consumer Research, John F. Sherry, Jr. and BrianSternthal, eds. Association for Consumer Research,Provo, UT: 19, 299–303.

Jolson, Marvin, Ralph E. Anderson, and Nancy J. Leber(1981), “Profiles of Signature Goods Consumers andAvoiders,” Journal of Retailing, 57 (Winter), 19–38.

Kohn, M.L. (1969), Class and Conformity: A Study inValues. Homewood, IL: Dorsey Press.

____________ and C. Schooler (1983), Work and Per-sonality: An Inquiry Into the Impact of Social Strati-fication. Norwood, NJ: Ablex.

Levy, Sidney J. (1959), “Symbols for Sale,” HarvardBusiness Review, 37 (July/August), 117–24.

Lynn, Michael and Judy Harris (1997a), “The Desire forUnique Products: A New Individual DifferencesScale,” Psychology & Marketing, 14 (6), 601–16.


____________ and ____________ (1997b), “IndividualDifferences in the Pursuit of Self-Uniqueness ThroughConsumption,” Journal of Applied Social Psychol-ogy, 27, 1861–83.

McAlexander, James H. and John W. Schouten, “HairStyle Changes as Transition Markers,” Sociologyand Social Research, 74 (October), 58–62.

McCracken, Grant (1988), Culture and Consumption.Bloomington, IN: Indiana University Press.

Miller, Lynn Carol and Cathryn L. Cox (1982), “ForAppearances Sake: Public Self-Consciousness andMake-up Use,” Personality and Social PsychologyBulletin, 8 (December), 748–741.

Munson, J. Michael and W. Austin Spivey (1981), “Prod-uct and Brand User Stereotypes Among SocialClasses,” Advances in Consumer Research, Kent B.Monre, ed. Ann Arbor, MI: Association for Con-sumer Research, 8, 696–701.

Nunes, Joseph C. (2009), “Status: Why ConsumersEngage in Conspicuous Consumption,” Advances inConsumer Research, Association for ConsumerResearch, 32, 12–15.

Peterson, Richard and Albert Simkus (1992), “How Musi-cal Tastes Mark Occupational Status Groups,” inCultivating Differences, Michael Lamont and MarcelFournier, eds. Chicago: University of Chicago Press,152–86.

____________ (1997), “The Rise and Fall of HighbrowSnobbery as a Status Marker,” Poetics, 25 (2/3),(November), 75–92.

Phau, I. and G. Prendergast (2000), “Consuming LuxuryBrands: The Relevance of the Rarity Principle,”Journal of Brand Management, 8 (2) 122–38.

Richerson, P.J. and R. Boyd (2005), Not by Genes Alone.Chicago: University of Chicago Press.

Richins, Marcia L. (1994), “Special Possessions and theExpression of Material Values,” Journal of Con-sumer Research, 21, 522–33.

Schor, Julia (1999), Why We Buy What We Don’t Need.New York: Harper.

Schouten, John W. (1991), “Selves in Transition: Sym-bolic Consumption in Personal Rites of Passage,”Journal of Consumer Research, 17 (March), 412–25.

Simmel, Georg (1957), “Fashion,” American Journal ofSociology, 62, 541–58. [orig. published 1904].

Simonson, Itamar and S.M. Nowlis (2000), “The Role ofExplanations and Need for Uniqueness in ConsumerDecision Making,” Journal of Consumer Research,27, 48–68.

Snyder, C.R. and Howard L. Fromkin (1980), Unique-ness: The Human Pursuit of Difference. New York:Plenum Press.

Solomon, Michael R. (1983), “The Role of Products asSocial Stimuli: A Symbolic Interactionist Perspec-tive,” Journal of Consumer Research, 10 (Decem-ber), 319–29.

Tian, Kelly Tepper, William O. Bearden, and Gary L.Hunter (2001), “Consumers’ Need for Uniqueness:Scale Development and Validation,” Journal of Con-sumer Research, 1 (June), 50–66.

Ulver-Sniestrup, Sofia (2008), Status Spotting: A Con-sumer Cultural Exploration of Home Aesthetics. Lund,Sweden: Lund Business Press.

Warner, W.L. (1960), Social Class in America. NewYork: Harper and Row.

Veblen, Thorstein (1899), The Theory of the LeisureClass. New York: Sentry Press.

Wright, Newell, C.B. Claiborne, and M. Joseph Sirgy(1992), “The Effects of Product Symbolism on Con-sumer Self Concept,” Advances in Consumer Re-search, John F. Sherry, Jr. and Brian Sternthal, eds.Provo, UT: Association for Consumer Research, 19,311–18.

www.landscapesandgardenstoday.com, accessed June 22,2011.

For further information contact:Elizabeth C. Hirschman

Rutgers University School of BusinessRockefeller Road

Picataway, NJ 08901Phone: 276.206.8675

E-Mail: [email protected].


HOW CLOSE BRANDS ARE INCLUDED IN THE SELF:PSYCHOLOGICAL AND NEURAL PROCESSES

Martin Reimann, University of Southern CaliforniaRaquel Castaño, EGADE Business School

Judith L. Zaichkowsky, Simon Fraser UniversityAntoine Bechara, University of Southern California

SUMMARY

In three experiments, this research provides newinsights into branding by studying the psychological andneurophysiological mechanisms of how consumers relateto their beloved brands. The authors propose that emo-tional arousal decreases over the brand relationship span,while inclusion of the brand into the self increases overtime. Results of Experiment 1 indicate greater self-re-ported emotional arousal for recently formed brand rela-tionships, as well as decreased emotional arousal andincreased inclusion of close brands over time. Addition-ally, the moderating role of usage frequency of the brand

brings out an interesting nuance of the way these effectsoperate. Experiment 2 measures skin conductance re-sponses and reveals increased emotional arousal for re-cently formed close relationships but not for establishedclose brand relationships, corroborating the results basedon self-reported data. In Experiment 3, a functional mag-netic resonance imaging study reveals an associationbetween established close relationships and activation ofthe insula, a brain area previously found to be a crucialmechanism in diverse but related psychological phenom-ena such as urging, addiction, loss aversion, and interper-sonal love.

For further information contact:Martin Reimann

Department of Psychology / Brain & Creativity InstituteUniversity of Southern California3620 South McClintock Avenue

Los Angeles, CA 90089Phone: 408.813.8449



THE REFERRAL BACKFIRE EFFECT: THE IDENTITY THREATENINGNATURE OF REFERRAL FAILURE

Bart Claus, Iéseg School of Management, FranceKelly Geyskens, Maastricht University, The Netherlands

Kobe Millet, VU Amsterdam, The NetherlandsSiegfried Dewitte, KU Leuven, Belgium

SUMMARY

Although the average consumer engages in referringothers on a daily basis (Keller 2007), stimulated bycompanies that even sometimes reward consumers fordoing so (Ryu and Feick 2007; Schmitt, Skiera, and Vanden Bulte 2011), we are not aware of any prior researchthat addressed the effect of referral failure on the personwho engaged in the referral. However, as the outcome ofthe referral becomes more and more transparent in theemerging online environment, this is an issue of growingimportance. Literature indicates that consumption itselfcan often be considered a non-verbal form of identity-expression (Belk, Bahn, and Mayer 1982; Reed 2004),and that engaging in referral makes consumption evenmore publicly visible (Brown, Barry, Dacin, and Gunst2005; Hennig-Thurau, Gwinner, Walsh, and Gremler2004). Therefore, we propose that referral failure – thesituation in which one’s advice is rejected - may in somecircumstances threaten consumers’ identities. Literatureprovides ample evidence that identity threats motivateconsumers to bolster their self-concept (Dunning, Perie,and Story 1991; Wentura and Greve 2005). When con-sumers’ identity is threatened, they choose products thatsupport their self-concept (Gao, Wheeler, and Shiv 2009)and become more motivated to firmly advocate theirthreatened self-beliefs (Gal and Rucker 2010). More ingeneral, theoretical literature indicates that self-deter-mined behavior is an important route to preserve and buildthe self-concept (Blanton and Christie 2003; Deci andRyan 1985). As a result, we propose that referral failuremotivates consumers to make self-determined decisions.In doing so, they discard external information becauseincorporating it in one’s decision may further question theconsumer’s ability to decide for herself. We focus on thereduced likelihood to comply with external influence asone instantiation of self-determination. We call this reluc-tance to take into account external information as trig-gered by referral failure, the referral backfire effect. Wetested the existence of this effect and its underlyingprocess in four experimental studies. Study 1 used ascenario study to demonstrate self-determination as amain effect of referral failure. In Study 2, we used ameasure of implicit egotism – i.e., participants’ prefer-ence for a name resembling their own first name (Joneset al. 2002) – to show that the referral backfire effect is

mediated by implicit egotism, supporting the explanationof this effect in terms of a defense against self-threat.Study 3 demonstrated that the effect occurs within andacross the domain of referral, but is conditional on someknowledgeability of that domain, in line with an explana-tion that referral failure should threaten a central part ofthe identity to be perceived as threatening at all. Study 4used a moderation design to show – in line with the fluidcompensation principle (Steele 1983) – that an externalsource of self-affirmation in an unrelated domain canattenuate the referral backfire effect. We therefore pro-vided process evidence and boundary conditions to thereferral backfire effect: The referral backfire effect (a) isdriven by a need to bolster the self (Study 2), (b) isattenuated when the topic of referral is not central to theidentity of the sender (Study 3), and (c) does not longerhold when the threat is compensated by self-affirmation(Study 4). In addition, the referral backfire effect isindependent of the match between domain of the referraland domain of the subsequent product decision (Study 3).In combination, these studies suggest that referral failureposes a threat to consumers’ self-concepts, leading tomore self-determined behavior as an attempt to preserveand build the self-concept. We believe that our findingsextend previous research in two important ways. First, tothe best of our knowledge we are the first to acknowledgethe fact that the outcome of giving advice to others mightbe consequential to the sender of advice. Although severalauthors have identified motivations to engage in givingadvice (Dichter 1966; Hennig-Thurau et al. 2004;Sundaram, Mitra, and Webster 1998), no research hasinvestigated what happens when the needs behind thosemotivations are not met. Second, the potential effects ofreferral failure on subsequent behavior questions theoften heralded efficacy of WOM (Reichheld 2003; Cheva-lier and Mayzlin 2006). Considering that consumers whomake referrals are usually the company’s most satisfiedcustomers (Anderson 1998), our results imply that espe-cially those consumers might no longer accept externalinformation upon noticing referral failure, including thatof the company. From a managerial point of view, we alsosee two important implications. Stimulating referral, forexample by referral reward programs (Ryu and Feick2007; Schmitt, Skiera, and Van den Bulte 2011), is oftenmentioned as the solution to the declining effectiveness oftraditional advertising methods (Kumar 2010; Van denBulte and Wuyts 2009). Ironically, failing referral might


very well contribute to that same decline, as consumers inthe face of referral failure might feel a need to make theirown decisions and ward off external information in theform of marketing messages. Interestingly, our findingsnot only point at a risk in the current practice of stimulat-ing consumer referral, they also offer ways of dealing withconsumers facing referral failure. Study 4 demonstratedthat self-affirmation in an unrelated domain can compen-sate for the threat caused by referral failure. Future

research might test other relevant implications. Forinstance, senders who just incurred a referral failure maybe more interested in offers that promise to bolster the self,such as “compensatory” conspicuous consumption (cf.,Rucker and Galinsky 2008). In conclusion, we emphasizethe necessity for more research into the effects of referraloutcome on the sender, as the implications could besubstantial for both consumer theory and at the market-place. References are available upon request.

For further information contact:Bart Claus

Iéseg School of Management40, passage de l’Arche – Collines sud de l’Arche 92037

Paris-La Defense cedexFrance

Phone: +33(0)155911010E-Mail: [email protected]


SELF-OTHER DIFFERENCES IN PURCHASE UNCERTAINTY ANDCONTINGENT DECISION STRATEGIES

Demetra Andrews, Florida State University, TallahasseeStephanie Oneto, University of Wyoming, Laramie

SUMMARY

Your customer wants to make a purchase but isuncertain how to proceed. How best should your front-line employee proceed? Findings from this exploratoryinvestigation suggest that the employee’s strategy shoulddepend on whether the ultimate recipient of the purchaseis the customer or another person.

Conceptualization

Drivers of Uncertainty: Because consumers havegreater familiarity with (Prentice 1990) and knowledge of(Jones and Nisbett 1971) their own goals versus those ofothers, they also possess more knowledge of the goals thatare at stake in a given purchase instance. This is expectedto lead to greater variety in the drivers of uncertainty forself- vs. other-focused purchases. Activation of multiple(as opposed to single) goals also alters the consumer’sappraisal of the purchase situation (van Osselaer et al.2005) which is also expected to lead to differences in theconsumer’s perception of the drivers of uncertainty.

Coping Tactics

Coping behaviors are selected based on the consumer’sappraisal of the situation (Payne, Bettman, and Johnson1988; Sujan et al. 1999). Differences in perception of thedrivers of uncertainty will lead to commensurate differ-ences in decision strategies employed to cope with thatuncertainty. Congruent with other marketing researchefforts, this research focuses on the decision strategies(Sujan et al. 1999) that consumers employ when engagingin confrontative coping, or behaviors focused on address-ing the problem at hand (see Holahan and Moos 1987;Mick and Fournier 1998; Viswanathan et al. 2005). In thisresearch, the need for coping arises in response to aconsumption-related issue as opposed to a life event (e.g.,divorce) (Moschis 2007). Additionally, the focus is onpre-purchase coping rather than post-purchase coping(see Duhachek 2005 and Yi and Baumgartner 2004).

Methods

This investigation employed exploratory proceduresthat were adapted from Greenleaf and Lehmann’s (1995).As with prior research, a phenomenological approach wasemployed that relies on retrospective recall of purchase

instances (see Duhachek 2005; Mick and Fournier 1998;Sujan et al. 1999; Yi and Baumgartner 2004).

In the first phase, participants were asked to write adescription of an instance in which he or she made apurchase despite feeling uncertain. Data were used todevelop measures to assess the drivers of uncertainty andcoping tactics that were employed the second phase of theresearch.

In the second phase, 245 undergraduate businessstudents (45% male) participated in an online study forextra credit. As per their assigned condition, participantsprovided information and answered questions regarding apurchase situation in which they experienced uncertaintywhen making a purchase for themselves or for someoneelse. One hundred twenty consumers (49%) providedinformation on self-focused purchases and 125 (51%)provided information on other-focused purchases.

Major Findings

A principal components analysis was conducted inSPSS, by condition of purchase recipient. SPSS automati-cally standardizes variables when using the correlationmatrix. Given that consumers may experience multipleforms of uncertainty simultaneously (Sujan et al. 1999)and the ability of oblique rotation to “provide a more validrepresentation of the phenomena under study” (Stewartet al. 2001, p. 78), items representing drivers of uncer-tainty were subjected to oblique rotation.

The analysis revealed notable differences in the driv-ers of purchase-related uncertainty as a function of pur-chase recipient. While total explained variance was simi-lar across the two purchase-recipient conditions (75%

Self-

focused, 71%

Other-focused), more components were extracted in

the self-focused purchase condition than in the other-focused condition. In other words, as anticipated, self-focused purchase was characterized by a greater variety ofuncertainty drivers than was other-focused purchase. Thisfinding is congruent with the supposition that greaterknowledge of one’s own goals may increase the numberof elements that induce uncertainty such that self-focusedpurchases are fraught with more challenges than other-focused purchases. This is also congruent with researchasserting that self-focused decisions are more difficult(Kray and Gonzalez 1999) and generate a greater magni-tude of anxiety or worry (Loewenstein et al. 2001) than


other-focused decisions. Additionally, while the largestproportion of variance for self-focused purchase wasexplained by insufficient search (29%), the biggest issuefor other-focused purchase was social concerns / risks(23%). This difference highlights the need for front-linesalespeople to ascertain the purchase recipient when try-ing to understand a consumer’s hesitance or uncertainty inmaking a purchase. The driver of the uncertainty is likelyto be very different.

The analysis also revealed differences in the decisionstrategies employed to contend with uncertainty as afunction of the purchase recipient. Note that five compo-nents were extracted in the self-focused condition vs. onlyfour in the other-focused condition. Moreover, one of the

components extracted in the other-focused condition actu-ally reveals an unwillingness to rely on external assistanceor cues, which reduces the number of decision strategiesin that condition down to only three. This suggests that thefront-line employee has fewer effective tactics to suggestto aid consumers who are trying to contend with uncer-tainty in other-focused purchases. Moreover, a markeddifference in usage of external assistance or informationcues was observed, such that external information wasrelied upon for self-focused purchases, but was eschewedfor other-focused purchases. Similarly, reliance on mar-ket-based heuristics (e.g., buying the most expensive ormost popular item), which would also, provide justifica-tion for choice, was only associated with self-focusedpurchases. References are available upon request.

For further information contact:Demetra Andrews

Marketing DepartmentThe College of BusinessFlorida State University

P.O. Box 3061110Tallahassee, FL 32306–1110



CORPORATE SOCIAL RESPONSIBILITY AND CONSUMERRELATIONSHIPS: AN APPLICATION OF THE

MOTIVATOR-HYGIENE THEORY

Russell Lacey, Xavier University, CincinnatiPamela Kennett-Hensel, University of New Orleans

SUMMARY

While corporate social responsibility (CSR) hasemerged as an important tenet of contemporary businesspractices, the impact of CSR efforts on consumer relation-ships remains ambiguous. In general, stakeholder groups,such as employees, suppliers, customers, and communitymembers perceive value in CSR but despite additionalknowledge gained from recent academic literature a keyquestion remains; namely, to what extent does CSR moti-vate consumer relationships? In order to increase thecollective knowledge of the influence and limitations ofCSR as a motivation factor, this consumer behavior studydevelops and tests a conceptual model influenced byHerzberg’s motivator-hygiene theory (Herzberg, Mausner,and Snyderman 1959) and explores the role of CSR as arelationship motivating factor and/or as a hygiene busi-ness requirement that if perceived to be subpar wouldinstead only weaken the consumer’s relationship with thecompany.

Consumer views on CSR’s impact on their support ofcompanies and products differ. Researchers have demon-strated that certain consumers see CSR as a motivatingfactor for strengthening company/product relationships(Sen, Bhattacharya, and Korschun 2006). To the extentthat CSR serves as a relationship motivator, consumersare more loyal to the company or product as a result oftheir positive evaluations regarding the company’s socialresponsibility activities (Marin, Ruiz, and Rubio 2009).However, it is plausible that the absence of CSR endeav-ors might primarily impact consumer dissatisfaction.Already in industries where CSR practices are common-place, CSR may be more likely to serve more so as ahygiene factor among many of its customers, rather thanas a motivating factor to enhance existing company (orbrand) allegiances. Thus, from the perspective of ongoingrelational exchanges, avoiding consumer dissatisfactionmay become the key benefit of CSR (Meijer and Schuyt2005).

In this study, competing moderating effect hypoth-eses explore whether the consumer perceives the role ofthe company to actively demonstrate social responsibilityas a motivator or hygiene factor and how this impacts theintensity of the relationship between CSR and the qualityof the consumer’s relationship with the company. Follow-ing from Agustin and Singh (2005) who make a further

distinction between basic lower-order (hygiene) andhigher-order (motivator) needs in their study of customerloyalty determinants for ongoing relational exchanges, itis posited that in relational exchanges, CSR can functionas a basic, lower-order, hygiene-type business require-ment or act as a growth, higher-order need and in eithercase will have a significant moderating effect on therelationship between CSR and relationship quality.

Multi-group analysis was conducted to assess thehypothesized moderating effects on the structural modelusing LISREL 8.80. The study’s hypotheses are tested byestimating a structural equation model with two waves ofsurvey data from consumers who had purchased tickets toattend home games of a National Basketball Association(NBA), captured at both the beginning of the 2009–2010(n = 1000) and 2010–2011 (n = 653) NBA seasons. UsingLikert-type scales, the researchers employed a combina-tion of new and existing scales. Most notably, theLichtenstein, Drumwright, and Braig (2004) measure ofCSR was used to assess perceptions of the company’sefforts regarding corporate giving and support. The rela-tionship quality scale was adapted from Garbarino andJohnson (1999). New measures were developed in orderto capture the respective potential roles of “CSR as aMotivator” and “CSR as a Hygiene Factor.”

Analysis of the study’s results reveals a significantmoderating affect of CSR in terms of how it is assessed asa hygiene factor. However, the association between CSRand relationship quality is not significantly impacted byCSR as a motivating factor. In other words, companiesthat engage in CSR are not rewarded for doing so throughenhanced relationship quality, but they are punished fornot doing so through declining relationship quality.

These key findings underscore the critical relation-ship between CSR, economic goals and marketing efforts.Research has shown that dissatisfiers are prioritized oversatisfiers (Vargo et al. 2007) and that people are moresensitive to negative performances (Audrain-Pontevia2004). When hygiene factors become core competencesof the company, they play as important of a role asmotivation factors since hygiene factors provide the fun-damental environment for motivation factors to be able tooperate. Motivating factors are less effective when hy-giene needs are lacking. A company wishing to leverageits marketing dollars will be handicapped if CSR effortsare lacking. References are available upon request.


For further information contact:Russell Lacey

Xavier University3800 Victory ParkwayCincinnati, OH 45207Phone: 513.745.3049



A NEW TOOL FOR CUSTOMER SEGMENTATION: DEFINING ANDMEASURING RELATIONSHIP PRONENESS

Stephanie M. Mangus, Louisiana State University, Baton RougeJacob L. Hiler, Louisiana State University, Baton Rouge

Benjamin D. McLarty, Louisiana State University, Baton Rouge

SUMMARY

Relationship proneness has been studied in customerand employee contexts and prior research has defined theconstruct as a personality trait reflecting a consumer’sstable and conscious propensity to partake in relation-ships with sellers of a specified product group. In order toenhance this work, a more sophisticated measurement ofa generalized relationship proneness tool is needed tomore precisely explain the various components of thisconstruct. Our conceptualization of relationship prone-ness identifies the construct as a compound personalitytrait relating to how consumers react to other people andorganizations regarding relationship development. Fol-lowing established scale development procedures(Netemeyer, Bearden, and Sharma 2003; Gerbing andAnderson 1988), 122 items were compiled from existingmeasures and generation, which were then examined byseven expert judges. Using student and adult samples, thescale was narrowed to a final count of 14 items to reflect

a second-order relationship proneness construct. The sec-ond-order construct represents four types of customerrelationships – relationships with firms, relationshipswith brands, relationships with representatives of firms,and relationships with other consumers. The resultingmeasurement tool offers researchers more generalizedusage of the construct in multiple important organiza-tional contexts (e.g., B2B, B2C, service, product, retail,and employee settings) and more accurately representsthe many types of relationships a consumer engages inwhen developing business relationships. This new mea-surement tool is valuable to researchers investigatingconsumer responses to relationship marketing efforts andcontributes to theory by explaining a mechanism consum-ers use in determining business relationships. Practitio-ners will find value in applying this tool to not only assessconsumer relationships, but to identify employees thatmay be more relationship prone, and therefore, especiallysuccessful in frontline sales or service roles. Referencesare available upon request.

For further information contact:Stephanie M. Mangus

Louisiana State University3127 Patrick F. Taylor Hall

Baton Rouge, LA 70803Phone: 513.748.2154



DETERMINANTS OF CUSTOMER LOYALTY: THE ROLE OFRELATIONAL BENEFITS IN THE CONTEXT

OF CUSTOMER CLUB

Kevin Yu, University of South Australia, AdelaideSong Yang, University of South Australia, Adelaide

ABSTRACT

This research found that most of the relational ben-efits identified in the literatures emerged in the context ofcustomer clubs. Economic-based special treatment ben-efits were found to have a greater influence on customersatisfaction, commitment and customer loyalty while shar-ing benefits was found to have an impact on commitment.

INTRODUCTION

Research on relational benefits has drawn muchattention in relationship marketing. Researchers haveidentified seven types of benefits which serve as theantecedents to loyalty-building projects, namely social,confidence, special treatment, symbiotic, identity-related,comfort, and interactive benefit (e.g., Gwinner, Gremler,and Bitner 1998; Hennig-Thurau, Gwinner, and Gremler2002; Lacey, Suh, and Morgan 2007; Algesheimer,Dholakia, and Herrmann 2005; Sweeney and Webb 2002;Spake et al. 2003; De Wulf, Odekerken-Schroder, and vanKenhove 2003). However, the contribution of these rela-tional benefits has rarely been examined in the context ofcustomer club. A customer club is an effective platformand a tool in customer relationship management initia-tives. Although previous research did find that somerelational benefits have stronger influence than others, theunderlying rationale is not clear (Gwinner, Gremler, andBitner 1998). Little is known why some customer clubsoutperform others. Managers of customer club still needconcrete guidelines for their operations. How to make acustomer club more attractive remains an unansweredquestion in marketing literature. Given the importance ofcustomer clubs, as a marketing vehicle, in enhancingcustomer loyalty and retention, it is imperative to identifythe critical success factors which count for the variance inthe attractiveness and success of customer clubs.

The primary objective of this study is to identify thekey relational benefits contributing to efficiency of acustomer club by answering the following three researchquestions:

• What are the possible structures underlying the rela-tional benefits identified in the literatures for cus-tomer club?

• Does impact of a relational benefit vary betweencustomer-club members and non-customer-club mem-bers?

• To make a customer club more attractive whichrelational benefits need to be enhanced?

LITERATURE REVIEW

The field of relational benefits from customers’ moti-vations to participate in relationship with companies hasbeen the focus in relationship marketing for a decade (e.g.,Bendapudi and Berry 1997; Gwinner, Gremle, and Bitner1998; Hennig-Thurau, Gwinner, and Gremler 2000;Sweeney and Webb 2002; Weng et al. 2010; Dimitriadis2010; Conze et al. 2010; Chen et al. 2010). However,relational benefits have rarely been the focal concern inrecent research in customer club. Customer clubs areplatforms where the social context will be broadened tonot only a relation between organizations and customers,but also facilitating networks amongst customers. Thus,customer clubs may lead to increase perceptions of socialand any other benefits. Organizations who offered cus-tomer club program is to establish switching barrier,retention and customer loyalty. Gwinner et al. (1998)pointed out that customer satisfaction and loyalty strate-gies can be built around relational benefits.

Gwinner et al. (1998) suggested that for a long-termrelationship to exist, both the firm and the customer mustbenefit. They studied the benefits to the customer forbeing loyal to a service provider and findings from theirstudies indicated that consumer relational benefits havecategorized into three distinct benefit types: confidencebenefits, social benefits, and special treatment benefits;and amongst these benefits, confidence benefits receivedmore attention and rated as more important than the otherrelational benefits by consumers.

In an attempt to examine the benefits from the per-spective of both buyers and suppliers, Sweeney and Webb(2002) studied across 20 buyer-supplier pairs and identi-fied seven benefit categories: symbiotic, psychological,operational, social, economic, strategic, and customization.Symbiotic benefits, according to Sweeney and Webb(2002), are derived from a sense of sharing, mutuality,


reciprocity, commonality, partnering, alliance, co-ism,togetherness and common understanding. This constructis rarely tested in the service context as well as thecustomer club context.

In an empirical study among members of theVolkswagen Customer Club, Germany’s largest automo-tive customer club, Strauss et al. (2001) indicate thatcustomer club satisfaction has a remarkable impact on thecustomer’s relationship satisfaction and customer reten-tion. A customer club is regarded as a suitable platform toincrease the interaction frequency between company andcustomer (customer interaction effect) by creating contactand feedback opportunities. By doing so, a close contactis built around the client throughout the entire customerlife cycle. Thus, interaction effects have played an impor-tant role in customer club and it is a prerequisite for thecustomer club organizer to obtain and manage customerinformation. On the other hand, customers can obtainproducts and company’s information through such aninteraction process too.

More recently, in a study of European car clubmembers, Algesheimer et al. (2005) found that strongerbrand community identification (identity-related benefits)leads to greater community engagement and brand loyaltyintentions. This is in line with Hennig-Thurau et al.’s(2000) advocate on identity-related benefits.

Consumer comfort also plays a key role in servicerelationships. Comfort benefits are positive emotions andare defined as feeling at ease or reduced anxiety associ-ated with alleviating mental discomfort (Spake et al.2003). Consequences of comfort can include increaseddisclosure, confidence, trust, commitment, self-esteem,reduced perceived risk, satisfaction, and improved rela-tional exchange. Comfort may be one of the criteria usedwhen selecting providers. Comfort is assessed also duringinteraction with service provider. It may also serve as abarrier to exit (Spake et al. 2003). Hennig-Thurau et al.(2000) found confidence benefits, partly measured interms of reduced anxiety, to be the most important rela-tional benefits across segments and industry types. Accor-ding to Spake et al. (2003), although previous scales havebeen developed to measure comfort, they are largelysingle-item measures, components of other scales (inclu-ding confidence benefits) not designed to assess comfortwith service provider or viewed as opposite of anxiety.

As a summary, it is logical to posit that relationalbenefits in the service relationships may also appear in thesetting of customer club programs. But their relativeimportance and relevance may vary. As a result, studiesare yet to verify the status and to integrate the researchstreams on relational benefits which include the identity-related benefits, interactive benefits, comfort benefits and

symbiotic benefits in the setting of customer club. Theprimary objective of this study is to identify the keyrelational benefits in the setting of customer club from theconsumers’ perspective.

CONCEPTUAL MODEL DEVELOPMENT

An initial model was developed from previous re-search which has been discussed in the previous section.This model was based on work by Gwinner et al. (1998),Hennig-Thurau et al. (2002), Sweeney and Webb (2002),Lacey et al. (2003), Algesheimer et al. (2005), Spake et al.(2003), Stauss et al. (2001) and DeWulf et al. (2003) andcan be seen in Figure 1.

The model represents a conceptualization of rela-tional benefits from consumers’ perspective (both clubmembers and non-club members). This model combinesthe relational benefits and the relationship marketingoutcomes (commitment, customer satisfaction and cus-tomer loyalty). It also specifies how relational benefitsmay influence the three important relational marketingoutcomes.

To refine the model and find out the possible structureunderlying the relational benefits, a series of focus groupinterviews have been conducted. Twenty-seveninterviewees were randomly selected from the Friso cus-tomer club (Friso, a brand of Infant milk powder from theNetherlands). Analysis of the qualitative interview datafollowed the guidelines of Lincoln and Guba (1985), andGwinner et al.’s (1998) research. Through the interviewresults, three new relational benefits are found, namelyInformation Benefits, Sharing Benefits and Communica-tion Benefits. Based on the strength of the links (in termsof the frequency of the key words) and directions descri-bed by the interviewees, we re-structure the model toreflect the multi-level structure of the relational benefits.The modified conceptual model was displayed in Fig-ure 2.

The following research propositions are put forth andwill be tested through the data analysis process below.

P1: There is a hierarchic structure underlying the rela-tional benefits.

P2: The strength of the impact of the relational benefitson loyalty outcomes will differ significantly betweencustomer-club members and non-customer-club mem-bers.

QUESTIONNAIRE SURVEY

Quantitative data were collected in the second phaseof the research. A questionnaire was designed usingestablished scales from previous studies, as well as scales


developed from the focus group interviews. Extending theresults from the interviews and drawing on the previouslycited literature, the structural equation modeling empiri-cally tests the dimensionality and relative importance ofcustomers’ relational benefits in the context of formulamilk powder category and in the context of customer club.

In empirical analysis, customer club member cus-tomers and non-club member customers will be surveyedto empirically examine the dimensionality, existence, andrelative importance of the relational benefits (confidencebenefits, social benefits and special treatments benefitsand any other relational benefits that will be identified inStudy 1) and its relationship to customer satisfaction,commitments and customer loyalty. From previous con-ceptual discussions in the literature, it is expected thatdifferent benefit types may vary in their importance andthe degree to which they are received, depending on thetype of service that consumers are in. This study is mainlyfocusing on one business category (formula milk powder)– customer club of consumer products. As such, in the firstpart of the questionnaire a reminder will be highlightedand the respondents are asked to reconfirm whether theyare members or non-members before giving respond tothe remaining questions.

Sample

Three hundred consumers voluntarily participatedinto the questionnaire survey. The respondents, shoppersof milk powders, were approached as they left the store inwhich they had bought at least one tin of infant milkpowder in supermarkets in Shanghai China. Screeningquestions listed on a separate contact sheet were used toidentify shoppers who satisfied the selection criteria.Shoppers who satisfied the criteria were asked to partici-pate into the research. Names and telephone numbers ofrespondents were collected, in order for a follow up call bya market research company hired by the researchers. Asmall cash prize was offered to encourage responses. Theoriginal questionnaires have been constructed in English,and then translated to Chinese. The questionnaires hadbeen pre-tested with a group of 20 customers to ensure theEnglish meaning of various concepts; phrases and evenwords were equivalent in the Chinese language.

Respondents were called by the staff of the marketresearch company. They are asked to respond to thequestions based on their experiences with milk powdercompanies who are operating a customer clubs. Beforethey answer the questionnaire, it is carefully verified that

FIGURE 1Initial Conceptual Model


each respondent has previously purchased the milk pow-der products. As the research involved mainly two groupsof consumers: Members and Non-members, data werecollected from two different samples of active milk pow-der shoppers. For members’ group, the respondents haveto be the current members of the milk powder club. For thenon-members group, the respondents have to be thecurrent milk powder users but never be a member of anyformula milk powder customer clubs. The total number ofrespondents is 300, 150 club members and 150 non-clubmembers. As it is mothers who make decision to buy themilk powder in China. All respondents are female. Major-ity of them (89.6%) were aged between 26–35 years.

A two-step approach was employed to analyze thedata as recommended by Anderson and Gerbing (1988).That is, the measurement model was assessed by perform-ing a confirmatory factor analysis and then the structuralmodel was assessed to obtain the path coefficients.

To test assess the convergence validity, confirmatoryfactor analysis was conducted. As indicated by factorloadings, dimensionality of constructs, the indices ofcomponent reliability (CR) and average varianceextracted (AVE), the data fit the measure models (conge-

neric model) well. All figures in psychometrics propertieshave exceeded the proposed threshold (Bagozzi and Yi1998). These results suggest that the measures adopted inthe study proved to be adequate convergent validity. Thediscriminant validity was also tested by following theassessment practices suggested by Thompson (1997). Allthe results indicate that measures maintain sound dis-criminant validity.

AMOS software was used to estimate the path coef-ficients in the conceptual model. The output of the estima-tion showed that a lot of links were insignificant. To cometo the final model, several steps have been taken includingre-specification of the conceptual model by deleting theinsignificant links and establishing new links. The finalmodel is presented in Figure 3. The path coefficients ofinitial and final model were shown in Table 1.

RESULTS AND DISCUSSION

As Figure 3 shows, sharing benefits have a directpositive and significant effect on commitment (β = 0.19,p < 0.01). However, there is no statistically significant linkbetween sharing benefits and customer satisfaction; andbetween sharing benefits and customer loyalty. It means

FIGURE 2Modified Conceptual Model Based on Qualitative Research Resules

CommunicationBenefits


that sharing benefits have not made a significant contribu-tion to customer satisfaction and customer loyalty to thesurveyed participants in China in milk powder category.

Special treatment benefits (economic-based) is theonly construct with a direct significant contribution toboth commitment and customer satisfaction (β = 0.34, p <0.001 and β = 0.71, p < 0.001 respectively). However, thisconstruct has not made a significant direct contribution tocustomer loyalty. Customer satisfaction has a direct posi-tive effect on customer loyalty (β = 0.76, p < 0.001).However, there is no statistically significant link betweencommitment and customer loyalty.

For the constructs of identity-related benefits, inter-active benefits, comfort benefits and symbiotic benefits;there is no statistically significant link between theseconstructs and commitment, customer satisfaction andcustomer loyalty. However, sharing benefits serve as themediating role.

Result of Multi-Group Comparison

The moderating effects of membership and non-membership are significant. It implies that models are

significantly different in terms of their structural weightsbetween the club member customers and non-club mem-ber customers. Thus, the effect of relational benefits oncustomer actions varies between club member customersand non-club member customers. A new link from sharingbenefits to customer satisfaction is established. The finalmodel for club member is presented in Figure 4 and thefinal model for non-club member is presented in Figure 5.

The results support the relationships proposed in themodel. However, among all different relational benefits,only six out of ten are considered relevant in the contextof customer club. In particular, the role of sharing benefitsand economic-based special treatment benefits as media-tors between other relational benefits (identity-relatedbenefits, Interactive benefits, comfort benefits and symbi-otic benefits) and relational marketing outcomes (cus-tomer satisfaction, commitment and customer loyalty) isgenerally supported by the data.

Evidenced from the results, the two research propo-sitions were supported. Specifically, the sharing benefits,the new construct which identified in the focus groupresearch, have a strong positive effect on commitment.Economic-based special treatment benefits have the great-

TABLE 1Standardized Structural Equation Estimation (Initial Model and Final Model)


FIGURE 4Path Coefficients of the Final Model for Club Member Samples

FIGURE 3Path Coefficients of Final Model for All Samples

InteractiveBenefit

InteractiveBenefit


est effect on customer satisfaction and indirectly to com-mitment and customer loyalty. Both sharing benefits andeconomic-based special treatment benefits also play amediator role. Sharing benefits serve as a mediator foridentity-related benefits and interactive benefits indi-rectly to commitment while economic-based special treat-ment benefits serve as a mediator for comfort benefits andsymbiotic benefits indirectly influence customer satisfac-tion commitment and customer loyalty.

Between sharing benefits and economic-based spe-cial treatment benefits, economic-based special treatmentbenefits have a greater effect on customer satisfaction andcommitment; and economic-based special treatment ben-efits are also found to have an indirect effect on customerloyalty mediated by customer satisfaction.

When considering the moderating effect of customerclub membership, both sharing benefits and economic-based special treatment benefits are important constructs;and economic-based special treatment benefits is theconstruct that most influenced the relationship marketingoutcomes (customer satisfaction, commitment and cus-tomer loyalty). However, while sharing benefits existamong club member customers, it does not have a director indirect influence on commitment among the non-clubmember customers. Economic-based special treatmentbenefits have a direct influence on customer satisfaction

and commitment among non-club member customers andaffect customer loyalty via the mediator of customersatisfaction; and have indirect influences on commitmentvia the mediator of customer satisfaction. One the otherhand, economic-based special treatment benefits play adirect influence on customer satisfaction and indirectinfluence on commitment and customer loyalty via themediator of customer satisfaction among club membercustomers. This suggests that it is vital for brand ownersto develop sharing benefits only for club member custom-ers while it is vital also for brand owners to developeconomic-based special treatment benefits for both clubmember customers and non-club member customers.

For club member customers, as both identity-relatedbenefits and interactive benefits are the antecedents ofsharing benefits, it is vital for brand owners to developfavorable identity-related benefits and interactive ben-efits while also generating sharing benefits perception.Although both comfort benefits and symbiotic benefitsare the antecedents of economic-based special treatmentbenefits, it is important to note that in the context of clubmember customers, symbiotic benefits impacted throughthe model’s mediating variable, economic-based specialtreatment benefits, relating to customer satisfaction whilecomfort benefits have insignificant impact to economic-based special treatment benefits. However, in the contextof non-club member customers, comfort benefits impac-

FIGURE 5Path Coefficients of the Final Model for Non-Club Member Samples

InteractiveBenefit


ted through the model’s mediating variable, economic-based special treatment benefits, relating to customersatisfaction and commitment while symbiotic benefitshave insignificant impact to economic-based special treat-ment benefits. Thus, it is suggested that the customer clubmanagers should differentiate its offers to various cus-tomer groups. It is vital for customer club operators tohighlight symbiotic benefits for club member customers.To make a customer club more attractive, managers shouldendeavor to enhance the sharing benefits.

LIMITATION AND FUTURE RESEARCH

Formula Milk Powder was the product category inthis study. Formula Milk Powder product required high-involvement by the mothers with the babies. The productsare known to be purchased under substantial consider-ations due to one child policy in China. Thus, the modelobtained may be more suitable to formula milk powder orproducts required high-involvement than to goods of lowinvolvement. Thus, managers should use these resultswith caution. The customer clubs in this study are thecustomer clubs of consumer goods (Formula Milk Pow-der). Other servicing customer club such as airlines com-panies with loyalty programs, hotel group with loyaltyprograms and restaurant groups are not included in thisstudy. Thus, the implications of relational benefits onservicing customer club would be different.

This research focused on the products of formulamilk powder. For this product category, the dimension ofrelational benefits, which is defined in context of cus-tomer club was explored. There are altogether ten con-structs of relational benefits discovered either in theprevious studies or in the focus group research in thisstudy. However, results in this study showed that only

limited choice are relevant in the context of pre-selectedproduct category or in the context of customer club.

A direction for further research is to apply the modelof this research to other products or services sectors. Theimplications may be quite different in different contexts.Future research could explore:

1. The context of customer club in services sector (e.g.,airlines, travel agents especially internet travel agent,hotel groups or chain stores).

2. Another context of interests could be the customerclub of low-involvement products.

3. The impact of relational benefits on the context ofconsumer sponsored brand community (e.g., CarOwners’ Club, Fan Club of celebrity, etc.).

So far only a limited number of studies have shedlight on the interrelationships of relational benefits (e.g.,Han and Kim 2009; Kim 2009). As there are differentrelational benefits existed in the relationship marketingcontext, future research could explore the nature of thepotential associations among different relational benefitsin different contexts. In addition, it is worth investigatingthe changes in relational benefits under the influence ofcultural values. Future research could explore and test themodel in cultures other than Chinese. Finally, differentrelational model may exist in different stages of relation-ship building process. Future research could test therelational benefits in different stages of relationshipsbetween customers and the company. Also, it is interest-ing to capture the change in relationship in the samerelationship context (e.g., behavior change in the samecustomer club in different stages of relationship).

REFERENCES

Algesheimer, Rene, Utpal M. Dholakia, and AndreasHerrmann (2005), “The Social Influence of BrandCommunity: Evidence from European Car Clubs,”Journal of Marketing, 69 (July), 19–34.

Anderson, J.C. and D.G. Gerbing (1988), “StructuralEquation Modeling in Practice: A Review and Rec-ommended Two-Step Approach,” PsychologicalBulletin, 103 (May), 411–23.

Bagozzi, R.P. and Y. Yi (1998), “On the Evaluation ofStructural Equation Models,” Journal of Academy ofMarketing Science, 16, 74–94.

Bendapudi, B.N. and L.L. Berry (1997), “Customers’Motivations for Maintaining Relationships with Ser-vice Providers,” Journal of Retailing, 73 (1), 15–37.

Chen, Po-Tsang and Hsin-Hui Hu (2010), “The Effects ofRelational Benefits on Perceived Value in Relation toCustomer Loyalty: An Empirical Study in the Austra-lian Coffee Outlets Industry,” International Journalof Hospitality Management, 29, 405–12.

Conze, Oliver, Thomas Bieger, Christian Laesser, andThomas Riklin (2010), “Relationship Intention as aMediator Between Relational Benefits and CustomerLoyalty in the Tour Operator Industry,” Journal ofTravel & Tourism Marketing, 27 (1), 51–62.

De Wulf, Kristof, Gaby Odekerken-Schroder, and Patrickvan Kenhove (2003), “Investments in ConsumerRelationships: A Critical Reassessment and ModelExtension,” The International Review of Retail, Dis-tribution and Consumer Research, 13 (3), (July 2003),245–61.


Dimitriadis, Sergios (2010), “Testing Perceived Rela-tional Benefits as Satisfaction and Behavioral Out-come Drivers,” International Journal of Bank Mar-keting, 28 (4), 297–313.

Gwinner, K.P., D.D. Gremler, and M.J. Bitner (1998),“Relational Benefits in Services Industries: TheCustomer’s Perspective,” Journal of the Academy ofMarketing Science, 26 (2), 101–14.

Han, Heesup and Wansoo Kim (2009), “Outcomes ofRelational Benefits: Restaurant Customers’ Perspec-tive,” Journal of Travel & Tourism Marketing, 26,820–35.

Hennig-Thurau, T., K.P. Gwinner, and Dwayne D.Gremler (2002), “Understanding Relationship Mar-keting Outcomes – an Integration of Relational Ben-efits and Relationship Quality,” Journal of ServicesResearch, 4 (3), 230–47.

Kim, Wansoo (2009), “Customers’ Responses to Cus-tomer Orientation of Service Employees in Full-Service Restaurants: A Relational Benefits Perspec-tive,” Journal of Quality Assurance in Hospitality &Tourism, 10, 153–74.

Lacey, Russell, Jaebeom Suh, and Robert M. Morgan(2007), “Differential Effects of Preferential Treat-ment Levels on Relational Outcomes,” Journal of

Service Research, 9 (3), 241–56.Lincoln, Y.S. and E.G. Guba (1985), Naturalistic Inquiry.

Beverly Hills, CA: Sage.Spake, Deborah F., Sharon E. Beatty, Beverly K.

Brockman, and Tammy Neal Crutchfield (2003),“Consumer Comfort in Service Relationships: Mea-surement and Importance,” Journal of ServiceResearch, 5 (4 May), 316–32.

Stauss, B., K. Chojnacki, A. Decker, and F. Hoffmann(2001), “Retention Effects of a Customer Club,”International Journal of Service Industry Manage-ment, 12 (1), 7–19.

Sweeney, Jillian C. and Dave Webb (2002), “Relation-ship Benefits: An Exploration of Buyer-SupplierDyads,” Journal of Relationship Marketing, 1 (2),77–91.

Thompson, B. (1997), “The Importance of StructureCoefficients in Structural Equation Modeling Con-firmatory Factor Analysis,” Educational and Psy-chological Measurement, 57, 5–59.

Weng, Rhay-Hung, Jin-An Huang, Ching-Yuan Huang,and Shih-Chang Huang (2010), “Exploring theImpact of Customer Relational Benefit on Relation-ship Commitment in Health Service Sector,” HealthCare Manage Review, 35 (4), 312–23.

For further information contact:Kevin Yu

University of AustraliaDivision of Business

The International Graduate School of BusinessWL5–38, City West Campus

Adelaide, South Australia 5000Australia

Phone: +86.13501702383Fax: +86.28.86668426



TURNING A LION INTO A KITTEN: THE ROLE OF APOLOGY TIMINGAND RELATIONAL EXPECTATION

Kyeong Sam Min, University of New OrleansJae Min Jung, California State Polytechnic University at Pomona

Kisang Ryu, Sejong University, Korea

SUMMARY

An apology is a speech act that seeks forgiveness.Even though much research has consistently shown thatan apology plays a crucial role in restoring damagedrelationships, there has not been agreement on the expla-nations for how the timing of making an apology operates.Specifically, literature on service failure or crisis manage-ment suggests that a wrongdoer should promptly apolo-gize to victims because the wrongdoer’s speedy recoveryhelps to enhance consumers’ perception of proceduraljustice (Kellerman 2006; Smith, Bolton, and Wagner1999). However, conflict management literature suggeststhat a wrongdoer’s delayed apology can be more effectivethan an immediate apology because the former givesvictims more opportunity for self-expression, comparedto the latter (Coleman 1997; Frantz and Bennigson 2005).Therefore, the goal of this research is to improve ourunderstanding about how a service person’s apologytiming influences consumer satisfaction.

According to Frantz and Bennigson (2005), victimsare more willing to accept an apology after, rather thanbefore, they have expressed their concerns and an offen-der has made them feel understood. Drawing on theirfinding, we predict that consumers who value a relationalself will react more positively if they have an opportunityto have their concerns expressed and heard before (i.e.,delayed apology), rather than after (i.e., immediate apol-ogy), receiving an apology. It is anticipated that consum-ers’ motivation to voice their concerns is more likely toincrease if they expect, rather than do not expect, tointeract with the same wrongdoer in the future. Thus, wehypothesize that a service person’s delayed apology willincrease consumer satisfaction than an immediate apol-ogy if consumers’ relational expectation is relativelyhigh. In contrast, we predict that an immediate apologywill produce a more positive outcome than a delayedapology if consumers do not value a relational self. In thiscase, consumers’ perception of a wrongdoer is likely toimprove if they feel that they have been fairly treated. Inparticular, consumers’ motivation to voice their concernsis more likely to decrease if they do not expect, rather thanexpect, to interact with the same wrongdoer in the future.Thus, we hypothesize that a service person’s immediate

apology will increase consumer satisfaction than adelayed apology if consumers’ relational expectation islow.

Using service failure and recovery scenarios, wetested the conditions that influence the effectiveness ofapology timing. We employed a 2 (apology timing) x 2(relational expectation) between-subjects design. As pre-dicted, we find that the apology timing effect depends onconsumers’ relational expectation to interact with thesame service person. Across different service failurecontexts, we consistently show that a delayed apologyleads to greater satisfaction than an immediate apologywhen consumers’ relational expectation is relatively high.In contrast, as predicted, the apology timing effect isreversed when consumers’ expectation to interact is low.An immediate apology enhances consumer satisfactionmore than a delayed apology in this case. Moreover, weobserve that the apology timing effect is differentiallymediated by consumers’ cognition and affect, dependingon their relational expectation. The advantage of makinga delayed apology over an immediate apology occurs onlythrough consumers’ beliefs that the service person under-stands consumers’ views when their relational expecta-tion is high, whereas the advantage of making an immedi-ate apology over a delayed apology occurs only throughconsumers’ feelings during the failed service experiencewhen their relational expectation is low.

In sum, even though much research has documentedthe important role of an apology in recovering damagedinterpersonal relationships, we know very little about howthe timing of an apology operates. The current researchextends our knowledge on apology timing in three ways.First, we consistently document the apology timing effectacross various service contexts. Second, we predict andconfirm that consumers’ relational expectation serves asa boundary condition for the apology timing effect.Finally, we shed light on the underlying processes throughwhich the apology timing influences consumer satisfac-tion, and demonstrate that a different process is at workdepending on consumers’ relational expectation. Thisresearch is expected not only to help illuminate the pro-cesses behind the apology timing effect but also to assistmarketing managers in better communicating their wrong-doings. References are available upon request.


For further information contact:Kyeong Sam Min

Department of Marketing & LogisticsCollege of Business Administration

University of New Orleans2000 Lakeshore Dr.

New Orleans, LA 70148Phone: 504.280.6195



THE IMPACT OF PARENT-CHILD BUYING COMMUNICATIONSTRATEGIES ON SELF-BRAND CONNECTIONS

Farrah Arif, University of Cambridge Judge Business School, United KingdomWayne D. Hoyer, University of Texas at Austin

Omar Merlo, Imperial College Business School, United Kingdom

SUMMARY

Several social scientists have highlighted the con-spicuous use of brands amongst young people anddescribe them as ‘bonded to brands’ and ‘branded babies’(Linn 2005; Schor 2005). They have criticized this com-mercialization of childhood, which has made young peoplematerialistic and consumption oriented individuals. Fur-ther, previous consumer research indicates that youngpeople’s association with brands is socially motivated,and they develop relationships with brands because brandsreflect who they are to others (Chaplin and John 2005;Chaplin and Lowrey 2010; Elliott and Leonard 2004;Robinson and Kates 2005). In addition, Chaplin and John(2005) suggest that young consumers, like adults, alsodevelop self-brand connections (SBC). SBC is a form ofbrand relationship that reflects the extent to which anindividual incorporates a brand into his/her self-conceptto communicate one’s self to others (Escalas and Bettman2003). However, Chaplin and John (2005) focus on thedifferential effects of age and suggest that young consum-ers begin developing SBC between middle childhood andearly adolescence.

Although the social motivation for owning brands isa recurring theme in previous research on young consum-ers, it has paid little attention to the role of predominantsocialization agents, such as parents and peers, who playan important role in shaping consumer attitudes andbehavior (John 1999; Moschis and Churchill 1978). Thecurrent study factors in the social context of consumptionto address this gap. Specifically, it focuses on the under-lying social mechanism of the development of SBC. Itsuggests a possibility that young consumers could sepa-rate their sense of self with popular brands, therebyreducing reliance on such brands for social approval. Itwould be possible, if parents give consumer educationand guidance to young people through monitoring theirbuying decision-making.

A socialization model is proposed and tested toelucidate the underlying mechanism of the developmentof SBC in young consumers (11–15 years old). In thecurrent research, SBC is considered a consumption motivethat enables adolescents to seek group affiliation. It issuggested that certain brands “symbolize group identityand a sense of belonging to distinct social groups” (John

2008, p. 236). Therefore, young consumers link brandconsumption to their self-identity and in the processdevelop SBC for social approval (Chaplin and John 2005).

It is argued that parental communication strategiestoward their children’s buying decision-making play animportant indirect role in the development of SBC. In thispaper, two parental communication strategies are dis-cussed: Empowered Buying Decision-Making (EB) ver-sus Monitored Buying Decision-Making (MB).

EB occurs when parents reduce their role in buying-decision making of young people. There could be severalreasons for giving this empowerment of decision makingto young consumers, including avoiding ridicule (Wooten2006); giving more independence in the expectation ofboosting children’s confidence (Schor 2005); substitut-ing for quality time spent with children (McNeal 1999).The reduction of parental input makes young consumersmore independent (from parents) but not necessarily frompeers, and so in the vacuum of parental input, they becomemore vulnerable to peer influence. When the need tobelong to peers increases, young consumers become moresusceptible to peer influences, and it would lead to astronger SBC for social approval.

MB occurs when parents monitor their children’sbuying decision-making. This strategy enables parents toreinforce their rules and expectations for young consum-ers’ purchases, which leads to social construction andthus, they engage in a dialogue with parents (Hunter1985). This parent-child communication is perceived as asign of caring by young people (Bugental and Grusec2006). It also affects the selection of peers and steers themaway from those peers who oppose parental views onconsumption (Collins et al. 2000). In the process, theybecome more susceptible to the views of their parents.Further, a strong relationship with parents provides asocial support, decreases the level of insecurity, andboosts self-esteem (Dumas et al. 2009). Some parents alsostress on a rational use of brands (Moschis 1985). In thepresence of a significant role of parents, young peoplemight not use brands as an identification mark (Rodhain2006), and it would lead to a lower degree of SBC.

Although prior research shows that socializationagents (parents and peers) have a direct influence on brand


loyalty (Moschis et al. 1984), SBC might play a role in therelationship. It is argued that socialization agents influ-ence SBC. In turn, SBC influences brand loyalty in youngconsumers.

To test the proposed model, a survey in a classroomsetting was conducted with 555 pupils (11–15 years old)from seventh to tenth grades of three state-run secondaryschools in Cambridge, U.K. The questionnaire was devel-oped with the items used in previous studies. However,the items were pretested and adapted according to the agegroup involved. The proposed model was run with struc-tural equations modeling. The data showed an acceptablefit to the proposed model.

In summary, the study links socialization researchwith the development of SBC in young consumers. Theprevious such research has largely considered the role of

peers as a predominant factor in young people’s consump-tion attitudes and behaviors (Chaplin and John 2005;Elliott and Leonard 2004; Pechmann et al. 2003; Rodhain2006). On the contrary, this research provides evidencethat parent-child buying communication strategy is ofhigh relevance to understand the mechanism of child-brand relationship development in a social environment.This research provides evidence for the differentialeffects of each of the parenting communication styles onthe susceptibility to interpersonal influences of parentsand peers. It is suggested that the consumption culture thatpromotes the empowerment of young consumers, therebyleading to higher peer influence. Potentially, this under-standing of the roles of parents and peers has implicationsfor educating young people about their choices of brands,when use predominantly for social approval. Referencesare available upon request.

For further information contact:Farrah Arif

University of Cambridge Judge Business SchoolTrumpington Street

Cambridge CB2 1AGUnited Kingdom

Phone: +44(0)7552617404E-Mail: [email protected]; [email protected]


MAKING BRAND ASSOCIATIONS AND BRAND ELEMENTSMEMORABLE IN ELDERLY CONSUMERS

Praggyan (Pam) Mohanty, Governors State University, University ParkS. Ratneshwar, University of Missouri, Columbia

Moshe Naveh-Benjamin, University of Missouri, Columbia

SUMMARY

Prior research in cognitive psychology has foundimportant differences between associative memory anditem memory (Clark 1992; Clark and Burchett 1994;Gronlund and Ratcliff 1989; Hockley 1991). While itemmemory is memory for individual items present in astimulus episode, associative memory is memory for acombination of two or more items together. Especiallygermane to the present research is the prior finding thataging affects item and associative memory differently.Larger age-related impairments are seen in associativememory than item memory (for a review refer to Old andNaveh-Benjamin 2008; Spencer and Raz 1995).

Item memory and associative memory for brandinformation are both important for brand building. Memoryfor item information such as individual brand names orbrand logos helps increase brand awareness and keeps thebrand salient in the minds of consumers. Associationsbetween the brand and favorable and distinct attributesand benefits work toward the creation of a positive brandimage (Aaker 1996, p. 25; Broniarczyk and Alba 1994;Keller 1993). Overall, brand equity and marketplacesuccess of brands depend upon the creation of high levelsof brand awareness as well as positive and distinctivebrand images (Warlop et al. 2005).

Despite the importance of item and associativememory for effective brand building, there has been verylittle consumer research on how these constructs aredifferent from one another. Further, to the best of ourknowledge, there is not even a single published study inconsumer research on the effects of aging on item vs.associative memory for brand information. This dearth ofresearch is perplexing, given the importance of the elderlyconsumer to marketers (Yoon et al. 2005). Indeed, currentdemographic trends suggest that the age segment of 65and above will increase dramatically over the next twodecades, growing to nearly 20 percent of the total U.S.population by 2030. The growth of the elderly consumersegment makes a compelling argument for studyingmemory issues in this age group.

Therefore, the overarching objective of this researchis to examine the effects of aging on associative and itemmemory for brand information. Further, we investigate

the very important issue of how marketer-controlled vari-ables can influence item and associative memory perfor-mance in the elderly consumer in a branding context.Building on prior theorizing and empirical research in thememory area (e.g., Naveh-Benjamin 2000; Naveh-Ben-jamin et al. 2007; Naveh-Benjamin et al. 2003), twostudies were conducted. In study 1, the main question ofinterest is whether using more versus less meaningfulbrand elements (i.e., brand logos) will improve item andassociative memory for brand information in elderly (vs.younger) consumers. Meaningfulness of a brand elementis the extent to which an individual has pre-existingsemantic knowledge of the element in memory. In Study 2,the same type of question is explored in regard to thedegree of relatedness between brand elements (i.e., brandlogos and brand names). See Figure 1 for a few real-worldexamples of the constructs of meaningfulness and related-ness in a branding context.

Study 1 (N = 50) used a 2 (elderly vs. youngerconsumer, between-subjects) x 2 (associative memory forbrand information vs. brand logo item memory, within-subject) x 2 (more vs. less meaningful brand logos,within-subject) mixed factorial design. The dependentvariable was recognition accuracy as measured by theproportion of hits minus false alarms (Law et al. 1998;Morrin and Ratneshwar 2003). The stimuli were pairs ofbrand logos and brand names. While brand logos weremanipulated to be more vs. less meaningful, the brandnames were meaningful ones that were counterbalancedacross the more and less meaningful brand logo condi-tions.

Participants were tested individually in enclosed labo-ratory rooms. There were four study-test blocks/trials inthe experiment. The study material in each trial comprisedof 14 less and 14 more meaningful brand logo-brand namepairs that were presented in a mixed random order. Allappropriate counterbalancing were done. At study, par-ticipants were instructed to memorize the brand logos andbrand names both individually and as a pair. Each pair waspresented for 4500 milliseconds in a computer-controlledtask. The study phase was followed by a 3-minute fillertask. Participants were then tested for both item andassociative memory (counterbalanced for order of presen-tation). Specifically, they were given 6000 millisecondsto press “V” if they had seen the item (for item memory


test) or pair (for pair memory test) during stimulus expo-sure and press “N” if they had not seen the item or pair.Please refer to Figures 2 and 3 for examples of stimuli andmemory tests.

Study 1 results showed that meaningfulness of brandlogos helped elderly consumers in remembering brandlogos better. Further, there was a two-way interactionbetween meaningfulness of logos and age in the case ofbrand logo (i.e., item) memory. Thus, the deficit in brandlogo memory between elderly (vs. younger) consumers inthe less meaningful brand logos condition was actuallyeliminated in the more meaningful brand logos condition(see Figure 4). In addition, meaningfulness of brand logosenhanced associative memory for brand logos and brandnames in elderly consumers. Some interesting resultswere also obtained in subsequent follow-up analyseswhere we examined hit rates and false alarm rates sepa-rately.

Study 2 (N = 48) used a similar design and procedureas study 1. Results showed that relatedness between brandlogos and brand names helped elderly consumers remem-ber the individual brand elements better (i.e., improveditem memory). Interestingly, there was also a significanttwo-way interaction between age and relatedness for itemmemory. Specifically, elderly (vs. younger) consumersbenefited more from relatedness between brand logos andbrand names in boosting their item memory performance(see Figure 5). Further, relatedness between brand logosand brand names also helped elderly consumers improveassociative memory for brand information. Finally, inaccord with past research, both studies 1 and 2 demon-strated that in the case of novel brand information, elderly(vs. younger) consumers display a larger deficit in asso-ciative memory than item memory, i.e., there was asignificant two-way interaction between age and type ofmemory. References are available upon request.

For further information contact:Praggyan (Pam) Mohanty

College of Business and Public AdministrationGovernors State University

1 University ParkwayUniversity Park, IL 60484–0975

Phone: 573.529.6782Fax: 708.534.8487



ON THE INTERPERSONAL TRANSMISSION OF LUCK

Chun-Ming Yang, Ming Chuan University, TaiwanEdward Ku, National Kaohsiung University of Hospitality and Tourism, Taiwan

Chung-Chi Shen, National Chiayi University, Taiwan

SUMMARY

Peculiar beliefs are beliefs “presumed (by scientists,at least) to not be veridical” and “do not have a rational,empirical, or scientifically established link to an outcomethey are intended to influence.” Despite it may be deniedon the conscious level, the influences of peculiar beliefs inconsumer’s daily life are prevalent and play an importantrole in individuals’ lives through its influence on the non-conscious level. While prior research on peculiar beliefsprovide considerable evidence as to its role in socialpsychology, it is surprising that relatively few efforts havebeen made to understand its influences on consumerbehavior and its role in marketing. An important butneglected research area in the field of peculiar beliefs isthe synergistic effects of various forms of peculiar beliefs.Scholars have identified two specific types of peculiarbeliefs that have recently received attention in the con-sumer psychology and marketing literature: superstitionand magical thinking. Although obviously related,research on these two as of yet disparate peculiar beliefsis scant. To fill this gap, this research introduces animportant but rarely examined concept-- the secondary-contamination effect (hereafter SCE) – and demonstratesits influences on consumer responses.

Based on the literature in social psychology andconsumer psychology, we define the SCE as the changesin consumer perceived luck and other downstreamresponses which are affected by physically contactingwith an object previously touched by another person (i.e.,a contaminated object or vehicle). The concept of the SCEhas its origins in medical and social psychology literature.However, it has never been examined in consumer behav-ior context. We contribute to the research on peculiarbeliefs in consumer behavior by introducing the conceptof the SCE and examining its influence on consumerresponses. This research also seeks to address the psycho-logical mechanism underlies the SCE. Based on the litera-ture on active-self concept, the authors argue thatconsumer’s self-concept will be temporarily changed bycontacting with a contaminated object.

This paper investigates the SCE with four studies.Study 1 provides initial support to the existence of SCE.By contacting with the source (our confederate), theneutral vehicle’s inner quality was changed. We madeparticipants physically touched this contaminated vehicleand captured their perceived luck and Likelihood Judg-ment. The results indicated that touching a contaminatedvehicle temporarily changed participant’s self-conceptand lead to higher perceived luck. Perceived luck then inturn had positive effects on likelihood judgment of win-ning a lottery game. Study 2 shows that SCE is nothomogeneous across participants. Touching a contami-nated vehicle had stronger effects on perceived luck andLikelihood Judgment for those with high trait supersti-tion. Another important finding of Study 2 is the replica-tion of meditating effect of perceived luck, which againdemonstrates that we can temporarily change participant’sself-concept, lead to different perceived luck level, andthen is reflected on their assessment toward likelihood ofwinning the lottery. In Study 3, the authors exploredanother important moderating factor: Participant’s aware-ness of influences of peculiar belief. Based on the FlexibleCorrection Model, the authors hypothesize that consum-ers who are aware of the influences of peculiar beliefs intheir daily lives will make adjustments with these possiblesources of influence. A lab-based experiment providessupports to our hypothesis. Our last study (Study 4)inquires an interesting question: Whether the occurrenceof first contamination should be visually observed byconsumers to activate the SCE? Research in consumercontamination suggests that direct observation is not anecessary condition for contamination effect. Empiricalevidences suggest that the SCE could be activated byconsumer imagination. In other words, what is reallyimportant is consumer’s belief that the neutral vehicle hasbeen contaminated.

Based on the empirical evidences, the authors alsoprovide theoretical and managerial implications. Limita-tions and future research directions are also mentioned.References are available upon request.


For further information contact:Chun-Ming Yang

Ming Chuan UniversityTourism School, No. 5, Deming Road

Taoyuan CountyTaiwan 333

Phone: +886.3.3507001, Ext. 5077E-Mail: [email protected]


PERSONAL AND HISTORICAL NOSTALGIA: INVESTIGATINGCONSUMPTION CONSEQUENCES OF TWO DIFFERENT

TYPES OF NOSTALGIA

Tina Kiessling, Chemnitz University of Technology, GermanySteffen Jahn, Chemnitz University of Technology, Germany

Cornelia Zanger, Chemnitz University of Technology, Germany

SUMMARY

Within the last twenty years we observe a trend thatcan be characterized as “nostalgia-boom” or “retro-revo-lution” (Brown, Kozinets, and Sherry 2003) leading to avariety of past-related market offers. In the media, nostal-gic advertising appeals are widely used to trigger consum-ers’ recollection and emotions (Muehling and Sprott2004). Documentation, celebration or analysis of the pastbecomes ubiquitous in TV reports and magazines andrequests increase for older products to be sold in antiqueshops, flea markets, and auctions or exhibited in museumsand galleries (Goulding 2001). Additionally, numerousalmost forgotten brands have successfully been relaunched(Brown, Kozinets, and Sherry 2003; Diamond et al. 2009).

Nostalgia is seen as a central concept for explainingconsumers’ demand for these past-related products (Belk1990; Davis 1979; Holbrook 1993; Kleine, Kleine, andAllen 1995; Loveland, Smeesters, and Mandel 2010;Wallendorf and Arnould 1988). Surprisingly, despite theobvious variety of nostalgic market appeals, mostresearch has an undifferentiated view on nostalgia(Holbrook 1993; Holbrook and Schindler 1994;Rindfleisch, Freeman, and Burroughs 2000; Routledgeet al. 2008; Wildschut et al. 2006). The myriad of past-related consumption, however, suggest that different kindsof nostalgic sentiments may be responsible for the diver-sity. It seems plausible that the liking of things with familytraditions may not be explained by the same nostalgicforce as the fondness for garments of the 1950s or the keeninterest in historic documentations. While research isbeginning to recognize that different types of nostalgiamay explain diverse portions of nostalgic behavior

(Goulding 2002; Stern 1992; Marchegiani and Phau 2010),no empirical study exists that investigates separate typesand links them with important consumer outcomes.

Advancing existing research, we offer a con-ceptualization of nostalgia that better reflects its multifac-eted nature and varying outcomes. We present personalnostalgia and historical nostalgia as two important typesof nostalgia. While the former has been subject to mostexisting research, investigation of historical nostalgiaeffects is a valuable addition to consumer research. Thepractical relevance stems from the notion that historicalnostalgia appeals are easier to evoke by advertising claimsand brand meaning. Personal connections can be addres-sed mainly at a generic nature (e.g., “Remember yourchildhood!”). We illustrate the benefit of this differenti-ated conceptualization of nostalgia by linking these twoimportant types with attitude toward past-related productsas well as purchase intention. Findings support our propo-sition that personal and historical nostalgia explain differ-ent parts of nostalgic consumption. Thus, practitionersshould target nostalgic market offers carefully.

In addition, the paper contributes to extant research ina methodological sense. We develop a new measurementinstrument to tap personal and historical nostalgia – thePersonal and Historical Nostalgia Inventory (PHNI) – andcompare the new instrument with the commonly usedNostalgia Proneness Scale (Holbrook 1993). Results showthat our conceptualization and measurement inventorygains advantage in predicting preferences for past-relatedproducts and services. References are available uponrequest.

For further information contact:Steffen Jahn

Department of MarketingChemnitz University of Technology

Reichenhainer Strasse 2909126 Chemnitz

GermanyPhone: +49.371.531.35604



WHEN I CAN’T TOUCH: AN EXAMINATION OF THE ROLEOF PURCHASE INVOLVEMENT AND REGULATORYORIENTATION IN ONLINE SHOPPING DECISIONS

Atefeh Yazdanparast, University of EvansvilleNancy Spears, University of North Texas, Denton

SUMMARY

Individuals make online purchases with varying lev-els of involvement in the shopping situation. Involvementwith a purchase task leads one to search for more informa-tion and spend more time searching for the right selection(Clarke and Belk 1979). The importance of involvementin consumer behavior is especially triggered by its effecton motivation and information processing. If individualslack the motivation or the ability to engage in effortfulprocessing of a message, they will engage in less thought-ful processes, such as simple inferences based on periph-eral cues, whereas individuals who are highly involvedare more likely to process the information systematically(Petty, Cacioppo, and Schumann 1983).

In addition to involvement, individuals’ regulatoryorientation also impacts their information processing.Higgins (Higgins et al. 1996; Higgins, Shah, and Fried-man 1997) proposed regulatory focus (RF) theory anddistinguished two forms of regulatory orientation, namelypromotion and prevention focus. A promotion focus isconcerned with advancement, growth, aspiration, andaccomplishment. In contrast, a prevention focus is con-cerned with security, responsibilities, and safety (Higgins1997). These two distinct goals prompt people to selec-tively pay attention to and rely on information that helpsthem attain their goals (Aaker and Lee 2001). Dependingon their regulatory focus, individuals are motivated to usedifferent strategies (approach or avoidance) to achievetheir desired states or distance themselves from theirundesired states. In other words, regulatory focus influ-ences how individuals pursue a goal or situate themselvesin relation to a desirable or undesirable state. Thus, whenpurchasing a product with a promotion approach, the goalis to make an ideal choice that makes the person happy andhelps him/her achieve desired results. However, a preven-tion approach in making a purchase is more concernedwith assuring that the choice is not wrong.

Indeed, prior research demonstrates that both invol-vement and regulatory orientation have the potential toinfluence product preferences when purchase decisionsare weighed (e.g., Boesen-Mariani, Gomez, and Marie-Laure 2010; Wang, Eric T.G., and Cheng-Kiang 2009).However, much remains to be learned about the role ofinvolvement and regulatory orientation in purchase deci-

sions, particularly in an online scenario where informa-tion about key product features obtained through the senseof touch is not available. Peck and Childers (2003a)defined the need to touch products as a preference forextracting and utilizing information through the hapticsystem (Peck and Childers 2003a). High-NFT individualshave a high degree of preference for extracting andutilizing information through the hapitc system (Peck andChilders 2003a). For these consumers, haptic informationis chronically more salient and thus, they are more likelyto use haptic information for product evaluations. In fact,if a haptic element is presented, these individuals may bemore persuaded compared with the situations in which nohaptic element is accessible (Peck and Wiggins 2006).

Overall, the present investigation examines theinteracton between the need for touch (NFT), situationalinvolvement, and momentary regulatory orientation onpurchase intentions in the context of a potential Internetpurchase. A 2 (NFT: high vs. low) x 2 (Involvement: highvs. low) x 2 (Regulatory Orientation: Promotion vs.Prevention) between subjects factorial design was imple-mented in which purchase involvement and regulatoryorientation were manipulated and NFT was measuredwhile controlling for participants’ prior online shoppingexperience. The results of the ANCOVA revealed that the3-way interaction suggested by the study design wassignificant. To further investigate the relationships, two-way interactions between purchase involvement and regu-latory orientation were assessed for high- and low-NFTindividuals. For high-NFT individuals, there was a sig-nificant two-way interaction between involvement andregulatory focus. Specifically, under a prevention focus,high-NFT individuals with higher levels of purchaseinvolvement had more purchase intentions than did thosewith low purchase involvement. On the other hand, high-NFT individuals with a promotion focus had higher pur-chase intentions in a low purchase involvement than in ahigh purchase involvement situation. In contrast to high-NFT individuals, there was no significant two-way inter-action between involvement and regulatory focus for low-NFT participants. Interestingly, according to the results,for high-NFT individuals, a prevention focus under highpurchase involvement situations can lead to higher pur-chase intentions, while a promotion focus works bestunder low purchase involvement for these individuals.Moreover, when the effect of involvement and regulatory


focus are examined together for low-NFT individuals,consistent with previous research (Wang and Lee 2006),regulatory focus does not have a significant influence onpurchase intentions under different levels of purchaseinvolvement. However, regulatory focus has a significantimpact on these relationships when individuals are high inNFT.

As Martin et al. (2011) argued, academic research hasplaced excessive emphasis on the study of general con-sumer responses and attitudes toward online shopping,without taking into account individual and situationalfactors that affect consumers’ online shopping behavior.The present research contributes to this call for researchby examining NFT (an individual factor) along withsituational involvement, and momentary goal orientation(both situational factors) and assessing their influence on

individuals’ online shopping decisions. The results haveimplications for electronic retailers. They should focus onthe format and types of product information they provideon their websites and match them with the way high versuslow involvement individuals seek information. Thiscustomization could be achieved by categorizing prod-ucts into high and low purchase involvement products.Even depending on the shopping occasions, productscould be grouped as high or low involvement. Customersmay be asked questions such as “for what occasion are youshopping today?” The pre-determined options could be“for myself, for a gift to a significant other, for donation,etc.” Moreover, electronic retailers can situationallyinfluence customers’ regulatory orientation by presentingrelevant phrases and promotional messages. Referencesare available upon request.

For further information contact:Atefeh Yazdanparast

Schroeder Family School of Business AdministrationUniversity of Evansville

1800 Lincoln AvenueEvansville, IN 47722Phone: 940.594.9384



DIFFERENCES IN THE INFLUENCE OF CHOICE CONFIDENCE ANDOUTCOME QUALITY ON SATISFACTION AS A FUNCTION OF

INFORMATION DIAGNOSTICITY

Demetra Andrews, Florida State University, TallahasseeAlexis M. Allen, Florida State University, Tallahassee

Edward Blair, University of Houston, Houston

SUMMARY

People want to feel good about their choices. Theywant to feel confident. When they do, choice confidence,or the extent to which a choice is believed to be correct(Heitmann, Lehmann, and Herrmann 2007; Tsai andMcGill 2011), can yield desirable consequences such asreductions in purchase delays (Greenleaf and Lehmann1995), increased consistency between attitude and behav-ior (Bearden, Hardesty, and Rose 2001), and potentially,higher satisfaction (Heitmann et al. 2007).

The present research was motivated by the possibilitythat choice confidence may offer a conduit via whichmarketing communications can facilitate desirable con-sumer reactions such as consumption satisfaction.Because it is readily influenced by the nature and contentof presented information (Farley, Katz, and Lehmann1978), choice confidence may offer a conduit via whichmarketing communications can be converted into desir-able consumer reactions such that enhancing the formerwill increase the latter. However, a deeper and morecomprehensive understanding of the effectiveness ofexternal tactics to increase choice confidence is neededbefore such actions can be fully recommended.

Choice confidence can be increased by informationthat is more diagnostic or useful in making (Lynch,Marmorstein, and Weigold 1988) or justifying (Feldmanand Lynch 1988) a choice. Presentation of highly diag-nostic information, such as cues indicating the superioralternative in a choice set, increases choice confidence(Tsai and McGill 2011; Yoon and Simonson 2008).However, findings from two experimental studies dem-onstrated that this type of confidence manipulation alsoweakens the correlation between choice confidence andsatisfaction that was observed in prior research (seeHeitmann et al., 2007).

In this research, information diagnosticity was ma-nipulated by providing attribute-level ratings of threeavailable products that either did (highly diagnostic) ordid not (non-diagnostic) indicate which of the three wasthe dominant option. Participants made a selection andstated their confidence in the correctness of their choice.Following receipt of either a better or worse quality

outcome, participants indicated their satisfaction with theoutcome.

Study findings demonstrated that high choice confi-dence weakens the influence of outcome quality on theconsumer’s experience of satisfaction. Contrary to intu-ition, however, this effect did not manifest when choiceconfidence is enhanced by strong external cues, but ratherwhen the information on which the confidence was basedoffered little indication as to the correct, or justifiable,choice. Moreover, when information diagnosticity waslow, consumers with high confidence are able to deriveequivalent satisfaction better or worse quality outcomes.While a weaker influence of external information (such asfeedback on outcome quality) may be inferred from priorliterature (see Sieck and Arkes 2005), the present researchmakes a unique contribution to literature by providingevidence that this effect varies as a function of thediagnosticity of the information basis of confidence. Thus,it enhances knowledge about the post-choice influence ofchoice confidence and gives a view to possible conse-quences of attempts to raise choice confidence via alter-ations in diagnosticity of marketing communications.

Interestingly, when consumers derived high confi-dence from information that was low in diagnosticity,they appear to underweight or disregard cues indicatinglow outcome quality (i.e., three bad pictures out of six).Such resistance may be indicative of biased informationprocessing toward information favoring prior beliefs (Lordet al. 1979) or a form of confirmatory hypothesis testing(Klayman and Ha 1987). Because realized outcomes areoften multi-valent in nature, i.e., containing both positiveand negative elements, evaluation of outcomes is open tosubjectivity.

The findings of this research suggest that confidencethat feels more intuitive may promote biased informationprocessing so that the consumer experiences high levelsof satisfaction even from imperfect outcomes. However,heavy-handed tactics to externally enhance confidencemay completely undermine this “protective” effect andeven increase the consumers’ scrutiny of the outcome. Inshort, the findings presented demonstrate that, while it ispossible to externally enhance consumer choice confi-dence, at least one overt tactic is ineffective at bringingabout in increase in satisfaction. References are availableupon request.


For further information contact:Demetra Andrews

Marketing DepartmentThe College of BusinessFlorida State University

P.O. Box 3061110Tallahassee, FL 32306–1110



SELF-EFFICACY AND SELF-PROPHECY EFFECTS ON PREVENTIVEHEALTH BEHAVIOR

Carmen-Maria Albrecht, University of Mannheim, GermanyHans H. Bauer, University of Mannheim, Germany

Kai Bergner, University of Mannheim, GermanyTamara Gogia, University of Mannheim, Germany

Daniel Heinrich, University of Mannheim, GermanyDavid E. Sprott, Washington State University, Pullman

SUMMARY

One key objective of marketing research in the fieldof health care is to develop mechanisms to promotebehaviors and lifestyles of people that maximize theirlong term health. In this regard, the main task is to increasepeople’s adherence to different prevention- and treat-ment-related regimens (Kazarian and Evans 2001). Dif-ferent approaches have been developed to help to influ-ence health behaviors such as the theory of plannedbehavior (Fishbein and Ajzen 1975; Ajzen 1991), thehealth belief model (Janz and Becker 1974; Becker 1974),and the protection motivation theory (Maddux and Rogers1983; Rogers 1983), to name only a few. However, whenhealth care professionals try to implement these frame-works in practice, they face the challenge that theseframeworks are rather complex and not easily applicable.In addition, these frameworks often fail to provide prac-titioners with specific procedures on how to alter healthbehaviors (Sprott et al. 2006).

Self-prophecy, however, can be regarded as a simpleway to influence health-related behaviors and can easilybe transferred to practice (Sprott et al. 2006). The self-prophecy effect suggests that asking people to predictwhether they will perform a socially normative behaviorleads to the increased likelihood of them performing thatbehavior.

While self-prophecy research demonstrated its theo-retical explanation to be based on the theory of cognitivedissonance (Spangenberg et al. 2003) and found somemoderators of the self-prediction effect, it still requires

identification of further conditions under which self-prediction is most effective (Sprott et al. 2006).

Since self-efficacy has generally been shown to influ-ence health behaviors (e.g., Jayanti and Burns 1998;Keller 2006), self-efficacy is such a potential factor.Except for one study (Longstreet et al. 2011) which testedthe self-prediction effect on computer-related task perfor-mance, the question of interaction between self-prophecyand self-efficacy has not yet been empirically studied inprior research.

The current research investigates self-prophecy inthe health behavior context to accomplish two objectives:(1) to provide evidence that self-prophecy has a signifi-cant effect on performance of a preventive health behav-ior and (2) to test whether self-efficacy moderates the self-prophecy effect on the focal behavior.

The results of an experiment indicate that self-proph-ecy has an effect on preventive health behavior, thusproviding further evidence that making a self-predictionchanges people’s behavior. Moreover, we are able toshow that this effect is moderated by a person’s level ofself-efficacy. In particular, people low in self-efficacy aremore likely to respond to the effect of self-prediction thanare people high in self-efficacy. This outcome suggeststhat the self-prophecy question can be particularly helpfulfor people with lower self-efficacy beliefs in bringingtheir level of exercising, for example, to the same level aspeople with higher self-efficacy beliefs who actually donot need a self-prediction question. References are avail-able upon request.

For further information contact:Carmen-Maria AlbrechtUniversity of Mannheim

L 5, 1Mannheim, 68131




THE ROLE OF GOAL SPECIFICITY IN INITIAL AND SUBSEQUENTDECISIONS REGARDING CONSUMPTION OF HEALTHY

OR UNHEALTHY FOODS

Meredith E. David, University of South Carolina, ColumbiaWilliam O. Bearden, University of South Carolina, Columbia

SUMMARY

The present research examines whether goal speci-ficity can assist in explaining why consumers, especiallydieters, so often fail to meet their goals. Goal specificity,defined as the “ambiguity or diffuseness in the exact levelof performance desired” (Wright and Kacmar 1994,p. 243), can differ among individual goals. For example,goals can be framed as “specific” goals which are restric-tive and have a concrete end point (e.g., lose 20 pounds insix weeks), or as “do-your-best” goals which are vague,less restrictive, and lack a set completion point (e.g., loseweight) (Locke and Latham 2002). A specific goal is “theobject or aim of an action, for example, to attain a specificstandard of proficiency, usually within a specified timelimit” (Locke and Latham 2002, p. 705). Do-your-bestgoals, on the other hand, still call for action, but, are morevaguely defined without a specific level of performance,and, often without a specified time limit.

Extant research on goal specificity offers mixedfindings, suggesting that, in fostering performance, spe-cific goals can be both more or less effective than do-your-best goals (Carter, Patterson, and Quasebarth 1979;Hollenbeck and Kleine 1987; Kirschenbaum, Humphrey,and Malett 1981; Ulkumen and Cheema 2011). To extendthis literature, we investigate how goal specificity impactsdieters’ goal activation (hedonic vs. health) and theirinitial food decisions after being faced with a food temp-tation. In addition, since consumption decisions oftenfollow one another, it is important to understand howpurchasing a healthy or unhealthy product impacts subse-quent decisions (Dewitte, Bruyneel, and Geyskens 2009;Laran and Janiszewski 2009; Soman and Cheema 2004),so we also examine the role of goal specificity in dieters’sequential decisions for healthy versus unhealthy prod-ucts. To our knowledge, the present research is the first toexamine the role of goal specificity in sequential decisionsand the role of goal specificity in the relationship betweena temptation and goal activation.

Our study suggests that, when setting personal goals,consumers should consider the level of specificity of thegoals they set. Specifically, we find that, facing a foodtemptation is likely to activate hedonic goals, as demon-strated in the extant literature (e.g., Fedoroff et al. 2003;Harvey et al. 2005; Stroebe 2008; Stroebe et al. 2008), but

this effect is stronger for dieters with specific goals.Interestingly, however, we find that, although when facedwith a temptation, dieters with specific goals are muchmore likely to have an activation of hedonic goals, dieterswith specific goals are indeed also more likely to resistthose hedonic thoughts and, ultimately, make a decisionthat is congruent with their dieting goal. Thus, our find-ings assist in extending two goal-related theories, theTheory of Temptation-Elicited Goal Activation (Fishbachet al. 2003) and Goal Conflict Theory (Stroebe 2008) byshowing that dieters with do-your-best goals are morelikely than dieters with specific goals to give into thehedonic thoughts stemming from a food temptation andorder an unhealthy food item.

We extended our analyses beyond initial decisions toinvestigate how goal specificity may interact with one’sinitial decision, be it goal congruent or incongruent, todetermine the likelihood of one’s subsequent decisionbeing goal congruent. In tests for subsequent decisionmaking, our findings indicate, once again, that specificgoals are more effective than do-your-best goals. Amongconsumers with do-your-best goals, we find support forthe “what the hell effect” (Herman and Polivy 1984),which refers to times when consumers fail in their self-control, consume an item not consistent with their goal,and, instead of immediately attempting to get back ontrack to reaching the goal, they feel as if the goal is a lostcause and thus continue with the lack of self-control(Herman and Polivy 1980). While the extant literature hasfound support for this “what the hell” effect and suggeststhat consumers prefer to engage in highlighting ratherthan balancing when it comes to food consumption goals(Soman and Cheema 2004), other research shows that,when faced with two goals (i.e., to eat a tasty food and tonot to eat an unhealthy food), consumers prefer to balancethe two goals rather than highlight one goal in a consump-tion episode (Dhar and Simonson 1999). Our findingsassist in explaining these contradictory conclusions byshowing that, the “what the hell” effect is likely to existamong consumers with a do-your-best goal, as theseconsumers prefer to highlight rather than balance goals.

Our findings show that, in a consumption episode,dieters with a do-your-best goal either prefer to either goall out and eat unhealthy (highlighting hedonic goal), or,eat a completely healthy meal (highlighting health goal).


Dieters with a specific goal, on the other hand, do notnecessarily have a preference for balancing or highlight-ing of goals. Instead, we find that dieters with specificgoals are equally as likely to select the subsequent un-healthy item, irrespective of their initial decision. That is,dieters with specific goals either prefer to highlight theirdieting goal and eat healthily throughout a consumptionepisode, or, make it a point to get back on track to finisha consumption episode on a positive note, with a goalcongruent decision.

In summary, we find that specific dieting goals resultin more goal congruent decisions than do-your-best goals.Although consumers with specific goals are more likelythan consumers with do-your-best goals to have a hedonicgoal activation when faced with a temptation, consumerswith specific goals are better able to resist such hedonicthoughts, and, ultimately make a goal congruent decision.In terms of subsequent decisions, following a goal incon-gruent decision, dieters with specific goals are more likelyto get back on track and purchase a healthy item, than are

dieters with do-your-best goals. The only instance inwhich do-your-best goals outperform specific goals is inmaking a subsequent decision, following a goal congruentdecision. In this particular situation, dieters with specificgoals are more likely than dieters with do-your-best goalsto select a subsequent unhealthy item. However, ourfindings suggest that, dieters with specific goals wouldlikely get back on track to meeting their dieting goal.

Future research could examine other variables thatimpact the success and performance of specific versus do-your-best goals. For example, specific goals may be moreeffective when construed at a higher level, and nonspe-cific goals may be more effective when construed at alower level (Ulkumen and Cheema 2011). The interactionbetween construal level and goal specificity may indeedalso impact the relationship between a temptation andgoal activation. In addition, goals vary in their levels ofspecificity, so future research could examine conditionsin which goals with different levels of specificity are moreor less effective. References are available upon request.

For further information contact:Meredith E. David

University of South Carolina1705 College StreetColumbia, SC 29208Phone: 803.413.6899



CHECKOUT FRUSTRATION: INVESTIGATING CONSUMERREACTIONS USING FRUSTRATION THEORY

Eric Van Steenburg, University of North Texas, DentonNancy Spears, University of North Texas, Denton

Robert O. Fabrize, California State Polytechnic University, Pomona

SUMMARY

Over the years, a scholarly stream of research hasinvestigated the interplay between goal pursuit, affectiveprocesses, and problem-solving processes in terms offrustration (e.g., Amsel 1958, 1962; Freud 1958; Meyer1956; Meyer and Ellen 1959; Shorkey and Crocker 1981).Despite the insights gained, questions remain about thenature of goal pursuit and the frustration processes whengoals are blocked. It is this area of research that presentsopportunities for marketers seeking theoretical and prac-tical means with which to address frustration experiencedin a retail checkout environment.

Research

Previous studies identified two response patterns in afrustrating circumstance: (1) an adaptive response thatworks through a facilitating process to attain the desiredgoal; and (2) a maladaptive response that works througha debilitative process that impedes successful goal attain-ment (Alpert and Haber 1960; Butterfield 1964). Thefrustrating situation, in conjunction with individual psy-chological characteristics, determines resulting behavior(Freud 1958).

Shorkey and Crocker (1981) defined three frustra-tion-elicited adaptive response strategies: (1) overcomethe obstacle; (2) circumvent the obstacle; and (3) avoid theobstacle. Individuals choosing adaptive response strate-gies exhibit a facilitating process aimed at problem-solving to address the frustration object. A maladaptiveresponse pattern works through a more affect-laden ap-proach aimed at addressing the stress, leading to any offour response strategies: (1) aggression toward the ob-stacle; (2) regression back to a less mature behavior; (3)fixation or repetitive behavior; and (4) resignation leadingto inertia or apathy (Shorkey and Crocker 1981).

The present research expands the current understand-ing of frustration response patterns by investigating therole it plays as an underlying process mechanism whengoals are blocked, and investigates the sequence follow-ing a frustrating event in a retail checkout environment.

Because frustration responses are associated with thesource of blame for the frustration event (Rosenzweig

1934; Shorkey and Crocker 1981), this should apply toboth adaptive and maladaptive responses. Moreover,maladaptive response strategies follow a debilitative pro-cess and hinder goal pursuit (Alpert and Haber 1960),while negative attainment of goals decreases desire forachievement and any associated behavior related to goalattainment (Weiner 1986). In addition, frustration toler-ance is the acceptance of the undesirable in an effort toachieve future goals, while intolerance is an individual’sattempt to make reality fit desire (Harrington 2007).Because individual factors influence the frustration pro-cess (Smith and Lazarus 1990), the ability to tolerate, ornot tolerate, frustrating events is related to one’s tolerancelevels. Therefore:

H1: Adaptive and maladaptive frustration response strat-egies are associated with the source of blame (exter-nal or internal).

H2: In the case of a maladaptive response, newly createdblocks are associated with the source of blame (exter-nal or internal).

H3: Individual differences in frustration tolerance/intol-erance will predict felt frustration in a checkoutsituation.

Results

Cognitive responses examined by a panel of judgesshowed that 71.8 percent (n = 79) of the study’s partici-pants chose an adaptive response to the frustrating situa-tion, with the remaining (n = 31) choosing a maladaptiveresponse. Coders followed the definitions provided byShorkey and Croker (1981) and were able to identify allthree adaptive response strategies adopted by respon-dents. In each case, respondents returned to goal-seekingbehavior, thus supporting previous research (Amsel 1958,1962; Butterfield 1964; Shorkey and Crocker 1981).

Results of coding for the maladaptive process re-vealed that a resignation strategy was used 100 percent ofthe time, supporting the claim that resignation is the mostcommon maladaptive response (Shorkey and Croker1981). Results also revealed the presence of three newlycreated blocks toward achieving the checkout goal: (1)helplessness; (2) anger (including impatience), and (3)


self-preoccupation (includes self-presentation, self-pres-ervation, self-recrimination, and self-advancement).

A chi-square test of association between adaptiveresponse strategies and source of blame was significant(χ2 = 22.11, df = 2, p = .00) providing support for H1.(Maladaptive responses were not tested because only onewas found.) A chi-square test to examine external vs.internal blame against each type of newly identifiedsecondary blocks (helplessness, anger, and self-preoccu-pation) was significant (χ2 = 6.68, df = 2, p = .04),supporting H2. Finally, a regression analysis to determineif individual differences in frustration tolerance predictfelt frustration in a retail checkout context was significant(F = 11.90, t = 3.45, p < .001), supporting H3.

Discussion

This study confirmed that consumers choose eitheran adaptive or maladaptive response to frustration, fol-lowed by an initial resolution strategy in an effort to returnto goal-seeking behavior. The results also suggest thatfrustration responses are associated with source of blame,with anger most often associated when the self is to blame,and self-preoccupation most when often when the store isto blame. Initial evidence suggested that individual differ-ences in frustration tolerance are related to felt frustrationin a frustrating checkout scenario. Findings also con-firmed the presence of all three adaptive resolution strat-egies. References are available upon request.

For further information contact:Eric Van SteenburgCollege of Business

University of North Texas1155 Union Circle #311160

Denton, TX 76203–5017Phone: 214.334.4813



ANTECEDENTS AND CONSEQUENCES OF SITUATED LEARNINGIN STRESSFUL SERVICE EXPERIENCES:

A CROSS-CULTURAL PERSPECTIVE

Mousumi Bose, Fairfield UniversityLei Ye, Frostburg State University

SUMMARY

As consumers, we are constantly learning from ourenvironment. However, not all learning experiences maybe positive and fruitful; some may be stressful because ofproduct or service failures, unsatisfactory or unexpectedexperiences or because of the nature of the service expe-rience. Theories on problem solving and decision-makingrely on mechanisms of learning and adaptation and assuch, stressful service encounters may call for problemsolving and learning. Learning in the context of theservice experience helps consumers’ better handle thesituation. Thus, the goal of this research is to explore theconcept of situated learning or in situ learning in stressfulconsumption experiences.

An essential aspect of situated learning relates to itsdynamic nature; it is not only based on our existingschema but also derived from the situations that we aresurrounded by, which can be conceived as “a continuingreconstruction of experience” (Dewey 1897, p. 79).Nidumolu et al. (2001) argue that situated, as opposed tostatic learning, is an emergent form and is an ongoingimprovisation enacted by consumers trying to make senseof and acting judiciously in dealing with stressful con-sumption experiences. A static view of learning assumesa decontextualized and simplified environment that isrigid and incomplete. By taking into considerations envi-ronmental transformations, one appreciates the more dyna-mic “how” or the process view of learning compared to themore limited “outcomes only” view of learning. Based onthe theories of situated cognition, social learning andmental models (Goel et al. 2010), one can suggest that theinterplay with the environment and other individuals, andthe role of individual cognition play an important role insituated learning. Such interplay is pivotal in mediatingthe ability of individuals to cope with stressful serviceepisodes.

In all, the goal of this research is to highlight theimportance of situated learning in stressful serviceencounters and understand the antecedents and conse-quence of such learning. Past research has discussedsituated learning from the viewpoints of organizations,web-based situated learning and learning in service

encounters. To the best of our knowledge, there is littleresearch that has considered the situated aspect of con-sumer learning in the context of stressful service encoun-ters and studied its antecedents and consequences. Asecond goal is to understand situated learning in stressfulservice encounters in two different cultural contexts: thatof U.S.A. and China.

Study 1

To understand the situated experiences of consumersdealing with stressful services, two studies were con-ducted. Since little research exists regarding situatedlearning and its factors with respect to stressful serviceencounters, we embarked on the first study via a qualita-tive method of enquiry. Emergent themes from semi-structured, in-depth interviews conducted in the U.S. andChina formed the basis for survey-based data collectionconducted in Study 2. In Study 1, interviews were con-ducted by using snowballing technique. Individuals fac-ing stressful service experiences were interviewed; inter-views were recorded, transcribed and analyzed by con-stant comparison method.

Participants revealed the importance of “short-term”and “long-term” situated learning that were used to dealwith current and future stressful service encounters. Justas they discussed the process of in situ learning, so toothey deliberated on the emotional aspects of their learn-ing; learning to stay calm was a common method used tocombat stressful situations. Past research has also alludedto such emotional learning in the face of stressful cancerdiagnosis (Pavia and Mason 2004). Thus, such emotionallearning in combination with cognitive learning is impor-tant to better cope with stressful services. Several anteced-ents that enhanced situated learning emerged: need forclosure (desire for a definite answer to a question), psy-chological closeness (staying close to the problem inquestion) and positive mindset. Additionally, the U.S.participants suggested that the impetus to provide upfrontfeedback and trust on the service provider enhanced theirsituated learning. In contrast, the Chinese participantsdiscussed about uncertainty avoidance as a motivator ofsituated learning. As a consequence of situated learning,all participants alluded to enhanced coping as a way todeal with stress.


Study 2

The goal of this study was to determine whethersituated learning mediates the relationship between itsantecedents and its consequence. Another goal was toexplore whether significant differences exist in situatedlearning and its antecedents in the two cultural contexts.To this end, an online survey of nonstudent customersfrom U.S. and China was used. ANOVA and regressionwere used to analyze the results. Results from an ANOVAanalysis showed that U.S. and Chinese consumers dif-fered in the process of situated learning with Chineseconsumers more inclined to emphasize on the processthan their U.S. counterparts.

Results from hierarchical regression revealed thatpositive mindset was significant for both cultures, needfor closure was significant for China but not for U.S. andpsychological closeness was significant for U.S. but notfor China. For U.S. consumers, greater trust on the serviceprovider and upfront evaluation positively influencedsituated learning while uncertainty avoidance helpedChinese consumers enhance situated learning. The dataalso revealed that situated learning positively influencedcoping. Finally, Baron and Kenny’s (1986) mediationanalysis demonstrated that situated learning mediates theeffects of positive mindset, psychological closeness andtrust on coping for U.S. consumers while it mediates theeffect of positive mindset only in case of Chinese consum-ers. References are available upon request.

For further information contact:Mousumi Bose

Charles F. Dolan School of BusinessFairfield University

1073 North Benson RoadFairfield, CT 06824

Phone: 203.254.4000, Ext. 2828E-Mail: [email protected]


THE PURSUIT OF EXTRAORDINARY EXPERIENCES:AN EXPLORATION OF TRIGGERS

Colleen Harmeling, Saint Louis UniversityMark Arnold, Saint Louis University

SUMMARY

Extraordinary consumption experiences have theability to transform and define our identities. Along thispremise, consumers, knowingly or unknowingly, pursueextraordinary consumption experiences that motivate tran-scendence to a desired possible self. In doing so, consum-ers seek or create certain triggers of extraordinary experi-ences. Evidence from related research streams is pre-sented to support this assertion and is used to develop fivedimensions of triggers. These dimensions are furtherdeveloped through a one and a half year qualitativeresearch study in the context of recreational road races.Research has explored the effects of extraordinary expe-riences on consumer behavior, however, a systematicexploration of the pursuit of these experiences and theirantecedents is lacking in the consumer behavior literature.Thus, the purpose of this research is to examine howconsumers, intentionally or unintentionally, pursue extra-ordinary consumption experiences through the construc-tion or use of triggers. The following is a summary of ourresearch design and findings.

Extraordinary experiences are “powerful, positivelyvalenced emotional experiences that create personal mean-ing and offer the possibility of transformation of the self”(Arnould and Price 1993). Due to the complexity ofextraordinary experiences as well as the desire to identifytriggers independent of the consumption activity, thisresearch began with an analysis of the existing studies ofextraordinary experiences published in peer-reviewedjournals. This method allowed for the examination of awide variety of consumption activities (e.g., Jeep Jambo-rees, white water rafting, sky diving) and promoted theidentification of triggers that spanned varied consumptioncontexts. From this initial analysis, a tentative frameworkof triggers of extraordinary experiences was established,which guided the resulting qualitative study. A multiplecase qualitative field study was conducted in the contextof recreational road races. The experience served as theunit of analysis. In total, 19 informants provided accountsof their experiences within five different contexts (i.e.,different road races). Informants as well as contexts werepurposively selected to maximize the number of contex-tual factors as well as the number of individual factors thatcould influence each participant’s experience and to al-low some degree of comparison between experiences.The definition of extraordinary experiences discussed

above is comprised of two main elements, (1) extremepositive emotions, and (2) self-transformation. This crite-ria was used to distinguish extraordinary experiencesfrom mundane experiences and was operationalizedthrough semi-guided interview questions as well as infor-mants’ self-categorization of the experience. This catego-rization promoted comparison between mundane andextraordinary experiences and further illuminated thedimensions of triggers.

Ultimately, five dimensions of triggers were identi-fied in the data: agon, aleatory, ilinx, mimetics, and gaia.Agonistic triggers are based on competition and motivateindividuals to seek self-actualization by encouraging themto challenge their ability (Caillois 1961). A criticallyimportant assumption of agonistic triggers is the equalityof competitors. Aleatory triggers, unlike agonistic trig-gers, make no effort to establish equality of adversariesand are defined by chance. Ilinx triggers are a feeling ofvertigo and are defined by disorder and destruction ofreality. Mimetic triggers allow individuals to create asense of oneness with the group or object through imita-tion or fantasy, but at the same time provide an escapefrom the everyday (Caillois and Halperin 1955). Gaia isthe belief in an interconnectedness of all beings and allobjects on Earth (Lovelock 1982). Gaia triggers differfrom mimetic triggers, where individuals feel connectedto the group, in that Gaia triggers provide a means fortranscendence through the sudden acknowledgment of ahigher power, an awareness of the universe, and theindividual’s connection to it (Maslow 1964).

This research has important implications for boththeory and practice. From a theoretical standpoint, thisresearch increases our understanding of how consumerspursue extraordinary experiences and how this pursuitinfluences consumer behavior. This research also pre-sented a comparative analysis between characteristics ofmundane experiences and extraordinary experiences andprovided preliminary evidence of the relationshipbetween triggers and extraordinary experiences and trig-gers and the individual.

This research has important implications for market-ing practice in at least two areas, (1) imagery in advertis-ing, and (2) service delivery. First, current consumerresearch examining the effects of imagery in advertisingsuggest that certain imagery can create false experiences,


or “the mistaken belief that an event that did not occur didoccur (Rajagopal and Montgomery 2011).” These falseexperiences influence consumer behavior in the same wayas true experiences. Building off of this, future researchcould test the effective use of triggers in advertising instimulating “false extraordinary experience.” Second,

understanding what influences extraordinary experiencescan dramatically improve how service providers createthese experiences. This research hopes to stimulate fur-ther inquiry into the role of extraordinary experiences inconsumer behavior. References are available uponrequest.

For further information contact:Colleen Harmeling

Saint Louis University3674 Lindell Blvd., CK 30

St. Louis, MO 63108Phone: 314.977.3810



CONSUMER GOAL DYNAMICS: AN EMPIRICAL ILLUSTRATION

Shilpa Iyanna, Abu Dhabi University, United Arab Emirates

SUMMARY

A body of extensive literature has compellingly em-phasized that goals play an important role in influencingand directing consumer behaviour. However, Baumgartnerand Pieters (2008) posit that systematic research on goaldirected behaviour is still lacking. Thus, the purpose ofthis paper is to expand on previous studies and take thediscussion a step further by providing empirical evidencethat (1) illustrate how goals at different levels impact oneach other as consumers’ progress through the variousstages of consumption; and (2) examines the interactionand integration of resources in goal formation. A quasi-longitudinal qualitative study involving semi-structuredinterviews was employed in this study.

The results suggest a shifting pattern in the goals asconsumers’ progress in their relationship with the serviceprovider. The analysis revealed that in the pre-consump-tion stage, consumer behaviour was guided by whatBagozzi and Dholakia (1999, p. 23) refer to as focal goals:“What do I want to achieve?” The results from theconsumption stage suggest that consumer behaviour wasguided by subordinate goals. Such goals answer thequestion “How can I achieve what I want to achieve?”

(Bagozzi and Dholakia 1999). Overall, most of the re-spondents in the post-consumption stage spoke in terms ofsuper-ordinate goal attainment that are at the top of thegoal hierarchy and answer the question: “Why do I wantto achieve what I want to achieve?”

With reference to resource integration in goal forma-tions, results indicate that, in the pre- consumption stage,focal goals were mostly co-created through the integra-tion of consumer resources and external resources. In theconsumption stage, subordinate goals were mainly formedor rationalised through focal goals. In addition, subordi-nate goals are also influenced by consumer physicalresources (intellectual capacity, personal interests andambition), consumer social resources (co-consumers) andexternal resources (economic conditions, prospectiveemployers and competition). Findings from the post-consumption stage provide empirical support to the claimthat attainment of super-ordinate goals is made possiblethrough the attainment of subordinate goals (Bagozzi andDholakia 1999).

The paper concludes with a discussion on the impli-cations of these results for theory, practice and futureresearch. References are available upon request.

For further information contact:Shilpa Iyanna

College of Business AdministrationAbu Dhabi University

Abu Dhabi, United Arab EmiratesPhone: +971.2.5015664

Fax: +971.2.5860184E-Mail: [email protected]


CSR-RELATED COMMUNICATION IN DIFFERENT INDUSTRIES:A QUALITATIVE AND QUANTITATIVE STUDY BASED ON

CORPORATE ANNUAL REPORTS

Thomas Kilian, University of Koblenz-Landau, GermanyNadine Hennigs, Leibniz University Hannover, Germany

SUMMARY

Introduction and Background

Along with the public’s increased demand for busi-nesses to operate responsibly, an increasing number ofcompanies proactively publish their CSR-related prin-ciples and activities. Legitimacy theory argues that com-panies engaged in industries that are more likely to affectthe environment are particularly more inclined to commu-nicate their CSR status. Our research aims are twofold:First, based on a content analysis of annual reports of allGerman DAX-30 companies from 1998 to 2009, maincategories of CSR-related communication are defined.Second, these categories are used in a quantitative analy-sis to compare the CSR-related communication in differ-ent industries. This study uses a mixed qualitative-quan-titative approach. Based on the annual reports from theGerman DAX-30 companies from 1998 to 2009, weconducted a content analysis to code CSR communica-tion. In the second step, based on the coding scheme, weanalyzed the data to compare CSR communication indifferent industries.

Data Analysis and Key Findings

In our qualitative study, we derived a category systemthat not only accounts for CSR-related activities but alsofor CSR philosophies and motives as the normative basisof CSR communication. Thus, this study deviates fromsimilar research because we account not only for CSRactivities but also CSR-related philosophy and motives.Both variables form the normative basis that encouragessocially responsible behavior and its reporting (Chen andBouvain 2009). In companies that act based on the prin-ciple of “doing good,” this behavior should spread through-out the organization and thus provide a basis for CSRpractices among both managers and employees (Jones1995). Therefore, we believe that CSR communication inannual reports should be analyzed within the normativeframework of the underlying corporate philosophicalprinciples and their related motives.

Based on a sample comprising the annual reports ofall German DAX-30 companies from 1998 to 2009, in ourquantitative analysis, we specifically concentrated on thequestion of whether companies in industries that are morelikely to affect the environment communicate more inten-

sely than companies in industries that operate withoutimmediate environmental risk. The quantitative analysisby different industries gives evidence for the assumptionthat an industry sector which is regarded to be morecontroversial, such as chemicals and pharmaceuticals,reports more CSR-related activities than other industries.In this respect, the main research question posed in theintroduction can be answered affirmatively: Companiesin different industries report CSR-related activities differ-ently. This finding supports the idea from legitimacytheory that companies in industries that have higherenvironmental and/or social impacts than others commu-nicate more environmental and/or social information toproactively comply with the higher expectations of stake-holders.

Recently, voluntary social and environmental report-ing has been subject to criticism from scholars who havecharacterized such efforts as “often partial, incomplete,and self-serving public relations exercises that seek orga-nizational legitimacy through appearance rather thanchanged behavior” (Kuruppu and Milne 2010), or, moresimply put, “greenwashing” (Greer and Bruno 1996). Ourresults cannot entirely dispel these ideas but do supportlegitimacy theory, which argues that companies in indus-tries such as energy, chemicals, and pharmaceuticals,perform more CSR communication, in line with the public’sincreased demand for these businesses to operate respon-sibly. This argument certainly does not prevent compa-nies in these industries from wrong behavior in the futurebut puts them under more pressure to keep their promises.However, CSR communication can also be seen as amanagement strategy to prevent the introduction of rigidregulation and to distract from factually poor social orenvironmental performance (Moerman and Van der Laar2005).

Limitations and Research Implications

This research is based on qualitative data, and it canbe argued that content analysis generally lacks reliabilityand validity as an interpretative means of analysis. How-ever, as Deegan and Gordon (1996) have noted, thesedifficulties are more significant when a rather small sampleof documents is examined. In our study, we examined areasonably large sample and therefore conclude that theoverall measurement error is negligible. In summary, with


a combination of qualitative and quantitative analyses,this study builds upon and extends past research efforts inthe field of CSR and CSR reporting, as it provides deeperinsights into qualitative and quantifiable effects encom-passing a rich category scheme that also accounts for thestated CSR-related philosophy and motives to engage inCSR related to different fields of CSR activities. Althoughour results are only initial empirical hints, they should beexplored in further research in different ways. Because,apart from the industry sector, the national culture of the

reporting company is considered a determinant of CSRreporting (e.g., Aerts et al. 2008; van der Laan Smith et al.2005), a study comparing different industries in differentcountries may lead to interesting results. Additionally, inour analyses, we did not separate the home-market activi-ties from the international activities of companies indifferent industries; given their multinational activities,there may be differences in the CSR commitments ofcompanies with international markets or production sites.References are available upon request.

For further information contact:Thomas Kilian

Institute of ManagementUniversity of Koblenz-Landau

Universitaetsstrasse 156070 Koblenz

GermanyPhone: (+49)511.762.4862Fax: (+49) 511.762.3142



DOES CORPORATE SOCIAL RESPONSIBILITY SAVE FIRMS? ANEXPLORATION OF CORPORATE SOCIAL RESPONSIBILITY,

FIRM CAPABILITY, ENVIRONMENTAL INFLUENCES,AND FIRM DEFAULT RISK

Wenbin Sun, Rockhurst University, Kansas City

SUMMARY

Corporate social responsibility (CSR) receives a grow-ing attention from both academic researchers and busi-ness managers. Prior research has confirmed that CSR, byits ability of building strong corporate image and reputa-tion, effectively improves firm’s performance. However,in the resource-based framework (RBT) and dynamiccapability theories (DCT), few researches have exploredthe interplay between CSR and firm attributes. Further,although the ongoing financial crisis spurs scholars torethink extant theories and search new avenues to locatenew drivers that help firm regain financial well-being, animportant financial indicator of firm, default risk, hasbeen largely neglected. The current research, by empha-sizing the critical role of CSR, bridges these gaps in RBT,DCT, and default risk studies. It hypothesizes a relation-ship in which CSR, by its ability of reputation-building,helps firms reduce the risk of default. In addition, thispaper formulates the moderating effects between CSRand firm capability, CSR and environmental dynamism/complexity and depicts a more complete pattern of CSR’sfunctions in different internal and external conditions.

Data, Measures, and Analytical Model

The data are collected from multiple sources. Wecollected CSR variable from Fortune Magazine (America’sMost Admired Companies), Default risk variable fromStandard & Poor’s Corporate Credit Rating, and othervariables from Compustat. We used a Stochastic FrontierEstimation to measure firm capability, and followed Keatsand Hitt (1988) to get environmental dynamism andcomplexity. To account for the heteroscedasticity andautocorrelation resulted from the panel structured data,we adopted two robust regressions to test the model. Oneis an estimation based on Newey-West standard error.Another is based on the White standard error (deals withheteroscedasticity) and clustering firms (deals withautocorrelation).


The results strongly confirm that CSR significantlyreduces firm default risk. “In the long run . . . social and

economic goals are not inherently conflicting but inte-grally connected . . .” (Porter and Kramer 2002, p. 5).While economic activities push social changes, firm’sengaging social responsibility often yields economic gains.Our paper further confirms such a proposition by illustrat-ing “Doing well by doing good” in the firm’s default riskaspect. Recent marketing researchers find CSR creates aninsurance-like firm asset and protects the firm from dras-tic turbulence during financial downturns. Our researchextends this stream and suggests that this insurance-likeasset not only assures investors, as the previous researchconfirms, but also generates confidence for debt-holders.A firm engaging in higher socially beneficial activities,although bears cost, generates an overall gain reflected bylower cost of debt and higher credit rating.

The Dynamic Capability Theories (DCT) suggeststhat firm capabilities drive firm performance. Our resultsconfirm this theory. A firm focusing on nurturing itsresource-deploying ability brings economic value reflec-ted by its boosted internal efficiency and external relation-ships and hence prevents itself from debt risk.

Supporting the above reasoning, the interactionbetween CSR and firm capability shows that when asuperior capability is missing, the CSR performs strongerin protecting firm from default risk. This pattern demon-strates a substitute effect of CSR and firm capability,which further reinforces the theoretical development ofthis paper in that both CSR and firm capability are specialfirm assets. When one is missing, another will help. Thismechanism has a strong practical implication. As the DCTstates that capabilities are firm specific, path dependent,and cannot be established in a short-terms or in a readilymanner. This is why many firms have trouble in theircompetition because of those missing or lagged capabil-ity. In this sense, one feasible strategic option for them, asour results suggest, would be engaging in more CSRactivities and building better corporate images. The pay-off would be improved debt holder’s confidence andhigher credit rating, which in turn will change the firm’sfuture micro-financial environment in a desirable way.

There is little surprise that firms in a turbulent indus-try are more likely to have default risk as the fast-changingenvironmental makes it harder to plan firms’ activities.


This unpredictability poses a major threat for a lot of firmssuch as the smart phone market. To better deal with thisindustry volatility, our results show that CSR’s effect onthe risk reduction will be stronger in high dynamic indus-try. This finding further extends CSR’s benefits for a firmto the business environmental literature. From anotherangle, it demonstrates an insurance-like effect that bettercopes with fast-changing surroundings. In this sense, thedebt holders in turbulent market pay more attention to the

fact if a firm is “doing good” and assign their confidenceon the firm’s future financial rigor accordingly.

In closing, the results confirm that CSR has a strongeffect on default risk reduction. It also exerts more powerwhen firms don’t have superior capabilities and when theenvironment becomes more volatile. These findings con-tribute to CSR studies, resource-based research and firmrisk literature. References are available upon request.

For further information contact:Wenbin Sun

Helzberg School of ManagementRockhurst University1100 Rockhurst Road

Kansas City, MO 64110Phone: 816.501.4016



COMMUNICATING CORPORATE SOCIAL RESPONSIBILITY USINGSOCIAL MEDIA: IMPLICATIONS FOR MARKETING STRATEGISTS

Amy Lyes, Massey University, New ZealandNitha Palakshappa, Massey University, New Zealand

Sandy Bulmer, Massey University, New Zealand

ABSTRACT

We investigate how Corporate Social Responsibility(CSR) strategy related messages are communicated toconsumers via Facebook, and what consumer perceptionsand behavioral responses to them are. Guided interviewsfocusing on the Facebook pages of 12 firms were con-ducted. The findings highlight five broad themes illustrat-ing how participants respond to CSR messages throughFacebook. Implications for marketing strategists offerinsights into how best CSR activities might be communi-cated using Facebook.

Keywords. Corporate Social Responsibility;Facebook; Marketing Strategy; Communication. Track 5:Ethical, legal, social, and public policy issues.

INTRODUCTION

Marketers are faced with both challenges and oppor-tunities presented by the increased connectivity and dis-persion of social media across the globalizing world.Facebook is one such internet-based social media tech-nology that has only recently emerged as a potentialmarketing tool. Similarly, firms are increasingly recog-nizing the importance of integrating social responsibilityinto their marketing strategy. While considerable researchhas been conducted on corporate social responsibility(CSR) and its influence on consumers, little research hasconsidered the use of social media as an effective market-ing tool. Research on social media as a means for commu-nicating CSR messages to consumers is scant. Insightsinto CSR social media strategies have the potential tocontribute to theory and would be useful to marketingpractitioners. Our study addresses this research gap anddevelops an understanding of consumer responses to CSRactivities communicated through Facebook. It investi-gates how CSR strategy related messages are communi-cated to consumers via Facebook, and what consumerperceptions and behavioral responses to them are. Webegin with a literature review on CSR and communicationusing social media. The qualitative research methodologyis detailed and findings are presented using selected textunits to highlight key points. We discuss our study withrespect to CSR and focus on implications for marketingstrategy.

LITERATURE REVIEW

Corporate Social Responsibility

CSR is a broad concept which has long been consid-ered difficult to define. CSR activities include ethicalmanufacturing of products, community involvement, con-tributing to not-for-profit organizations, environmentalsustainability such as recycling and reducing emissions(Berger et al. 2007; McWilliams et al. 2006). CSR hascommonly been defined as accepting greater responsibil-ity toward society, which encompasses the pursuit of non-economic as well as economic goals (Carroll 1979). Morerecent studies have seen firms adopting a “triple bottom-line” approach which ensures that a firm’s success isdependent not only on financial bottom-line results, butalso on social, environmental and ethical performance(Berger et al. 2007; Norman and MacDonald 2004).Mainstreaming CSR throughout the entire organizationand embedding CSR values into corporate culture is anincreasingly common practice, central to successful mar-keting strategy (Berger et al. 2007).

The benefits that socially responsible business prac-tice can bring to a firm such as customer loyalty, long-termsustainability and competitive advantage have been wellestablished (Bhattacharya and Sen 2004). These includemore positive consumer attitudes and purchase intentionstoward the firm (Bhattacharya and Sen 2004; Du et al.2010; Lichtenstein et al. 2004; Mohr et al. 2001), enhan-ced profitability (Cochran and Wood 1984; Lev et al.2010), willingness to pay premium prices (Creyer andRoss, Jr. 1997), boycotting irresponsible firms, higherresilience to negative information about that firm andpositive word-of-mouth (Bhattacharya and Sen 2003;Klein and Dawar 2004; Mohr et al. 2001).

While firms are recognizing their duty to act respon-sibly and contribute positively to society, consumers arealso increasingly concerned about their health, the well-being of society and the sustainability of the environment.Consumer preferences for more ethical and sustainableproducts have encouraged firms to respond to these de-mands (Jahdi and Acikdilli 2009; Webb et al. 2008).Furthermore, growing concern for global issues such as


climate change, human rights and labor standards hasencouraged firms to self-regulate as firms are faced withincreased pressure to engage in socially responsible busi-ness practice (Muller 2006; Pomering and Dolnicar 2009).It has been suggested that more positive evaluations offirms are apparent after consumers review CSR informa-tion about the firm (Bhattacharya and Sen 2004). Failureto deliver on a CSR promise can damage a firm’s reputa-tion.

There is evidence that CSR impacts on purchasedecisions, brand-switching, and customer loyalty(Bhattacharya and Sen 2004; Webb and Mohr 1998).Consumer perceptions of CSR depend on how they viewthe firm’s motives, i.e., whether they regard a firm’s CSRmotives as extrinsic increasing organizational profits orintrinsic having genuine concern for social issues (Duet al. 2010). Few consumers view CSR as purely self-interested behavior, believing that there are mixedmotives for CSR activity (Mohr et al. 2001). Consumersjudge the durability of a firm’s commitment to CSRcauses and perceive that long-term, frequent supportindicates genuine concern for society (Du et al. 2010;Webb and Mohr 1998).

Corporate Social Responsibility and Communication

Communication to consumers is essential to raiseawareness of the contribution a firm makes to society.While there are many possible communication channels,the most commonly utilized traditional communicationvehicles for CSR are point-of-purchase and informalchannels, such as word-of-mouth (Dawkins 2004). How-ever, according to the literature, consumers do notproactively seek information on firm’s CSR behavior(Dawkins 2004). In general, consumer awareness of CSRis low because information about such issues is scant andpoorly recalled (Mohr et al. 2001). While it is clear thatfirms need to increase awareness of CSR initiatives usingappropriate communication channels there has been littleresearch to indicate which are most effective.

While there are conflicting views about communicat-ing CSR efforts some researchers suggest that it does notmake a positive impact on a firm’s reputation (Sen andBhattacharya 2001). Conversely, Swaen and Vanhamme(2004) find that firms communicating a socially respon-sible image are viewed more positively and credibly thanthose who do not. This makes CSR communication strat-egy challenging (Morsing et al. 2008).

Social Media

Social media presents new ways for firms to commu-nicate with their customers and deliver marketing mes-sages. Interactivity, entertainment value and the ability to

maintain social ties with friends are key features whichattract consumers to social media. With these characteris-tics in mind, firms have transformed the way that theyconnect and establish relationships with their customers.Increasingly, firms are using social media platforms suchas Facebook to respond to consumer needs more effi-ciently (Fieseler et al. 2010). Social media provide aneffective resource for gathering useful information aboutcustomers, targeting and connecting with specific con-sumer groups and providing up-to-date information aboutproducts and services. Facebook users are able to custom-ize their pages so that they only receive information ofinterest to them. This enables marketers to target specificaudiences as Facebook allows advertisements to be deliv-ered to users who have expressed an interest in a particularproduct or service.

Firms can connect with consumers by creating theirown Facebook page and inviting other Facebook users to“like” or become “fans” of their page. Each Facebookpage has a “wall” where the firm and other users can postmessages which can be “liked,” “commented on” by thefirm and other users, or shared to another user’s page.Each firm’s page also contains an information page,photos, events and discussion links on a left-hand sidemenu bar, which can also be altered to include other linksunique to the firm. When a firm posts onto their “wall” itappears in newsfeeds of users who have “liked” theFacebook page. This provides marketers with an opportu-nity to offer information, develop relationships with theircustomers and engender a sense of communitybelongingness (Foster et al. 2010). Consumers are alsoable to provide feedback by sending messages or “wallposts” on the firm’s page.

An increasing number of corporate websites havelinks to Facebook, such as “join our Facebook page.” Thismeans that consumers browsing the net can immediately“like” or comment on the material and link it to theirFacebook page. This gives consumers a sense of empow-erment, because they are in control of the pages they“like.” Individuals can use this to portray a certain per-sonal image that is visible to other users. By identifyingthemselves with brands, videos, and other information,users can create their own unique identities. It is possiblethat users may be using social media as a means ofsignifying their ethical values or social responsibility toothers – e.g., Facebook users may “like,” share or joincampaigns such as “Shave for a Cure.”

Researchers offer various recommendations for howmarketing managers can use social media to best commu-nicate their messages – e.g., Lee (2010) asserts that themost effective Facebook posts are interesting and encour-age users to share information, stimulating communica-tion. Lee (2010) also emphasizes the importance of


Facebook posts being short, concise and relevant to users.Kaplan and Haenlein (2010) also address the importancefor firms to be active, sociable and interesting whencommunicating messages through social media. Sharingand interaction are central to social media, thereforeengaging and interacting with users and ensuring that newcontent is communicated regularly is important for socialmedia communicators (Kaplan and Haenlein 2010).

In summary, new social media are changing the waysin which consumers can connect with firms and becomeaware of their CSR activities, and it appears that theemergent corporate practice of using social media tocommunicate CSR to consumers is under researched.Therefore, the purpose of this study is to examine howconsumers respond to firms’ CSR efforts throughFacebook.

METHODOLOGY

A qualitative depth interview methodology was cho-sen for this study conducted in a major New Zealand city.Audio-recorded interviews lasting approx 90 minuteswere carried out in a private computer enabled room at amajor university campus. Participants signed on to theirFacebook account and took part in an interview thatincluded visiting selected corporate Facebook pages withCSR messages. A purposive sampling technique wasemployed because a predefined group was sought basedon the premise that interviewing those people wouldmaximize chances of uncovering insights on CSR andsocial media communications. Eight participants wereselected using a snowball recruitment technique designedto recruit male and female adult consumers who wereusers of Facebook that were easily able to verbally articu-late and express their feelings. In order to gain depth andrichness, the study also aimed to select contrasting pointsof view. Thus, one third of our participants were chosenbecause they professed to be highly socially responsibleand the others were either moderately or not at all con-cerned about social responsibility. Twelve well-knowncorporate Facebook pages were identified for use in thisstudy. A range of local and multi-national firms from avariety of industries was selected to appeal to consumersof different ages and both genders. The firms included AirNew Zealand, ASB Bank, Coca-Cola, Dove, Johnson &Johnson, Kellogg’s, Revlon, Starbucks, The Body Shop,Toyota, Westpac Bank, and EcoStore (a local leadingmanufacturer and retailer of sustainable household andbody care products).

Each participant was asked to choose four firms fromthe list of 12. They were then invited to log onto theirFacebook account and navigate through each of the fourfirms’ Facebook pages. The researcher invited commen-tary from the participants as they progressed through the

pages, generating a stream of consciousness type of nar-rative. Prompts were used to probe what they were think-ing about certain features of the page and the content ofcertain posts. Participants were directly asked about theirfeelings and responses as they browsed through the pages.They were also asked to comment on how they wouldadvise a firm that was considering using Facebook as acommunication tool for CSR activities. Finally, the par-ticipants were questioned about their understanding ofCSR and its importance to them personally, with anemphasis on Facebook usage patterns and social or ethicalissues. Written transcripts of the interviews were themati-cally analyzed.

FINDINGS

Participant and Facebook Profiles

Participants included three males and five women,aged from 19–53 years. Of these, three people wereclassified as active according to their professed level ofsocial responsibility; two were moderate and two were notactive. The focal firms had local and international CSRactivities. Facebook strategies included making specificwall posts about their CSR strategies and inviting com-ment; designing stand alone information pages; postingphotos; providing links to and information about thecampaigns that firms were involved with; and encourag-ing Facebook friends to make page likes. As an examplewe note that The Body Shop had wall posts about fair tradeproducts and ingredients, preventing sex trafficking andanimal testing, and support for a local earthquake appeal.They had an information page with a focus on protectingthe planet and selling ethically produced products. Theirinteractive quiz was entitled “Are You an Earth Lover?”and they displayed photos celebrating community fairtrade and in support of their stop sex trafficking campaign.They also included page likes and support for SaveAmericas Wolves, SPCA, and Fairtrade.

Consumer Responses to CSR Facebook Strategies

The five broad themes drawn from the data andpresented below illustrate how participants respond toCSR messages through Facebook.

Insatiability. This theme deals with participantssearching for more information, clarity and evidence.When faced with a CSR message, participants oftenexpressed frustration and confusion. This was becausesome of the messages were not very clear—they did notspecify exactly what the initiative was about or how theywere accomplishing it. Participants preferred informationto be obvious and easy to access—simple messages aboutsimple CSR activities had a greater impact on participantscompared to those that were complicated and used jargon


or complex terms. “It makes it really easy to see what it isabout, like water drop for water conservation and stuff sothat’s good “‘cos” it makes it more easily understandablefor someone who maybe doesn’t, like, understand buzzwords” (Hannah).

Participants also preferred it when the CSR messagewas short and contained a link for further information. Itwas evident that “actively socially responsible” partici-pants in particular, wanted to know more about firms’CSR activities. They were disappointed when the infor-mation provided did not meet expectations: “This is crap.There’s no information here! Rubbish! . . . It doesn’t tellme how Air NZ and Star Alliance are doing this” (Gerald).Furthermore, when firms did not provide enough visibleinformation about their CSR initiatives, participants wereconvinced that they were not socially responsible firms,contributing to skeptical perceptions: “It’s just a post ‘costhey want to look like they’re being socially responsible”(Gerald).

When insufficient information was provided, partici-pants would question the firm’s initiatives. This left themsearching for some kind of evidence or visible outcomesto back up their claims and show that the firms wereactually doing what they said they were doing. Some firmsposted photos of their CSR activities, such as volunteeringin the community, or messages about their relationshipswith community and not-for-profits. Participants re-sponded to this verification more positively. They alsorequested more proof of support for other Facebook pagesthat the firm “liked” – participants felt that if the firm wasactively involved in support of such other causes thensimply “liking” a page wasn’t enough.

Some participants were impressed by CSR messagesthat mentioned large dollar sums relating to donations orfunds raised for charities: “I suppose a dollar figure reallydoes put a price on how much they have prepared to putin and it shows that obviously it’s a successful companyand that they are contributing towards a pretty memo-rable or big cause” (Jane). However, participants alsoconsidered that firms who mentioned a large dollar figurewere bragging which made them question whether thefirm’s motives were genuine. Positive testimonials, feed-back and comments from others such as communitymembers, employees or those directly affected by theCSR activity (e.g., a tornado victim) provided furtherevidence of a firm’s support for a cause. Participants wereeven more impressed when an outside party mentionedsomething that was not communicated on the Facebookpage. “Hearing and seeing what someone does is moreimportant than what someone says . . . if you hear . . . thatthey are getting the support that they need, or fromemployees or other companies, then you know they’redoing a good job” (Matt).

Altruism Does Not Exist. This theme emerged fromthe cynical disposition that participants had towards thefirms and their motives with regard to CSR initiatives,highlighting the idea that for these participants, altruismdoes not exist in the corporate world. Participants re-garded CSR as an insincere act – the firms were trying tobe something they were not and were only pretending tocare. They would often be quite suspicious when facedwith a CSR message, as if the firm was trying to trick them,using a marketing ploy disguised as CSR. Participantsappeared quite pleased with themselves when they couldidentify these “marketing tricks”: “There was probablyno need to do it, but they did it to enhance their image. Itdidn’t work on me though, I can see through that”(David).

Participants felt that firms see CSR as a transaction,where they are essentially “buying goodwill”, allocatingresources and hoping to achieve certain benefits in return.They viewed CSR as something “fashionable” that firmswould only adopt in order to put a positive spin on theirbusiness. Participants believed that firms recognize theappeal that CSR has to consumers and argue that they onlyengage in CSR because it encourages consumers to buytheir products or services. “They make a big song anddance about giving stuff away to kids at a charity auctionor teaming up with other organizations, but that positivemedia will feed back into their revenue streamanyway...they’re only doing it to make themselves lookbetter so they get more business” (Gerald). “It’s aboutgetting your image up, so people perceive you as a coolcompany so they’re gonna start using your products”(David).

These perceptions of CSR were aggravated by nega-tive views about big corporations. However, even thoughparticipants were skeptical and believed that CSR isessentially money driven, they understood that makingmoney is a business’ ultimate purpose. CSR providesbenefits for parties other than the firm, therefore, partici-pants accepted that it is a positive thing and is “better thandoing nothing” – the end justifies the means: “If they’reusing it as a marketing ploy who gives a crap, ‘cos in theend, they are giving money to all these causes . . . the factthat they are doing something is good enough for me”(Matt).

Participants were more convinced of the authenticityof firms’ CSR motives when they took part and got othersinvolved as well, such as facilitating consumer recycling.Helping smaller firms or not-for-profits, and having long-term community partnerships also made participants per-ceive firms as more genuine. Anything that demonstratedallocating additional time, effort and resources towards aCSR activity was seen as more believable; for examplevolunteering and community work was seen as more


practical and genuine because it takes a lot more effortthan a simple donation. “It’s the idea that the company isactually participating in the community rather than, “OhStarbucks made a donation.”Starbuck’s employees areactually out there making a difference in the communityand interacting with the community rather than justwriting checks” (Gerald).

Size Matters. The theme “size matters” involves thenotion that bigger firms are better able to afford CSRactivities. It was apparent that size was a significant factorin determining the extent to which participants believed afirm should engage in CSR. All participants emphasizedthe size of the focal firms and the amount of money theymake. They seemed to imply that bigger organizationshave more responsibility than smaller ones to engage inCSR because they make a lot more money and can affordit: “They have so much money and still charge a lot so itwouldn’t cost them that much to help” (Kelly). Theymaintained that only firms with the financial capacity toengage in CSR should do so – ultimately realizing a firmneeds to make money in order to survive. Therefore, theyargued that it doesn’t really matter what small firms do, aslong as they try: “Big companies can do big things; theyhave the means to do it and more money available,whereas small companies can only do small things”(Rachel).

This perception resulted in more critical evaluationsof bigger firms’ CSR messages – their retention of profitswas considered selfish and reaffirmed participant precon-ceptions that large impersonal firms only care aboutmaking a profit. They argued that in the end, money haspriority over social responsibility for these firms. Theyalso felt that the “people at the top,” the “big guys”, don’tcare and are the ones who get all the money: “I don’t thinkthey genuinely care to be honest, I think it’s all just a front,most companies don’t really, do you honestly think thepeople at the very top give a shit about that sort of stuff,they’re just doing it to get more money” (David). There-fore, participants were pleased to see “big companies”engaging in CSR and seemed to feel more satisfied to seetheir profits going towards something other than theCEO’s pay check: “‘Cos everyone thinks banks makesuch profit and to see them doing something for thecommunity . . . it just makes you think that they’re not quiteas selfish as them seem” (Rachel). Participants also sawbigger firms as role models, setting an example for others,and obliging consumers also to be socially responsible.

Reminiscing. This theme deals with how certain CSRmessages evoked memories that influenced the way par-ticipants (particularly older ones with some life experi-ence) judged the messages. They would often tell a shortstory, drawing on certain memories or experiences linkedto the firm, or related to the CSR activity that the firm was

engaging in. Participants also occasionally referred toexperiences with the firm’s competitors. These storiesinduced powerful emotions which ultimately had a sig-nificant effect on the way participants responded to andperceived both the CSR message and the firm. Theyreacted favorably to funny, happy, and exciting messagesand had more positive impressions of the firm or CSRinitiative. “That’s cool! Oh yeah I did see that, saw themdo that on TV (laughs) He [TV presenter] shaved his headfor money, that’s good” (Kelly). “[Johnson & Johnson]still have the pink baby lotion that I used to use on my kids”(Jane).

Recalling a negative memory or experience on theother hand (e.g., annoying, frustrating, scary, sad, ordisgusting), had the opposite effect: “You know that AirNZ uses non-wax paper cups that are not recyclable! It’sa sham” (Gerald). Emotive stories would make the par-ticipant more appreciative of the firm’s CSR efforts. Theyalso contradicted themselves, changing their previousskeptical perceptions of the firms. “The Westpac helicop-ter appeal . . . that’s good . . . our son had an accident andthe Westpac rescue helicopter flew him to Wellington, wewere personally touched by that so that made us moreinclined to deal with them and sort of see a human side ina huge corporation” (Rachel).

Relevance Has Meaning. This theme considers theCSR message relevance to participants, linked to a senseof familiarity and personal meaning. Overall, CSR mes-sages relating to something the participant was familiarwith or had heard of before were regarded more favorablythan messages they were not familiar with. Participantsalso responded more positively to CSR messages whenthe firm they were analyzing was one they were familiarwith, that they liked, or purchased from. CSR messagesthat related to participants’ interests (e.g., cars, animals,and children) and that had personal importance to themappealed to them significantly more than messages thattalked about things they were not interested in. “It’stalking about Third World countries if you like, countriesthat haven’t got such great healthcare, and being amother, that appeals to me” (Claire). “That catches myattention because I was involved with that, the SPCA”(Kelly). This also applied when there was a celebrity orfamous person that the participant admired or liked: “Idon’t really care about this to be honest. I like John Key[Prime Minister] though! He’s the man, I would read thatand I’d like it!” (David). Participants were not interestedin messages that were not aligned with their interests andoften considered them to be a “waste of money” that couldbe used for something “more worthwhile.”

Participants also preferred the CSR messages to berelevant to the industry or to the product or service that thefirm sold. For example, they considered it important, that


if a firm’s products targeted women, that they supportwomen’s issues; or, if they sold drinks in plastic bottles,that their CSR activities promoted recycling. Participantsresponded strongly to this because they felt that firms whooperate in a way that negatively impacts on, for example,the environment, have a duty and responsibility to dosomething to counteract that damage.

Occasionally firms would make donations using theproducts/services that they specialized in (e.g., Kellogg’sgave away free breakfasts to hungry children). Partici-pants preferred this method over simply donating moneybecause they felt it was more practical and efficient: “It’sa better idea giving something they can actually use,because money, people can just blow it and you have nocontrol” (David). Local CSR initiatives took priority forparticipants, as they were more locally familiar and rel-evant. Participants preferred seeing CSR initiatives thatwere regionally specific. “When they do have local stuffyou’re more familiar with it, it would have a closer to yourheart type feeling” (Rachel). Participants were not asinterested in CSR messages that talked about internationalcharities or causes that they were not familiar with andwere disappointed when they could not get involvedbecause they lived abroad. “The only thing that worriesme about international pages . . . you can’t take partunless you’re in the States, and that’s disappointing”(Rachel). “I would do that to get a free cup of coffee onEarth Day! Oh, but it says US Canada only so not keen .. . I wouldn’t click on it anymore” (David). Unfortunatelymany of the firms analysed did not have a country specificFacebook page for their firm, but rather an internationalpage which presented CSR messages related to the coun-try the firm is most associated with (often the U.S.).

DISCUSSION

The results from this study established that thirdparties play an important role in the way that consumersrespond to firms’ CSR messages through social media.Liking, commenting and sharing activity from otherFacebook users intrigued participants and attracted theirattention to the CSR message. Participant intrigue wasfurther enhanced when the Facebook user was someoneknown to the participant, such as a famous person or aFacebook friend. This response can be attributed to thesignificance of familiarity – participants were attracted bythe fact that someone familiar “liked”, commented on, orshared the post. These results signify that creating inter-esting CSR messages that stimulate third party user activ-ity, is an effective means of getting consumers to noticethe message as suggested by the literature.

The findings also suggest that testimonials and com-ments from community members, partners or employeesappealed to participants. External stakeholders providedevidence attesting to the firm’s support for a particular

CSR cause, which made the CSR messages more believ-able. The findings confirm the literature on how thirdparty endorsement enhances message credibility (Jahdiand Acikdilli 2009; Pomering and Dolnicar 2009). Encou-raging external stakeholders to provide feedback aboutthe firm and CSR activities on their Facebook page shouldelicit more positive responses from consumers.

Participants often had a cynical disposition towardlarge firms, perceiving them as selfish, impersonal anduntrustworthy. Although the act of engaging in CSR wasregarded positively, the reasons for which participantsbelieved firms’ engaged in CSR were regarded nega-tively. While the literature emphasized positive percep-tions of firms engaged in CSR, our study differed, withmany negative views about self-interested behavior. Par-ticipants believed that the main motivation behind a firm’sengagement in CSR was to improve the reputation of thefirm in order to increase profits, expecting that large firmsshould engage in CSR because of their financial strength.Our findings confirm the literature on this point.

Overall, the negative responses that participants hadtoward the motives of large firms may be attributed toincreasing awareness of firms’ marketing strategies. Skep-tical perceptions were reduced when CSR informationwas easy to access and clearly. Therefore, it is necessaryfor firms to assume that consumers are well aware of thebenefits of CSR to firms and that they are wary of“marketing tricks.” Our findings highlight the need forfirms to be honest in all their dealings with consumers andto communicate their CSR initiatives in an open andtransparent way, supporting the findings of researcherssuch as Du et al. (2010). The ultimate image and percep-tion that consumers have will largely depend on whetherthe firm follows through with CSR claims (Jahdi andAcikdilli 2009; Webb and Mohr 1998).

The results highlighted ways in which firms canreduce consumers’ negative perceptions and can demon-strate authenticity and humanity – e.g., participantsresponded positively to CSR messages when there wascredible evidence of genuinely putting time and resourcestoward a cause. Photos elicited positive responses fromparticipants, particularly when accompanied by messagesfrom staff who personalized the impersonal firm. The firmappeared more friendly and humane. Facebook has anadvantage over traditional media in communicating CSRmessages in this respect, as firms are able to provide visualevidence of their CSR activities.

Participants attached personal meaning to certainCSR messages, which impacted how participants respon-ded to these messages. CSR messages evoked both posi-tive and negative memories or experiences and when theywere positive participants responded more positively,whereas negative memories or experiences would elicit


negative responses. By relating the messages to their ownpersonal experiences, participants felt closer to the firmand were able to relate more easily to what the firm wasdoing. Consumers derive value from the particular CSRcauses to which they are emotionally attached (Green andPeloza 2011). This means that CSR messages will be mostinfluential when the messages themselves elicit positiveconsumer memories.

The “relevance has meaning” theme signified thatlocal CSR initiatives were regarded more favorably byparticipants. This can be attributed to the importanceparticipants place on supporting their local environmentand the significance of familiarity and personal meaning.Local causes also offer consumers the opportunity to getinvolved, which delivers value to consumers by contrib-uting to their sense of well-being (Bhattacharya and Sen2004; Green and Peloza 2011). The “relevance has mean-ing” theme also highlighted personal meaning as a signifi-cant contributor toward positive responses to CSR mes-sages from participants. As mentioned previously, par-ticipants favored messages that were relevant to theirinterests. Participants took notice and responded posi-tively to these messages because they were interesting,familiar, and thus, easier to understand and relate to.

IMPLICATIONS AND CONCLUSIONS FORMARKETERS

The findings of our study have significant implica-tions for marketing strategists. In this section we highlightfour points that could have an impact on firms as they lookfor insights into how CSR activities can be best commu-nicated using Facebook. Firstly, we suggest that firmskeep CSR messages simple but informative. Visitors tocorporate Facebook pages have short attention spans,thus, messages need to be concise and easy to understand.More in-depth information can be provided via a link if

necessary. Secondly, we note that being honest and com-municating openly is vital. Consumers are well-informedand conscious of the benefits that CSR brings to firms.Therefore, honesty and transparency is critical in order tominimize consumers’ skeptical perceptions of firms’ CSRmotives, and to change consumers’ negative perceptionsof large firms. This can be achieved by using an activeCSR approach such as posting photos of staff membersvolunteering in the community to provide evidence of thefirms’ CSR activities.

Thirdly, we highlight the importance of stimulatingactivity from third parties. Liking, commenting and shar-ing activity from other Facebook users attracts atten-tion to messages and encourages consumers to attend tomessages for longer and in more detail. This can beaccomplished by posting likeable, interesting, and easy toread messages that create buzz and elicit discussion fromusers. Furthermore, feedback from external stakeholdersshould also be encouraged to enhance the credibility of theCSR message, and engender more positive responsesfrom consumers.

Finally, we emphasize the significance of customiz-ing CSR messages to different consumers or groups. Itwas clear that each participant is unique with differentinterests, definitions of CSR, levels of “social responsibil-ity” and memories and experiences with the firm. We alsounderline the value of creating CSR messages with per-sonal meaning as this significantly contributed towardpositive responses toward the firm. Therefore, consumersor groups of consumers need to be approached and tar-geted in different ways in order to develop closer relation-ships with their customers. By creating CSR messagesthat elicit positive stories or experiences from consumersand by communicating more relevant and appealing CSRmessages, more meaningful communication of a firm’sCSR strategy may be facilitated.

REFERENCES

Berger, Ida E., Peggy Cunningham, and Minette E.Drumwright (2007), “Mainstreaming CorporateSocial Responsibility: Developing Markets for Vir-tue,” California Management Review, 49 (4), 132–57.

Bhattacharya, C.B. and Sankar Sen (2003), “Consumer-Company Identification: A Framework for Under-standing Consumers’ Relationships with Compa-nies,” Journal of Marketing, 67 (2), 76–88.

____________ and ____________ (2004), “Doing Bet-ter at Doing Good: When, Why, and How ConsumersRespond to Corporate Social Initiatives,” CaliforniaManagement Review, 47 (1), 9–24.

Carroll, Archie B. (1979), “A Three-Dimensional Con-ceptual Model of Corporate Performance,” Academyof Management Review, 4 (4), 497–505.

Cochran, Philip L. and Robert A. Wood (1984), “Corpo-rate Social Responsibility and Financial Perfor-mance,” Academy of Management Journal, 27 (1),42–56.

Creyer, Elizabeth H. and William T. Ross, Jr. (1997),“The Influence of Firm Behavior on Purchase Inten-tion: Do Consumers Really Care about BusinessEthics?” Journal of Consumer Marketing, 14 (6),419–32.

Dawkins, Jenny (2004), “Corporate Responsibility: TheCommunication Challenge,” Journal of Communi-cation Management, 9 (2), 108–19.


Du, Shuili, C.B. Bhattacharya, and Sankar Sen (2010),“Maximizing Business Returns to Corporate SocialResponsibility (CSR): The Role of CSR Communi-cation,” International Journal of Management Re-views, 12 (1), 8–19.

Fieseler, Christian, Matthes Fleck, and Miriam Meckel(2010), “Corporate Social Responsibility in theBlogosphere,” Journal of Business Ethics, 91, 599–614.

Foster, Mary K., Anthony Francescucci, and Bettina C.West (2010), “Why Users Participate in Online So-cial Networks,” International Journal of E-BusinessManagement, 4 (1), 3–19.

Green, Todd and John Peloza (2011), “How Does Corpo-rate Social Responsibility Create Value for Consum-ers?” Journal of Consumer Marketing, 28 (1), 48–56.

Jahdi, Khosro and Gaye Acikdilli (2009), “MarketingCommunications and Corporate Social Responsibil-ity (CSR): Marriage of Convenience or ShotgunWedding?” Journal of Business Ethics, 88, 103–13.

Kaplan, Andreas and Michael Haenlein (2010), “Users ofthe World, Unite! The Challenges and Opportunitiesof Social Media,” Business Horizons, 53 (1), 59–68.

Klein, Jill and Niraj Dawar (2004), “Corporate SocialResponsibility and Consumers’ Attributions andBrand Evaluations in a Product-Harm Crisis,” Inter-national Journal of Research in Marketing, 21 (3),203–17.

Lee, Dong-Hun (2010), “Growing Popularity of SocialMedia and Business Strategy,” SERI Quarterly, 3 (4),112–17.

Lev, Baruch, Christine Petrovits, and SureshRadhakrishnan (2010), “Is Doing Good Good forYou? How Corporate Charitable Contributions En-hance Revenue Growth,” Strategic Management Jour-nal, 31 (2), 182–200.

Lichtenstein, Donald R., Minette E. Drumwright, andBridgette M. Braig (2004), “The Effect of CorporateSocial Responsibility on Customer Donations to Cor-porate-Supported Nonprofits,” Journal of Market-ing, 68 (4), 16–32.

McWilliams, Abagail, Donald S. Siegel, and Patrick M.Wright (2006), “Corporate Social Responsibility:Strategic Implications,” Journal of Management Stud-ies, 43 (1), 1–18.

Mohr, Lois A., Deborah J. Webb, and Katherine E. Harris(2001), “Do Consumers Expect Companies to BeSocially Responsible? The Impact of Corporate So-cial Responsibility on Buying Behavior,” Journal ofConsumer Affairs, 35 (1), 45.

Morsing, Mette, Majken Schultz, and Kasper Ulf Nielsen(2008), “The ‘Catch 22 ’ of Communicating CSR:Findings from a Danish Study,” Journal of Market-ing Communications, 14 (2), 97–111.

Muller, Alan (2006), “Global versus Local CSR Strate-gies,” European Management Journal, 24 (2/3), 189–98.

Norman, Wayne and Chris MacDonald (2004), “Gettingto the Bottom of ‘Triple Bottom Line’,” BusinessEthics Quarterly, 14 (2), 243–62.

Pomering, Alan and Sara Dolnicar (2009), “Assessing thePrerequisite of Successful CSR Implementation: AreConsumers Aware of CSR Initiatives?” Journal ofBusiness Ethics, 85, 285–301.

Sen, Sankar and C.B. Bhattacharya (2001), “Does DoingGood Always Lead to Doing Better? Consumer Re-actions to Corporate Social Responsibility,” Journalof Marketing Research, 38 (2), 225–43.

Swaen, Valerie and Joelle Vanhamme (2004), “See How‘Good’ We Are: The Dangers of Using CorporateSocial Activities in Communication Campaigns,” inAdvances in Consumer Research, Barbara E Kahnand Mary Frances Luce, eds. Vol. 31. Valdosta, GA:Association for Consumer Research.

Webb, Deborah J. and Lois A. Mohr (1998), “A Typologyof Consumer Responses to Cause-related Marketing:From Skeptics to Socially Concerned,” Journal ofPublic Policy & Marketing, 17 (2), 226–38.

Webb, Deborah J., Lois A. Mohr, and Katherine E. Harris(2008), “A Re-Examination of Socially ResponsibleConsumption and its Measurement,” Journal of Busi-ness Research, 61 (2), 91–98.

For further information contact:Nitha Palakshappa

Private Bag 102 904North Shore Mail Centre

Auckland 0745New Zealand

Phone: +64.9.414.0800, Ext. 9454Fax: +64.9.414.8426



CUSTOMER OUTCOMES OF CORPORATE SOCIAL RESPONSIBILITYIN SUPPLIER-CUSTOMER RELATIONSHIPS

Christian Homburg, University of Mannheim, GermanyMarcel Stierl, University of Mannheim, Germany

Torsten Bornemann, University of Mannheim, Germany

SUMMARY

The notion of corporate social responsibility (CSR)has gained momentum and is now of strategic importancein today’s business practice. Marketing research has ech-oed this development. A number of studies – all in abusiness-to-consumer (B2C) context – have established apositive link between a firm’s CSR activities and severalimportant consumer outcomes. In practice, however,CSR, defined as a firm’s activities and status related to itsperceived societal or stakeholder obligations, is an issuenot only in B2C industries, as business-to-business (B2B)companies are often at the forefront of engaging in CSR.However, studies dealing with customer outcomes ofCSR activities in a B2B context are virtually nonexistent.

Owing to the distinctive decision-making behaviorof organizational customers, findings regarding the impactof CSR actions in a B2C context do not readily transfer toa B2B context. In particular, compared to B2C buying,organizational buying follows a more rational decision-making process that uses information on a particularsupplier in a different manner. In this kind of environ-ment, CSR may act more as a signal for positive firmattributes. This heightens the importance of investigatinghow CSR actions foster favorable organizational sup-plier-customer relationships.

Addressing this issue, we developed a frameworkthat describes a causal chain leading from a firm’s actualCSR engagement via its CSR reputation to customers’trust and customer-company identification and ultimatelyto customer loyalty. Within our framework, we argue thata dual mechanism exists through which a positive CSRreputation leads to customer loyalty. First, relying oninformation economics, we postulate a functional mecha-nism in which CSR operates as a signal for trustworthi-ness. Second, even in industrial buying, individuals ulti-mately act and decide, and on the basis of social identitytheory we propose a second and more emotionally focusedmechanism by which CSR fosters increased customer-company identification.

In two empirical studies, we test this framework. Westart by demonstrating that CSR works distinctly in B2Bmarkets through an initial experimental study with youngprofessionals. The results show that in contrast to itsfunctioning in a B2C environment, in a B2B setting CSR

works primarily under conditions of high uncertainty, afinding that underlines the importance of CSR as aninformational cue in organizational buying. In our mainstudy, we test our model by applying dyadic data drawnfrom 200 existing supplier-customer relationships from avariety of B2B industries (e.g., machine building, elec-tronics, logistics, chemicals). Dyadic data came from asurvey of purchasing managers (customer firm) and theirrespective marketing/sales contact person (supplier firm).

Employing structural equation modeling, our resultsreveal positive effects of two distinct facets of a supplier’sCSR engagement on customer loyalty through a dualmechanism as hypothesized. Business process CSR, henceCSR activities within a firm’s core business operationstargeting primary stakeholders, fosters customers’ trustwhereas non-business process CSR, hence activities out-side a company’s core business operations involvingsecondary stakeholders, strengthens customer-companyidentification. In turn, trust and customer-company iden-tification drive customer loyalty. In addition, we considerseveral contingency factors that moderate the effective-ness of CSR. In high-uncertainty markets and for highlyimportant products, the effect of business process CSR oncustomers’ trust is stronger. However, for highly rela-tional business relationships this effect is weaker. Inmarkets with intense competition and where customersare strongly CSR-oriented, the positive effect of non-business process CSR on customer-company identifica-tion is amplified.

The findings of this study advance academic knowl-edge in five ways. First, we empirically show that CSRactivities can, in a distinct manner, increase customerloyalty in a B2B environment. Second, we holisticallyexamine the entire causal chain from a supplier’s actualengagement in CSR, through measuring its CSR reputa-tion, to considering customer outcomes simultaneously.Third, we explicitly allow for CSR’s complexity by con-sidering two CSR facets and show their distinct effects ontrust and customer-company identification. Fourth, byderiving a dual mechanism through which CSR actionslead to favorable customer outcomes we consolidate argu-ments for a signaling value of CSR actions with thosehighlighting an emotional path. Lastly, we identify condi-tions under which the different facets of CSR gain or losetheir effectiveness.


For further information contact:Christian Homburg


68131 MannheimGermany

Phone: +49.621.181.1555Fax: +49.621.181.1556



MEASURING CONSUMER CONFUSION EVOKED BY DESIGNCOPYCATS: THE INFLUENCE OF EVALUATION MODE

Steffen Herm, Technische Universität Berlin, GermanyJana Möller, Freie Universität Berlin, Germany

SUMMARY

By imitating the product design of a leading brand,copycats gain advantages such as acceptance from con-sumers. These advantages are based on the similaritybetween the product designs of original brands and copy-cats entailing consumers to confuse one brand with theother. Managers and judiciary need to measure the extentof confusion in order to inhibit competitors to take unfairadvantages. First, this research investigates to what de-gree different confusion measures are biased by the evalu-ation mode they evoke. Two experimental studies, across

three different product categories and in three countries(USA, Spain, Germany), reveal that participants confusea design copycat more with the original design when theidentification task brings them in a non-comparative com-pared to a comparative evaluation mode. This result isrobust across all studies even when product categoryinvolvement is entered as a covariate. Second, this re-search introduces a new procedure examining how accu-rately and how quickly individuals can recognize a brandby its product design while determining how much andwhich information (design cues) is needed for correctbrand recognition. References are available upon request.

For further information contact:Steffen Herm

Technische Universität BerlinStrasse des 17. Juni 135 (H76)

10623 BerlinGermany

Phone: 0049.30.314.22614E-Mail: [email protected]


CONSUMERS VIEW NUTRITION FACTS THROUGHCULTURAL LENSES

Annie Cui, West Virginia University, MorgantownM. Paula Fitzgerald, West Virginia University, MorgantownKaren Russo France, West Virginia University, Morgantown

SUMMARY

Obesity has become a serious health issue world-wide. While at one time unique to industrialized nations,obesity rates are increasing in developing countries, inclu-ding China (Chen 2000). In response to this health threat,the Chinese Health Ministry issued Nutrition LabelingRegulations for Pre-Packaged Food which mandates thatall pre-packaged food sold in the country to properlydisplay nutrition information (such as calorie, protein, fat,carbohydrates and sodium contents) on the packages.With this context as a backdrop, the present study com-pares how Chinese and U.S. consumers process nutritioninformation on food packages.

While the U.S. Food and Drug Administration (FDA)has been working to promote healthy food choices amongconsumers via proper labeling of nutrition information onpre-packed food since the late 1960’s (Heimbach andStokes 1982), providing nutrition labels on food packagesis a relatively new practice in China. The Chinese HealthMinistry issued the Nutrition Labeling Regulations forPre-Packaged Food in November 2011, requiring all pre-packaged food sold in the country to provide nutritioninformation in the correct format by January 1, 2013.Manufacturers who violate these rules will be suspendedfrom producing and distributing their products (NutritionLabeling Regulations for Pre-Packaged Food). ThoughChinese consumers have become increasingly concern-ing about the nutrition and safety of food products, thepreviously unregulated nutrition labeling practices mayhave confused and complicated consumers’ food pur-chase decisions.

Furthermore, to Chinese consumers, nutrition infor-mation may not play as important of a determinant ofpurchase as other food-relevant factors. For example,evidence suggests that safety cues are important in deter-mining food choices, accentuated, perhaps, by themelamine-contaminated milk incident in 2008 whichresulted in a serious nation-wide food scare among Chi-nese consumers (note, of course, that food safety concernsdo not only present in developing countries. Food-bornediseases cause approximately 76 million illnesses, 325,000hospitalizations, and 5,000 deaths in the United Stateseach year; Mead et al. 1999). Chinese consumers alsooften make quality judgment from other cues such as

expiration date, brand name, government safety seals andcountry of origin (COO). Indeed, consumers frequentlyuse COO information as a cue to guide their productevaluations and purchase preferences (cf., Bilkey and Nes1982; Hong and Wyer 1989, 1990; Li and Wyer 1994;Maheswaran 1994).

In this study, 80 Chinese citizens living in Americaand 96 US citizens residing in America provided useabledata via an on-line questionnaire using Amazon Mechani-cal Turk. Two products were chosen for this study; respon-dents were randomly assigned to a milk (Chinese or U.S.COO) and cookie (Chinese or U.S. COO) product.

A chi-square test examined the Chinese and U.S.identification of the most important piece of informationwhen making a decision for food for themselves. Chineseadults are significantly more influenced by brand nameand less affected by the nutrition information than U.S.adults when choosing a product for themselves (χ2 =20.19, df = 4, p < 0.01, phi = 0.42). For both the cookieproduct and the milk, safety seal importance ratings forChinese citizens are higher than those for U.S. citizens(3.25 vs. 2.92 for the cookie product and 3.28 vs. 2.98 forthe milk, both p < 0.02, eta-squares of approximately0.04).

MANOVA models revealed that Chinese consumersperceive cookies to be more nutritious, to be of higherquality, and to be less likely to cause harm than U.S.consumers. Consumers have less favorable attitudes aboutcookies from China and lower quality perceptions aboutChinese cookies. Main effects for the milk product showU.S. consumers feel that milk is more nutritious and lesslikely to cause harm than Chinese consumers. In addition,main effects show that consumers have more positiveattitudes toward a U.S. milk product, feel that the U.S.milk is healthier, of higher quality and is less likely to beharmful.

However, the real story lies in the interaction effects.These interactions indicate that for Chinese expats livingin the U.S., COO is relatively unimportant; yet, U.S.consumer attitudes are much more a function of COO.U.S. consumers tend to have more favorable attitudesabout the U.S. products and less favorable attitudes regar-ding the Chinese products than the Chinese consumers.


There is a similar pattern for quality perceptions: Chineseexpats are indifferent to COO, but U.S. consumers’ qual-ity ratings are significantly influenced by COO with U.S.products evaluated as being of higher quality and Chineseproducts evaluated as lower in quality.

Several interesting implications fall out of these find-ings. First, our results suggest that nutritional labeling indeveloping nations may not be a particular effectivelymethod for combating the worldwide increase in obesityrates, since nutritional information is less important amongthe Chinese expats than among the U.S. citizens. Second,we find a more complex COO effect than was anticipated.

Why were U.S. consumers so greatly influenced byCOO, but COO was basically irrelevant among the Chi-nese expats? Could it be that, as Mark Twain so eloquentlystated, “Travel is fatal to prejudice?” Or could it be thatexpats are, by their nature, less likely to be ethnocentric?ANOVA analysis indicates a significantly higher ethno-

centrism score from the U.S. citizens (2.68) than amongthe Chinese consumers (2.33; p = 0.01).

Alternatively, is this effect specific to China, for wehave seen a recent strain in U.S./China relationships and,indeed at least some anecdotal evidence of “Chinaphobia.”Consider, for example a reaction to recent concern overfungicide in orange juice. The fungicide is used in Brazil.One expert commentator, Marion Nestle at New YorkUniversity, in reacting to Coke’s (makers of Minute Maidand Simply Orange brands) proactive response stated thatCoke should be applauded, “But as a matter of nationalpublic health policy, this country badly needs an indepen-dent regulatory agency – as the FDA is supposed to be –to keep companies honest,” she said. The FDA “is com-pletely overwhelmed by imported foods. Is FDA rou-tinely testing foods from China? Don’t we wish,”(USAToday, January 13, 2012). References are availableupon request.

For Further Information contact:M. Paula Fitzgerald

West Virginia UniversityP.O. Box 6025

Morgantown, WV 26505Phone: 304.293.7959



WILLING TO PAY FOR A BETTER BRAND: CONSUMER RESPONSESTO CSR PERFORMANCE SCORES

Laurel Aynne Cook, University of Arkansas, FayettevilleRonn J. Smith, University of Arkansas, Fayetteville

Yao Jin, University of Arkansas, Fayetteville

SUMMARY

As competitive pressures in a company’s externalenvironment escalate, brand differentiation becomesincreasingly important. In response, strategies incorpo-rating a variety of combinations of the marketing mix maybe employed. However, using corporate social responsi-bility (CSR) initiatives as a measure of strengthening thebrand is a recent and under explored development inmarketing strategy (Kitchin 2003). For example, recentresearch suggests that the positive brand implicationsresulting from CSR investments are a valid but rarely usedsource of competitive advantage (Melo and Galan 2011).In response, the present research examines the impact ofCSR initiatives from a consumer behavior perspective. Inthe context of the apparel industry, we build upon priorexaminations of CSR (Klein and Dawar 2004; Becker-Olsen, Cudmore, and Hill 2006) and provide insight intohow a business’s socially responsible activities may (ormay not) be effectively communicated to enhance a brandfor which consumers are willing to pay more.

Grounded in the Elaboration Likelihood Model (ELM;Petty and Cacioppo 1979, 1986a, 1986b), this study takesan experimental approach to examine the role of multidi-mensionality in CSR rating and its impact on brandattitude, brand equity, and two measures of purchasingbehavior. Specifically, this research explores the effectson purchase decisions when only one dimension of CSR(e.g., “overall” or “human rights”) is favorably rated, andthus attempts to answer a call for specific focus on the “defacto enslavement of workers” (Byrne 2011, p. 507). Wefind that the human rights sub-dimension of CSR hasinfluence on behavior that is unique from an overall CSRperformance score. Given the connotations elicited whenconsidering the apparel industry coupled with the humanrights sub-dimension of social responsibility, we expecteffects on brand perceptions and purchase decisions to besignificantly lower for a low human rights rating than fora low overall CSR rating. Additionally, we demonstratethat the path of CSR performance to consumer behavior ismediated by CSR’s impact on brand equity. Altogether,the CSR è Brand equity è Consumer behavior path sug-gests that companies should consider the industry inwhich they operate, and selectively invest firm resourcestoward appropriate sub-dimensions of CSR in order tomaximize its effectiveness and impact on consumer pur-chasing outcomes.

Results indicate that even while holding the firm’soverall CSR rating constant, consumers respond nega-tively to implied human rights violations within the ap-parel context. While this finding is entirely congruentwith Sen and Bhattacharya (2001), we specifically iden-tify three ways that the human rights sub-dimensions mayadversely affect a firm: perceived brand equity, purchaseintention, and price willing to pay. In addition, we findthat those who are more conscious of human rights issuesindicate a higher level of motivation to engage in productevaluation. For example, when participants are shown theexperimental stimuli for the first time, response latenciessuggest that consumers with high levels of HR sensitivity(MN = 118.4 seconds) spend more time viewing thestimuli than those consumers with low levels of HRsensitivity (MN = 98.0 seconds). In response to humanrights abuses, these consumers also tend to react morenegatively in their purchase decisions. The cause for theirlowered purchase intention and price willing-to-pay isdue to the harm inflicted upon the firm’s brand equity bynegative human rights ratings. That brand equity fullymediates purchase intention indicates that human rightsperformance in the apparel context is especially impor-tant. Across both levels of HR sensitivity, the effect of ahigh HR score was viewed more favorably by consumersthan the effect of a high overall CSR score. This result andthe overall pattern of findings offer substantial support forour hypotheses.

Our results suggest interesting and useful implica-tions for firms, consumers, and public policy. Althoughthe link between CSR and profitability has been ques-tioned, this article provides evidence that consumers havehigher purchase intentions and are willing to pay more forapparel products that are more ethically manufactured.The resultant “human rights” premium may prove to be asource of company profit if firms are able to uphold a highstandard of ethical labor practice – a consideration that isespecially important for the apparel industry. The primaryimplication for companies is to invest in CSR activitiesthat are part of the dimension (human rights, diversity,community, employee relations, and the environment)most salient and important to their consumers. This find-ing also serves to temper conclusions reached in priorresearch that suggest a uniform reaction to changes inCSR performance. References are available upon request.


For further information contact:Laurel Aynne Cook

University of Arkansas302 Business BuildingFayetteville, AR 72701Phone: 479.575.8748



ENVIRONMENTALLY FRIENDLY CONSUMER BEHAVIOR: A SCALEREVIEW, MODIFICATION, AND VALIDATION

Lynn Sudbury-Riley, Liverpool John Moores University, United KingdomFlorian Kohlbacher, German Institute for Japanese Studies, Japan

Agnes Hofmeister, Corvinus University of Budapest, Hungary

SUMMARY

Despite significant additions to knowledge pertain-ing to contemporary issues in ethical and environmentalmarketing, four notable gaps remain: (1) a paucity ofcross-cultural studies; (2) an absence of research intoseniors, despite their importance; (3) an attitude-behaviorgap, between what people say about ethical consumerbehavior and what they actually do in store; (4) a severeshortage of reliable and validated scales that measureactual – as opposed to intended – ethical and environmen-tally friendly purchasing behavior. A major piece ofinternational research was designed on the basis of thesegaps. The current paper concentrates primarily on thefourth omission. The study used the Ecologically Con-scious Consumer Behavior (ECCB) Scale (Roberts 1996)because it measures actual as opposed to intended behav-ior. However, extensive research and testing across fournations (Japan, Germany, U.K., Hungary) resulted in ascale so different to the original that a new title is sug-gested: The Environmentally Friendly Consumer Behav-ior Scale (EFCB).

The ethical consumer movement was already worth$230 billion by 2000 (Cortese 2003). Ecologically-friendlyproducts and packaging, Fair-trade, Rainforest Alliance,and other ethical goods are now found in mainstreammarkets (Doherty and Tranchell 2007). Yet, there remainsa well-documented ‘attitude-behavior gap’ that is possi-bly due to a lack of perceived consumer effectiveness(Laskova 2007), or the delegation of responsibility to thecorporation (Sudbury and Böltner 2010), or an unwilling-ness to pay a price premium (McEachern et al. 2010).Whatever the reason, positive attitudes toward environ-mentalism do not equate to the purchasing of ethical andenvironmentally-friendly products. Thus, many of thescales available to measure attitudes toward such productsare not good indicators of actual consumer behavior. Atthe same time, research has found socio-demographiccharacteristics to also be poor predictors of ethical pur-chasing (Schlegelmilch et al. 1994), with the exception ofage, where several studies have found older adults to bemore positive than their younger counterparts (DePelsmacker et al. 2006; Trendbüro 2009). Given thatpopulation ageing has emerged as a powerful megatrendaffecting a large number of countries around the world,(UN 2007) and the fact that seniors are becoming an

increasingly important segment (Sudbury-Riley et al.2012), the decision to focus on older consumers (aged 50+was made).

Method

The ECCB scale (Roberts 1996) was chosen becauseit clearly measures a “variety of consumer behaviorswhose intent was to protect the environment” (p. 223).The scale has subsequently been used in many countries.It was therefore with some confidence that the scale wasincorporated into the questionnaire for the current study.Additionally, the questionnaire comprised a selection ofother well-know scales pertaining to attitudes towardethical and environmental practices, the Marlowe-CrowneSocial Desirability scale, an assortment of scales pertain-ing to older consumers, and a battery of questions relatingto their environmental and ethical behavior.

The questionnaire was translated and back translatedby teams in each country, and then piloted to a minimumof 10 members of the population in each nation usingde-briefings (Webb 2002). Across all nations, feedbackrevealed there to be too much similarity and overlapbetween many items, which resulted in several itemsbeing merged. Respondents were confused over somequestions regarding choice and price, thus these wereadapted. Respondents also wondered why there were 2items which related to detergent and aerosols specifically,when there is such a wide choice of environmentally-friendly products available, and these were already cov-ered in the more general questions. These items were alsodropped, on the basis that environmentally-friendly prod-ucts are now widespread, and are more freely-availablethan in the 1990s when the ECCB scale was first devel-oped. Finally, respondents felt there was an omissionregarding other socially-irresponsible behavior such aschild labor and poor working conditions, thus an item wasadded. Some terminology was also amended, e.g., ‘pollu-tion’ was replaced with “environmental damage” andrespondents preferred the term “environmental” to “eco-logical.” These are interesting changes, which appear togo beyond mere semantics or language preferences. Rather,respondents tended to view the term ‘environmental’ to bewider and felt the term ‘ecological’ was too specific andnarrow. Thus many pilot respondents did not make adistinction between ecologically conscious consumer


behavior, and ethical or socially responsible consumerbehavior. Both Roberts (1995) and later Webb et al (2008)note that socially responsible consumption needs con-tinual refinements of its measurement due to its dynamicnature.

This process resulted in an extensively adapted scalecomprising 11 items. Questionnaires were then adminis-tered to representative samples in each country. The finalsample comprised 412 Japanese adults (mean age 64years), 450 U.K. adults (M 66), 213 Germans (M 63), and200 Hungarians (M 59).


The 11 items of the adapted scale were subjected toprincipal components analysis (PCA) with Oblimin rota-tion in all 4 countries individually. One item was removedon the basis of low loadings. The Kaiser-Meyer-Olkinmeasure verified the adequacy of the sample for theanalysis, exceeding the recommended value of .6 (Kaiser1970) in all countries (U.K.: KMO = .898, Germany: .877,Japan: .836, Hungary: .895), Bartlett’s test of Sphericity(U.K.: X2 (45) = 7759.92, p < .001, Germany: 1179.32,p < .001, Japan: 1704.17, p < .001, Hungary: 1387.49,p < .001). The PCA on the 10 remaining items resulted ina single factor solution explaining 60.65 percent of thevariance in the U.K., 54.21 percent in Germany, 44.66percent in Japan, and 60.69 percent in Hungary. The

reliability of the modified scale was checked usingCronbach’s Alpha and item-total correlations and foundto be “exemplary” (Robinson et al. 1991, p. 13). Alphaswere .925 in the U.K., .904 in Germany, .859 in Japan, and.927 in Hungary, revealing high internal consistency in all4 countries. Corrected item-total correlations were allabove .4 and in most cases even above .6. Convergentvalidity was tested by measuring the relationship betweenthe scale and behaviors relating to environmentalism.Theoretically, we predicted that those people who havetaken action about any ethical or environmental issuewould demonstrate higher scores on the scale than dothose who had never taken any action. This predictedoutcome emerged in all countries. Finally, a simple regres-sion analysis showed social desirability effects to be verysmall across all countries (1.2% of the variance in the U.K.sample, 4.3% for Germany, 6.3% for Japan, and 8.2% forHungary).

In terms of construct definition, and because therequired modifications to the original ECCB scale were sosevere, we suggest a new scale has emerged: the environ-mentally friendly consumer behavior scale. It has severaladvantages over its predecessor, in that it is shorter andeasier to administer, it uses the terminology preferred bypeople today, it has been shown to be reliable and valid infour nations, and a potential source of bias in the form ofsocial desirability is so small it can be considered to beeliminated. References are available upon request.

For further information contact:Lynn Sudbury-Riley

Liverpool John Moores University,98 Mount PleasantLiverpool, L3 5UZUnited Kingdom

Phone: +44.151.2313671E-Mail: [email protected]


AFRO-DESCENDENTS IN CHILDREN’S TELEVISIONADVERTISEMENTS: A CONTENT ANALYSIS STUDY

IN THE BRAZILIAN CONTEXT

Claudia Rosa Acevedo, Universidade de São Paulo/Universidade Nove de Julho, São PauloMarcello Muniz, Universidade de São Paulo, São Paulo

Jouliana Jordan Nohara, Universidade Nove de Julho, São Paulo

SUMMARY

Investigations related to representations of Afro-descendants in the media in Brazil are welcome because,in spite of this group constituting 50.7 percent (being thatthe Mulattos are 43.1% and Negros 7.6%) of the popula-tion of the country (Varella 2011), most studies haveshown that compared to the composition of the popula-tion, it is still, percentage wise, little portrayed in themedia (Araújo 2000; Barbosa 2004; Guimarães 2004).

In general, the investigations show that the rolesassociated with the Afro-descendants are impregnatedwith social stigmas (Barbosa 2004; Carvalho 2003;Rodrigues 2001). The analyses of speech in the mediareveal depreciation of that ethnic group. In fact, researchon the subject has identified that the new stigmatizationstrategies of Afro-descendants in the media are moresubtle and complex. Such forms of discrimination havebeen called both as neo racism as well as, Brazilian racism(Silva and Rosemberg 2008). Studies on the subjectpointed out that the messages in the media reflect theracism that is ingrained in Brazilian society (Araújo 2000;Carone and Bento 2003; Carvalho 2003; Rodrigues 2001).It is on this context of concern that the present investiga-tion is founded. The goal of the research is to examinewhat, and how, the representations of Afro-descen-dants in advertisements are directed at children.

The vast majority of research indicates that Afro-descendants are still under-represented in comparison totheir proportion in the population. Such investigationshave also shown that these speeches are impregnated withracist ideologies (Barbosa 2004; Bowen and Schmid1997; Domingues 2002; Rodrigues 2001; Rosemberget al. 2003; Roso et al. 2002). In relation to the importanceof the roles played by different ethnic groups, most studieshave revealed that, in general, Afro-descendants are play-ing, either secondary roles or as extras (Bowen andSchmid 1997; Bristor et al. 1995; Domingues 2002; Haeand Reece 2003; Licata and Biswas 1993; Mastro andStern 2003; Pinto 1987; Seiter 1990). The investigationsalso show that generally, the interactions between Afro-descendants and whites involve work or business situa-tions. Interactions are rarely found concerning familyenvironments. It was also noted that it is more common

that the interactions occur between children of the twoethnicities, or between children or Afro-descendant ado-lescents and white adults. The research also revealed thatthere are few scenes in which Afro-descendants andCaucasians interact (Bowen and Schmid 1997; Hae andReece 2003; Taylor et al. 2005; Taylor and Stern 1997).From the literature review this study presents the follow-ing assumptions: (H)

1: In comparison to Caucasians,

Afro-descendents are under-represented in relation totheir composition in the population; H

2: In comparison to

Caucasians, Afro-descendants tend to be less representedin leading roles; H

3: In comparison to Caucasians, Afro-

descendants tend to be represented in non-family inter-relationships; (H)

4: In comparison to Caucasians, the

Afro-descendants tend to be represented more as childrenand teenagers than as adults or elderly.

The data were analyzed using the technique of con-tent analysis as proposed by Berelson (1952) andKassarjian (1977). The analysis period in this study wasfrom 2002 to 2010. The unit of measure is commercialtelevision (advertising). The research universe has beenestablished as television advertisements that advertiseproducts for children between six and twelve years of ageand have human beings as characters. To compose thesample, the following products were defined as clippings:toys, footwear, clothing, and food. From these criteria 503advertisements which used human characters were iden-tified. The commercials for this research were selectedfrom the database of the company “Propaganda Arquives,”a company which has systematically filed archives ofBrazilian advertisements since 1972. Data were analyzedusing qui square statistical technique. Of the 503 adver-tisements analyzed, only 86 (17%) had Afro-descendants.For this reason, only 86 advertisements that contain Afro-descendants were studied. It was identified that among the86 advertisements there were a total of 913 characters. Ofthese, 700 (76.7%) were Caucasian and 173 (18.9%) wereAfro-descendants.

The composition of Caucasians and Afro-Brazilianpopulation is respectively of 47.7 percent and 50.7 per-cent (being that the Mulattos are 43.1% and Blacks 7.6%).Thus, comparing the data, one can see that the proportionof Caucasians in the population (47.7%) is less than inadvertisements studied (76.7%) and the percentage ofAfro-descendants is larger in the population (50.7%) and


lower in the advertisements (18.9%). In other words, theadvertisements do not reflect the reality of the composi-tion of ethnic groups in Brazilian society. This way, thefirst hypothesis was confirmed. In relation to the impor-tance of the roles, the results of the survey showed thatCaucasians have been featured in leading roles morefrequently than other ethnic groups. Of all 913 characterspresented in analyzed advertisements, 20.7 percent areCaucasian in leading roles, while only 3.7 percent of Afro-descendants occupy these roles.

In relation to the interaction between the characters,the relationships mostly represented are social interac-tions. The ethnicity that appears more in family interac-tions is Caucasian. 6.6 percent of the family interactionsare represented by Caucasians, while for the Afro-descen-dant that number is 1.1 percent. Caucasians accounted for8.6 percent of family interactions. However, in this type ofinter-relation the Afro-descendants make up just 5.8 per-

cent. However, in working relationships Afro-descen-dants are 9.8 percent and 1.4 percent are Caucasian. It isinteresting to note that the percentage of Afro-descen-dants that appear alone is greater than the percentage ofCaucasians.

In relation to the age of the characters (see Table 4)one can see that regardless of the ethnicity, most of thecharacters are children (52.1%, 11.7% Caucasian andAfro-descendant). This can be explained by the fact thatthe advertisements are targeted at children. Little morethan twenty percent (20.6%) of the adults in advertise-ments are Caucasian and only 6.9 percent of them areAfro-descendants. However, it appears that there is noelderly Afro-descendant, and that there are only threeteenagers of that ethnicity. Among the Caucasian, thesegroups make up 1.1 percent and 2.8 percent respectively.Hypothesis 4 was also confirmed. References are avail-able upon request.

For further information contact:Claudia Rosa Acevedo

USP (Universidade de São Paulo) / UNINOVERua Duarte da Costa, 1170São Paulo, SP, 05080-000

Phone: 5511.38342222E-Mail: [email protected]


HOW COMPANIES SHOULD REACT ON SOCIAL MISCONDUCTS:THE ROLE OF IN- AND EXTRA-DOMAIN COMPENSATION

Isabell Lenz, Georg August University Goettingen, GermanyHauke Wetzel, Georg August University Goettingen, Germany

Maik Hammerschmidt, Georg August University Goettingen, Germany

SUMMARY

Most firms encounter social misconducts, given thefact that it seems impossible to monitor every course ofaction and business process. Particularly, when firm’soperations are spread around the world, social miscon-ducts are often uncontrollable and thus, unpreventable.Importantly, along with nongovernmental organizations’eagerness to reveal firms’ misbehavior, shareholders areincreasingly sensitive to reported negative firm behavior(Wagner, Lutz, and Weitz 2009). As a consequence,shareholders respond with resentment, leading to rigor-ous harm for firm value. In order to compensate societyfor misconducts and to win the goodwill back, firms canengage in favorable corporate social responsibility (CSR)activities (e.g., pollution prevention, charitable giving) indifferent domains (e.g., environment, community).

However, firm’s reactions to social misconducts dif-fer strongly. Consider the examples of Nike and Coca-Cola. Nike has been associated with relocating productionto factories which employed labor under poor workingconditions and minimum wage. Nike’s reaction to offsetthe potential damage of firm value was a direct one:predominately Nike improved labor conditions. In addi-tion to resolve the problem directly, Nike reformed trans-parency in disclosures and implemented monitoring withrespect to labor’s working condition. Finally, all theseinitiatives were communicated to shareholders. In con-trast, Coca-Cola drew negative attention with beveragesthat were contaminated with pesticides – a social miscon-duct which represented a serious threat to Coca-Cola’sfirm value. Coca-Cola also eliminated the source of nega-tive attention. However, as opposed to Nike, in its com-munications Coca-Cola did not broach the social miscon-duct itself to avoid spurring further reactions to it. RatherCoca-Cola chose an indirect approach to reestablish good-will via charitable giving programs.

As the examples shall demonstrate, firms face differ-ent strategies to compensate for social misconducts. WhileNike built on favorable CSR activities in the domain of thesocial misconduct (i.e., in-domain compensation), Coca-Cola rather reacted on product related concerns withengagement in the community domain (i.e., extra-domaincompensation). However, it is unclear which approach ismore promising in order to safeguard firm value.

While the literature emphasizes that “engaging in theright CSR-initiatives enhances firm performance” (Luoand Bhattacharya 2009, p. 198), the urging questionremains what constitutes the “right” initiative. Yet, cur-rent literature on CSR has not provided an answer on thisquestion. By and large, this gap is ascribed to researchscholars’ disregard of multiple compensation strategies.While providing seminal insights, prior studies eithercommingle different options of compensation (e.g., Luoand Bhattacharya 2009) or concentrate on consumer-related impacts of a single CSR activity (e.g., Sen andBhattacharya 2001). In doing so, research has shown thatCSR can generally help to reap more financial benefits(Luo and Bhattacharya 2009). However, managers needto know the effects of different compensation strategieson firm value in order to allocate resources most effi-ciently. Against this background, we are interested inanswering whether and when certain compensation strat-egies are appreciated more by shareholders than others.

In an attempt to answer these questions, we contrib-ute to literature in several ways. First, we contribute toCSR research by contrasting different compensationefforts. Specifically, we introduce the concepts of in-domain compensation and extra-domain compensation tocapture the two possibilities that companies might use toreact to social misconducts. While in-domain compensa-tion refers to firms’ favorable CSR activities in the domainof social misconduct, extra-domain compensation cap-tures favorable CSR activities in other domains. Further,we extend current research by investigating the impact ofin-domain compensation and extra-domain compensa-tion on firm value. To our knowledge, we are the first tomove toward such a “finer-grained quest” (Luo andBhattacharya 2009, p. 199). Third, we examine howlevers can be used to leverage the impact of compensationstrategies on firm value. In doing so, we show that itdepends on firms’ compensation strategy whether thespecific lever (i.e., advertising or R&D) has an advanta-geous leveraging effect or not.

We test our framework with longitudinal data thatcombines metrics on CSR performance and financialperformance for the majority of S&P 500 firms. Theresults of linear mixed models show that extra-domaincompensation is capable to stimulate firm value meaning-fully, whereas in-domain compensation backfires andhurts firm value. Furthermore, our findings suggest that


the effects of extra-domain compensation are even morepronounced in presence of high advertising and R&Dspending, while in-domain compensation mitigates firmvalue for firms with high advertising and high R&Dspending. In sum, our findings indicate that the moderat-ing role of levers strongly depends on the compensationstrategy implemented.

In sum, our findings argue for extra-domain compen-sation as an unobtrusive and promising strategy for

rebuilding shareholders’ goodwill after social miscon-ducts, whereas in-domain compensation should be takenwith care. For managers, our results highly suggest thatCSR initiatives must be chosen with prudence in order tosuccessfully compensate for social misconducts. For stud-ies commingling compensation efforts in one construct,our findings indicate that results might be biased depend-ing on whether in-domain compensation or extra-domaincompensation is dominant in the overall construct. Refer-ences are available upon request.

For further information contact:Isabell Lenz

Economic SciencesGeorg August University Goettingen





HOW BIG A CONCERN? PRIVACY, TARGETED ADVERTISING ANDSOCIAL MEDIA: THE CANADIAN CORPORATE

AND CONSUMER VIEW1

Avner Levin, Ryerson University, TorontoBettina West, Ryerson University, TorontoMary Foster, Ryerson University, Toronto

SUMMARY

Using the experiences of advertisers who want toreach consumers, and consumers who may or may not bereceptive to receiving advertising messages, this paperinvestigates the intersection of the opportunities pre-sented by targeted online advertising and the privacy andsecurity concerns related to the use of personal informa-tion.

We used both qualitative (in-depth interviews, focusgroups, and town hall) and quantitative (survey research)methods to gain insight into behavioral advertising fromboth the consumer and the corporate perspective. Using asemi-structured format, we conducted six -depth inter-views with senior executives representing advertiserswho promote their goods and services in an online envi-ronment and with publishers who provide space or act asbrokers for space for online advertisements and ad net-works. We conducted two focus groups with consumersin the 18 to 30-year-old age range to gain insight into useof the Internet related to shopping behavior and attitudesand opinions about targeted online advertising. An audi-ence of 60 “town hall” observers for each focus groupadded comments and observations to enhance our under-standing of how young consumers operate in the digitalspace. Finally, we conducted an online survey aboutonline attitudes and behaviors, resulting in a final sampleof 1317 of undergraduates in a large, culturally, urbanCanadian university.

Privacy is not a major concern among corporaterespondents as they believe they are already compliantwith the existing regulatory environment. Further, theyview compliance as a competitive advantage as discuss-ing privacy considerations may uncover other issues thatneed to be addressed before launch. Consumers do notpresent coherent attitudes and behaviors with respect toonline advertising and privacy. While purporting to ig-nore or dislike all forms of advertising, they report click-ing on ads, and are not willing to pay to avoid them. Whilerespondents display some privacy concerns, these are notdeveloped enough to make them change their onlinebehavior.

These results underline the policy dilemma faced byagencies such as the Office of the Privacy Commissioner.While consumers talk the privacy talk, they do not walkthe privacy walk. The advertising industry does not seethe need for additional regulations on targeted onlineadvertising and while consumers value privacy, they donot take steps to protect their information. Instead theyseem to see privacy as a right and expect government toprotect it, rather than taking the appropriate protectivesteps themselves. References are available upon request.

ENDNOTE

1 The research on which this paper is based was funded bythe Office of the Privacy Commissioner of Canada.

For further information contact:Avner Levin


Toronto, ON M5B 2K3Phone: 416.979.5000



DEVELOPING A CORPORATE SOCIAL RESPONSIBILITY PROCESSSCALE OF INDIVIDUAL STAKEHOLDER’S PERCEPTION

Seongjin Kim, University of Seoul, KoreaClarissa Chaiy, Northwestern University, Evanston

Seoil Chaiy, Korea University, Korea

ABSTRACT

Most measurement of corporate social responsibility(CSR) focus on measuring only the final outcome of CSRactivities. Present research argues that because existingscales do not incorporate motivation or execution processinto measuring CSP, it is difficult to clearly measure theimpact of CSR on corporate performance. To mitigate thiscritical limitation in existing CSP scales, present researchproposes a new 45-item scale that incorporates motiva-tion, process and performance evaluation of CSR activi-ties to measure the performance of CSR activities. CSRmotivation is categorized into economic and profit-seek-ing dimensions, CSR process into legal and ethical execu-tion and CSR into maximizing profit, employee welfareand social wealth to yield an integrated process of modelevaluating CSR.

INTRODUCTION

Corporate social responsibility (CSR) has enjoyedrecent surge of scholarly attention in various managementfields. For many years, by juxtaposing measurements ofCSR activities with corporate performance, scholars havebeen trying to identify if a causal relationship existsbetween them. However past twenty years of researchindicates that the relationship is ambiguous at best. Thereason for this is because a firm’s CSR-related invest-ments can generate two distinct types of CSR perfor-mances. First type of CSR performance is the socialimpact caused by the firm’s CSR investments. Forinstance, if a firm invested in reducing childhood hunger,its performance should be measured by how many chil-dren were actually relieved of hunger as well as how muchthe issue itself has been resolved at a societal level. Basedon this measurement of social impact from CSR, corpo-rate performance from CSR such as reputation, brandasset and loyalty, are induced. However, because mostCSR research does not take into account the social impactside of CSR performance and attempt to link CSR invest-ments with corporate performance directly, the relation-ship between the two has been shown to be different byeach study. Therefore, it is necessary to measure and takeinto account the social impact of CSR to accuratelymeasure the impact of CSR on corporate performance.Unfortunately, social impact from CSR investments is

tough to measure objectively. There is no definitive set ofmetrics to measure how much CSR contributed resolvinga given social issue. Moreover, what really influences theCSR performance is not the actual social impact but thestakeholders’ perception of the social impact. Even if theCSR investment did in fact lead to significant socialimpact, if the stakeholders are not cognizant of this, thenthe CSR investment has failed to contribute to corporateperformance in any way. Therefore, measurement basedon the stakeholders’ perception of social impact fromCSR rather than just the social impact itself, needs to bedeveloped in order effectively to test the relationshipbetween CSR and corporate performance.

While measurement tools on CSR and its socialimpact do exist, they are not without flaws. First of all,many of these scales are developed with focus on specificissues. These scales are focused issues that include, butare not limited to, poverty, gender, workplace diversity,racial equality, and the environment. These scales that aredeveloped to measure corporate performance in specificareas are not adequate in measuring the wide range of CSRactivities of today and also pose a significant threat togeneralizability. Some studies evaluate CSR by analyzingthe public relations or publicly released information fromthe corporations. However, these documents released bythe corporations are limited in that they typically includethe firm’s CSR investments but often do not include theactual social impact derived from those investments.Therefore, while these documents may be ore objective,they are limited in the measurement of aforementionedsocial impact from CSR. Lastly, existing scales of CSRmeasurement are focused on CSR performance evalua-tion. According to Attribution Theory, when faced withunexpected or unpredicted information, instead of simplyaccepting this new information at face value, consumerswill try to deduce the reason and the motivation behind it.Attribution theory refers to when the motivations forindividuals’ actions are attributed to an external circum-stance or fundamental characteristics. That is, it is aprocess in which words and actions are observed todeduce their motivations. Attribution is also exhibitedunder unusual or abnormal circumstances continue for anextended period of time, such as when consumers arefaced with CSR-related information that run counter tocorporations’ profit-seeking behavior.


CSR is typically understood by as external stakehold-ers as a non-profit seeking activity that is far from thegeneral profit-seeking nature of a corporation. That is whywhen stakeholders are exposed to CSR-related informa-tion or messages, they question its underlying motivationand seek other related information and aggregate them tosubjectively evaluate the CSR activity, rather than solelyrelying on the provided information to objectively evalu-ate it. This is to say that regardless of how much socialimpact a given CSR investment yielded, if it is perceivedby the stakeholders as having ulterior motives or utilizedillegal or unethical methods in its execution, the stake-holders evaluation of CSR will suffer. For instance, let’sassume that a corporation invested in a CSR activity andhad positive outcome for the society at large. However ifthe resources used in this CSR activity was acquiredthrough illegal or unethical channels or if the corporationwas seen to have financial motivation for the CSR activ-ity, the stakeholders’ evaluation of the CSR activity willdeteriorate. Therefore, it is necessary to include evalua-tion of both motivation and results in measuring CSRperformance. Our research derives an integrated CSRprocess model that takes into account CSR motivation andactivity with CSR performance and develops a measure-ment scale that evaluates the integrated process. We firstbriefly review the existing literature on CSR, criticallyreview the existing CSR measurement scales and developa new scale based on Churchill’s (1973) scale develop-ment model.

LITERATURE REVIEW

Existing Literature on CSR Evaluation Scale

Maignan’s (2001) CSR scale for consumers is one ofthe most comprehensive CSR scales for external stake-holders that are faithful to the definition of CSR. Based onCarroll’s (1979, 1991) definition of CSR and four catego-ries of social responsibility, Maignan developed 16 scaleitems. Using four scale items each in Carroll’s classifica-tion of economic, legal, ethical, and philanthropic respon-sibilities, Maignan tested the scale in French, German andAmerican context. However Maignan’s scale lacks itemsthat measure the motivation behind why corporationsinvest in CSR. While the scale has items that measurelegal and ethical responsibilities, it is limited in that it ismeasured in the same dimension as economic and philan-thropic responsibilities. This is because economic andphilanthropic responsibilities items measure the results ofCSR activities whereas legal and ethical responsibilitiesmeasure the motivation and the process of the CSRactivities. Therefore legal and ethical responsibilitiesshould be conceptualized as preceding factors that influ-ence measurement of economic and philanthropic respon-sibilities.

In 2005, David, Kline, and Dai developed a new scaleaccording to their classification of CSR activities intofollowing dimensions: relational, moral and philanthropic.This scale also only focuses on the execution of CSR anddoes not address the motivation behind CSR activities.The scale also does not address evaluation of the legal andethical component of the CSR execution, which makes itdifficult to evaluate the overall process of CSR. More-over, other scales developed by Abbott and Monsen(1979), Brown and Dacin (1997), Davenport (2000),Mitnick (2000), Mattingly and Berman (2006) share sev-eral common limitations. First, many of these scales focuson measuring ethical or philanthropic activities, whichdoes not fully reflect the definition of CSR. Secondly,because these scales measure specific categories of CSRactivities, they do not cover comprehensive range of CSRactivities nor evaluate process-related factors such aslegal and ethical components of CSR execution. There-fore to mitigate these limitations, present study aims todevelop a new scale of CSR process evaluation thatfollows the proposed definition of CSR.

Existing Research on CSR Process Model

Wartick and Cochran (1985) and Wood (1991) pro-posed frameworks for firms to manage CSR. Both of thesestudies share common ground in that they categorize acorporation’s social responsibilities into a three-step modelincluding principles, processes and programs and policiesbut they differ in their content. Wartick and Cochran’s(1985) framework is based on the aforementioned prin-ciples, processes and policies steps. Principles are catego-rized by Carroll’s (1979) economic, legal, ethical andphilanthropic elements. Process is categorized into reac-tive, defensive, accommodative and proactive elementsdepending on how CSR is executed and policies thataddress the issues derived from these elements are institu-tionalized. On the other hand, Wood (1991) argues thatWartick and Cochran’s principles of CSR are not prin-ciples but actions and that their proposed policies are toorestrictive to incorporate all of the results from CSRactivities. Furthermore, Wood criticized that Wartick andCochran’ model is closer to a preparation for CSR activi-ties rather than am actual CSR management for a corpo-ration.

Building on these criticisms, Wood (1991) proposeda different three-step model of CSP composed of prin-ciples, processes and outcomes of corporate behavior.The model proposes different principles for institutional,organizational and individual levels. Individual principlespeaks to managerial discretion, organizational level topublic responsibility and institutional level to legitimacy.Based on these principles, environment assessment, stake-holder management, and issues management takes place


to reach outcomes of corporate behavior. Corporate be-havior is also then divided into social impacts, programsand policies.

However, Wood’s (1991) model is not without flaws;first of all, the three types of CSR processes proposed inthe model cannot be seen as distinct processes, but ratherdifferent terminology stemming from various CSR per-spectives. In fact, the term issues management is oftenused interchangeably with CSR. Also, social programsand policies, which are classified as outcomes of corpo-rate behavior, are difficult to conceptualize as resultsbecause they are closer to fundamental elements of CSRactivities. Therefore both Wood (1991) and Wartick andCochran’s (1985) models are limited in effectively man-aging CSP. Finally, these models have not been testedempirically, which limits their corporate application.

Not only does conceptualizing CSR as a processmodel provide insight into how to effectively manageCSR from a corporate perspective, it also provides astandard in how external stakeholders evaluate CSP.Stakeholders evaluate key issues in each CSR process,which contribute to the evaluation of the final CSR result.Therefore in order to construct an accurate CSP measurefor external stakeholders, it is important to clearly concep-tualize the CSR process and develop an appropriate scalebased on the process model.

New CSR Process Scale Development

We strictly followed Churchill’s (1979) 8-step para-digm for developing a new scale. To clarify the definition

of CSR, we conducted an extensive literature review togenerate a new definition of CSR. Based on the newdefinition, we defined and categorized each of the subse-quent elements of CSR. Based on these elements, weconducted another literature review to derive a pool ofsurvey items. Because most of the existing items wereoptimized for western cultures, we conducted a set ofinterviews with Korean stakeholders to verify and adjustpreviously items derived from literature review. Aftersupplementing the items to reflect Korean stakeholders’understanding, these survey items were purified througha refining process and the final scale was tested forreliability and validity with a new sample.

Conceptual Definition of CSR

Bowen (1953) defines CSR as “the obligations ofbusinessmen to pursue those policies, to make thosedecisions, or to follow those lines of action which aredesirable in terms of the objectives and values of oursociety.” This definition of CSR is accepted as the firstacademic definition of CSR and many definitions ensuedfollowing Bowen’s footsteps. Table 1 summarizes andcompares the key scholarly definitions of CSR. Despitemore than fifty years of research, scholars have yet toagree on a single definition of CSR. This divergence is dueto the different perspectives in approaching CSR. Fried-man, a classic economist, famously argued that the onlycorporate responsibility toward society is to maximizeshareholders’ profits as bound by law and ethics andpromoted shareholder primacy (1962). In contrast, Davis,a neoclassical economist, defined CSR as “an obligationto evaluate in its decision-making process the effects of its

TABLE 1Comparison of Key CSR Definitions

Friedman Davis Carroll Petkoski and Twose(1962) (1973) (1979) (2003)

Goal Profit generation Social benefits and Fulfilling social Improving quality of lifeeconomic profits expectations

StakeholdersRange Shareholders Social System (Stakeholders) (including future

generation)

CSR domains N/A N/A Economic, Legal, N/AEthical, Philanthropic

Limitations Narrow range Only focused on Overlap among Narrow range ofof corporate the effect of CSR domains corporate activities

responsibility corporations’decision-making


decision on the external social system in a manner that willaccomplish social benefits along with the traditional eco-nomic gains which the firm seeks” (1973). Davis’ defini-tion extends the responsibility of the corporation than thatof Friedman (1962) to the social system, but these defini-tions share the assumption that the corporation’s profitsare the priority.

While various perspectives yield differing defini-tions of CSR, Carroll’s CSR definition and pyramid(1979, 1991) is a widely accepted definition that concep-tualizes CSR based on social economic perspective andstakeholder management. Unlike other definitions of CSR,Carroll identifies four areas of CSR in his definition. Hedefines CSR as “the social responsibility of businessencompasses the economic, legal, ethical, and philan-thropic expectations that society has of organizations at agiven point in time” (Carroll, 1979) and articulates thateconomic, legal, ethical and philanthropic areas exist forCSR. According to Carroll, economic responsibility refersto making profit by producing goods and services that thesociety wants. Legal responsibility refers to “fulfill[ing]its economic mission within the framework of legalrequirements” (Carroll, 1979) and ethical responsibilityrefers to conducting business in ethical fashion. Finally,philanthropic responsibility refers to fulfilling the society’sexpectations for “businesses to assume social roles” toimprove society at large (Carroll, 1991). However,Carroll’s definition of CSR carries some limitations, withfirst being that it is difficult to clearly differentiatebetween ethical and philanthropic responsibilities.Schwartz and Carroll (2003) criticized that the tworesponsibilities overlap significantly and that they shouldbe unified as one. The second limitation is that Carroll’sdefinition views CSR from a passive and defensive pointof view. In his definition, Carroll (1979) conceptualizesCSR as fulfilling the society’s expectations, which impliesthat corporations are not responsible for actions that theysociety does not expect of them. However, a corporationtruly committed to its social responsibilities should beable to take on issues even when it is not expected of them.That is, while it is important for corporations to resolveexisting social issues, it is also important to prevent socialissues from arising. The third limitation is that Carroll’sfour areas of responsibilities are insufficient in measuringCSR. Scholars including Maignan and Ferrell (2002,2003) developed CSR scales based on these four areas,however, these scales do not measure motivation. More-over, while economic and philanthropic responsibilitiesare related to specific areas of CSR but legal and ethicalresponsibilities are related to the execution method ofCSR. Also, the four areas of CSR are not mutuallyexclusive or all inclusive which further adds to its limita-tions.

Present research aims to overcome the limitations ofexisting CSP scales by first deriving a new, integrated

definition of CSR. To do this, we first categorized thecommon factors between the existing definitions accord-ing to the 5W’s and 1H. First of all, CSR is without doubtconducted by the corporation (who). Second, CSR is oftenexecuted in communities and regions in which corpora-tions conduct their business activities (where). Third,CSR is conducted in the present (when). Different socialissues arise in different temporal perspectives, requiringdifferent types of social responsibility and executionmethods. Therefore, it is most important for corporationsto find the most relevant social issue at the given momentand address it. Fourth, objective of CSR is to maximizewelfare and profit for the society, including the corpora-tion (why). While some argue that corporations’ seekingprofit cannot be classified as part of CSR, since corpora-tions themselves are also members of the society, weargue that increasing the wealth of the corporations can beconsidered as part of CSR. Fifth, the areas of CSR (what)should be identified as fulfilling the economic and non-economic needs and expectations. As aforementioned,Carroll (1979) defines the areas of CSR as fulfilling thesociety’s expectations, however it is more appropriate toinclude both needs and expectations because corporationsshould seek social responsibility proactively and addressissues that the society has yet to expect of corporations.Also, in summation of the variety of definitions on CSRareas, the fundamental areas for corporate activities areclassified into economic and non-economic areas, and thesocietal expectation of the corporations also include botheconomic and non-economic areas. Therefore CSR areasshould include both the society’s economic and non-economic needs and expectations. Finally, executing CSR(how) should be within the limits dictated by the legal andsocial norms. This is because no matter how great a CSRactivity is, if it breaks the law or the social norms, it isbound to be criticized by the society. Summing thesecomponents into one statement we derive the followingdefinition of corporate social responsibility:

“Corporate social responsibility is a present obliga-tion of a corporation that meets economic and non-economic needs and expectations of stakeholderssuch as investors, employees, consumers, suppliers,government, and society in order to improve welfareof society including a corporation itself within thelimit of laws and social norms.”

Then how can corporations manage CSR effectively?The answer to this lies in the most fundamental manage-ment process. In order to effectively manage CSR, theremust exist an accurate evaluation of the entire CSR-related process. Therefore, developing a clear frameworkfor CSR evaluation yields a framework for effective CSRmanagement. Based on the most fundamental manage-ment process of planning, executing and evaluation, presentresearch aims to identify the key factors for evaluation ineach CSR process to develop a scale that can accuratelymeasure them (see Figure 1).


FIGURE 1Conceptual Adaptation of Management Processes to

Create CSR Evaluation Framework

Based on the aforementioned definition, key CSRevaluation factors in each process can be categorized intothree dimensions. The first dimension is the CSR motiva-tion. According to Becker-Olsen et al. (2006), whenconsumers are faced with a corporation’s CSR-relatedinformation, they attempt to predict if the motivation forCSR was profit motivated or socially motivated. Profitmotivation refers to when corporations engage in CSR asa tool to generate more profit while social motivationrefers to when corporations engage in CSR as part ofenhancing the wealth and welfare of the entire society. Itshould be noted that neither motivation is “better” than theother. Similarly, van de Ven and Graafland (2006) alsodifferentiate between strategic and moral motivation forCSR. Strategic motivation for CSR is similar to the profitmotivated CSR of Becker-Olsen et al. (2006) in that acorporation is motivated by utilizing CSR to increaselong-term financial performance, and moral motivation issimilar to social motivation in that a corporation have amoral duty to toward society. Together, a corporation’smotivation for CSR can be differentiated as having eitheran economic motivation that utilizes CSR as a strategictool in seeking profit or a social motivation that aims toenhance the wealth and welfare of the society.

The second dimension for evaluation addresses spe-cific CSR activities. In this dimension, individual stake-holders evaluate CSR based on whether it was executedwithin the confines of law and social norms. Following

Carroll’s (1991) definition of legal responsibility, acorporation’s legality can be defined as complying withthe regulatory system as elected by institutions. On theother hand, scholars disagree on ethical corporate activity.Consequentialists, following the tradition of Bentham andMill, argue that achieving good results is the morally rightand ethical behavior while deontologists, following thetradition of likes of Kant, define ethical behavior hashaving ethical motivations and process. Therefore wesummarize these perspectives to define ethical behavioras ethical motivations and process that lead to goodresults.

Finally, the last dimension of evaluation for indi-vidual stakeholders is the outcome of CSR. The CSRoutcome can be differentiated into economic and noneco-nomic outcomes. Schwartz and Carroll (2003) defineeconomic component of CSR as actions motivated bydirectly or indirectly maximizing profit. Therefore eco-nomic out come CSR can be understood as the outcome ofeconomic behavior and individual stakeholders will likelyevaluate on to what extent corporations maximized theirprofits. On the other hand, noneconomic outcome istypically understood as philanthropy. Carroll (1991) de-fines philanthropic responsibility as utilizing resourcesfor the community and enhancing the welfare. Thus,noneconomic outcome of a corporation refers to theoutcome of CSR activities except for its economic out-comes.


Scale Item Development

Based on the literature review conducted for devel-oping a new CSR measurement scale of individual stake-holders, Maignan’s (2001) scale of consumers emerged asthe most comprehensive existing scale. Also, because itwas developed under diverse cultural settings, we drewmany items developed in this scale. However, becauseMaignan’s scale only consists of items that measure CSRperformance and CSR activity, we collected additionalitems through interviews to also include CSR motivationmeasurement in the proposed scale. During these inter-views, we first presented our definition of CSR to theinterviewees and subsequently asked open-ended ques-tions on the corporate motivation for CSR and what wouldbe an effective CSR effort. On CSR motivation, werecorded a total of 225 items and 21 unique items. Theseunique items were classified as either profit motivated andsocially motivated. On CSR activity and performance, werecorded a total of 327 items, which were reduced to 21unique items and 16 items identified from literature re-view. These 37 items were then classified as legal, ethical,economic outcome and noneconomic outcome.

Measurement Item Refinement

To refine the 58 measurement items derived fromliterature review and interviews, we conducted a survey

on a sample of 238 Business-major undergraduate andgraduate students in Seoul, Korea. Among the sample,148 students did not have occupations other than being astudent, 88 had occupations in corporations or govern-ment agencies and two reported “other.” Eighty-one stu-dents answered that they have invested in corporation insome capacity and 157 students answered that they haveno corporate investments.

First, we conducted reliability analysis to test for thereliability of our measurement items. The analysisrevealed that five items in the profit-motivation decreasedthe reliability and were consequently removed (0.794 to0.828), and one item was also removed from social-motivation because it decreased the reliability (0.885 to0.895). Two items were each removed from legal dimen-sion (0.812 to 0.846) and economic outcome (0.882 to0.889) for the same reason. For the remaining 48 itemsafter the reliability analysis, principle component analysis(PCA) and exploratory factor analysis with Equimaxrotation was conducted. During this process, three itemsthat were grouped to irrelevant factor (1 in economicoutcome, 2 in social motivation) were removed.

From the refining process, we were able to identify atotal of seven factors and 45 items. Results of factoranalysis indicates that the items on CSR motivation andactivities were divided between profit motivation (dimen-

FIGURE 2Multi-Level Conceptualization of CSR Constructs


TABLE 2Final CSR Process Measurement Scale Items

To enhance corporate imageTo promote the corporation

Economic To advertise the corporationMotivation To provide service for prospective consumers

Part of public service marketingTo create good impression

CSR To achieve long-term successMotivation To contribute to the society’s development

To donate to various social causesTo fulfill thecorporations’ responsibility to the society

Social To maintain its conscienceMotivation To repay consumers’ loyalty

Part of noblesse obligePart of natural duty as an entrepreneurTo fulfill the duty of corporations

Always accepts regulations set forth by laws and institutionsDo not break laws for the sake of increased profit

Legal Try its best to uphold all contractsCSR Always abide lawsExecution Protect employees’ rights according to lawMethod Manages ethically

Promotes ethical management even at the expense of losing profitEthical Has a set of well-developed ethical standards

Prioritizes ethical standards above profitDo not violate ethical standards to achieve goals

Generates profitDevelops new productsAdvancing into international market through globalization

Profit Plans for long term successMaximizing Maximizes profit

Produces better products and servicesContributes to increasing national competitivenessIncreasing in investmentContinuously improving economic performanceImproves employee welfare

Employee Guarantees employment for its employeesCSR Welfare Guarantees fair competitionOutcome Improves employee welfare

Supports cultural and/or sports industriesHas a scholarship programDonates sufficiently

Social Committed to increasing the welfare of the society as a wholeWelfare Allocates resources to philanthropic activities

Actively participates in public policyPlay a role in our society that goes beyond the mere generation of profits


sion 6) and social motivation (dimension 1), and legal(dimension 7) and ethical (dimension 4) as expected,however items on CSR performance were split into threefactors. We found performance related items were groupedto profit-seeking item (dimension 6), employee welfareitem (dimension 5) and social welfare item (dimension 2).Analysis on the items characteristics of each factor showedthat while profit-seeking items were consisted of itemsthat are mainly of interest to individual stakeholders whowere investors, employee welfare items were consisted ofitems that are mainly of interest to individual stakeholderswho were employees. Similarly, social welfare itemsconsisted of items mainly of interest to individual stake-holders who are consumers.

Reliability and Validity Test

To test for the reliability and the validity of the newlydeveloped scale, we conducted an additional survey. Thesample included a total of 320 participants and afterremoving 14 incomplete surveys, total of 306 answerswere included in the analysis. Reliability analysis byfactors yielded Cronbach’s alpha or composite reliabilityof greater than 0.8 for all factors, thus ensuring internalconsistency. To test for convergent validity and discrimi-nant validity, confirmatory factor analysis was conductedon the 45 refined items in 7 dimensions. Since presentresearch conceptualizes CSR as consisting of CSR moti-vations, CSR activity process and CSR performance items,we conducted a bi-level confirmatory factor analysis. Thefitness of the model was revealed to be acceptable (χ2 =1691.642, d.f., = 912, p = 0.000, ROSEA = 0.0492, GFI =0.824, CFI = 0.970, RMR = 0.104).

All t-values were significant in convergent validitytest, thus yielding convergent validity. To evaluate dis-criminant validity, Fornell and Larcker (1981) suggest acomparison between the average variance extracted (AVE)for each factor and the square of the correlation betweenthe two factors. Discriminant validity is demonstratedwhen the lowest AVE is greater than the squared correla-tion. In our analysis, the smallest AVE value was 0.503,greater than the largest squared correlation of 0.483,which indicates our results have discriminant validity.Next, we tested for the relationship between CSR sub-constructs and the results show that profit motivation andsocial motivation grouped with CSR motivation, legaland ethical with CSR activity, profit maximizing,employee welfare and social welfare with CSR perfor-mance respectively, indicating that our scale has predic-tive validity.

Predictive Validity Test

To test for the predictive validity of the developedscale, regression analysis was conducted to compare withexisting scales. To evaluate whether existing scales or

present study’s scale has more explanatory power, corpo-rate image was used, a construct often used in existingresearch to test of the effect of CSR on individual stake-holders (Becker-Olsen et al. 2005; Brown and Dacin1997; Freeman 1991; Folkes and Kamins 1999; Klein andDawar 2004; Sen and Bhattacharya 2001). Corporateimage refers the overall impression made on the minds ofthe public about an organization (Barich and Kotler 1991;Dichter 1985; Finn 1961; Kotler 1982). To measurecorporate image three items developed by Nguyen andLeblanc (2001) were used. To compare with the newlydeveloped scale, we used Maignan’s (2001) CSR mea-surement scale for the consumers. Based on the four areasof CSR as presented by Carroll (1991), Maignan’s (2001)scale includes four items in each of economic, legal,ethical and philanthropic area with a total of 16 items.Regression analysis results indicate that present study’snew CSR measurement scale is significantly higher in itsR2 of 0.689 than in comparison with 0.402 of Maignan’sscale. Thus we can see that the predictive validity of thepresented study’s scale is superior to that of Maignan’sscale.

CONCLUSION

Discussion

Present study aimed to overcome the limitation ofexisting CSR scales that only focus on CSR performanceevaluation by developing a new scale that measure thethree components of CSR, CSR motivation, CSR activityand CSR performance. Our results show that individualstakeholders evaluate CSR activities with not only just theCSR result but as a combination of the motivation behindCSR, legal and ethical component of execution and CSRoutcome. CSR motivation was categorized into profitseeking motivation in which corporations engage in CSRto seek economic profit and social motivation in whichcorporations engage in CSR to increase social wealth.CSR activity was categorized by legal and ethical natureof the execution methods. CSR performance was catego-rized by profit maximization where corporations wereevaluated by how well they served their fundamentalpurpose of seeking profit, and social welfare in how muchsocial wealth they created and contributed to increasingsocial welfare. These results indicate that unlike extantresearches that conceptualize CSR as a single level con-struct, CSR is a multilevel construct. Moreover, resultsindicate that individual stakeholders do not simply evalu-ate the outcome in CSR performance evaluation andevaluate corporation’s inferred CSR motivation andwhether corporations execute their CSR activities throughlegal and ethical manner. Comparison with Maignan’s(2001) scale indicates that present scale yields superiorexplanatory power, thus suggesting that it is more opti-mized in measuring CSR performance.


Some notable characteristics emerged in developingthis scale for individual stakeholders. First, it should benoted that in evaluating CSR outcomes, social issues andproblems are strongly emphasized. For instance, the strongrepresentation of employee welfare dimension in the scalefinds its roots in current global economic turmoil andrising unemployment in Korea. Secondly, our researchfound that when individuals evaluate CSR, they evaluateusing their comprehensive identity as a multi-identitystakeholder rather than focusing on a single stakeholderidentity. The final scale developed in this study shows thatprofit maximization dimension consists of items thatreflect investors’ interests, employee welfare dimensionitems reflect employees’ interests, and social welfaredimension items reflect consumers’ interests. Ultimatelythe developed scale includes all stakeholder relationshipsbetween an individual and corporation. Third, we foundthat in their evaluation of CSR, stakeholders not onlyevaluate on noneconomic areas such as volunteering,supporting culture and education but also on the funda-mental corporate purpose of profit maximization. Thisindicates that when individuals expect socially respon-sible behavior from corporations, they expect corpora-tions to stay faithful to its fundamental purpose of creatingprofit while simultaneously contributing to enhancingsocial wealth.

Contributions and Limitations

Present research shows that CSR is a multilevelconstruct with many sub-constructs. That is, individualstakeholders not only evaluate the CSR outcome but alsoevaluate each of the subconstructs and aggregate them toform a comprehensive evaluation that leads to change inattitudes and behaviors. While extant research concep-tualize the subcontracts of CSR as areas of CSR outcomes,present research illustrates that the subconstructs of CSRconsist of CSR motivation, activities and outcome andthat each of these subconstructs in turn consist of profitseeking motivation and social motivation, legal and ethi-cal conduct, and profit-maximization, employee welfare,and social welfare, respectively. This result also showsthat unlike conceptualized by existing research, individu-als evaluate CSR-related information by inferring the

underlying motivation for CSR activities. Therefore, fu-ture studies that examine the relationship between variousperformance variables with the CSR subconstructs iden-tified in this study will prove to be more fruitful than thosethat try to directly relate the overall CSR evaluation andcorporate performance. Examining the relationship be-tween the CSR subconstructs themselves may also yieldvaluable insight in how individuals relate between them toform overall evaluation. Furthermore, in utilizing CSR asa strategic tool in corporate settings, strategic communi-cation of CSR motivation must be emphasized for effec-tive CSR performance and close monitoring of stakehold-ers’ perception of CSR motivation must follow.

Present study also shows that when individual stake-holders evaluate CSR, they base their evaluation on thevariety of relationship they hold with the corporation, notfocusing on a single type of relationship. The subconstructsof CSR outcomes derived from this study shows thatprofit maximization is investor-centric, employee welfareis employee-centric, and social welfare is consumer-centric factors. Therefore in order to increasegeneralizability of CSR research on individual stakehold-ers, multi-faceted measurement tool that accounts forthese varieties of dimension, instead of focusing on asingle dimension.

However, present research is not without caveats.The first is that the sample was selected by conveniencesampling and over represents students and workers intheir 20s and 30s. Second limitation is that the social needand expectation for CSR may vary by culture and nation-ality. These pose threats to generalizability and it can bemitigated in the future by rigorously testing the scale withstakeholders in various cultures. Cross-cultural testing ofthe scale can also confirm whether the developed scale’seffectiveness is only limited to Korea or holds the same inother countries as well. Lastly, present scale was devel-oped with only individual stakeholders’ perceptions. Forfuture research in efficient and effective management ofCSR, developing a CSR measurement scale from theperspective of corporations could yield valuable insight incomparing the different in evaluation of CSR motivationand outcomes between corporations and individuals.

REFERENCES

Abbott, W.F. and R.J. Monsen (1979), “On the Measure-ment of Corporate Social Responsibility: SelfReported Disclosure as a Method of Measuring Cor-porate Social Involvement,” Academy of Manage-ment Journal, 22, 501–15.

Barich, H. and P. Kotler (1991), “A Framework forMarketing Image Management,” Sloan ManagementReview, 32, 94–104.

Becker-Olsen, K.L., B.A. Cudmore, and R.P. Hill (2006),“The Impact of Perceived Corporate Social Respon-sibility on Consumer Behavior,” Journal of BusinessResearch, 59, 46–53.

Brown, T. and P. Dacin (1997), “The Company and theProduct: Corporate Associations and Consumer Prod-uct Responses,” Journal of Marketing, 61, 68–84.

Carroll, A.B. (1979), “A Three-Dimensional ConceptualModel of Corporate Performance,” Academy of Man-agement Review, 4, 497–505.


____________ (1991), “The Pyramid of Corporate So-cial Responsibility: Towards the Moral Managementof Organizational Stakeholders,” Business Horizons,(July/August), 39–48.

Churchill, Jr., G.A. (1973), “A Paradigm for DevelopingBetter Measures of Marketing Constructs,” Journalof Marketing Research, 16, 64–73.

Davenport, K. (2000), “Corporate Citizenship: A Stake-holder Approach for Defining Corporate Social Per-formance and Identifying Measures for AssessingIt,” Business and Society, 39, 210–19.

David, P., S. Kline, and Y. Dai (2005), “Corporate SocialResponsibility Practices, Corporate Identity, and Pur-chase Intention: A Dual-Process Model,” Journal ofPublic Relations Research, 17, 291–313.

Davis, K. (1973), “The Case for and Against BusinessAssumption of Social Responsibilities,” Academy ofManagement Journal, 16, 312–22.

Dichter, E. (1985), “What’s in an Image,” Journal ofConsumer Marketing, 2, 75–81.

Folkes, V.S. and M.A. Kamins (1999), “Effects of Infor-mation About Firms’ Ethical and Unethical Actionson Consumers’ Attitudes,” Journal of ConsumerPsychology, 8, 243–59.

Fornell, C. and D.F. Larcker (1981), “Structural EquationModels with Unobservable Variables and Measure-ment Error: Algebra and Statistics,” Journal of Mar-keting Research, 18, 382–88.

Freeman, R.E. (1994), “The Politics of Stakeholder Theory:Some Future Directions,” Business Ethics Quarterly,4, 409–29.

Friedman, M. (1962), Capitalism and Freedom. Chicago:University of Chicago Press.

Klein, J. and N. Dawar (2004), “Corporate Social Respon-sibility and Consumers’ Attributions and Brand Evalu-ations in a Product-Harm Crisis,” International Jour-

nal of Research in Marketing, 21, 203–17.Maignan, I. (2001), “Consumer’s Perceptions of Corpo-

rate Social Responsibilities: A Cross Cultural Com-parison,” Journal of Business Ethics, 30, 57–72.

____________ and Ferrell (2002), “Nature of CorporateResponsibilities Perspectives from American, French,and German Consumers,” Journal of BusinessResearch, 56, 55–67.

____________ and ____________ (2003), “CorporateSocial Responsibility and Marketing: An IntegrativeFramework,” Journal of the Academy of MarketingScience, 32, 3–19.

Mattingly, J.E. and S.I. Berman (2006), “Measurement ofCorporate Social Action: Discovering Taxonomy inthe Kinder Lydenburg Domini Ratings Data,” Busi-ness and Society, 45, 20–46.

Mitnick, B.M. (2000), “Commitment, Revelation, and theTestaments of Belief: The Metrics of Measurementof Corporate Social Performance,” Business andSociety, 39, 419–65.

Schwartz, M.S. and A.B. Carroll (2003), “CorporateSocial Responsibility: A Three-Domain Approach,”Business Ethics Quarterly, 13, 503–30.

Sen, S. and C.B. Bhattacharya (2001), “Does Doing GoodAlways Lead to Doing Better? Consumer Reactionsto Corporate Social Responsibility,” Journal of Mar-keting Research, 38, 225–43.

van de Ven, B. and J. Graafland (2006), “Strategic andMoral Motivation for Corporate Social Responsibil-ity,” Journal of Corporate Citizenship, 22, 111–23.

Wartick, S. and P.L. Cochran (1985), “The Evolution ofCorporate Social Performance Model,” Academy ofManagement Review, 10, 758–69.

Wood, D.J. (1991), “Corporate Social Performance Re-visited,” Academy of Management Review, 16, 691–718.

For further information contact:Seongjin Kim

University of Seoul163 Siripdaero

Dongdaemun-gu, SeoulSouth Korea, 130–743

Phone: 82.10.5437.2857E-Mail: [email protected]


THE ROLES OF MARKETING AND SALES IN NEWPRODUCT DEVELOPMENT

Wim G. Biemans, University of Groningen, The NetherlandsAvinash Malshe, University of St. Thomas

SUMMARY

Numerous studies have identified market factors ascritical to the success of new products. The obviouscandidate for capturing the voice of the customer andtranslating it into successful new products is the firm’smarketing department. But market input does not alwaysrequire a marketing department. Traditional marketingactivities are increasingly allocated to different functionalgroups inside the firm and this dispersion of marketingactivities increases firm performance. One of the depart-ments frequently involved in market-related activities anddecisions is sales. Several researchers investigated theadoption of new products by the sales force, as well as theeffect of sales force adoption on new product perfor-mance. Others looked at related issues such as salesstrategy, the role of control mechanisms and differencesbetween sales persons and sales managers. Still otherresearchers looked at the other end of the NPD processand investigated the role of salespeople in identifying newbusiness opportunities and generating new product ideas.

A growing body of literature on the sales-marketinginterface suggests that sales and marketing contribute tofirm performance through complex interactions and coor-dination. Both salespeople and marketing employees iden-tify customer needs and the respective roles of marketingand sales in NPD depend on the sales-marketing dynamic,as well as the broader dynamic within the NPD team. Forinstance, salespeople may communicate new customerneeds upwards in the sales organization, to marketing orto other departments that are more directly involved inNPD. This suggests that any investigation of the roles ofmarketing and sales in NPD should account for the mecha-nisms and effectiveness of the marketing-sales dynamic.Thus, the following key observations can be made:

1. Market-related input is key to successful NPD andmay be provided by both marketers and salespeople.

2. Existing studies focus only on parts of the contribu-tions of marketing and sales to NPD and fail tosystematically investigate their role throughout thecomplete NPD process.

3. The contribution of both marketing and sales to NPDis shaped within the context of the sales-marketinginterface.

This study takes a closer systematic look at the rolesof marketing and sales during the complete NPD process,within the context of the marketing-sales interface. Theprimary data came from depth-interviews with 36 salesand marketing informants from the US healthcare indus-try, both pharmaceutical and device manufacturing firms.Discovery-oriented depth-interviews were conducted andinformants were encouraged to share examples and anec-dotes to illustrate their points. The involvement of market-ing and sales in the NPD process of US healthcare firmscan be summarized as follows.

1. Stage 1: Discovery. Marketing is an important mem-ber of the cross-functional team and responsible forensuring that the team remains market-oriented. Al-though marketing is ultimately responsible for valuecreation, communication and delivery to physiciansand patients, marketing is not in the driver’s seatduring discovery. Barring the exception of salesleadership in some firms, the involvement of sales-people during discovery is indirect and mediatedthrough marketing. There are several reasons for this:e.g., salespeople lack of overview of the whole mar-ket and the fact that physicians do not see them asvaluable discussion partners.

2. Development. Once again, marketing does not havedirect control, but acts as process enabler to insureoptimal coordination across the various activities.Just as during the discovery stage, sales is not directlyinvolved in the development stage, but only indi-rectly through marketing. Salespeople may be instru-mental in providing marketers with access to influen-tial, well-known customers who can provide criticalinput into the drug design process.

3. Commercialization. In sharp contrast to marketing’ssupporting role during discovery and development,marketing plays the leading role during commercial-ization. Marketing makes sure that an appropriatelaunch strategy for the new product/indication is inplace, provides the necessary inputs to other depart-ments (sales forecasts, pricing and logistics details,cost/revenue projections) and coordinates communi-cation strategy, product availability and sales train-ing. There is also a dramatic shift in the nature andextent of salespeople’s involvement compared to theprevious stages: during commercialization sales-


people are extensively involved in new product/indication training.

In addition, the findings show that the nature andextent of marketing and sales involvement in NPD is verymuch shaped by idiosyncratic characteristics of thehealthcare industry and the marketing-sales interface dyna-mics. In addition, it clearly changes over the course of theNPD process. At the start of the process, marketing triesto capture as much market information as possible, butsalespeople are often protective about their customers;thus, requiring a cordial, effective marketing-sales rela-tionship. When launch approaches, marketing’s interac-tion with sales undergoes a complete metamorphosis.Marketing no longer holds back and makes every effort to

provide salespeople with details about the new product.Marketers must manage this transformation from care-fully controlling information to freely everything veryskillfully.

This study offers insight into the nature of the roles ofmarketing and sales during the entire NPD process, de-scribes the role of industry characteristics and offersseveral managerial implications. In addition, it suggestsseveral avenues for further research; e.g., the roles ofmarketing and sales in other industries and the roles inNPD of different levels and different types of marketingemployees, such as brand management, customer serviceand customer applications. References are available uponrequest.

For further information contact:Wim G. Biemans

Economics and BusinessUniversity of Groningen

P.O. Box 8009700 AV Groningen

The NetherlandsPhone: +31.50.3633834



SALES INTEGRATION AND ITS PERFORMANCE EFFECTS IN NEWPRODUCT DEVELOPMENT: THE MODERATING ROLE OF

INNOVATIVENESS

Malte Brettel, RWTH Aachen University, GermanyAndreas Waschto, RWTH Aachen University, Germany

SUMMARY

Prior research on new product development has re-vealed cross-functional integration to be one of its keyperformance drivers (cf., Troy, Hirunyawipada, andPaswan 2008). Existing literature has, in particular,focused on the extent to which integration between mar-keting and research & development (R&D) drives newproduct performance (e.g., Griffin and Hauser 1996).Only more recently, other functions like manufacturingand IT have been integrated in the analysis (e.g., Nakata,Zhu, and Izberg-Bilgin 2011; Olson et al. 2001).

While recent research on new product developmenthas examined the role of sales in innovation processes(e.g., Hultink and Atuahene-Gima 2000), extant litera-ture – with a few exceptions only (Ernst, Hoyer, andRübsaamen 2010) – has largely ignored the role of cross-functional integration of the sales department. This dearthis problematic for several reasons: First, while extantresearch may have implicitly subsumed marketing andsales as a form of “marketing function,” recent studiesindicate that “they are separate functions within an orga-nization” (Kotler, Rackham, and Krishnaswamy 2006,p. 68; Workman, Homburg, and Gruner 1998). The salesdepartment, in particular, is the most effective in provid-ing the rest of the firm with necessary customer-relatedinsights, hence representing “the voice of the customer” indevelopment activities (cf., Ernst, Hoyer, and Rübsaamen2010). In contrast, the marketing department takes a morestrategic role, contributing, among others, market trendanalyses, market research, product positioning, and com-munication activities (Griffin and Hauser 1996; Rouzièset al. 2005). In a similar vein, Biemans, Malshe, andBrencic (2010, p. 12) state that “in contrast to the tradi-tional marketing literature, a reversal of roles may occur,with sales taking the initiative by developing new cus-tomer approaches, which are only later recognized andformalized by marketing.” Moreover, sales has been foundto be more influential than marketing in certain strategicissues, such as customer service, distribution and pricingdecisions, and the expansion to foreign markets (Hom-burg, Workman, and Krohmer 1999).

The present study contributes to sales, marketing,and new product development research by examining theimpact of cross-functional integration from the sales

department perspective. More concretely, we draw acomprehensive picture of performance effects – incorpo-rating a broad performance measure covering sales growth,market share, and profitability – by examining effects ofsales-marketing, sales-R&D, and sales-finance integra-tion. To the best of our knowledge, this is also the firststudy including finance in a respective research setup. Toshed more light on sales integration, going beyond theinitial contribution of Ernst, Hoyer, and Rübsaamen (2010),we take a contingency perspective and examine the mod-erating effects of market- and technology-basedinnovativeness, representing different types of uncer-tainty. We use survey data from 304 firms to empiricallytest our research hypotheses. To validate our results, weinclude answers from second respondents as well asobjective performance data.

Based on resource dependency theory (Pfeffer andSalancik 1978) and the concept of different thought worlds(Dougherty 1992; Homburg and Jensen 2007) of sales,marketing, R&D, and finance departments, we developedour hypotheses.

H1: Cross-functional integration between (a) sales andmarketing, (b) sales and R&D, and (c) sales andfinance is positively related to new product perfor-mance.

H2: The effects of cross-functional integration between(a) sales and marketing, (b) sales and R&D, and (c)sales and finance on new product performance arestronger when market-based innovativeness is highrather than low.

H3: The effects of cross-functional integration between(a) sales and marketing, (b) sales and R&D, and (c)sales and finance on new product performance arestronger when technology-based innovativeness ishigh rather than low.

We applied multiple regression analysis with interac-tion terms to test our hypotheses. Our results show highlysignificant positive effects for the relationship betweensales-marketing, sales-R&D, and sales-finance integra-tion and new product performance, confirming H1a, H1b,and H1c. Moreover, significant positive interaction effectsof market-based innovativeness with sales-marketing andsales-R&D integration are shown, confirming H2a and


H2b. In contrast, a high degree of sales-finance integra-tion under high levels of market-based innovativenessresults in a significant negative interaction effect, stillrevealing a significant positive relationship for the maineffect. Thus, the results lead to a rejection of H2c. Theinclusion of technology-based innovativeness reveals simi-lar moderating effects. H3a and H3b regarding sales-marketing and sales-R&D integration are confirmed.However, highly technologically innovative developmentimplies a significant negative moderating effect on therelationship between sales-finance integration and newproduct performance, leading to a rejection of H3c.

Our results confirm the pivotal role of the salesfunction in innovative new product development andcorroborate the hypothesis that it brings unique resourcesto cross-functional development teams. Sales’ customerorientation and listening to the “voice of the customer” is

complemented by marketing’s strategic market and prod-uct perspective, R&D’s technical knowledge and finance’sanalytical skills and cost perspective. While market- andtechnology-based innovativeness positively moderate theperformance relationship for sales-marketing and sales-R&D integration, surprisingly and contrary to our expec-tations, our results show negative moderating effects forhigh levels of innovativeness on the performance impactof sales-finance integration. This may be the case ashighly innovative, creative processes might not allowfinance to fully apply its superior financial, risk, and costmanagement skills. In this case, strong sales-finance inte-gration is likely to consume resources that are not used aseffectively and to create significant opportunity costs thatreduce overall performance. Our results indicate that thetypical mantra “the more the better” does not necessarilyhold true for cross-functional integration under all cir-cumstances. References are available upon request.

For further information contact:Andreas Waschto

RWTH Aachen UniversityKackertstr. 7

52072 AachenGermany

Phone: +49.241.80.96359Fax: +49.241.80.92371



RESOURCES IN NEW PRODUCT DEVELOPMENT: ANINVESTIGATION OF RESOURCE FLEXIBILITY,

AUGMENTATION, AND LEVERAGING

Sanjay R. Sisodiya, University of Idaho, MoscowJean L. Johnson, Washington State University, Pullman

Stephen M. Wagner, Swiss Federal Institute Technology, Switzerland

SUMMARY

Nearly one half of firm sales and profits come fromproducts launched in the past five years (Schmidt andCalantone 2002), suggesting that product innovation isthe key to a firm’s success (Henard and Szymanski 2001).With successful product development having a strongimpact on the profitability of firms, research on newproduct development (NPD) is of continued interest. Inorder for a firm to innovate, a focus must be placed on firmresources and capabilities that facilitate enhanced NPD(Rumelt 1987). For the purpose of exploring the effect ofresources and capabilities on NPD, we examine the NPDprocess outcomes of speed and cost. We define projectspeed as being based on the time taken during specificphases of the product development process from projectconceptualization and initiation to product commercial-ization. Process costs are here defined as the costs asso-ciated with the various phases of NPD, including variousaspects of development and commercialization.

Firms have various levels of technology and intellec-tual property at their disposal when developing products.Resource flexibility, as defined by the ability to deployresources (e.g., Sanchez 1995; Johnson, Lee, Saini, andGrohmann 2003), is a characteristic that we incorporate inour study. Firm resources are considered flexible whenthey can be used in a variety of new product projects.Likewise, in order to succeed, firms must constantly pressto expand and grow their limited base of resources. Wetherefore cast resource augmentation as the process ofdeveloping and expanding the existing firm’s resourcebase. Resource flexibility and augmentation may not beenough for a firm to successfully complete new productdevelopment projects, as many argue that managers mustbetter use their resource base and not just acquire newresources (Calantone, Harmancioglu, and Droge 2010).Thus, we consider resource leveraging, the ability of thefirm to utilize resources to their benefit, as a criticalcapability for a firm. We examine leveraging as it isimportant to have access to resources and to use themefficiently (Sirmon and Hitt 2003).

Using project costs and speed as dependent measureswe examine the role of these resources in enhancing NPDoutcomes. Although resource augmentation and flexibil-ity add to the stocks and options needed to develop new

products, they might not be enough to successfullydevelop improved products. An often overlooked aspectof resource management is that simply controlling them isnot enough to achieve a satisfactory market position(Dierickx and Cool 1989). Resources in and by them-selves are not strategically significant, nor are theyadequate to create wealth. Thus, we must study what firmsdo with their resources (Calantone, Harmancioglu, andDroge 2010). The firm must be innovative in takingadvantage of the limited number of resources available(Prahalad and Hamel 2005) and effectively use theirresource stocks to their benefit. Therefore, we positresource leveraging as a distinctive capability that a firmcan use to gain a competitive advantage (Sirmon and Hitt2003).

The sampling frame was based on a list of industrialand service firms in the European Union (EU). Afterpurification of the list, because of incomplete company orinformant data, email was sent to 729 firms, with follow-up email sent within a week. A total of 71 usable responseswere obtained, resulting in a response rate of 9.74 percent.

Sample firms represented machinery, automotive,construction, chemicals, hi-tech, and other manufacturingindustries. To test for nonresponse bias, we comparedearly respondents to late respondents on firm characteris-tics and found no differences, thus concluding that therewas not a threat to nonresponse bias in the data.

Upon analysis of the data, some support was foundfor the hypothesized relationships and some contradictoryresults were found. Since NPD often involves the processof problem solving (Sheremata 2000), resource flexibilityshould benefit the process of NPD as it should increase thenumber of options where resources can be applied whenneeded. While testing the effect of resource flexibility toincrease product development speed and to decreasecosts, we found a significant relationship with speed, butin the opposite direction, and no support for cost savings.This suggests that resources with multiple uses might nothave process costs advantages to firms and could evenhinder development speed. The negative effect ofresource flexibility on speed could be due to the time ittakes to reconfigure and redeploy the resource for anotherapplication, since this type of resource is used in more


than one product (Chatterjee and Wernerfelt 1988). Addi-tionally, the abundance of options could be problematicfor managers who need to find quick solutions to chal-lenges in the development process, but managers couldbecome mired in the selection process and require moretime to make decisions.

Consistent with our hypotheses, resource augmenta-tion had a positive effect on product development speed.However, it had a negative effect on costs. Since resourceaugmentation is often directed by managers to developand expand existing resource bases, it requires managersto pursue a focused path in augmenting the resource baseto develop products quickly. The expansion of theresource base is known to come at a price to the firm andis thus consistent with Kessler and Bierly (2002), whosuggest that a speed-based strategy comes at a cost because

of the additional resources needed to gain faster develop-ment speeds.

Resource leveraging reduced NPD project costs buthad no effect on speed, suggesting that firms developingnew products can see cost benefits by increasing the useof available resources but see no improvements in speed.While testing the moderating role of resource leveragingon resource flexibility and resource augmentation withNPD process speed and costs, we found partial support formoderation of the link between resource flexibility andNPD process costs. Unfortunately, our findings do notsupport the moderation of the flexibility and NPD speedlink. This partial support further reinforces the differentialeffect due to the interaction of a resource characteristicand capability on NPD. Measures, tables, figures, andreferences are available upon request.

For further information contact:Sanjay R. SisodiyaUniversity of Idaho875 Campus DriveMoscow, ID 83844

Phone: 208.885.0267Fax: 208.885.5347



INCREMENTAL SERVICE INNOVATION, SERVICE EMPLOYEES’INNOVATIVE ACTIVITIES, AND THE ENABLING FACTORS

Jun Ye, University of Oregon, Eugene

SUMMARY

Both practitioners and academics agree that serviceinnovation is one of the key drivers for continual growth(Berry et al. 2006; Michel et al. 2008) and superiorcompetitive advantage (Bitner, Brown, and Meuter 2000;Johne and Storey 1998). However, little is known aboutthe role of service employees in a firm’s service innova-tion. This lack of attention is inspite of the recognition thatservice innovations often emerge in the daily practice ofservice employees (Bohmer 2010; Roth and Jackson1995) rather than through deliberate R&D activities(Menor and Roth 2008; Toivonen and Tuominen 2009).Tucker and Edmondson (2003) found that 93 percent ofall problems identified and solved by healthcare serviceemployees during their daily practice were not elevated toa level in the organizations where a more generalizedsolution could be found, tested, and implemented.

Despite the potential contribution service employeescan make to a firm’s service innovation, relatively littleattention has been paid to service employees in the emerg-ing literature of service innovation (Johne and Storey1998; Cadwallader, Jarvis, Jo Bitner, and Ostrom 2010).This study addresses the gap in the literature by examiningtwo related questions. First, how do service employeescontribute to service innovation? And second, what mana-gerially controllable factors influence service employees’engagement in service innovation. We focus on small-scale innovations related to incremental improvement inservice concepts or work processes as opposed to radicalservice innovation, because the nature of incrementalservice innovation is well-suited to service employees’role of customer service provider (De Brentani 1991; DenHertog 2000; Avlonitis et al. 2001).

To understand how service employees contribute toa firm’s incremental service innovation, we adopt anorganizational learning perspective and delineate serviceemployees’ knowledge generation and knowledge articu-lation as the two learning mechanisms that capture thenew knowledge created by service employees, and trans-form it into incremental service innovation. We realizethat although service employees’ knowledge generationand articulation are imperative for incremental serviceinnovation, this is not taking place in many service orga-nizations. In fact, there are considerable risks and disin-centives for service employees taking on such innovativeactivities in organizations (Tyre and Orlikowski 1994;

Crossan et al. 1995; Edmondson 1999). Drawing fromKahn’s (1990) framework of work engagement, we pro-pose three factors, intellectual stimulation, open commu-nication, and monitoring that promote employees partici-pation in service innovative activities.

Hypotheses

H1: Knowledge articulation mediates the effect of knowl-edge generation on incremental service innovation.

H2: Incremental service innovation is positively relatedto profitability.

H3: Intellectual stimulation is positively associated withservice employee knowledge generation and knowl-edge articulation.

H4: Open communication is positively associated withservice employees’ knowledge articulation.

H5: Monitoring is positively associated with knowledgegeneration and knowledge articulation.

Data came from multiple sources in a big healthcareorganization including: (a) self-reported data from front-line employees regarding knowledge generation, (b) keyinformant data from frontline employees for knowledgearticulation, intellectual stimulation, open communica-tion, and monitoring, (c) key informant data from front-line unit managers regarding service innovation, and (d)longitudinal financial archives for frontline unit profit-ability. After matching survey and financial data andexcluding observations with missing values, 41 units with411 responses (362 employee responses and 49 managerresponses) provided usable data.

We used standard procedures including EFA andCFA to assess the psychometric property of the studiedconstructs. These constructs demonstrated acceptable levelof construct reliability, convergent validity and discrimi-nant validity. To account for the multi-level structure ofdata (frontline employees nested within units), we utilizeda random-parameters model (Greene 2008), whichaccounts for individual-specific heterogeneity and allowsfor between- and within-unit effects.

The findings suggest that incremental service inno-vation is primarily driven by service employees’ indi-vidual knowledge generation and group knowledge arti-


culation, and incremental service innovation, in turns,enhances a frontline unit’s profitability. We also foundthat intellectual stimulation and monitoring promote bothknowledge generation and articulation, while open com-munication served as an enabling factors for knowledgearticulation.

The findings of the study equip service organizationswith new insights and guidelines to promote serviceemployee-driving incremental service innovation in orderto provide superior service offerings and maintain com-petitive advantage. First, the study highlights the impor-tant role of service employees in a firm’s incrementalservice innovation. Our findings suggest that service

employees’ knowledge generation and knowledge articu-lation, if activated and nurtured, can be valuable sourcesfor promoting incremental service innovation, which inturn, enhances a firm’s bottom line. Second, the studyidentifies enabling factors including intellectual stimula-tion, open communication, and monitoring, that promoteemployees engagement in knowledge generation andarticulation. Finally, this study provides a framework forunderstanding how a firm can transform its market-basedintelligence into financial outcomes. A firm can transformmarket-based knowledge into financial outcomes throughencouraging and involving service employees in incre-mental service innovation. References are available uponrequest.

For further information contact:Jun Ye

University of Oregon1208 University of OregonEugene, OR 97403–1208

Phone: 541.346.2839Fax: 541.346.3341



EXAMINATION OF THE LINK BETWEEN NEW PRODUCTPREANNOUNCEMENTS AND FIRM VALUE: THE

CASE OF THE U.S. AUTOMOTIVE INDUSTRY

M. Billur Akdeniz, University of New Hampshire, DurhamM. Berk Talay, University of Massachusetts Lowell

SUMMARY

Over the last decade, there has been a spurt on thediscussion about the value of, and return on, marketinginvestments. Stringent economic conditions have ren-dered companies even more concerned about marketinginvestment returns. Calls by executives about their need tomeasure the return on marketing investment (ROMI)coupled with efforts to advance marketing as a moreprominent function in firms have triggered a strong schol-arly interest on the link between marketing actions andfirm performance, with particular emphasis on financialmetrics. The magnitude of this interest is such that assess-ment of marketing returns and financial metrics have beenamong the top research priorities of the Marketing Sci-ence Institute in 2002, 2004, 2006, and 2008; and Journalof Business Research, Journal of the Academy of Market-ing Science, and Journal of Marketing have devotedspecial issues to the measurement of the ROMI in 2000,2005, and 2009, respectively.

Vast majority of the studies on ROMI are based on theefficient market hypothesis (EMH), which asserts thatfinancial markets are “informationally efficient” and hencethe stock price of a company fully reflects all publiclyavailable information about the company. According tothe EMH, favorable prospects of a company will drive itsstock price up, and vice versa. Based on this premise,several studies based on the EMH have tried to predict thefuture performance of various marketing actions likebrand extensions (Lane and Jacobson 1995), celebrityendorsements (Agarwal and Kamakura 1995), productplacement in movies (Wiles and Danielova 2009), and

new product introductions (Chaney et al. 1991; Sorescuet al. 2007).

Nevertheless, both the validity of the EMH and –more importantly – its relevancy has been cast aspersions.For instance, several scholars in behavioral finance (e.g.,Barberis and Thaler 2003) have conceptually and empiri-cally questioned EMH and argued that irrationality ofinvestors may cause the stock prices to deviate from theirmost efficient values. On the other hand, Srinivasan andHanssens (2009) caution the marketing scholars thatstock returns are measured at firm or corporate level,while marketing actions are most of the time at product orbrand level. Furthermore, Hanssens et al. (2009) call forfuture research to challenge the efficient market hypoth-esis.

Our study attempts to address this call by a multi-method analysis of the link between the stock price of acompany and the future success of its marketing actions.On contrary to the previous studies, which examine howa marketing action triggers a change in the stock price ofa company, we analyze the extent to which changes in thestock price can accurately augur the future success (orfailure) of a marketing action. Specifically, we focus onnew products, which have been identified as a key con-tributor to long-term firm sales, as well as to financial andstock market performance (Pauwels et al. 2004). Webelieve this study is highly relevant for both marketingacademics and practitioners since it will help them betterunderstand how and when the stock price fluctuations cansignal the potential of a marketing investment. Referencesare available upon request.

For further information contact:M. Berk Talay

University of Massachusetts LowellOne University Avenue FA207C

Lowell, MA 01854Phone: 978.934.2810



LEARNING MARKETING MANAGEMENT AND LEADERSHIPPRACTICES FROM PEERS: THE CASE OF INDUSTRY

PEER NETWORKS

Ada Leung, Penn State Berks, ReadingHuimin Xu, The Sage Colleges, New York

Kyle Luthans, University of Nebraska, KearneySusan Jensen, University of Nebraska, Kearney

SUMMARY

In this research, we examine a unique resource forsmall business owners, industry peer networks (IPNs),and study how IPNs impact the performance of smallbusinesses, in terms of implementation of marketing man-agement and leadership practices.

Industry peer networks are a unique form of “parallelpeers” in which the members of the network belong to asub-segment in a given industry that draws on similarinputs to provide similar goods or services targeted todifferent sets of customers. These non-competing (andnon-colluding) members gather regularly in small groups(typically 20 or fewer carefully selected members), in anatmosphere of significant trust, to share knowledge, ex-change information about industry trends beyond theircore markets, and discuss issues related to companyperformance.

In their own way, IPNs embody many importantcharacteristics of mentors, advisory boards, and tradeassociations. Individual members often discuss their man-agement issues one-on-one with specific peers whom theyidentify as mentors. Moreover, during a typical IPNmeeting, a facilitator guides the members to present theiroperation/financial data, discuss their management,finance, and marketing issues, and provide constructivecriticism of their business. In other words, the groupmembers act as the advisory board for each of the mem-bers. It is also not uncommon for members to collaboratetogether to achieve a common goal, such as submittingbusiness proposals together. Through face-to-face meet-ings and electronic communications in between the meet-ings, IPN members stave off problems of myopia andinertia by staying current with industry changes, learningvicariously from the experiences of their peers, and col-laborating on mutually beneficial projects (Sgourev andZuckerman 2006).

We studied the IPN phenomenon among the mem-bers of a “technology industry reseller” industry peernetwork whose member companies are located in theUnited States, Canada, the United Kingdom, and Austra-lia. The members in the peer groups gather regularly each

quarter for several days of intense face-to-face meetings.Peer group members also communicate with each otherextensively throughout the year via electronic means. Inaddition, the IPN under investigation holds bi-annual“ALL” conferences in which all members of the peergroups convene together.

Data gathered for this study were generated fromobservations and interviews conducted during face-to-face peer meetings and sessions held during the ALLconference, along with numerous communications withthe IPN founder, committee chairpersons, and peer groupfacilitators. Based on our understanding of the context, alongitudinal survey was developed to track the socialinteractions and the level of learning of the members. Inaddition, IPN members were asked to give permission torelease their operation/financial data via a third party andthese data were then merged with the survey database. Thefindings reported in this study are based on the survey andoperation/financial data collected in January and April2009.

In this research, we found that industry peer networksserve as an important source of new knowledge for thesmall business owners. New knowledge source diversitywas identified as the best predictor in the baseline model,indicating that as IPN members are exposed to a morediverse set of knowledge sources, they achieve a higherlevel of perceived marketing and management learning.Nevertheless, as social embeddedness is added subse-quently in the regression, the values of beta coefficients ofsource diversity decrease and those of sociallyembeddedness become statistically significant. Such find-ings suggest that industry peer networks provide a facili-tating platform for learning in between meetings (such assetting up meeting agendas and working on IPN-relatedprojects and approaching peers for professional adviceabout work-related problems).

As for the mediation models that test the implicationsof perceived level of learning, we found that the imple-mentation of transformational leadership practices is par-tially mediated by the perceived level of managementlearning. The width of product portfolio, however, is notmediated by the perceived level of marketing learning.This research suggests that learning needs to be articu-


lated and internalized before transformational leadershippractices are implemented successfully in their respectivefirms.

While a significant gain in knowledge about supplierinformation may be evident after just a few peer groupmeetings, the implementation of newfound leadershipmanagement knowledge demands significantly moreefforts from the members. IPN members also must be

aware that joining an IPN does not automatically improvefirm’s performance. In order to learn and implement smartbusiness practices, not only do members need to preparefor and provide critical advice to their peers during thequarterly face-to-face meetings, they must also networkwith their peers between the meetings by collaboratingwith them on a variety of activities so that the memberscan ask for advice whenever they need. References areavailable upon request.

For further information contact:Ada Leung

Penn State BerksTulpehocken Road, P.O. Box 7009

Reading, PA 19610Phone: 484.797.8389



THE IMPACT OF PROACTIVE MARKETING AT THE FUZZYFRONT END OF INNOVATION

Fiona Schweitzer, Upper Austria University of Applied Science, Wels

SUMMARY

The central idea of market orientation is the depen-dence of companies’ success on their ability to satisfy theneeds of target markets better than their competitors (e.g.,Kotler and Keller 2006). Corresponding to this under-standing, many studies find that market orientation leadsto new product success (e.g., Gatignon and Xuereb 1997;Kohli and Jaworski 1990; Lukas and Ferrell 2000; Slaterand Narver 1994). As the role of marketing departmentsin companies is to understand and satisfy customer needs,an early integration of these individuals into the newproduct development team increases the team’s marketorientation and consequently yields innovation success(e.g., Nakata and Im 2010; Olson et al. 2001).

This view is challenged by a number of recent studiesthat find a negative correlation between marketing orcustomer integration and innovation performance (e.g.,Atuahene-Gima 2005; Baker and Sinkula 2005; Spanjolet al. 2011). Explanations for this phenomenon includeincreased coordination effort (Song, Thieme, and Xie1998), cultural differences and goal incongruity (Xie,Song, and Stringfellow 2003), the conservative mind ofthe customer and his or her inability to conceptualize ideasthat go beyond their own experience (Knudsen 2007)which in turn hampers the realization of radical innova-tions (e.g., Christensen and Bower 1996).

In this article, a contingency perspective is taken toanalyze whether the amount of proactivity of the market-ing department impacts the interaction between the mar-keting department and the research and developmentdepartment in a company (R&D-marketing interaction)on new product performance. More precisely, the rel-evance of taking a proactive marketing approach is inves-tigated for collaborating in the early phases of productinnovation. This front end of the product innovationprocess comprises the time between the first mentioningof the idea within the company and its approval or rejec-tion to enter the development phase (Kim and Wilemon2002) and is often referred to as the fuzzy front endbecause uncertainty and equivocality are high (Smith andReinertsen 1998). Decisions taken in this period are saidto have the highest impact on the overall innovation

process (Cooper 2001; Reid and de Bretani 2004) and thetasks and characteristics of the front end are different fromlater innovation phases (Koen et al. 2001). For this reason,the impact of cross-functional integration can differ to thatin other phases. This reasoning is in line with resentresearch that highlights the importance of more detailedresearch into success factors for the different phases of theinnovation process (e.g., Brettel et al. 2011). Hence, thisarticle focuses on the impact of R&D-Marketing interac-tion on market uncertainties and technical uncertainties atthe front end of innovation and introduces proactivemarketing as a moderator of this interaction.

To gather empirical evidence, 190 innovation projectsin various industrial goods industries are investigated andthe data was analyzed using hierarchical multiple regres-sion analyses. The results indicate that R&D-marketinginteraction does not per se reduce market uncertainties nortechnical uncertainties, while proactive marketing doesreduce uncertainties. Moreover, a moderating effect ofproactivity is found which demonstrates that an integra-tion of R&D and marketing at the fuzzy front end ofinnovation projects is beneficial when marketers take aproactive approach.

Therefore, innovation managers who set up an inno-vation team should carefully investigate whether the teammembers take a proactive approach. They should take aprudent approach toward R&D-Marketing interactionand should not foster cross-functional integration by allmeans at the fuzzy front end of innovation. Managersshould assess the proactive orientation of the marketingdepartment first. If marketers lack proactivity, they mightoppose expedient innovations which are considered athreat to their current product lines or customer base. Ifthey are proactive, they can contribute to the innovation ina savvy way by identifying latent needs of current andfuture customers and assuring that these needs are well-accounted for throughout the phases of idea generation,idea evaluation, product conceptualization and beyond.In addition, managers may need to identify ways tomanage marketing in a proactive way and to motivateR&D and marketers to think proactively. References areavailable upon request.


For further information contact:Fiona Schweitzer

School of Engineering and Environmental SciencesUpper Austria University of Applied Science

Stelzhamerstraße 23Wels, 4600

Phone: +43(0)50804.43870Fax: +43(0)50804.943870



I DON’T NEED IT AND I DON’T WANT TO - EXAMINING EFFECTS OFSITUATIONAL AND COGNITIVE RESISTANCE TO INNOVATIONS

Sven Heidenreich, EBS Business School, Oestrich – WinkelMatthias Handrich, EBS Business School, Oestrich – Winkel

SUMMARY

Since decades, the innovation literature reports highfailure rates for innovations, ranging between 50 percentand 90 percent (e.g., Andrew and Sirkin 2003; Cierpicki,Wright, and Sharp 2000; Sivadas and Dwyer 2000). Yet,a closer look in the adoption literature reveals that mostresearch focuses on positive outcomes of the adoptionprocess and thus is subject to a so-called “pro-change”bias (e.g., Rogers 1976; Sheth 1981; Speier and Venkatesh2002). In order to overcome “pro-change” bias we pro-pose to incorporate the concept of passive innovationresistance. Passive innovation resistance is widely seen asgeneric innovation resistance evolving from an individual’sinclination to resist changes (cognitive resistance) and astatus quo satisfaction (situational resistance). Despitescientific acknowledgement of the relevance of resistanceto change and status quo satisfaction for new productevaluation, empirical research into this topic is surpris-ingly scarce (Bagozzi and Lee 1999; Ellen, Bearden, andSharma 1991; Nabih, Bloem, and Poiesz 1997). Morespecifically, to the best knowledge of the author noempirical evidence currently exists on whether and howdifferent types of passive innovation resistance affectconsumer evaluations of new products. Hence, the aim ofthis research is to examine whether and how cognitive-,situational- and dual-passive resistance may influencenew product evaluation.

We selected 307 individuals to participate in ourStudy, a 4 (type of passive resistance: low (control),situational, cognitive, dual) x 2 (degree of newness: low(INP), high (RNP)) between-subjects design. We used thescenario method, which ask participants to imagine them-selves in hypothetical constellations and roles, to investi-gate the impact of varying the type of passive innovationresistance and the level of degree of newness on newproduct evaluation. In our specific case, each scenarioasked participants to envision themselves inside an elec-tronics retail store, in which they had decided to buy a newmobile phone that was available either with a traditionallithium-ionic battery or with the new tin-sulfuric-lithium-ionic battery (INP) / kinetic-energy-harvester battery(RNP). Participants were randomly assigned to one of theeight consumer scenarios in line with Dreze and Nunes’s(2007) recommendations. At the beginning of the experi-ment, participants in all eight experimental groupsreviewed background information describing the sce-

nario, followed by scenario-specific manipulations. Toprovide realistic and effective experimental manipula-tions, we adapted our manipulations from prior experi-mental scenario studies and implemented several manipu-lation checks (e.g., Churchill and Surprenant 1982; Hess2009; Phillips and Baumgartner 2002).

As the obtained data of our study demonstrated a non-normal distribution (D (307) = .210, p < .001) andheteroscedasticity (F (7, 299) = 5.416, p < .001), weutilized adjusted rank transform (ART) two-wayANOVAS for our statistical analysis. The results indicatea significant main effect for each type of passive innova-tion resistance on adoption intention (F (3, 306) = 16.94,p < .001). In the low passive innovation resistance condi-tion, consumers are more willing to adopt new products(M

PIRlow = 6.76) than consumers in cognitive- (M

PIRcognitive =

5.35), situational- (MPIRsituational

= 5.38) or dual-passiveresistance conditions (M

PIRdual = 4.06). While no signifi-

cant difference for the effects of situational- and cogni-tive-passive resistance could be confirmed (F (1, 155) =.34, n.s.), dual-passive innovation resistance exhibited byfar greater negative effects on new product evaluationthan cognitive- (F (1, 154) = 9.68, p < .01) and situationalpassive resistance (F (1, 161) = 6.94, p < .01). Moreover,the proposed interaction of passive innovation resistanceand degree of newness on new product evaluation wassignificant (F (3, 306) = 2.92, p < .05). According to ourresults, consumers in cognitive-, situational- and dual-passive innovation resistance condition prefer low levelsof stimuli and thus INPs while consumers in the lowpassive innovation resistance condition prefer high levelsof stimuli and thus RNPs.

Understanding how different types of passive inno-vation resistance affect new product evaluation offersimportant implications for both theory development andmanagerial practice. From a theory development perspec-tive, our controlled experimental scenario study enabledus to test causal effects of different types of passiveinnovation resistance on new product adoption. The find-ings reported herein increase our sparse knowledge onhow cognitive-, situational- and dual-passive innovationresistance affects new product evaluation. All three typeswere shown to substantially decrease attitude formation,whereas dual-resistance was shown to be the most crucialtype of passive innovation resistance. From a managerialperspective, innovation managers have to be aware that it


is not sufficient to offer new products with a superiorrelative advantage as it has long been suggested in theliterature of innovation adoption (e.g., Ostlund 1972;Rogers 2003). Our findings provide evidence that passiveinnovation resistance emerges from either cognitive-pas-sive resistance or situational-passive resistance, inhibit-ing favourable new product evaluation. Hence, consum-ers often act irrational in the innovation adoption processand over-rate products they already possess while under-

estimating the new benefits offered by an innovation dueto perceived changes in behavior necessary to use theproduct. A possible solution to passive innovation resis-tance would be to reduce negative effects from bothcognitive and situational resistance. Innovation managersmay use mental simulation or categorization cues toreduce effects of cognitive-passive innovation resistance,while using benefit comparison to reduce situationalresistance. References are available upon request.

For further information contact:Sven Heidenreich

EBS Business SchoolRheingaustraße 1

Oestrich – Winkel, Hesse 65375Phone: +49.611.7102.1335Fax: +49.611.7102.10.1335



THE PRICING OF SOFT AND HARD INFORMATION: LESSONSFROM SCREENPLAY SALES

Milton Harris, University of ChicagoS. Abraham Ravid, Yeshiva University, New York

Suman Basuroy, The University of Oklahoma, Norman

SUMMARY

Aesthetic evaluation is central to the motion pictureindustry. However, the industry does not usually create artfor art’s sake – it processes complex inputs from manydifferent fields of art with the ultimate goal of making aprofit. Our major findings highlight the dual role of softand hard information in the successful sale of intellectualproperty – the screenplays. The screenwriter’s experienceand past success, which can be easily expressed in mea-surable terms, are very important, and increase the pricespaid. However, the presence of soft information in thesales pitch, as indicated by our proxies, depresses prices.That is, screenplays characterized in “softer” terms, par-ticularly if they are written by lesser-known writers,command substantially lower prices. Various manifesta-tions of soft information are also shown to increase theprobability of a contingent contract. This suggests thatcontingent contracts may allow contracting in an environ-ment where uncertainty and asymmetric information inte-ract. However, reputation increases the probability ofreceiving cash upfront.

We also link the economic performance of films withscreenplay characteristics. We find that pricing is effi-cient, even in an industry with a complex productionfunction relying fundamentally on soft information. Our

findings suggest a few managerial implications for themovie industry as well as other industries, in particularones which produce intellectual property. First, we areable to justify the movie industry emphasis on shortpitches as an effective selling tool. Our analysis suggeststhat when one pitches a product, a short, concise, andsimple description increases the sales price, in particularfor intangible products. This could be applicable forselling anything from a new drug to a new book. In ourempirical work we show that this high pricing for shortpitches may be justified--the screenplays that cost morewere predominantly the ones that resulted in more suc-cessful movies. In other words, it seems audiences tooprefer simple “high concept” stories. In a more generalsense, our results suggest that “soft” elements in market-ing products are important and have a significant impacton the price paid. The finding that more experiencedwriters fare better is less surprising.

Finally, it matters to whom you sell. We affirm thetheoretical predictions and evidence from the bankingsector that small organizations can handle “softer” pitchesbetter, perhaps because the hierarchy is much less of anissue and sellers may interact directly with the people whomake the final decision. References are available uponrequest.

For further information contact:Suman Basuroy

Price College of BusinessThe University of Oklahoma,

307 West BrooksNorman, OK 73019



THE ENTRANT’S PRICING DILEMMA: LINKING LOW PRICE ENTRYSTRATEGIES TO AGGRESSIVE PRICE RESPONSES BY INCUMBENTS

Alexander J. Mrozek, EBS Business School, Oestrich – WinkelTomas Falk, EBS Business School, Oestrich – Winkel

SUMMARY

We investigate the entrant’s dilemma created bypenetration pricing. Penetration pricing is a commonstrategy where firms enter into a market with relativelylow prices. Setting a low price offers several advantagesfor entrants. For example, beating competitor’s pricesreduces consumers’ risk of trial and gives consumers amonetary incentive to switch brands (Blattberg and Neslin1990). Low prices thus help entrants to gain a foothold inthe market and to build scale, especially when demand isprice elastic (Moore 1992). However, there are also draw-backs of low price market entry. Low prices send anobvious signal of aggressiveness and can pose a substan-tial threat to incumbent firms (e.g., Ailawadi, Lehmann,and Neslin 2001). Consequently, low entry prices likelytrigger hostile pricing responses by incumbents (Leeflangand Wittink 1992).

If incumbents reduce prices in response to entry, theentrant faces a severe dilemma: If the entrant does notreadjust its prices, it will lose its competitive advantageand risk being pushed out of the market by the incumbent.If the entrant does readjust its price to reclaim its priceadvantage, this might trigger downwards spiraling prices,ultimately resembling a price war (Heil and Helsen 2001).In this price war, experience-curve cost-effects are likelyto favor the incumbent (Kerin, Varadarajan, and Peterson1992). As a result of the entrant’s pricing dilemma, newproduct failure rates are “notoriously high” (Ernst, Hoyer,and Ruebsaamen 2010, p. 81).

In our research, we explore this dilemma and suggestways to both reap the benefits of low price market entryand avoid aggressive price reactions by incumbents. Spe-cifically, we extend knowledge in three important ways.

First, we particularly focus on market entry pricingand expand the currently narrow understanding of marketentry with relatively low prices. Previous research haslimited low price entry strategies exclusively to penetra-tion pricing (e.g., Hultink et al. 1998; Liu 2010; Noble andGruca 1999). This low level of complexity is surprisingfor several reasons. Most importantly, it does not mirrorcommon practice in many markets that are characterizedby largely homogenous products and a price elasticdemand. Additionally, in non-entry settings a wide arrayof low-price strategies is discussed. Examples include

exclusive price promotions (Barone and Roy 2010), non-linear pricing (Gu and Yang 2010), and reward pointpromotions (Zhang and Breugelmans 2012). Yet, itremains unclear to what extent these insights can betransferred to a market entry context given significantdifferences in the management of new and existing prod-ucts (e.g., Homburg, Fuerst, and Prigge 2010). In view ofthe above, we classify three especially relevant forms oflow price entry strategies based on in-depth interviewswith 32 experts from various industries: bonus pricing,introductory pricing, and price-matching guarantees(PMGs). With bonus pricing firms do not lower theabsolute price level but lower the relative price for valuecompared to incumbents by increasing product valuethrough a “bonus.” This bonus can be an increased pack-age size or free after-sales services. Firms entering withintroductory pricing set an exceptionally low price duringa preannounced time frame in order to test customeracceptance and enhance customer trial. When the timeframe ends, prices are typically raised. PMGs areannouncements that the entrant will match any lowercompetitive price.

Second, we contribute to literature on competitiveinteraction between entrants and incumbents. Pricingdecisions are especially important to competitive interac-tion, as price actions trigger competitive reactions moreoften than other instruments of the marketing mix (Kuester,Homburg, and Robertson 1999; Leeflang and Wittink1992). Yet, empirical evidence on competitor assess-ments of price attacks remains inconclusive. On the onehand, competitors seem to generally interpret penetrationpricing as an aggressive move calling for reciprocalretaliation (Hultink and Langerak 2002). On the otherhand, some authors explicitly ask for more research on the“underlying causes of an observed absence of reaction” toprice promotions attacks (Steenkamp et al. 2005, p. 51).Based on an experiment with 358 graduate and MBAstudents, we find that entrants can “camouflage” theiraggressive entry signal when choosing innovative lowprice entry strategies compared to entry with classicalpenetration pricing.

Third, we study how incumbents react to low priceentry strategies. Previous literature has extensively inves-tigated determinants of incumbent reaction, includingentrant’s launch decisions (Shankar 1999). Pricing deci-sions are especially important to competitive interaction,


as price actions trigger competitive reactions more oftenthan other instruments of the marketing mix (Leeflang andWittink 1992). Low prices have been consistently foundto cause aggressive and retaliatory responses (e.g., Hultinkand Langerak 2002). We contribute to academic researchby showing that incumbents do not react in the same wayto all low price strategies, but unique low price entry

strategies trigger unique competitive reactions. Particu-larly, we find that market entry with bonus pricing, intro-ductory pricing, and a PMG significantly reduces thestrength of competitive reaction compared to classicalpenetration pricing. Consequently, we show that theseinnovative low price entry strategies help avoid theentrant’s dilemma. References are available upon request.

For further information contact:Alexander J. MrozekEBS Business School

Rheingaustrasse 165375 Oestrich-Winkel

Phone: +49.611.7102.1494E-Mail: [email protected]


ACCEPTING OR FIGHTING PIRACY: CAN FIRMS REDUCE PIRACYFOR DIGITAL MEDIA PRODUCTS BY OPTIMIZING

THEIR MARKETING?

Felix Eggers, University of Hamburg, GermanyAlexa Burmester, University of Hamburg, GermanyMichel Clement, University of Hamburg, Germany

Tim Prostka, University of Hamburg, Germany

SUMMARY

The challenge for the media industry is to find strat-egies to optimally compete with pirates. The major play-ers in the media industry actively lobby via their associa-tions MPAA and RIAA for strict copyright enforcementsand stronger punishments of pirates. However, consum-ers (and pirates) argue that the offered products are notoptimally designed to meet their preferences (Clement etal. 2012). The pressure to compete against pirates is high,especially with the advent of file sharing and, recently,locker services that allow users to efficiently and mas-sively share their media files illegally (IFPI 2011). Thesenew technologies have substantial effects on consumerbehavior (Papies et al. 2011; Bhattacharjee et al. 2007).Interestingly, prior research mainly focused on under-standing piracy behavior and addressed the effects ofdigital rights management systems (DRM, Sinha et al.2010), legal actions (Sinha and Mandel 2008), or freeservices on downloading intention (Gopal and Gupta2010; Clement et al. 2012). However, the key question forthe industry players remains whether and how firmsshould change their marketing strategies in order to bettercompete against pirates. Thus, should firms fight piracyby providing specific offers to piracy segments or shouldthey simply accept a certain level of piracy because theywill not be able to gain more business due to cannibaliza-tion effects?

In order to address this research question we developa conceptual framework and model for media choice inwhich we account for pirated alternatives. More specifi-cally, we analyze the effects of different (1) timing and (2)pricing strategies on consumers’ choices while control-ling for piracy effects. We focus on pricing and timingdecisions as the main two strategic instruments in apply-ing a sequential release strategy. Both play a major role inmany entertainment industries such as the motion picture(Elberse and Eliashberg 2003) or book industry (Clerides

2002) in which products are sold for different prices atdifferent times (e.g., hardcover versus paperback books).We empirically test the validity of the model in two largerepresentative samples of 2,521 (1,623) consumers in themotion picture (book) market collected in Germany in2011. Subsequently, we use the empirical results to pre-dict consumer choices and simulate (piracy corrected)revenues for more than 17 million potential market con-figurations, which differ in terms of timing and pricing ofthe market options, and identify those scenarios with thehighest sales for the industry, both with and withoutaddressing piracy.

Although the motion picture and book industries arequite different (e.g., the digital distribution channel isrelatively new in the book market, while movie down-loads are established in the motion picture market) con-sumer reactions toward piracy are very consistent in bothcontexts. This is even more surprising, since the pricelevels (e.g., higher in the book market), the target audi-ence (older for the book market) and the novelty of digitalproduct versions (newer in the book market) differ be-tween the two markets. The results indicate that about 1/3 of the revenue potential is lost accordingly, thus, piracyhas a large effect in both industries. Addressing thisproblem explicitly and trying to optimize the piracy cor-rected revenue with timing and pricing decisions has onlylimited impact. The optimal results that can be achievedare not much different than maximizing the revenuewithout piracy consideration and only lead to a 1–2percent revenue gain. Thus, from a managerial perspec-tive, piracy can be neglected because maximizing revenuealso leads to higher piracy corrected revenue. Conse-quently, although timing and pricing decisions have alarge effect on the overall revenue, they cannot reduce therelative loss due to piracy, which remains a constantproblem in all scenarios we have examined. Referencesare available upon request.


For further information contact:Felix Eggers

Institute of Marketing and MediaUniversity of Hamburg

Welckerstraße 820354 Hamburg

GermanyPhone: +49.40.42838.8714



FRAMING EFFECTS ON EVALUATION OF COUPON OFFERS:ASSESSING THE PERFORMANCE OF ALTERNATIVE

FACE VALUE FRAMES

George Baltas, Athens University of Economics & Business, GreeceGrigorios Painesis, Athens University of Economics & Business, Greece

Paraskevas Argouslidis, Athens University of Economics & Business, Greece

SUMMARY

This paper is concerned with coupon effectivenessand in particular framing of coupon face value. Couponface value has been identified as the most critical determi-nant of coupon redemption rate but very little is currentlyknown about how different face value frames influencecoupon attractiveness, evaluation and redemption.

Considered generally, the term framing refers toindividuals responding differently to alternative descrip-tions of the same decision problem. The framing ofalternatives is found to affect choices in many and diversedomains, such as health precautionary behavior, creditcard usage, product and service evaluation, donations tonon-profit organizations, and new product adoption. Fram-ing of discount promotions has been the focus of severalstudies, which addressed the presentation of the savingsof promotional offers, also known as deal semantics.However, very little attention has been paid to the relatedissues of coupon design and framing.

The present study considers three alternative facevalue frames: “cents off,” “percentage off” and “reducedprice.” In all three cases, consumers save the same amountof money, acquire the same amount of product units andreceive the same reference price information. Thus thedeals are totally equivalent and the “pure” framing effectcan be detected. The “cents off” face value frame informsconsumers about the actual monetary savings received byredeeming the coupon. The “percentage off” frame is anindication of the relative attractiveness of the offer. The“reduced price” frame is equally popular in businesspractice but it has not so far been considered in the framingliterature. This frame stresses the final reduced price andnot the potential savings. In some respects, the “reducedprice” frame has a more negative connotation, as it empha-sizes what is left to pay and not what is saved. Theconsumer has to calculate the discount value or use anassimilation-contrast process. Thus, the “reduced price”frame may underperform relative to the “cents off” frame.In addition, it has been suggested that consumers tend toprocess price reductions relatively rather than in absoluteamounts. We also know from empirical studies that thepsychological pleasure derived from a discount maydepend not only on the absolute amount of discount but

also on the price level of the promoted product. Conse-quently, framing effects may depend on the price level ofthe promoted product.

Existing studies suggest that framing and price levelinfluence perceived savings but do not affect purchaseintention. This may be so because of the small incentivesince a certain discount level is needed to trigger anybehavioral response. The present study attempts toresolve this issue, by employing a moderate as well asa high discount size. Existing studies also highlight therole of discount size in coupon attractiveness and redemp-tion behavior in the sense that greater face values nor-mally lead to more favorable evaluations and responses.Thus, we expect that the discount size will influenceconsumer perceptions, attitudes and redemption intentionbut it will not affect the observed framing effects. Finally,the interrelationships of perceived savings, attitudes to-ward the offer and redemption intention as antecedents ofbehavior are well-established in the literature and weexpect that the same pattern will appear in this setting.

The pretests reveal that a laptop computer is per-ceived as the most familiar, involving, likely to purchase,high-priced product and a USB flash drive is the low-priced product with the same traits. In addition, the 25%discount is considered moderate and the 40% discountplausibly high.

After conducting the pretests, we implement a 3(cents off, percentage off, reduced price) X 2 (LaptopComputer, USB flash drive) X 2 (25%, 40 % discount) fullfactorial between-subjects experimental design involving247 participants. Age and gender has no significant effectson any of the dependent variables, suggesting that simpleindividual demographics are poor predictors of couponuse. The manipulation checks are successful and no con-founding effects (familiarity, involvement, noveltyeffects, believability) or potential covariates (offer andprice believability, coupon proneness, price conscious-ness, familiarity, involvement and prior brand attitudes)are detected.

The results of the experiment are revealing. Differentframes are effective in terms of perceived savings, atti-tudes toward the offer, and redemption intention when the


price level of the promoted product is considered. The“reduced price” frame can be considered as a peer of the“cents off” frame, both for high and low-priced products.In the case of high price products, the latter framesoutperform the “percentage off” frame. It is conceivablethat discounts of high price products are more noticeableand influential framed in absolute terms. In the case of lowprice products, we observe the opposite. It is equally clearthat discounts of low price products are more noticeableand influential framed in relative terms. It is also useful tonote that contrary to our expectations, the “reduced price”frame performs unexpectedly well. The results suggestthat potential difficulties in consumer evaluation of suchframes have been overestimated. We respect to the dis-count size, it is demonstrated that moderate and high sizesdo not distort framing effects. Note also that discount sizeimproves promotional effectiveness as larger savingsenhance the desirability of the coupon offer. One mayhere parenthetically note that sufficiently large discountsizes may fill the attitude-to-intention gap observed inprevious studies and be able to trigger a behavioralresponse. Finally, by performing meditational analysis,the study sheds light on the effects of frames on theantecedents of behavior. It appears that transaction value

influences attitude toward the offer, which in turn deter-mines redemption intention.

The results lead to some useful managerial implica-tions. Marketers can design more efficient coupon cam-paigns by matching the right frame with the right pricelevel. It has been demonstrated that framing and pricelevels influence every stage of the coupon evaluationprocess. In particular, coupon face values framed inpercentage terms are more effective for low-priced prod-ucts. Reduced-price and cent-off frames appear moreeffective for high-priced goods. Finally, managers canuse face value as a means of improving coupon effective-ness as long as the incentive is plausibly high. Referencesare available upon request.

ACKNOWLEDGEMENTS

This research has been co-financed by the EuropeanUnion (European Social Fund - ESF) and national fundsthrough the Operational Program ‘Education and Life-long Learning’ of the National Strategic Reference Frame-work (NSRF) - Research Funding Program: “HeracleitusII: Investing in knowledge society through the EuropeanSocial Fund.”

For further information contact:George Baltas

Athens University of Economics & Business76 Patission Avenue

Athens, 10434Greece

Phone: +30.210.8203714Fax: +30.210.8223802E-Mail: [email protected]


I THOUGHT IT WAS ALL OVER AND NOW IT IS BACK AGAIN:CUSTOMER REACTIONS TO TIME EXTENSIONS

OF SALES PROMOTIONS

Ina Garnefeld, University of Wuppertal, GermanyEva Muenkhoff, University of Paderborn, GermanyAndrea Bruns, University of Paderborn, Germany

SUMMARY

Sales promotions are a ubiquitously used marketingtool. Firms often rely on promotions in order to directlyimpact customer behavior. A common practice in manyindustries is the time extension of sales promotions. Here,firms decide to postpone the initial deadline of a promo-tion after the announced promotion period has beenexpired. Such extensions of sales promotions are notrestricted to certain industries but are heavily used by,e.g., telecommunication providers, media or leisure travelfirms as well as (online) retailers. For example, the elec-tronic manufacturer Panasonic has postponed the initiallyannounced deadline of offering a free Blu-ray Disc forevery purchase of a HD Blu-ray Disc Player from the endof February to end of April.

Announcing a short promotion period and then exten-ding the deadline of the sales promotion seems to bebeneficial for firms for two reasons. First, past researchfound promotions with short periods to affect customersto accelerate their purchases (Aggarwal and Vaidyanathan2002). Second, according to Inman and McAllister (1994)customers show a tendency to buy shortly before a promo-tion expires. Consequently, firms extending the deadlineof a sales promotion expect to profit (1) from two shorterpromotion periods, as well as (2) twice from the revenuepeaks shortly before the expiration dates.

However, so far research on sales promotions hasneglected the issue of extending a sales promotion on(potential) customers and empirical evidence of the con-sequences for firms is lacking. However, researchingthese effects is important as based on regret theory wecannot only assume positive but also negative effects ofpostponing the deadline of a sales promotion. We there-fore strive to shed light on the effects of extending a salespromotion on potential customers’ as well as existingcustomers’ attitude and behavior.

We draw on regret theory to develop our hypotheses.Regret is an emotion individuals feel when they “realizeor imagine that [their] present situation would have beenbetter had [they] made different decisions” (Zeelenbergand Pieters 2006, p. 210). Building on regret theory, weassume that existing customers who bought within theinitial promotion period will experience higher levels of

regret when the deadline of a promotion is postponed,whereas potential customers’ regret of not having boughtthe product within the first promotion period will decreasewhen a promotion period is extended. Moreover, weassume the negative effect of extending the sales promo-tion on existing customers’ attitude to be stronger whencustomers have accelerated the purchase than when theyhave not accelerated the purchase.

To analyze the effects of extending a sales promotionon potential customers’ as well as existing customers’attitudes and behaviors, we conducted two laboratoryexperiments. In the first experiment, we employed a 2x2factorial design and manipulated customer type (potentialcustomer, existing customer) and sales promotion timeextension (no time extension, time extension) using ascenario approach. We measured attitude toward the firm,recommendation and purchase behavior as dependentvariables. A total of 192 students with an average age of21 participated in the study. Eighty-one point eight per-cent of the participants were female. Supporting ourhypotheses, the results of the two-way ANOVA show thatexisting customers’ and potential customers’ attitudetoward the firm are affected differently by a time exten-sion of a promotion, that is, existing customers’ regretincreases and potential customers’ regret decreases if apromotion is extended. We could also confirm mediatingeffects of attitude toward the firm on customers’ purchaseand recommendation behavior.

In our second experiment, we concentrate on thepotential negative effects of sales promotion extension onexisting customers and analyze purchase acceleration as acontingency factor. Again, we conducted a laboratoryexperiment with a 2x2 factorial design. We manipulatedsales promotion time extension (no time extension, timeextension) and purchase acceleration (no acceleration,acceleration). In total, 256 participants with an averageage of 30 participated in the study. Fifty-one point sixpercent of the participants were female. We conducted atwo-way ANOVA with attitude toward the firm as depen-dent variable. The interaction effect between extension ofsales promotion and purchase acceleration is significant.However, the direction is opposite to our assumption.When a sales promotion is extended, we do not see adifference in customers’ attitude toward the firm among


customers who have accelerated their purchase comparedto those who have not. In contrast, when the sales promo-tion is not extended, customers who have accelerated theirpurchase have a more negative attitude toward the firmcompared to those who have not forwarded their purchasedate.

The results of our two experiments allow for thefollowing implications: First, when firms are extendingtheir sales promotion they should consider the positive aswell as the negative effects. Extending the deadline of asales promotion can mean losing some of the newly

acquired customers as their action regret increases. Con-sequently, firms should carefully evaluate the return-on-marketing of an extension of a sales promotion. Second,firms could attempt to fence potential customers from theexisting customer base in their communication and onlyinform potential customers about a postponed deadline.Third, future research should develop and test creativemeans to reduce or avoid negative effects of extending theduration of a sales promotion. For example, researchcould compare the effects of extending the same promo-tion with the effects of offering a reframed promotionaloffer. References are available upon request.

For further information contact:Andrea Bruns

Marketing DepartmentUniversity of PaderbornWarburger Strasse 100

33098 PaderbornGermany

Phone: +49.5251.60.21.11Fax: +49.5251.60.34.33



PURCHASE BEHAVIOR AND PSYCHOPHYSIOLOGICAL RESPONSESTO DIFFERENT PRICE LEVELS

Outi Somervuori, Aalto University School of Economics, FinlandNiklas Ravaja, Aalto University and University of Helsinki, Finland

SUMMARY

The reference price concept and how consumersreact to price changes from a reference price has beenwidely studied in economics and marketing. This infor-mation is important for companies in planning pricingstrategies and timing pricing changes.

Prospect Theory, introduced by Kahneman andTversky (1979), suggests that consumers react more tolosses (high price levels) than to gains (low price levels).The phenomenon has been extensively studied by statis-tically modeling scanner panel data of frequently pur-chased grocery products. However, empirical results aremixed; some concluding that consumers are more respon-sive to losses and others that consumers are more respon-sive to gains, while some of studies report symmetricbehaviour (Bell and Lattin 2000; Mazumdar and Papatla1995; Putler 1992; Terui and Dahana 2006).

Our study will extend the research on the discussionabove. Our aim is not only to look at the behavior aroundreference price (normal selling price) but also to under-stand the emotional responses elicited by low and highprices, and brand. In this study we apply psychophysi-ological measures to study the emotional responses.

Even though recent research on emotions has identi-fied, for example, that emotions have a significant role indecision making (Vohs, Baumeister, and Loewenstein2007), the current pricing literature has paid only limitedattention to the role emotions play in how people respondto prices and price information. The advent of psycho-physiological measures in psychology has made it pos-sible to include reliable measures of emotions also inpricing research. Psychophysiological measures can poten-tially add a new dimension to our understanding of emo-tional processes – a dimension that we cannot necessarilytap if we only record behavioral responses. Their use maylead to a more complete and objective understanding ofconsumer desires, and may consequently assist compa-nies to adjust their strategies.

The research was conducted as a laboratory experi-ment. The idea in the experiment was to study partici-pants’ purchase behavior and psychophysiological reac-tions when product’s price levels were changed fornational brand and store labeled products in seven differ-ent product categories. Altogether 33 right-handed healthy

business students participated in the experiment. A 7(Product Category) × 2 (Brand) × 15 (Price) within-subjects design was employed.

The participants were asked to imagine them groceryshopping in a local supermarket and having 40 to spend(their endowment). In each trial, they were shown apicture of a product with a price and they were askedwhether they want to buy the product or not. All partici-pants were presented with 224 trials in a random order.While the participact conducted the experiment theirfacial electromyographic (EMG) activity and electroder-mal activity (EDA) were recorded. In return for theirparticipation, the participants could keep the purchasedproducts and that part of the endowment they had notspent when leaving the experiment.

All data were analyzed using the Generalized Esti-mating Equations (GEE) procedure in SPSS. In the GEEprocedure, the dependent variable is linearly related to thefactors and covariates via a specified link function. TheGEE approach requires the specification of the correlationstructure of the repeated observations of the dependentvariable, distribution of the dependent variable, and linkfunction. The GEE models were introduced by Liang andZeger (1986), and the method has received wide use inmedical and life science research (Ballinger 2008).

In 38% of the trials (n = 224), the participants choseto purchase the product and, in 62 percent of the trials, notto purchase the product. The frequency of purchasesvaried across product categories and brands. When ana-lyzing the results, we considered the direct influence thatemotions have on purchase decision and the influence thatprice and brand has on the elicitation of emotions. Wefound that increased zygomatic EMG activity (an index ofpositive emotions and approach motivation) predicted anaffirmative decision to purchase a product. However,emotional arousal as indexed by EDA did not have asignificant impact on purchase decision. In this study, thismay be due to the low value of items purchased (theaverage price of products was 1.72 ) as previously EDAhas been found to be an important construct for theexplanation of buying behavior (Groeppel-Klein, 2005).When we looked at the elicitation of zygomatic EMGacitivity, we found that low prices elicit significantly morezygomatic EMG activity than high prices. Peine et al.(2009) had similar finding in their research where partici-


pant in self-reports expressed that a price increase lead tochanges in price affect. In our study, the increased zygo-matic EMG activity was grater for national brand productsthan for private label products. As brand products are seento provide comfort, security, and value (Hankuk andAggarwal 2003) the greater emotional attachment seemsnatural.

Price and brand have also direct influence on pur-chase decision. The results suggest that a low price leveland private label product predict affirmative purchasedecision. As private label products are cheaper they mayinduce direct positive influence on purchase decision.However, the national brand products seem to elicit morepositive emotions. It may be that via increased positiveemotions/approach motivation the reaction to price changesis stronger for national brand products than for privatelabel products.

The identified behavior around a reference price inour study supports some of the previous findings thatconsumers react more strongly to price decreases of

national brand products compared to private label prod-ucts (Bronnenberg and Wathieu 1996) and that consumerbehavior around a reference price is mixed (Halme andSomervuori 2009; Hankuk and Aggarwal 2003; Klapperet al. 2005). Our experiment highlights that not only lowprice levels but also high price levels have a larger effecton demand for national brand products compared toprivate label products. That being so, all price changes aremore critical for national brand products than for privatelabel products. In addition, the results indicate that theconsumer purchase behavior around a normal price ismore gain seeking for national brand products, whereas itis more loss aversive for private label products. Theresults suggest that emotions may have a role in lossaverse/gain seeking behavior.

This research highlights the importance of emotionsin purchase decision process and on the role emotions playin how people respond to prices and price information.Our work is a good example of how psychophysiologicalmeasures may be applied in marketing and pricingresearch. References are available upon request.

For further information contact:Outi Somervuori

Aalto University School of EconomicsRuneberginkatu 22–24

00100 HelsinkiFinland

Phone: +358.50.60295E-Mail: [email protected]


SOCIAL COMMONALITIES AND SUBJECTIVE DISCOUNTING

Mark S. Rosenbaum, Northern Illinois University, DeKalbRichard Wozniak, Northern Illinois University, DeKalb

Carolyn Massiah, University of Central Florida, Orlando

SUMMARY

In many retail transactions, such as those involvingjewelry, furniture, furs, antiques and estates, automobiles,pawn sales, and artwork, sales associates are usually ableto offer their customers subjective price accommodationswithout needing to obtain managerial approval for doingso. The literature is replete with articles showing thatcustomers often expect financial discounts from salesassociates with whom they maintain relational ties, basedon a working history, or with organizations with whichthey have an established purchasing history, such as byparticipating in the organizations’ loyalty programs orbrand communities (Berry 2002; Gwinner, Gremler, andBitner 1998; Mulhern and Padgett 1995; Verhoef 2003).In addition, in many cases, families and close friends havecome to expect financial discounts from some retail salesassociates under the auspices of “family-and-friend” spe-cials, such as those offered to General Motors’ and Macy’semployees.

However, a dearth of knowledge remains aboutwhether customers who maintain weak social ties to salesassociates also expect to receive price accommodationsduring marketplace exchanges. Nearly 20 years ago,Goodwin (1996) speculated that commercial friendships,or so-called commonalities between service providersand customers, require time to develop, but when in place,they profoundly influence working relationships so that,quite often, customers exhibit organization loyalty andresistance to switching when they sense commonalitieswith particular employees. Furthermore, customers whoshare close commercial friendships with service provid-ers often expect demonstrations of appreciation, perhapsin the form of unexpected discounts or even birthdaypresents (Price and Arnould 1999). Although these inves-tigations remain valid, we speculate that the recognitionof some shared social commonalities between customersand retail employees may occur within a single market-place occurrence and, similar to long-term commercialfriendships, may provoke customers to believe that theyare entitled to resources in the form of discounts.

We draw on marketing and anthropological theory toargue that customers who realize that they have common-alities with employees (e.g., through Facebook or LinkedInlinkages), shared life experiences (e.g., chronic diseases,physical handicaps, university attendance, fraternity mem-bership), or shared lifestyles (e.g., affinity to a sportsteam, sexual orientation) may believe that they are entitled

to price accommodations from these employees merelyby virtue of shared commonalities. Furthermore, we extendthe commercial friendship paradigm and service nepo-tism theory (Rosenbaum and Walsh 2011) by showing theextent to which customers believe that they are entitled todiscounts from retail employees simply because theyshare weak, nonfamilial, social commonalities with them.We also show that this sense of entitlement derives fromcustomers’ perceptions of shared commonalities as animpetus that propels them into an organization’s so-calledbest status.

We speculate that the recognition of anysociocollective commonality between customers andretail employees during marketplace exchanges may besufficient to encourage customers and like employees toband together in consumption settings, in turn resulting inthe receipt and offer of discounts, respectively. To under-stand this speculation, we turn to group nepotism theory(Jones 2000), which derives altruistic behaviors from asociocollective level – for example, people who sharecommonalities, such as Facebook, Twitter, or LinkedInconnections, fraternity/sorority memberships, sharedlifestyles, or hobbies. Surprisingly, group nepotism theoryposits that at this level, even weakly bonded kin areinclined to share resources with each other.

We explore three research questions in this study.The first asks whether the mere knowledge of sharedcommonalities between customers and retail employeesresults in customers believing that they are entitled torelational benefits, such as price discounts. The secondasks whether the knowledge of shared commonalitiesbetween customers and employees results in customersbelieving that they are entitled to discounts because theyview themselves as favored by employees as best custom-ers. The third research question asks whether the knowl-edge of shared commonalities between customers andemployees results in customers believing that they areentitled to discounts, even when they know that employ-ees may confront job termination when providing themwith the discounts.

To address the research questions, the study employssurvey methodology. As a result, 199 respondentsanswered the survey in the first condition, 229 answeredthe survey in the second, and 153 answered the survey inthe third, for a total of 581 respondents (49.5% male,average age 21.8 years). Respondents read a scenario


explaining that Pat was a sales associate in a jewelry store.In the first condition, respondents were told that Pat couldoffer customers a discount on jewelry without managerialapproval. This condition explores the extent to whichconsumers believe that they are entitled to price accom-modations because of their shared commonalities. In thesecond scenario, respondents were told that Pat couldoffer customers a discount without managerial approvalbut that the discount was reserved for the store’s bestcustomers. The third condition mirrors the second, exceptthat in this condition, Pat told the customer that he couldget fired for disregarding the organization’s discountingpolicy. After reading each scenario, respondents consid-

ered whether they as customers felt entitled to receive adiscount from Pat because they shared 12 different com-monalities with him (yes or no).

Overall, the findings reveal that respondents believedthat they were entitled to monetary discounts simplybecause they shared social commonalities with the retailemployee. The results also show that respondentsbelieved that commonalities catapulted them to “best-customer” status. Finally, many respondents believed thatthey were entitled to discounts despite placing theemployee’s job at risk. References are available uponrequest.

For further information contact:Mark S. Rosenbaum

College of Business, Department of MarketingNorthern Illinois University

DeKalb, IL 60115Phone: 815.753.7931



PRICE-QUALITY RELATIONSHIPS AND THE PRICE HEURISTIC

Mark J. Kay, Montclair State University

SUMMARY

Correlations between price and quality (PQ) havetraditionally been noted to be low. Yet consumers stub-bornly persist in utilizing price as a highly important cueto quality. Arguably, price is an especially important anddeterminative choice heuristic since it is doubly “usable”to customers: price is both available at the point ofpurchase and has immediacy to the choice decision at theprecise time the decision is being made.

Price is especially relevant in instances where buyingdecisions are made quickly. A “good buy” is often contin-gent on getting a discounted price, especially for manybranded fashion goods offered in limited quantities.Retailers often purposely configure buying decisions tobe made in a short time frame by offering marked down“specials” and placing items temporarily “on sale.”

Finding a discounted high quality brand can certainlybe an exciting “find” for consumers. Hence price cuts canbe an effective marketing tactic. Consistent with dualprocess models of cognition (Kahneman and Fredrick2005), store managers may effectively motivate quickbuying decisions that are intuitively appealing and impul-sive. They may thus be able to overcome much morecareful and deliberative cognitive choice processes.

Managers may use price deals to stimulate positiveand memorable shopping experiences and thus bolstertheir brand. While discounted prices can be an especiallyimportant motivator to these “quick choice” situations,dropping prices can also be a problematic decision forbrand managers. Lowering prices significantly affectsperceptions of quality. Yet store managers regularly dropprices in response to lower than expected sales volumes.Prices need to be strategically assessed by managers inthis context. Managers clearly need better tools toimprove branding decisions.

This study revisits the extensive history of the PQliterature to serve this goal. Several researchers havecontended that low PQ correlations are evidence thatmarkets are inefficient (e.g., Ratchford and Gupta 1990).This study takes a different approach.

The effort was not to assess the so-called “efficiency”of markets as they are related to PQ relationships. Instead,price and quality were examined in the context of devel-oping a better understanding the changing spectrum mar-

ket offers. PQ correlations and scatterplot data wereexplored as potentially simple and useful analytic tools toaid what are often fast and immediate decisions of market-ing managers.

These tools are especially useful in conditions whenprices are subject to quick managerial revision. The issueis especially relevant to many situations that have emergedin which prices are subject extraordinarily rapid change.For example, in the merchandising strategy of what hasbeen called “fast fashion,” rapid price change is an espe-cially important. PQ relationships would naturally behighly unstable in this context.

Utilizing data from Consumer Reports, PQ relation-ships were examined in several categories, and longitudi-nally where possible. Looking at a wide range of data overthe past three decades, PQ relationships were seldomstable or persistent over time. PQ relationships appear torapidly and competitively change. Additionally, somesurprisingly high PQ correlations were occasionally found.In this context, PQ measures were considered a “snap-shot” of short term market performance.

While PQ correlations alone are simple and usefulquantitative measures, they have their limitations in cat-egories in which a limited number of brands compete. Thenumber of products competing in many product catego-ries (and subcategories) may consist of a rather smallnumber. This commonly produces an “n” value of dataobservation points that are below the standard that arigorous statistical analysis would require. Counter-intu-itively, a low level of PQ correlation is not always exhib-ited in categories with few competitors.

As illustrated by the example of the Anscombe (1973)quartet, datasets with identical statistical properties can bedifferent when visibly graphed. In fact, the visual inspec-tion of PQ data in a scatterplot form reveals that evenstrong correlations may not be linear. Particularly if thecost of achieving higher levels of quality in a categoryvaries, a curving PQ relationship would be expected.

Scatterplots are discussed as a complementary ana-lytic PQ tool. Scatterplots are able to visibly picture orrepresent the range of competitive offerings that composea market. Scatterplot charts also have the advantage ofallowing a quick and easy identification of the “non-efficient” outliers in the price to quality spectrum. Addi-tionally, scatterplots often suggest to managers some


immediate competitive opportunities. This is illustrated insituations in which PQ product cluster into high and low-priced offers. Such competitive clusters are suggested byPorter’s “generic” approach to cost and differentiationstrategies.

Finally, there are times that the PQ relationship couldreasonably be expected to be quite low. During times oftechnology change, during times of scarcity, and duringnew product introductions, prices could be expected to beat least temporarily high. A strong PQ relationship wouldtheoretically appear to be conditional. PQ correlations

and scatterplots were found to be valuable tools to exam-ine changing competitive strategies.

Finally, PQ decisions are particularly relevant asconsumers are increasingly empowered by digitaldevices. Online consumer reviews and social network-ing increases information diffusion substantially. Poten-tially important are convenient mobile devices and appsthat can provide fast access to product reviews. These canaid immediate shopping decisions and alter these PQrelationships in some categories. References are availableupon request.

For Further Information Contact:Mark J. Kay

Marketing DepartmentMontclair State University

1 Normal AveMontclair, NJ 07043Phone: 973.655.7445



PRODUCT MASKING: EFFECTS OF CONSUMER EMBARRASSMENTON SHOPPING BASKET SIZE AND VALUE

Bridget Satinover Nichols, Northern Kentucky University, Highland HeightsDavid Raska, Northern Kentucky University, Highland Heights

Daniel J. Flint, University of Tennessee, Knoxville

SUMMARY

Consumers experience feelings of embarrassment inmany social situations, including shopping in a retailsetting. Embarrassment is a socially-dependent emotioncapable of motivating people to engage in behaviorsaimed at reducing or avoiding social rejection and judg-ment, some of which affect what and how consumers buyproducts. Wilson and West (1981) were early in formallyaddressing the topic of “unmentionables” as it relates tomarketing by categorizing these products as (1) thoseconsidered unpalatable by society, but tolerated and soughtafter, and (2) those that are accepted by society, but thatthe buyer is reluctant to acknowledge. Despite socialclassifications, both carry the ability to create feelings ofembarrassment for some consumers, especially when thepurchase must be made in a social setting like a conve-nience store, pharmacy, or similar retail establishment.According to Goffman (1959), when individuals are in-volved in any kind of social interaction, they engage incertain practices in order to avoid being embarrassed orembarrassing others. In this view, people modify theirbehavior to control the outward impression given toothers.

Because embarrassment is a strong socially-depen-dent emotion, it is capable of motivating people to engagein behaviors aimed at reducing or avoiding social rejec-tion, judgment, or increased levels of the emotion. Forexample, Lewittes and Simmons (1975) conducted natu-ralistic field studies of men buying “girlie” magazines.They found that college-aged men who purchased girliemagazines were more likely to request a bag than menwho purchased other types of magazines. They were alsomore likely to purchase other items like gum or candy.They concluded that requesting the bag and purchasingother items was an attempt to avoid negative externalevaluations, avoid embarrassment, and disguise or “mask”the true nature of their shopping trip. Along these lines,Moore et al. (2008) reported on coping strategies thatconsumers use when purchasing embarrassing products.They found that embarrassment felt when purchasingcondoms is counteracted by both cognitive and activecoping strategies.

The present set of studies focus on these behavioraleffects of, or active coping strategies for, embarrassment

in a retail setting. In Study 1 we were interested indiscerning the level of embarrassment that would beanticipated in association with purchasing a wide set ofpersonal care products that could be considered sensitive,unmentionable, or controversial, and if consumers wouldreadily acknowledge masking behaviors. The studyincluded 357 undergraduates (M

age = 22) from two large

public universities who indicated their degree of antici-pated embarrassment with a variety of both sensitive (e.g.,condoms, fungal creams) and non-sensitive type products(e.g., toothpaste, sunscreen). Data were collected via anonline survey that displayed images of such products. Theresults suggest that participants’ anticipated embarrass-ment was significantly higher for sensitive products thanfor non-sensitive products (p < .05). Additionally, morethan two-thirds of participants acknowledged the use ofmasking as a coping strategy. These two-thirds wereasked to describe the masking behavior and circum-stances. Responses suggested that products related tosexual health and feminine care might be strongly relatedto masking behavior as a coping strategy. Study 1 pro-vided us with initial evidence that college students acknow-ledge participating in masking behavior when purchasingproducts that cause them to experience embarrassment.Further, there seems to be some product categories thatelicit greater embarrassment than others.

The main purpose of Study 2 was to empiricallyexamine relevant retail outcomes of masking behaviors.Specifically, our intentions were to examine how (1)masking behaviors influence the total basket size andbasket value of drug store shopping visits, and how (2)embarrassment influences checkout behaviors aimed atreducing embarrassment. Using an experimental design,participants were randomly assigned to the task of pur-chasing either an embarrassing product (hemorroid cream)or a non-embarrassing product (bandages) from a drugstore. Then they were asked to shop for any other itemsthey would like to buy. Three hundred eighty undergradu-ate students (M

age = 21) completed the online study, in

which they were asked to imagine that they were shoppingin person at a well-known retail drug store. Participantswere able to browse product assortments in a variety ofproduct categories (e.g., drinks, snacks, magazines), andselect items they wished to purchase. Each product waspriced accordingly to the retailer’s actual item prices.When finished, participants were asked to answer a vari-


ety of questions, including how they would check out withtheir items. Those in the embarrassment condition (n =186) expressed greater embarrassment with their pur-chase task than those in the non-embarrassment condition(n = 194), p < .0001. Additionally, a marginally signifi-cant difference in the number of products purchased (p =.07), and a significant difference in the value of theshopping basket (p < .05) was observed such that those inthe embarrassment condition purchased more items andspent more money than those in the non-embarrassmentcondition. Participants did not indicate any gender con-gruence with respect to the checkout cashier, nor were anysignificant differences present for preferring a self-check-out lane. Additional analyses include assessments of

socially-relevant personality variables and post-hoc testsregarding the types of products used to mask the embar-rassing purchases.

While preliminary, this research suggests that a reduc-tion in the emotion of embarrassment could be detrimentalto the retailer. Specifically, the present research suggeststhat the retailer stands to gain a significant revenue in-crease when shoppers enter the store tasked with anembarrassing purchase. Ethically speaking, while theretailer might not attempt to increase the felt embarrass-ment, they may also decide to avoid trying to reduce it.More research is needed to streamline and generalizethese findings. References are available upon request.

For further information contact:Bridget Satinover Nichols

Northern Kentucky UniversityHaile/US Bank College of Business, BAC #215

Highland Heights, KY 41099Phone: 859.572.6558



EXAMINING THE ROLE OF COGNITIVE DISSONANCE AFTERPURCHASE ON PRODUCT RETURN INTENTIONS

Devdeep Maity, Oklahoma State University, Stillwater

SUMMARY

The U.S. retail industry loses billions of dollars a yearbecause of product returns, but scholarly articles examin-ing consumers’ product return intentions have been some-what limited. This study explores how cognitive disso-nance after purchase affects consumers’ product returnintentions. Using cognitive dissonance theory, it exam-ines product return intentions in the context of both lenientand strict return policies. The results indicate that cogni-tive dissonance after purchase positively influences prod-uct return intentions but the relationship remains unaf-fected at strict and lenient return policy situations. Thisquestions the previously held notion of retailers to controlproduct return by imposing strict return policies.

Product returns cost U.S. retailers $100 billion peryear in lost sales (Petersen and Kumar 2009). The elec-tronics industry alone spent around $13.8 billion in 2007to restock returned products (Lawton 2008). Return pro-portions for electronic retailers are at least 6 percent(Strauss 2007) and for catalog retailers as high as 35percent (Rogers and Tibben-Lembke 1998). For productssuch as personal computers return proportions are as highas 25% (Mixon 1999), eroding retailers’ profit margin by4 percent (Petersen and Kumar 2009) and sometimescausing them to file bankruptcy. Retailers often imposestrict return policies to control such disturbing productreturn rates (Bower and Maxham 2006; Hess et al. 1996;Wood 2001) without an in depth understanding of theeffect of such policies and more importantly why consum-ers form product return intentions in the first place. Thus,to address the above issues, the role of cognitive disso-nance as an antecedent to product return intentions and therole of return policy leniency as a moderator are examinedin this study. It is proposed that consumers will formproduct return intentions to reduce their cognitive disso-nance after purchase and that the relationship will vary atdifferent levels of leniency of a retailers’ return policy.

In the pre-purchase phase, consumers have an idealproduct or purchase requirement in mind, but they oftenpurchase a product that deviates from this ideal require-ment (Shulman et al. 2010). This deviation may occurbecause of something as simple as a wrong choice (Petersenand Kumar 2009) resulting from the difficulty in choosingamong closely competitive alternatives. Such deviationsmay give rise to uncomfortable post-purchase feelingsknown as cognitive dissonance. According to cognitivedissonance theory (Festinger 1957), consumers exhibit a

motivation to reduce their psychological discomfort.Festinger (1957) discussed various ways that an indi-vidual can reduce such psychological discomforts. Adop-tion of a specific dissonance reduction strategy will de-pend on the reversibility of the purchase decision(Cummings and Venkatesan 1976; Korgaonkar andMoschis 1982; Mowen 1995). If the purchase decision isreversible (i.e., when there are product return facilities),consumers will reduce dissonance by undoing the behav-ior in order to maintain the attitude – behavior consis-tency. On the other hand, if the purchase decision isirreversible (i.e., when there are no product return facili-ties) consumers will display irrevocable commitment tothe decision they made, reducing their dissonance bychanging their attitude toward the product (Cummingsand Venkatesan 1976; Korgaonkar and Moschis 1982;Mowen 1995). It is argued that in the US retail market dueto its fierce competition among retailers to satisfy con-sumers, almost all retailers offer some kind of productreturn options (Davis et al. 1998). In any such situationthat involves some type of return options, it is reasonableto assume that consumers will not show such irrevocablecommitment toward their purchase decisions. Instead,when they perceive cognitive dissonance after purchase,they will be much more likely to reduce dissonance byforming return intentions using the offered return facili-ties, thus restoring psychological balance and comfort(H

1). Additionally, it can be argued that a lenient return

policy offers a higher degree of reversibility of a purchasedecision than a strict return policy, since it imposes fewerrestrictions and hassles to the consumer than a strict returnpolicy. Thus, a lenient return policy will encourage con-sumers more to undo the buying event and cause them toform higher product return intentions than a strict returnpolicy, in order to reduce discomfort and restore psycho-logical balance (H

2).

The data for this study was collected from 120 under-graduate students enrolled in marketing courses at a largestate university in the central part of United States. As apart of the survey, each student was initially required toprovide a written description of a recent product purchaseexperience which involved a difficult choice between twoor more close alternatives (Sweeney et al. 2000). Thewritten description was followed by three sections con-sisting of many survey items. The first section of thequestionnaire contained the cognitive dissonance afterpurchase scale (Sweeney et al. 2000), which is a twentytwo item, seven point Likert scale designed to capture themulti-dimensional construct cognitive dissonance after


purchase. The second section of the questionnaire con-tained the return policy leniency index. This six-itemmeasure was adapted from Davis, Hagerty, and Gerstner(1998) and Bonifield, Cole, and Schultz (2009). The thirdsection of the questionnaire contained the product returnintentions scale, which is a three item, seven point seman-tic differential scale adapted from the “Online purchaseintention” scale (Schlosser et al. 2006).

The findings suggest that cognitive dissonance afterpurchase positively influences product return intentionsbut remains unaffected at varying levels of return policyleniency. In other words, the study empirically verifies thepreviously assumed relationship between cognitive dis-sonance after purchase and product return intentionsimplying that in order to reduce dissonance resulting froma wrong choice, consumers will form product returnintentions. Retailers can try to minimize such dissonanceby introducing comparison charts for competing productsin their store wherever feasible.

Insignificant moderation effect of return policy le-niency, points toward important possibilities such as (1)inadequacy of the present return policy practices in theU.S., and (2) need to shift to a more centralized returnmanagement system. As long as the retailers providereturn options to the customers through some kind ofreturn policy, the issue of the degree of strictness (orleniency) of the return policies perhaps becomes unim-portant to the customers as per this study. The focal issuefor a dissonant consumer becomes then whether she/he isable to return the purchased product and correct his/her

“purchase mistake.” The U.S. retailers, while formulatingtheir return policy’s careful factor in the risky implica-tions of a strict return policy on purchase rate, long-termcustomer relationship and store patronage issues. That iswhy in most situations the “strict” return policies are notstringent enough. As a result, the restrictions imposed onconsumers upon return such as mode of refund (cash vs.non-cash refund), the product and packaging condition orreceipt requirements, or imposition of time limits oradditional fees are not stringent enough to prohibit con-sumers from forming product return intentions. To in-crease the stringency of return policies, retailers can comeup with tailor-made strict return-policies (such as accept-ing return only if the product is defective) which maysuccessfully weaken consumers’ product return inten-tions in many situations.

Additionally, perhaps the retailers might considershifting to a more centralized return management systemthat involves recording consumers’ return practices andshare the data across retailers. Such return data can beused to arrive at a score-based return system which can befurther used to profile consumers based on their returnpractices (similar to the present credit score practices inthe U.S.). Furthermore, such data can be used to offertailor-made return facilities, specific to the individualconsumer profiles (Janakiraman and Ordóñez 2011).Moreover, such data can be integrated with the existingcustomer relationship management systems to arrive atappropriate levels of return restrictions. Future researchcan investigate the feasibility of such systems. Referenceswill be available on request.

For further information contact:Devdeep Maity

Oklahoma State University625 E Redbud Dr. Apt D43

Stillwater, OK 74075Phone: (956) 802-0643

Fax: (405) 744-5180E-Mail: [email protected]


ASSESSING CUSTOMER SATISFACTION AND UNPLANNED BUYINGBEHAVIOR IN GROCERY RETAILING

María Pilar Martínez-Ruiz, University of Castilla-La Mancha, SpainAlicia Izquierdo-Yusta, University of Burgos, Spain

Phil Megicks, University of Plymouth, United Kingdom

SUMMARY

Retail managers must seek the most effectiveresources and capabilities to satisfy their customers, espe-cially in difficult economic contexts that encourage con-sumers to control their spending. For example, in thecurrent economic downturn, food retailers must find newways to maintain or improve their performance, such asthrough more unplanned purchases by consumers, whichthey make on the basis of decisions at the point of sale(e.g., Bell et al. 2008, 2011; Inman et al. 2009). However,some research suggests that shoppers may make fewerunplanned purchases than previously believed. There-fore, we have attempted to determine which factors con-tribute more to customer satisfaction and assess if thereare differences, depending on shopping lists usage byconsumers, a variable that we use to measure theirunplanned purchase behavior. This research objective inturn reveals which store attributes and factors food retail-ers should consider most carefully to enhance customersatisfaction.

Bearing these goals in mind, we conducted anin-depth review of relevant literature pertaining to storeattributes, unplanned purchase behavior, and their rela-tionship with customer satisfaction. By conducting afactorial principal components analysis of 422 consumersfrom a representative Spanish city, we identified threefactors underlying consumer perceptions of the value ofdifferent store attributes. Then our parametric linear re-gression analysis revealed that by running a regressionanalysis for different consumer segments with diverseunplanned purchasing behavior, we could uncover uniquesatisfaction responses, depending on consumers’ specificbehavior.

This study thereby offers conclusions and ideas forfurther research into store attributes and their underlyingfactors, as well as their influence on overall satisfaction.Moreover, we demonstrate the possibility of determiningvarying perceptions of store factors, according to con-sumers’ unplanned purchase intentions, operationalizedby a shopping list usage variable. This important contribu-tion highlights which store attributes and factors retailmanagers should address first if they want to enhancecustomer satisfaction across different types of purchaseintentions. Consumers who exhibit planned buying want

services and convenience, as do consumers who exhibitunplanned behavior. These consumers are also very con-scious of the products and services they demand, and theirdecisions are not affected by the quality image of the store.Price discounts and promotions even cause concern amongthis consumer group. Finally, consumers who exhibitunplanned behavior sometimes are affected most by ser-vices and convenience, but they also consider the qualityimage and economic value. Thus this last consumer seg-ment clearly differs from the other segments, and retailmanagers have more room to maneuver their various storeattributes and factors to enhance customer satisfaction.

These findings emerged from data collected at thebeginning of a significant economic downturn in theSpanish economy, which seemingly should have increasedthe importance of the economic factor. In contrast, theresults show it was valued positively (though least) onlyby customers who sometimes use a shopping list. Amongthe other groups, the factor either did not contribute tooverall customer satisfaction or had a negative impact.Perhaps the most compelling of the possible explanationsfor this finding is the chance that many consumers simplywere not aware of the economic downturn when wecollected the questionnaires. As previous studies havenoted though (e.g., Freshplaza 2009), in 2008 Spanishconsumption patterns had begun to change, comparedwith the years prior to 2008, including greater valueplaced on aspects related to the purchase scenario (e.g.,new services) and grocery products (e.g., quality, innova-tion). Grocers therefore began to issue more appealsrelated to services, convenience, and quality, instead ofjust costs and/or production. Moreover, because morethan half the questionnaires were collected in EDLPpositioned supermarkets that emphasized customer ser-vices and convenience, perhaps the economic value factorhad less importance than other factors in these storesspecifically. We also acknowledge that a shopping listmay be a relatively poor surrogate for unplanned buyingbehavior, in that it is overly simplistic to claim that onlyconsumers who use shopping lists plan their trips to thegrocery store. Finally, the findings are consistent withacademic research that indicates consumers tend to avoidpromotions involving uncertainty, so such promotionswould not be particularly effective to achieve their in-tended results.


The results reinforce the importance of services andconvenience. To keep customers satisfied, regardless oftheir unplanned purchasing behavior, retail managersshould develop and communicate value propositions thatfocus on value-added services, such as wider productassortments, increased distribution intensity, high-qualitycustomer attention programs, additional services, enhancedstore atmospheres, and longer operating hours. A widerassortment also benefits consumers by providing manychoices, as does increasing the availability of stores orextending into online channels (especially for fast movingconsumer goods, for which an online presence can stimu-late sales by fostering wider consumer recognition andimpulse buying). An effective customer satisfaction man-agement program, as part of the retailer–customer rela-tionship management strategy, should create a positive,fair working environment to motivate employees to takebetter care of customers. Services such as advice, homedelivery, or free parking can increase store attractiveness;once inside, a positive atmosphere can improve thecustomer’s shopping experience. Extending operatinghours also might influence the level of consumer demand.

Next, grocery retailers should develop a coherentprice and promotion strategy to avoid the potential forcustomer reactance. Managers should determine the priceimage they want to communicate and the promotions theywill offer (e.g., type, magnitude), ensuring they are totallyconsistent with both the desired price image and thepricing policy. For example, price discounts and salespromotions are more aligned with a high-low pricing

policy. The low valuation our respondents expressedregarding the economic value factor might indicate thegreater convenience of everyday low pricing strategiesthough.

Several questions also remain to be addressed. Fur-ther work should analyze the influence of the identifiedfactors on the degree of loyalty consumers exhibit towardcertain commercial formats or specific establishments.Satisfying consumers is particularly important for compa-nies in this industry, because grocery retailers aim toachieve long-term, loyalty-based relationships. As Oliver(1999) points out, satisfaction is a necessary step inloyalty formation, but other mechanisms also enter intothe construction of loyalty. We call for research to assesshow store factors might influence the degree of selectionin certain product categories, particularly in terms ofpurchase frequency. Noting the relatively low varianceexplained by the models, we also recommend furtheranalyses that incorporate other potential explanatory vari-ables related both to store attributes, such as the speed ofcheckout, accuracy of scanned prices, or availability offresh food, as well as respondents’ economic vulnerabil-ity, attitudes, and demographics. These factors may havegreat relevance for assessments of customer satisfactionand should be included in further investigations of this keyarea of importance for retail service managers. Finally,alternative measures of unplanned buying behavior,beyond shopping list usage, could expand research intogrocery retailing and customer satisfaction. Referencesare available upon request.

For further information contact:María Pilar Martínez-Ruiz

University of Castilla-La ManchaAvenida de los Alfares, 44

16071 CuencaSpain

Phone: +34.902.204.100Fax: +34.902.204.130



FLOW EXPERIENCE IN PHYSICAL RETAIL STORES

Liz C. Wang, West Chester University of Pennsylvania,Daniel F. J. Hsiao, University of Minnesota Duluth

SUMMARY

Introduction

Internet marketing research has found that onlineconsumers with a flow experience tend to be less price-sensitive, and develop more favorable attitudes towardonline stores (Novak et al. 2000; Wang et al. 2007).Surprisingly little research has been conducted to investi-gate similar flow experiences in brick-and-mortar stores,which contribute more than 90 percent of retail industrysales (Anderson et al. 2011).

The objectives of this paper are twofold. First, itprovides a theoretical foundation to develop the anteced-ents (challenges and skills) associated with in-store shop-ping activities. Secondly, it aims to further investigate theflow theory in physical retail environments.

Theoretical Background and Research Methods

According to the flow literature for online shopping,the two antecedents of flow (navigational challenge andInternet/computer skill) are specifically related to Internetsurfing and not directly applicable to shopping activitiesin traditional brick-and-mortar stores. The theoreticalfoundation for antecedents is lacking to extend flow to in-store shopping. The first study used focus group inter-views to explore consumers’ shopping challenges andtheir shopping skills.

In addition, prior research suggests that consumers’different states of shopping experiences may influencetheir future shopping intention. An empirical study wasfurther conducted to examine if the two antecedents (in-store shopping challenge and skill) could influence con-sumers’ shopping experiences (such as flow, or bore-dom), which then impact their shopping intention.

Research Results

The results suggest that a shopping activity can bechallenging depending on the level of complexity andinherent risk associated with a product. Consumers’ skillsmay vary based on their level of knowledge, familiarity,or experience with a particular product category. Con-sumers may experience different states of shopping expe-rience (such as anxiety, apathy, boredom, or flow) inphysical retail stores, resulting from their perceived chal-lenges and skills. In addition, consumer’s future shoppingintention was positively associated with their shoppingexperience.

Implications

The current research extends the flow theory tophysical retail stores. It suggests the two antecedents(shopping challenge and skill) for the shopping context.Retailers can develop marketing strategies to enhanceconsumers’ skill to deal with shopping challenges. Refer-ences are available upon request.

For further information contact:Liz C. Wang

Department of MarketingWest Chester University of Pennsylvania

700 South High StreetWest Chester, PA 19380

Phone: 610.436.2259Fax: 610.436.2592



EFFECTS OF ILLUMINATION ON STORE ATMOSPHERE, PRICE ANDQUALITY PERCEPTION, AND SHOPPING INTENTION

Stephan Zielke, Aarhus University, DenmarkThomas Schielke, Darmstadt University of Technology, Germany

SUMMARY

Although, store design and atmosphere are popularresearch fields, only a few studies focus on store lightingin particular (e.g., Areni and Kim 1994; Babin et al. 2003;Briand and Pras 2010; Summers and Hebert 2001). Thisis astonishing, as lighting is an important design param-eter that retailers can change easily with subsequentconsequences on energy costs and customer behavior aswell.

However, especially predicting the impact on cus-tomer perception and behavior is a challenging task.Firstly, lighting can be characterized by a number ofparameters, such as brightness, the number and arrange-ment of light beams or the color and saturation of the light.Secondly, lighting influences a number of related vari-ables, such as store atmosphere, price and quality percep-tion, and shopping intention. Predicting consequences oflighting parameters becomes even more challenging, as insome cases conflicting theories suggest opposite effectscanceling each other out. Bright light might, for example,increase perceived pleasure and thereby serve as a cue forhigher prices. At the same time, bright light is also acharacteristic of discount stores. Contradicting effectsalso exist for color. Previous color research indicates thatblue is preferred over orange, while research on colorappropriateness indicates that customers rate orange asmore appropriate for sensory-social products, such asapparel (e.g., Bottomley and Doyle 2006).

These examples illustrate that it is not trivial todevelop hypotheses about the effects of store lighting.This might be a reason, why a comprehensive analysis ofdifferent lighting parameters and their interactions on abroader range of dependent variables is still missing in theliterature. This paper closes this research gap by reportingresults from two experimental studies. In the first experi-ment, we analyze the impact of the number and brightnessof light beams on store atmosphere, price and qualityperception, and shopping intention. In a second experi-ment, we investigate the impact of color type and satura-tion on the respective dependent variables.

In the experiments, lighting conditions were variedthrough digitally manipulated pictures of a fashion store.Respondents were randomly assigned to the stimuli. Theyrated the pleasure, price level perception, quality percep-

tion, and shopping intention for the selected store com-pared to a second store that was identical in all treatments.All variables were measured with single items on seven-point scales. Each stimulus was rated by 40 to 50 personsin an online survey. The total sample consisted of 338respondents, relatively equally distributed across thestimuli. To guarantee homogeneity between the groups,the majority of respondents were students of a largeEuropean university. Data was analyzed using MANOVAand separate ANOVAs. Interrelations between dependentvariables and conflicting effects were considered by medi-ation analyses through the inclusion of covariates.

The results reveal interesting insights. They showthat lighting influences several dependent variables, suchas the pleasantness of the store atmosphere, price andquality perception, and shopping intention.

Several significant effects of lighting on pleasure andshopping intention were found. The pattern of effects forboth dependent variables is similar, as most effects onshopping intention are fully mediated by pleasure. Inparticular, several interesting interaction effects wereobserved. Increasing brightness through the number oflight beams has positive effects on pleasure and shoppingintention, when beams are bright, but not for dimmedbeams (here, a smaller number of light beams result ingreater pleasure). Hence, it seems that customers preferextreme stimuli configurations versus mixed ones. How-ever, comparing both extreme configurations, many brightbeams are still preferred against a small number of dimmedbeams. This can be explained by vision theory and antide-pressant effects of light. Interesting interaction effectsoccurred also for color. The results show that orange lightresults in greater pleasure and shopping intention com-pared to blue light, supporting research on color appropri-ateness for sensory-social products. However, the posi-tive effect of orange exists only for saturated, but not forpastel colors.

The results show also that the total effects on priceand quality perception are small, which was somewhatdisappointing at first glance. For dimmed light, increasingthe number of light beams resulted in an inferior qualityperception, indicating that inconsistent stimuli configura-tions negatively affect quality perception. Furthermore,orange light resulted in a better quality perception com-pared to blue light. This underlines the preferability of


orange versus blue. The small total effects on price andquality perception can be explained by different effects,canceling each other out. The results show, for example,that bright light has a positive effect on quality perceptionvia inferences drawn from a more pleasant store atmo-sphere. At the same time, bright light has a negative effecton quality perception when controlled for pleasantness.This is an indication that bright light is also used as a cueto categorize a retailer as a discounter, resulting in a lower

quality perception. Hence, atmospheric inferences andcategorization effects cancel each other out.

The observed effects make a contribution to theliterature, as some of them were not reported in priorresearch. However, and more importantly, the resultsshed also light on the relevance of different theoriesexplaining the observed effects. References are availableupon receipt.

For further information contact:Stephan Zielke

Department of Business Administration and MAPPBusiness and Social Sciences

Aarhus UniversityHaslegaardsvej 10

8210 Aarhus VDenmark



THE HYBRID ONLINE SHOPPING ENVIRONMENT

Jeffrey Hu Xie, University of Oregon, Eugene

SUMMARY

Brick and mortar store atmospherics have long beenconsidered as a way for retailers to influence customers’mood and behaviors, creating a competitive advantagethrough differentiation. Several atmospheric models havebeen introduced to retail marketing such as the Mehrabian-Russell environmental psychology model which buildsthe relationship between environmental stimuli and cus-tomer affect and behavioral responses (Donovan andRossiter 1982; Donovan et al. 1994). However, fewstudies have addressed the effects of the online shoppingenvironments on customer behaviors. A single exception,Eroglu et al. (2001) examined online atmospherics sys-tematically but limited discussion to computer mediateddesigns.

This paper investigates this gap in the literature andproposes the Hybrid Online Shopping Environment(HOSE) framework to help investigate online shoppingatmospherics and effects. The HOSE framework is thefirst to identify the online shopping environment as con-stituted of both physical and virtual environments, andalso the first to introduce the Customer Control con-struct into atmospheric literature. The HOSE frame-work defines the online shopping environment as ahybrid system where online customers experience twotypes of environment simultaneously: a Virtual environ-ment that refers to the computer-mediated online storeswhere customers shop for products; and a Physical envi-ronment that refers to the place where a customer isphysically located while participating in online shopping.The dual environments interact with customers as well aswith each other, competing or enhancing eachenvironment’s impact on customers.

Introducing the hybrid environment concept pro-vides a more comprehensive understanding of onlineshopping environments and enables holistic investigationof all potential atmospherics and their influential effects.Compared to traditional brick and mortar store research,the HOSE framework not only incorporates the onlinevirtual world but also considerably extends the boundaryof physical locations and atmospherics by recognizing theflexibility and viability of online shopping. Meanwhile,compared to previous online shopping research whichmerely focuses on medium, website design, and computermediated process, the HOSE concept encompasses thephysical environment as well, and consequently allows

for the examination of the atmospherics from physicallocation and related space attachments.

Previous research shows that people can have emo-tional or affective ties to a place as the result of connectionwith a certain environment (Low and Altman 1992), andthe place with which a person has an association cantrigger associated memories, emotions and feelings uncon-sciously (Hidalgo and Hernandez 2001). Respectively,the HOSE framework proposes that customer associatedlocations for online shopping could be virtual and physi-cal, and also could influence customers’ online shoppingmood and behaviors. The stronger an association is, thestronger influence the environment (physical or virtual)may have on an online customer.

In environmental psychology, control is one of thefocal constructs to influence people’s behaviors, present-ing the need to demonstrate one’s competence, superior-ity, and mastery over the environment (White 1959).Existing research suggests that perception of controlcan have a positive impact on psychological and behav-ioral outcomes (Wortman 1975; Glass et al. 1969). Sur-prisingly, customer control has been consistently left outof the literature on retail atmospherics, and it is likely aresult of customers having little control and influence onthe environment setting in a brick and mortar store. On thecontrary, online shopping environment enables highercustomer control, and therefore demands the relativeresearch.

In the HOSE framework, the role of customer controlhas two levels, actual customer control and perceivedcustomer control. The Actual Customer Control (ACC) ofthe shopping environment presents the extent to which acustomer can choose, change, and act on both physicaland virtual environments; and the Perceived CustomerControl (PCC) is the extent to which a customer perceivesand believes his or her control of shopping environment.ACC recognizes online customers’ ability and authorityto predetermine what atmospheric combination from thehybrid system has potential to affect customers. Forexample, customers choose to turn on/off music eitherfrom a virtual web or a physical player, and the decisiondetermines whether music will participate into atmo-spheric influence. Meanwhile, increased PCC has apositive influence on customers’ cognitive and affec-tive responses to atmospherics.


The atmospherics in the hybrid online shopping en-vironment are expanded into both virtual and physicalcomponents and categorized as ambient, design, andsocial factors which are consistent with prior research(Baker et al. 2002). Meanwhile, multi-media interactionhas been emphasized and differentiated from other envi-ronmental factors due to media’s profound capacity toentertain and inform. In that regard, the HOSE frameworkdiscusses online shopping atmospherics’ emotional influ-ence based on a mediated stimulus-response model alongwith major cognitive effects including perception andattention. Eventually the customer behavioral responses

and business outcomes influence customer satisfaction,time spent, sales performance and customer loyalty.

The proposed HOSE conceptual framework is aimedto help gain a deeper understanding of the online shoppingenvironment and its potential effects on customers,encouraging researchers on empirical validation forthe framework’s theoretical constructs and proposi-tions. Such research will help enable industry practi-tioners to enhance business performance by leveragingonline atmospheric influences on customer behaviors.References are available upon request.

For further information contact:Jeffrey Hu Xie

Lundquist College of BusinessUniversity of Oregon

1208 University of OregonEugene, OR 97403–1208

Phone: 541.346.9083Fax: 541.346.3341



TEMPORARY STORES: CONSUMER RESPONSES TO A LIMITEDASSORTMENT AND LIMITED TIME HORIZON

Doreén Pick, Freie Universitaet Berlin, GermanyMarko Schwertfeger, Freiberg University of Technology, Germany

SUMMARY

For years, the fast-moving consumer goods (FMCG)industry and its retailers have had to address high satiationlevels in several product categories. With firm offers inboth offline and online contexts, it became much moredifficult to attract the consumer’s attention and interestand motivate him/her to purchase more and more often.Especially during the past few years of global economicdownturn, we have observed both a high industry depen-dency on consumers’ trust in economic growth and con-sumers’ adapted purchase behavior. Thus, in Europe,many apparel retail firms have had to accept hugedecreases in sales and profit. Consequently, retailersare searching for innovative marketing concepts andretail formats to attract consumers and to increase theirsales.

In the consumer goods industry, a popular strategy isto make product offers scarce by limiting the supply, e.g.,by weekly product promotions. Firms often emphasizethese limitations with slogans such as “for a limited timeonly,” sometimes even when the offer is not limited bytime. Particularly in the automobile industry and in luxuryretail, offering limited editions is very common. Theselimitations are often combined with anniversaries, such asthe 25th anniversary of the BMW M3, marked with thelaunch of a limited edition model available for the Chinesemarket. Consumers strive to buy luxury and limited itemssuch as the limited edition of the Chevrolet Camaro, 100of which were sold within three minutes at a luxurydepartment store chain (Wahba 2010). Other limitededitions have social and charity objectives, such as thespring 2011 Limited Edition Bag by Tommy Hilfigerin Aid of Breast Health International. According toeconomic-based scarcity theory, limited offers influ-ence several aspects of the FMCG industry and its

retailers, such as the extent to which consumers like andare attracted to a product, consumer preference and desirefor a product, and the perceived uniqueness of an offer.Limited offers can lead to a higher perception of productquality and can consequently increase a firm’s sales andprofits (e.g., Amaldoss and Jain 2008).

The potential value of limited offers raises the ques-tion of whether these scarcity strategies can be transferredto the retail sector by setting up limited retail stores. Howcan time-limited stores, i.e., temporary stores, influenceconsumers by increasing their willingness to visit thestore, purchase there and spread positive word-of-mouth(WOM) about it? In this context, we define temporarystores as stores set up by retailers and manufacturers toremain open for a limited time at a specific location.

In this study, we will focus on addressing the abovementioned research questions by investigating the impactof time-limited stores on consumer intentional responses.In addition, we will investigate how limited editions ofgoods also impact consumer responses. Therefore, ourresearch will extend the knowledge about limited offersby investigating different roles of limited time and limitedassortment in temporary stores. We will contribute to theliterature on scarcity theory by broadening the focus of thetheoretical framework from products to retail formats.Furthermore, we will contribute to the marketing andretail research literature by introducing the concept oftemporary stores as a new and promising retail format, andwe discuss the consequences of this format for retailers’future. Additionally, we will position our results withinthe retail management literature by showing the impact oftwo main marketing mix variables (time and assortment)on consumer intentions toward the retailer. Referencesare available upon request.

For further information contact:Doreén Pick

Freie Universitaet BerlinOtto-von-Simson-Str. 19

Berlin, 14195Germany

Phone: +49.30.83854547Fax: +49.30.83854557



DO CUSTOMER EXPERIENCES ENHANCE RETAIL BRAND EQUITY?

Alexander Leischnig, Freiberg University of Technology, GermanyMarko Schwertfeger, Freiberg University of Technology, Germany

SUMMARY

Slow market growth rates and increasing marketsaturation challenge retailers to generate competitiveadvantages that rely on alternative approaches ratherthan offering innovative products and reducing prices(e.g., Hansen 2003). Two recent trends in retail indicatebusinesses’ efforts in tackling this challenging problem:(1) an increasing focus on customer experience manage-ment (Grewal, Levy, and Kumar 2009) and (2) a growinginterest in building and establishing strong retail brands(Ailawadi and Keller 2004). Practitioners and scholarsalike have paid increasing attention to the strategy ofcustomer experience management as an appropriate meansfor differentiation in competitive retail environments(Berry, Carbone, and Haeckel 2002; Naylor et al. 2008;Puccinelli et al. 2009). According to Verhoef et al. (2009,p. 31), “[c]reating superior customer experience seems tobe one of the central objectives in today’s retailing envi-ronments.” This study explores the relationship betweencustomer experience and retail brand equity. When cus-tomers come to a retailing environment they typicallyhold perceptions about two types of brands: the retailbrand (e.g., Wal-Mart, Macy’s) and the manufacturer orservice brands that are sold in the retail stores (e.g., RalphLauren, Tide) (Verhoef et al. 2009). The objective of thisresearch is to investigate the influence of customers’ retailexperience on retail brand equity.

Customer experience refers to consumers’ subjectiveinternal and behavioral responses to stimuli from retailingenvironments. More specifically, we conceptualize cus-tomer experience as a multi-facetted construct includingfour components: sensory, affective, cognitive, and behav-ioral experience. Sensory experience refers to thecustomer’s reaction on environmental stimuli throughsenses (i.e., sight, hearing, smell, taste, and touch). Affec-tive customer experience is defined as the customers’emotional response toward environmental stimuli duringthe shopping trip. Cognitive experience relates to custom-ers’ thinking or mental process during shopping. Finally,we define beha vioral experience as customers’ physicalresponse toward the retailer, including physical activitiesand labor work.

Retail brand equity is defined as a set of brand assetsand liabilities linked to a store brand, its name, and symbolthat add to or subtract from the perceived value of the storebrand by its customers (Arnett, Laverie, and Meiers 2003).Based on prior research (Arnett, Laverie, and Meiers

2003; Pappu and Quester 2006; Yoo, Donthu, and Lee2000) we specify the domain of content for retail brandequity in this study as consisting of four dimensions,namely retail brand awareness, retail brand associations,retail brand quality, and retail brand loyalty. While retailbrand awareness is the degree to which a retailer’s nameis familiar to consumers (Arnett, Laverie, and Meiers2003), retail brand associations are attributes and benefitslinked to the name of the retailer in the consumer’s mind(Keller 1993). Retail brand quality refers to customers’perceptions of quality associated with the retailer brand(Pappu and Quester 2006; Yoo, Donthu, and Lee 2000).Finally, retail brand loyalty refers to the intention to buyfrom one retailer as a primary choice (Yoo, Donthu, andLee 2000).

Within this study, we anticipate that customers’ retailexperience has a positive impact on retail brand equity.More specifically, we predict that customer experiencepositively affects retail brand awareness, associations,quality, and loyalty. This notion is supported by previousstudies revealing that customer experience has an impacton traditional performance measures and on a set ofintangible assets of the company (Gentile, Spiller, andNoci 2007). This notion is further supported from priorresearch in branding. For example, Keller (1993) notesthat customer experiences are a key determinant of brandequity. Finally, this notion is backed up by previousstudies in retailing research, which explore the interplaybetween customer experiences and brand-related out-come variables (e.g., Leischnig, Schwertfeger, andGeigenmüller 2011a,b).

To test the influence of customers’ retail experienceon retail brand equity, empirical examinations includingqualitative and quantitative studies were conducted. Thequalitative study involved interviews with retailing andmarketing professionals, as well as customers. The quan-titative study involved a survey with customers of tworetailers. In order to analyze the hypothesized relation-ships we employed structural equation modeling.

This research finds significant support for all of thehypothesized relationships. The findings of this studyshow that customers’ experiences do have significantpositive effects on retail brand awareness, associations,quality, and loyalty. Thus, we can extrapolate the notionthat the way customers respond to stimuli from retailenvironments influences their perception and evaluation


of the retail brand. The key message of this study formanagers is twofold. First, the creation of customer expe-riences requires thoughtful consideration of customers’subjective internal and behavioral responses to retailenvironments. Second, to build strong retail brands, man-agers should consider the customer experience as animportant determinant having the potential to positivelyinfluence retail brand equity. Against this background,we suggests that retailers should develop managementsystems that consider customers’ sensory, cognitive,affective, and behavioral responses when creatingunique customer experiences. These management sys-

tems require strategies and instruments to address each ofthe facets of the customer experience construct. A furtherimplication is that retail managers should develop moni-toring systems to survey customer experiences. The mea-surement instrument that was developed in this studymight serve as such a survey instrument. Taking intoaccount that customer experience has a significant posi-tive effect on retail brand equity, another implicationconcerns the fact that retail managers who wish to buildstrong retail brands should consider customer experienceas a potential determinant of customers’ brand perceptionand evaluation. References are available upon request.

For further information contact:Alexander Leischnig

Marketing DepartmentFreiberg University of Technology

Lessingstrasse 4509599 Freiberg




ANTECEDENTS OF PRIMARY STORE DISLOYALTY INA STAGNANT MARKET

Peter J. McGoldrick, Manchester Business School, United KingdomDaniel P. Hampson, Manchester Business School, United Kingdom

SUMMARY

Relationship marketing and loyalty programs havebeen major strategic tools for retailers but recession andstagnation impacted severely on loyalty. Customer loy-alty has generated direct economic benefits for retailers,loyal customers being more likely to purchase high mar-gin supplementary products (Gummesson1997). Customerretention lowers costs, Dhar and Galzer (2003) suggestingthat loyals require up to 90 percent less marketing expen-diture. Gremler and Brown (1999) refer to “the loyaltyripple effect” to suggest that loyalty can enhance customerbehaviors providing indirect benefits. Loyals generatefurther interest and patronage in the firm through positiveWOM, a most cost-effective form of marketing commu-nication (Zeithaml et al. 2006).

Given its importance, numerous research studiesexamine the antecedents of consumer loyalty and adichotomy has emerged between loyals and non-loyals.The former require relationship maintenance strategies,the latter relationship building. However, a disproportion-ately small amount of research has looked specifically atshoppers who remain patrons of their primary stores butbecome less loyal – i.e., primary store disloyalty. Todevelop strategies to restore previous loyalty, marketersneed to understand the causes of shifts toward disloyalty.

This need is particularly apparent in a stagnant eco-nomic climate. Consumers are faced with factors poten-

tially increasing their willingness and ability to be dis-loyal. Hampson and McGoldrick (2012) find that recenteconomic crises have undermined many retailers’ loyaltybases, as consumers have become more price-consciousand concerned with value. Greater Internet access alsoallows consumers to become more knowledgeable. Manyretailers’ recession responses contributed further to thesetrends, creating enduring levels of increased disloyalty.

Based upon a literature review, qualitative investiga-tions and quantitative primary research, this paper addres-ses the following research objectives: (1) to define andconceptualize primary store disloyalty; (2) to examineshifts in primary store disloyalty since the 2008/09 reces-sion; (3) to identify some antecedents to primary storedisloyalty; (4) to discuss the managerial implications andtheoretical contributions.

The qualitative investigations with managers andconsumers, combined with two surveys of 1,211 and 611shoppers at different phases of recession and stagnation,reveal significant influences upon primary store disloy-alty of (post) recession effects, service failures and intrin-sic motivations. It is argued that researchers need to givecloser scrutiny to recession effects and to shopper disloy-alty, not treat this merely as an antonym of loyalty. Thepaper suggests implications for researchers, retailers andother marketers. References are available upon request.

For further information contact:Peter J. McGoldrick

Manchester Business SchoolBooth Street West

Manchester, M15 9PBUnited Kingdom

Phone: +44.161.306.3478E-Mail: [email protected]


SIGNALING E-TAILER TRUST: EXPLORING SOURCE INFLUENCESOF INTERNET TRUSTMARKS

K. Damon Aiken, Eastern Washington University, SpokaneVincent Pascal, Eastern Washington University, SpokaneSohyoun Shin, Eastern Washington University, Spokane

SUMMARY

Marketing researchers have learned that context-specific trustmarks are effective (Aiken and Boush 2006);however, the field has yet to more rigorously investigatemany of the complex signal-based communications pro-cesses surrounding trustmarks. This study provides afoundational investigation into the source influences ofInternet trustmarks (i.e., any third-party mark, logo, pic-ture, or symbol presented in an effort to dispel consumers’concerns about privacy and security). Thus, the currentwork has two main objectives. First, the study explores theprocess of trust development through the use of trustmarks.Second, the study specifically investigates the impact ofsource influences of trustmarks in early stages of trusttransference.

A between-subjects experiment manipulated thesource characteristics of stimuli so as to come from (1) agovernmental source, (2) consumer reviews, (3) an objec-tive, expert third-party source, or (4) a control condition.Data collection took place in classroom settings ratherthan over the Internet in order to increase response ratesand to increase subjects’ involvement levels. A conve-nience sample of 400 subjects was taken from a largenorthwestern university.

The paper-and-pencil questionnaire consisted of threeparts. Part One provided an introduction to the study andinquired about generalized Internet commerce trust(GICT). The scale was adapted from Aiken, Mackoy, Liu,Fetter, and Osland (2007) and measured the five dimen-sions of GICT. Part Two opened with another shortintroduction, written in a font size three times that of therest of the questionnaire. The short paragraph providedfurther introduction to the stimuli and implored subjects topay special attention to the website they were about to see.Next, one of the four possible experimental stimuli waspresented in the form of a screen print from a mockwebsite. The trustmarks only differed in terms of theidentification of the certifying source at the bottom of themark. Phrases at the bottoms of the trustmarks were: (1)“Reviewed and Certified by Customers Like You,” (2)“PC Magazine – Editors’ Choice,” (3) “The Bureau ofConsumer Protection (A Division of the Federal Govern-ment).” Part Three of the questionnaire contained vari-ables adapted from Aiken and Boush’s (2006) measure-

ment of firm-specific trust (FST). Here, variables attem-pted to measure the affective, behavioral, and cognitiveelements of trust in the experimental website. After thefirm-specific trust questions, subjects were asked severaldemographic and Internet-usage questions.

Initial analysis of subjects’ GICT means revealed arelatively low overall trust of the Internet. This is consis-tent with previous research noting that people are stillhesitant to trust web information (Gosling 2004). Themean for the entire sample was 3.91 on a seven-pointscale. Across gender groups, the data showed that menwere more trusting than women in terms of generalizedInternet commerce, (male mean = 4.12, female mean =3.85; t = 4.05, p < .01). Further, an ANOVA revealed thathigh GICT subjects exposed to the government-sourcedtrustmark displayed significantly higher firm-specific trustmeans (F = 4.44, p < .01). High GICT subjects seemed tobe more influenced by the government-sourced trustmark.Low GICT subjects were not swayed by exposure to anyof the different conditions. Additionally, an ANOVAfound significant differences between GICT groups andtheir firm-specific trust scores (after controlling for expo-sure to experimental stimuli). Thus, generalized Internettrust was determined to be a significant moderating vari-able between the trustmark condition and firm-specifictrust (F = 11.02, p < .01) as well as the three basecomponents of firm-specific trust.

Next, a set of ANOVA and t-tests were utilized toinvestigate differences in FST means. When FST meanswere analyzed across experimental conditions, the meansof the group exposed to the governmental source wereconsistently higher than that of subjects in the other threegroups. The differences were statistically significant forthe combined measure of FST (F = 4.22, p < .01). Furtheranalyses showed that the government-sourced trustmarkfavorably influenced affective trust (F = 5.05, p < .01) aswell as cognitive trust (F = 4.46, p < .01), but it did nothave an effect on the behavioral FST component. Post hoctests utilizing Tukey’s HSD multiple comparisons furtherreinforced the power of the government-sourced trustmark.Finally, the sample showed significant FST differencesacross gender groups. Similar to differences in GICT,men showed higher trust levels than women at the firm-specific level.


This study provides several noteworthy conclusions.First, with respect to firm-specific trust, the results revealthat (amongst the three sources) the government-sourcedtrustmark was the most influential. Second, findings showthat the behavioral component of trust is still not usuallyhigher than the affective or cognitive components of FST.From an important strategic perspective, consumers’ be-havioral trust is not significantly influenced by trustmarks.Third, the data show that men not only have higher levelsof generalized Internet commerce trust, but that they also

(oftentimes) are more trusting at the firm-specific levelcompared to women.

Regardless of the interesting and significant findingsuncovered by this study, there is still a great deal of workto be done before researchers can fully understand thecomplexity of Internet communications in E-tailing. Ex-ploring the intricacies of communication processes andsignaling through trustmarks certainly requires furtherexploration and empirical validation. References are avail-able upon request.

For further information contact:K. Damon Aiken


Spokane, WA 99202Phone: 509.828.1255



A COMPARISON OF BRAND EXTENSIONS FOR PRIVATE LABELSAND NATIONAL BRANDS

Monica Grosso, Bocconi University, ItalySandro Castaldo, Bocconi University, Italy

SUMMARY

Private labels continue to grow in both market shareand penetration. Their evolution in the US has beenconstant over years and in August 2009 their average unitshare reached 22.8% (ACNielsen data 2010). Those prod-ucts now account for more than $86 billion in the U.S.,(ACNielsen data 2010) while in some UE countries, theyhave gained a market share similar or higher to that of theleading brands one.

The growth in private labels has spawned muchresearch on them that addressed a wide variety of issues:who buys and how private label products, whether andhow private labels provide leverage to retailers, the deter-minants of private label share (Ailawadi and Keller 2004;Dhar and Hoch 1997; Hoch and Banerji 1993); the inter-action between national brands and private labels (Cotterillet al. 2000; Sayman et al. 2002; Blattberg and Wisniewski1989; Allenby and Rossi 1991) and conditions for entry ofstore brand and its impact on retailer’s pricing and bar-gaining power with manufacturers (Raju et al. 1995;Chintagunta et al. 2002; Pauwels and Srinivasan 2004).

Although the growth of private labels has been inter-preted by some as a sign of the “decline of brands,” it couldeasily be argued that the opposite conclusion is morevalid, as their evolution could be seen in some ways as aconsequence of cleverly designed branding strategies(Ailawadi and Keller 2004, p. 336). Retailer brands haveindeed changed throughout time (De Wulf et al. 2005). Ifprivate labels were introduced in the market as products oflower quality offered at better prices (De Wulf et al. 2005),today retailers offer real brand systems, similar to themanufacturers ones, that show how their branding strate-gies became more and more articulate (Ailawadi andKeller 2004). Until recently, the general wisdom was thatprivate labels were essentially meant to serve price-sensitive market segments (Amrouche, Martín-Herrán,and Zaccourù 2008). Nowadays, retailers’ brands are seenby consumers as valuable substitutes to established manu-facturers’ brands (e.g., Amrouche, Martín-Herrán, andZaccourù 2008; Dunne and Narasimhan 1999; Hoch andBanerji 1993; Fitzell 1992; Keng and Ehrenberg 1984;Wilensky 1994; Quelch and Harding 1996; Uncles andEllis 1989).

Retailers’ private branding strategies are clearly basedon brand extension (Alexander and Colgate 2005), that is

the use of an established brand name to enter a newproduct category (Aaker and Keller 1990). Brand exten-sion is a heavily-researched and influential area in market-ing (Czellar 2003). Surprisingly research on brand exten-sion has only marginally considered the retail context byfocusing on manufacturers’ extensions sold within aretailer’s store, rather than analyzing retailers’ own brandextensions (Alexander and Colgate 2005). This is coher-ent with a trend in the branding literature that, in spite ofthe emergence and growing importance of store brands,mostly focused on national brands (Steenkamp andDekimpe 1997).

In this paper we rely on the brand extension frame-work to contribute to solve an unresolved debate inliterature on whether private labels should be investigatedor not as something different from the “traditional” brands.Richardson (1997) and subsequently De Wulf et al. (2005)support this identified gap, indicating that the questionwhether store brands are perceived to be just anotherbrand in the market or something different has receivedlittle attention in the marketing literature.

As a consequence of this, the key question literaturehas not yet succeeded to answer is: are private labelsbrands as manufacturers’ brands are in terms of howvarious marketing principles can be successfully appliedto them? Or, although many important branding prin-ciples apply to them, are retailers brands sufficientlydifferent from product brands that the actual applicationof those branding principles should vary for them asAilawadi and Keller (2004) suggest?

Uncles and Ellis state that consumers buy own labelsin the same way as a brand with a comparable market share(Uncles and Ellis 1989), thus supporting the first point ofview. However, at least three characteristics of privatelabels seem to support the latter view:

• Retail brand exclusivity. Indeed, while the nationalbrands are sold in various retailers, private labels aresponsored or owned by a retailer and are only sold inits stores (Bushman 1993; Burt 2000; De Wulf et al.2005; Hansen, Singh, and Chintagunta 2006).

• While advertising is the dominant factor in creatingnational brands that reveal strong brand equity, storeimage is the most important source of brand equityfor store brands (De Wulf et al. 2005). Consumers


therefore transfer brand connotations to the retailbrand products from their experiences of the retailstore (Burt 2000). This is also due by the fact that theretailer itself (e.g., Tesco) is seen as the brand(Swoboda et al. 2009) and sometimes the retailer’own brand and its products’ brand coincide.

• Private labels are advantaged by retailers’ closenessto the final market and privileged allocations on theshelves (Nenycz-Thiel et al. 2010; Nogales and Suarez2005). Indeed, an effective marketing of store brandscreates a captive clientele and makes the chain lessvulnerable to price pressures or aggressive attacks bythe competition (Dick et al. 1995; De Wulf et al.2005).

In this paper we will test if consumers perceive thesecharacteristics of private label differently. If so, thiswould thus justify the past ad hoc focus on PLs in thebranding literature. We will focus on the effects of brand

extensions since despite product and service extension isa fundamental feature of retail operational developmentand brand extension has become a fundamental feature ofretail commercial competition (e.g., Alexander and Colgate2005; Burt 2000), few studies are available on consumers’evaluation of retailers’ brand extension compared tonational brands’ ones.

The paper reports the results of four experimentalstudies analyzing similarities and differences in the im-pact on brand extension evaluation of the two key vari-ables identified within the brand extension literature:category fit and brand knowledge.

The results seem to support the view that privatelabels differ form national brands as both category fit andbrand knowledge have no significant effect on their exten-sion evaluations. References are available upon request.

For further information contact:Monica Grosso

SDA Bococni School of ManagementBocconi University

Via Bocconi 820136 Milano

ItalyPhone: +39.02.5836.2487Fax: + 39.02.5836.6893



AN EXPERIMENTAL EXAMINATION OF BUYERS’ RESPONSESTO RELATIONSHIP FAILURES

Donald J. Lund, University of Alabama – BirminghamJohn D. Hansen, University of Alabama – Birmingham

Thomas E. DeCarlo, University of Alabama – BirminghamLisa K. Scheer, University of Missouri, Columbia

SUMMARY

A 1997 survey of sales managers revealed that nearlyhalf (49%) of the managers believed their salespeople hadlied on a sales call, 34 percent believed their salespeoplehad made unrealistic promises, and 22 percent believedtheir salespeople had sold products the customer did notneed (Marchetti 1997). A 2002 survey of sales and mar-keting executives similarly revealed that approximately50 percent of the executives believed their salespeoplehad lied on sales calls and 74 percent believed the aggres-sive pursuit of sales goals encouraged their salespeople tolose focus on customer needs (Strout 2002). Though thesefindings have spurred research aimed at identifying theantecedents of ethical salesperson behavior (cf., Ferrell,Johnston, and Ferrell 2007), there is little evidence tosuggest the situation has improved much in recent years(Ramsey, Marshall, Johnston, and Deeter-Schmelz 2007).

This suggests that the prevention of unethical sales-person behavior is but one piece to the puzzle – firms mustdevelop recovery strategies to effectively respond toethical transgressions. While extant research has focusedon salvaging both business and consumer relationshipsfollowing a service failure (Gonzalez, Hoffman, Ingram,and LaForge 2010; Tax, Brown, and Chandrashekaran1998), we know little about recovering from an ethicalfailure. The overarching hypothesis advanced here is thatethical and service failures differ in that a more intenseemotional response is elicited by ethical failures. Thus,while previous research supports the notion that emo-tional responses resulting from service failures serve toheighten the level of emphasis customers place on recov-ery efforts (Smith and Bolton 2002), we examine howbuyers’ responses differ when facing an ethical transgres-sion relative to a service failure. We argue that the keydifference distinguishing the two is that a service failureis accidental in nature while the ethical transgression hasoccurred intentionally. This is the first research we areaware of that contrasts the two types of relationshiptransgressions and investigates the differential impacts inbusiness relationships.

This research also examines the effects of the buyer’sperception of the salesperson’s relationship role. Heideand Wathne (2006) maintain that individuals in

interorganizational relationships assume different rolesbased on how the relationship is governed. The authorsidentify two types of roles: a friend and a businessperson.A friend is motivated by the welfare of the relationship asa whole, while a businessperson is motivated by utilitymaximization and will defect from the relationship if sucha move is deemed economically desirable. Consistentwith research which reveals that norms influence cus-tomer responses to relationship failures (e.g., Aggarwal2004), we propose that ethical transgressions will elicit astronger emotional response when the buyer believes thatthe offending salesperson is a friend. Buyers who catego-rize the salesperson as a businessperson will be morecalculative in their response as for them the transgressioncreates a debt that must be repaid.

Research supports the notion that buyers view theirrelationships more through the salesperson than they dothe sales company (Palmatier, Scheer, and Steenkamp2007), as salespeople largely determine the level of valueand satisfaction provided the customer. While this sepa-ration of salesperson and company can prove challengingfor selling firms, might it also prove beneficial followingunethical salesperson behavior? Are there ways throughwhich the firm can disentangle itself from the salespersonfollowing an ethical transgression or service failure? Inresponse to these questions, this research addresses twoprimary research questions:

1. How are buyers’ responses to ethical transgressionsdifferent from their responses to service failures?

2. How do buyers’ role perceptions affect their responseto a relationship transgression?

A scenario based experimental design was employedto address these questions using an online survey deliv-ered to professional buyers. Respondents were recruitedfrom a national panel of business professionals and had tohave two years purchasing experience. Relationship role(friend vs. businessperson) and relationship failure (ser-vice vs. ethical) were manipulated through a scenario thatdescribed a relationship failure with a real salesperson thebuyer interacts with. After reading the scenario, partici-pants responded to survey questions asking about theirperceptions and reactions to the incident. We find that


ethical transgressions elicit more anger from the buyerthan service failures. We also find that salespeople cat-egorized as friends will produce more anger following anethical transgression than those categorized as abusinessperson. Additionally, salespeople categorized asfriends produced less anger when the buyer faced aservice failure than those categorized as a businessperson.

The prevalence of ethical transgressions evidencedin the popular press suggests that managers would benefitfrom a better understanding of the impact unethical sales-people can have on a relationship. We contribute toexisting research by (1) examining the differential impact

that service failures and ethical transgressions can have ona buyer’s attitudes; (2) developing a measure to capturethe friend vs. businessperson distinction; and (3) examin-ing how the relationship role impacts a buyer’s responseto each type of relationship failure. We contribute tomanagerial practice by (1) showing that building strongerrelationships with buyer firms insulates the firm againstservice failures and (2) that buyers expect more fromsalespeople perceived as businesspeople following a ser-vice failure. Future research directions include examina-tion of the most appropriate type of recovery effortsfollowing a failure and who leads that recovery effort.References are available upon request.

For further information contact:Donald J. Lund

University of Alabama – BirminghamBEC 207F; 1530 3rd Avenue South

Birmingham, AL 35294–4460Phone: 205.934.8837



WHEN CUSTOMERS SHOW DIVIDED ATTITUDINAL LOYALTY: USINGCHANNEL INTERMEDIARIES TO INCREASE BEHAVIORAL LOYALTY

Sridhar N. Ramaswami, Iowa State University, AmesS. Arunachalam, Iowa State University, Ames

Kirti Rajagopalan, Washington State University, Pullman

SUMMARY

The cultivation of customer loyalty is a challenge(and an important opportunity) that businesses face today. It is a challenge because there has been a steady erosionof loyalty in almost every product market because ofincreased competition. However, cultivating customerloyalty is also an opportunity since loyal customers arebelieved to contribute to increased revenue and profitabil-ity (Chaudhuri and Holbrook 2001; Gupta and Lehmann2003), provide more predictable sales and profit streams(Dick and Basu 1994), and show greater propensity tofavor the firm’s extensions and product enhancements(Reichheld 1996).

A firm’s competitors are constantly engaged in weak-ening customer loyalty and stunting value exploitationfrom loyalty efforts. In a brand context, Keller (2003) hasnoted that competitors are constantly engaged in not onlycreating points of differences (PODs) but also nullifyinga firm’s PODs by converting them to points of parity(POPs). The impact of such competitive efforts is two-fold: weakening or even eliminating perceived differ-ences among competing brands (Ehrenberg, Barnard, andScriven 1997), or signaling the presence of comparativedifferences (that is, competitor has comparative advan-tage in satisfying certain benefits). Further, Uncles,Dowling, and Hammond (2003) observed that “having afavorable set of beliefs about one brand does not precludehaving an equally favorable set of beliefs about otherfunctionally similar brands in the category” (p. 301). Either way, the end result is divided customer loyaltyamong competing brands. Interestingly, purchase pat-terns in many product categories show that customers aredivided in their loyalty (Ehrenberg and Scriven1999).While loyalty literature has shown the presence ofdivided loyalty using behavioral information (Ehrenberg,Goodhardt, and Barwise 1990), it has however not takenthe next step to develop and test a sound rationale or basisfor its presence. Using customer value theory (Woodruff1997; Ulaga and Eggert 2006) we examine the issue ofdivided loyalty within the context of customers exhibitingsuch loyalty to two closely-competing brands (which weterm as “dual loyalty”).

We define dual loyalty as the phenomenon whereincustomers have “equivalent levels of overall attitudinal

loyalty” toward two firms-a focal firm and its closestcompetitor-based on past exposure and experience. Itshould be noted that (a) the terms firm and brand are usedinterchangeably in the rest of the manuscript, and (b) thefocus is on loyalty toward a firm, where the firm is also abrand.

The primary purpose of the present study is to exam-ine the presence of, the reasons for and the outcomes ofdual loyalty. In doing this, the present study addressesthree important issues. First, it offers two explanations fordual loyalty-equivalence and comparative advantage-andexamines which explanation is more valid. The equiva-lence explanation rests on the logic that customers can beequally loyal to two firms if they do not see any differencebetween them, implying that the value drivers of loyaltyare the same for both firms, both in terms of level andimpact. The comparative advantage logic assumes thatcustomers may be loyal to two firms because they seekdifferent benefits from them. The comparative advantagepremise is supported if the drivers are perceived to beimportant and if different drivers are salient for loyaltytoward each firm. Second, the study examines how firmsrespond to divided loyalty of customers. Firms can esca-late value offerings to dual-loyal customers on the beliefthat higher value is needed to keep them satisfied; or firmscan focus on the more loyal customers and thus indirectlypenalize the dual loyal customers for showing loyalty tosomeone else. Third, the study examines the role playedby intermediaries that interface between the firm and thecustomer. Specifically, it evaluates to what extent inter-mediaries have the ability to induce customers that showdual loyalty at the attitudinal level to demonstrate behav-iors that makes them more valuable to the firm.

The conceptual framework was empirically tested ina supplier-dealer-customer triadic setting in a B-to-Bcontext using data from survey responses from customersand supportive objective information on customer lifetime value from the supplier firm. The results of the studysuggest that (a) consumers are loyal to more than one firmbecause they value different benefits offered by compet-ing firms, (b) dual loyal customers get more attention fromfirms than customers who are only loyal to them, and (c)intermediaries are successful in converting customers thatare loyal to a competing firm to purchase more from the


focal firm. Based upon the last result, a second study isbeing proposed to evaluate alternative strategies that thefocal firm (a manufacturer and also a finance provider)can use to encourage dealers to recommend the focal firmto their customers. In particular, the authors aim to evalu-ate the issue of maximizing monetary gain versus maxi-mizing value propositions (Verhoef 2003) related to thetask of recommending the focal supplier firm to endcustomers. Overall, the results of the first study are insupport of customer value theory and suggest that dealerrecommendation can increase behavioral loyalty towardthe supplier.

In sum, our study will make three key contributionsto the loyalty literature. First, the study will add to the

sparse literature on divided loyalty of customers. Second,it will investigate how firms respond to consumers thatexhibit divided loyalty. If firms recognize the existenceof such loyalty, they will be expected to escalate theirservice and commitment to divided-loyal consumers ascompared to those that show loyalty to one firm. Third,the study will throw light on how a firm can use itsintermediary network to handle customers who showdivided loyalty. This evaluation will have implications forthe design of dealer strategies and training programs todealer staff. Further, this evaluation will have generaliz-able implications for firms in product categories otherthan the one examined here that face similar customer andintermediary issues. References are available uponrequest.

For further information contact:Sridhar N. RamaswamiIowa State University

3216 Gerdin Business BuildingAmes, IA 50011–1350Phone: 515.294.5341



CUSTOMER REACTIONS TO MASSIVE WORKFORCE REDUCTIONS:WHEN IS SATISFACTION AFFECTED?

Martin Klarmann, Karlsruhe Institute of Technology, GermanyChristian Homburg, University of Mannheim, Germany

Johannes Habel, Booz & Company

SUMMARY

While firms usually downsize to improve perfor-mance, studies show that this attempt often fails. In searchfor explanations, researchers have begun to examine theeffects of downsizing on customer satisfaction and foundfirst evidence of a negative relationship. This would implythat downsizing is an undesirable practice, at least from amarketing perspective. It would also raise the question,why so many firms still rely on downsizing.

In our paper we argue that the issue may be morecomplex. Drawing on the theoretical argument on therelationship between firm productivity and customer sat-isfaction by Anderson, Fornell, and Rust (1997), wepropose that it will mostly depend on contextual factorswhether downsizing has a positive or a negative effect oncustomer satisfaction. Consider for instance firms that arehigh on organizational slack (Love and Nohria 2005).They have the opportunity to compensate for potentialnegative effects of work force reductions by improvingexisting business processes (Marks 2003), which mayeven increase customer satisfaction. At the same time,firms that were already working very efficiently beforethe downsizing will find it hard to preserve their servicequality. Thus, customer satisfaction is likely to go down.

Similarly, we expect that the negative effect ofdownsizing on customer satisfaction is more pronouncedif the firm has employed downsizing proactively and itslabor productivity is high. Further, we expect a negativeeffect of downsizing on customer satisfaction in servicefirms and in industries characterized by high (R&D)intensity and by high consumer involvement.

To test our hypotheses, we combine data from threesources. (1) We use data from the American CustomerSatisfaction Index (ACSI) to measure our focal variablecustomer satisfaction. (2) We measure organizationaldownsizing as well as most context factors usingComputstat data. (3) To measure consumer industry in-volvement, we collected survey data. As a result, we have

a longitudinal data set with data from 1994 to 2008 (beforethe financial crisis) from over 140 companies, coveringmore than 180 downsizing events.

Based on this data, we find that the negative effect ofdownsizing on customer satisfaction is more pronouncedif a company has low organizational slack, benefits fromhigh labor productivity, operates in an R&D-intensiveindustry, and markets a product that customers feelhighly involved with. Interestingly, we find no strongsupport for our hypothesis that downsizing is lessharmful to customer satisfaction if financial perfor-mance has been declining before the downsizing. Ourinitial reasoning was that under these circumstances,customers would perceive downsizing as fairer (Love andKraatz 2009). We propose that our ambiguous findingsare the result of different downsizing strategies dependingon the previous trend of financial performance.

By examining the effect of downsizing on customersatisfaction in the context of a large longitudinal dataset,our study contributes to academic knowledge in at leastthree ways. First, we identify conditions when downsizingreduces customer satisfaction and conditions whendownsizing may actually improve customer satisfaction.Second, our study is the first to use longitudinal data oncustomer reactions to downsizing, allowing us to makestronger causal claims than previous research. This con-tribution is particularly important, because a negativeassociation between downsizing and customer satisfac-tion can also arise in cross-sectional data through an effectof customer satisfaction on downsizing. Third, this studyis the first to link the customer-related outcomes ofdownsizing to firm performance. By showing thatdownsizing has an indirect effect on financial perfor-mance via customer satisfaction, our findings provide apossible explanation of why so many downsizing projectsfail. Thus, we also contribute to research on the so-called“hidden costs” of downsizing (Buono 2003) by providingevidence that these hidden costs actually translate intomonetary disadvantages. References are available uponrequest.


For further information contact:Martin Klarmann

Karlsruhe Institute of TechnologyKaiserstrasse 1276131 Karlsruhe

Phone: +49.721.608.43726Fax: +49.721.608.47765



A CUSTOMER-CENTRIC APPROACH TO B2BMARKET SEGMENTATION

Yihui (Elina) Tang, University of Illinois at ChicagoMurali Mantrala, University of Missouri, ColumbiaEsther Thorson, University of Missouri, Columbia

SUMMARY

In B2B markets, it is logical for the selling firm toconsider relevant characteristics of the buying firm’scustomers when developing its marketing strategies(Bradford et al. 2010; Kleinaltenkamp and Ehret 2006;Piercy and Lane 2003). This is especially true with deci-sions regarding market segmentation strategy which drivesall other marketing strategies. Scholars have called forresearch which focuses on downstream markets to gener-ate market intelligence and provide understanding ofcustomers’ customers’ preferences (Slater and Mohr 2006;Henneberg, Mouzas, and Naude 2009). Despite the im-portance of the concept on customers’ customers, we areunaware of any theoretical or empirical approach tosegmentation that explicitly incorporates this dimension(i.e., the relevant characteristics of the buying firm’scustomers).

In this paper, we focus on the implications of deriveddemand for segmenting business marketers’ focal cus-tomers and propose a segmentation method that distin-guishes between immediate and downstream customercharacteristics and explicitly accounts for both sets ofcustomer variables in deriving richer, more meaningful,and actionable segments for B2B marketers.

We propose a three-dimensional joint segmentationmodel that fit these criteria. The model is an extension ofRamaswamy et al. (1996). The model is first validatedthrough a series of Monte Carlo studies. Then we apply itin a real B2B market. In the empirical application of themodel, we collected survey data in collaboration with amedia company in the Midwest United States. This mediacompany’s main business line is a print newspaper. It isone of the top thirty newspapers in the United States, withover one million readers each week. As is the case for mostmedia companies, it has two business lines: one is aimedtoward readers and therefore B2C, the other one is gearedtoward advertisers and therefore B2B. Our objective is tosegment the advertisers (B2B) market based on three

criteria: previous ad purchase behavior, benefits desiredwhen buying the ads, characteristics of advertiser’s(customer’s) customers.

Data were collected by a phone survey administeredby trained interviewing staff. Altogether, 2,307 telephonenumbers were dialed. Five hundred sixty-six eligiblebusinesses were reached. Three hundred thirteen surveyinterviews were completed, for a response rate of 55percent. We applied the proposed model to the data. UsingAIC3 as the model selection criteria, we extracted twosegments on the purchase behavior dimension, three seg-ments on the benefits dimension, and two segments on thecustomer’s customers dimension. One of the conceptualcontributions of the 3D joint segmentation model is that itexplicitly considers benefits desired by customer’s cus-tomers in a B2B market. The analysis results showed thateven when advertisers or the direct customers’ desiredbenefits are the same, these accounts (advertisers) couldbelong to different segments because advertisers’ cus-tomers desire different benefits.

Executives at the collaborating media companyreviewed the results and found it meaningful. Theyselected two segments as the focus of selling effort.This is because these segments have ad spending behav-ior and desired benefits that well match what the focalcompany can offer. Also in these segments, the advertiser’scustomers have attributes that are reflected in thecompany’s reader base. So the company is able to offerreaders that are potential customers for these advertisers.

Conceptually, out model is the first segmentationmodel that explicitly takes customer’s customer needs/characteristics into account. Methodologically, it advancesthe current state of the art for finite mixture models in themarket segmentation literature. Substantively, it adds tothe sparse literature on empirical applications of B2Bmarket segmentation models. References are availableupon request.


For further information contact:Yihui (Elina) Tang

Department of Managerial StudiesUniversity of Illinois at Chicago601 S Morgan Street, MC 243

Chicago, IL 60607E-Mail: [email protected]


A STRATIFIED EXAMINATION OF THE DRIVERS AND OUTCOMESOF MARKETING CROSS-POLLINATION

Jeff S. Johnson, University of Nebraska, LincolnAmit Saini, University of Nebraska, Lincoln

SUMMARY

As competitive intensity and market dynamism con-tinue to escalate in today’s business environment, theneed for organizational innovativeness has never beengreater. The cross-pollination of marketing employees isa potent way to pursue innovation within the firm. Theimpact of HRM practices that facilitate the cross-pollination of knowledge in the marketing functionhave not been explored in an aggregate fashion. Assuch, little is known on the relative impact of differentcross-pollinating activities.

We conceptualize marketing cross-pollination (MCP)as the extent to which marketing intentionally integratespersonnel to cultivate new perspectives and ideas. Mar-keting departments utilize inter- and intra-functional rota-tion, cross-training, and cross-integration of employees tofacilitate cross-pollination. The facets of MCP are influ-enced by a company’s values and policies and are pro-posed to be impactful on performance. These activitiescreate greater functional and competence diversity andfacilitate explicit and tacit knowledge sharing within thefirm. The activities were selected subsequent to an exten-sive review of HRM practices utilized to promote learningin the organization. MCP is a mechanism essential tocreating the opportunity for the transfer of explicit andtacit knowledge stores through its components. Addition-ally, MCP can occur at different levels: between differentproduct groups within the marketing function, betweenmarketing and sales, among other functional areas andmarketing, or through acquisition of employees from andinteraction with different companies. We theorize thepotential differential impact of these four levels of cross-pollination on key marketing outcomes.

Organizational learning is a key motivation for cross-pollination in a marketing context. Knowledge acquisi-tion can accrue from many sources including learningfrom experience, learning by observing other organiza-tions, and grafting (acquiring necessary talent not pos-sessed). MCP creates the opportunity for the distributionof knowledge and the incorporation of different interpre-tations. We propose several drivers and outcomes of MCPin the marketing department.

Factors pertaining to culture and strategy are pro-posed to drive MCP and the structural aspect of marketingslack to moderate their impact. The cultural factor com-

prises the learning orientation. Organizations with a highlearning orientation will value and practice MCP as theknowledge created and transferred through this processwill be consistent with the organization’s values. Firmspursuing an exploration strategy are also more likely topractice MCP. Exploration focuses on new, emergingmarkets and radical changes and firms pursuing an explo-ration strategy seek to innovate and create new knowledgeand skills. MCP serves to cultivate different, new ideasthrough the rotation, cross-training, and cross-integrationof employees. Slack resources allow for increased flex-ibility and time to reflect that may result in more efficientand effective business practices. Slack can be utilized topromote learning through MCP. Marketing departmentsoperating with no slack may not have the time and re-sources for MCP. Regardless of their strategic or learningorientations, departments without slack resources willhave a difficult time enacting these behaviors. Conversely,an abundance of perceived slack resources in a marketingdepartment enhances the effect of these two variables asthe resources are available to cross-pollinate their em-ployees and facilitate learning.

While MCP could be argued to affect a wide array oforganizational outcome variables, we seek to indentifythe positive, negative, and contingent effects MCP canhave on the organization. In this pursuit, we proposerelationships between MCP and (a) innovativeness, (b)relative cost, and (c) market knowledge.

The components of MCP are likely to increase theinnovativeness of the marketing product group. Personneltransfer enhances the collective base of experiences amongbusiness-unit members and provides diverse ideas andnew ways of conducting business, cross-training pro-motes learning, and cross-functional teams are conduitsfor creativity and innovation. A drawback of MCP, how-ever, is increased cost. MCP activities entail both hard-dollar marketing expenditures and soft-dollar inefficien-cies which increase the relative cost of the product group.Efficiencies that are developed with experience in a mar-keting or sales position are often lost or reduced when jobtransfer occurs, cross-training requires a time investmentby both the trainer and trainee, and coordinating strategiessuch as cross-functional teams can give rise to inefficien-cies. Last, the effect of MCP on market knowledge isproposed to be contingent on the type of MCP occurring.For example, personnel rotation will affect market knowl-


edge differently depending on the whether it occurs withinthe marketing function, between marketing and sales,among other functional areas and marketing, or throughacquisition of employees from different companies. Theseeffects are proposed to be moderated by the use of knowl-edge integration mechanisms (KIMs). KIMS provideformal structures to capture, interpret, and integrate knowl-

edge within the firm that should leverage the positiveeffect of MCP. Last, the impact of MCP on innovativenessand relative cost will likely vary contingent on what levelit occurs. More intense MCP activities (rotation) per-formed at higher levels (from different companies) willhave higher innovativeness benefits, but also higher costs.References are available upon request.

For further information contact:Jeff S. Johnson

College of Business AdministrationUniversity of Nebraska, Lincoln

512 N. 12th StreetLincoln, NE 68588

Phone: 402.415.3639Fax: 402.472.9777 (f)



THE ROLE OF FRANCHISEE MARKETING COMMITMENT ONPROMOTION EXECUTION AND PERFORMANCE

Joseph Matthes, University of Nebraska, LincolnAmit Saini, University of Nebraska, Lincoln

SUMMARY

The primary purpose of this article is to introduce anddevelop understanding of the role that franchisee market-ing commitment plays in the relationship betweenfranchisors and franchisees in terms of executing market-ing promotions. Franchisors and franchisees represent asignificant and specific type of marketing channel rela-tionship. Extant research on the dynamics of businesschannel relationships, including franchisor-franchiseeinteractions, has concluded time and again that satisfac-tion is of vital importance for sustaining successful rela-tionships (Brown, Lusch, and Smith 1991; Geyskens,Steenkamp, and Kumar 1999; Ruekert and Churchill1984). While the concept of satisfaction has been exam-ined considerably in the literature, the role of commitmenthas received relatively less attention, especially in termsof how it affects contractual franchisor-franchisee inter-actions (Andaleeb 1996; Anderson and Weitz 1992;Morgan and Hunt 1994; Williams and Hazer 1986).Furthermore, while general commitment has been thesubject of research, franchisee marketing commitmenthas not despite its existence and importance in under-standing franchisor-franchisee marketing interactions.Marketing commitment is critical to examine because ofthe role it plays in determining how a business owner orgeneral manager will strategically and tactically imple-ment marketing activities, such as product promotions,that originate from an external source.

There are several examples of interactions betweenfranchisors and franchisees concerning marketing pro-motions. Burger King has seen public and litigious con-flict between these groups for over six years now due totwo different marketing promotions – $1 double cheese-burgers and later operating hours – which were mandatedto franchisees by corporate management (Gibson 2010).In recent years, Blockbuster has increased their willing-ness to listen to their franchisees. In attempts to make theirbrick-and-mortar locations more viable, Blockbuster hasbeen de-emphasizing its online service and experiment-ing with various store concepts that will ideally lead toincreased foot traffic (Maze 2008).

Franchisee marketing commitment is conceptualizedas comprising both strategic and tactical forms of commit-ment. It captures the extent to which a franchisee wants tocomply with marketing promotions that are initiated and

mandated by the franchisor. A conceptual frameworkis presented and assessed. Within this framework, theauthors examine how marketing promotions, whichare initiated by the franchisor, are strategically and/ortactically committed to by the franchisee. In addition,the roles of franchisee economic and social satisfactions(e.g., Gassenheimer and Ramsey 1994; Geyskens andSteenkamp 2000) as well as franchisor equitable promo-tion development are assessed in order to provide a clearerpicture of how franchisee strategic and tactical commit-ment affects promotion execution and promotion perfor-mance. Lastly, the influence of coercion as a moderatingfactor is also theorized in terms of how it actually alters theimpact of strategic and tactical marketing commitment.

Overall, it is argued that high levels of franchiseeeconomic and social satisfactions will positively affectstrategic forms of commitment; while high levels offranchisee social satisfaction and evaluation of equitablepromotion development will positively affect tacticalforms of commitment. These two types of franchiseemarketing commitment are expected to affect promotionexecution, and hence promotion performance. Two stud-ies are proposed to further investigate these franchisor-franchisee channel interactions.

The following research questions are examined: (a)What is franchisee marketing commitment? (b) How doesa franchisee’s level of strategic and tactical commitmenttoward a franchisor-instigated marketing promotion im-pact execution of the promotion as well as performance?(c) What antecedents impact a franchisee’s desire tocomply? (d) How does coercion moderate the relationshipbetween a franchisee’s marketing commitment and theirpromotion execution?

The conceptual framework discussed is applicablefor both franchisees and franchisors alike. For franchi-sees, it provides guidance to help them better understandtheir own levels of franchisee strategic and tactical com-mitment as well as what drives these motivations and howthey affect marketing promotion outcomes. For franchisors,the framework provides understanding of the driversbehind franchisee strategic and tactical commitment sothat they can improve relational input. This research alsooffers guidance on how coercive tactics can be used. Bybetter understanding how franchisors affect the affectivestate of their franchisees, it is possible to better manage


such relationships. If franchisees are satisfied with therelationship and evaluate promotions equitably, then thereexists a higher level of marketing commitment. Identify-ing the drivers and effects of franchisee marketing com-mitment, as well as determining the boundary conditions,

is especially important in terms of accurately capturingactual outcomes that accompany varying levels of mar-keting commitment in real world franchisor-franchiseemarketing interactions. References are available uponrequest.

For further information contact:Joseph Matthes

University of Nebraska at LincolnCBA 330a

P.O. Box 880492Lincoln, NE 68588–0492

Phone: 402.578.9936Fax: 402.472.9777



DETERMINANTS OF THE ADOPTION OF A MOBILE COMMERCESTRATEGY: THE PERCEIVED BENEFITS TO THE FIRM

Esther Swilley, Kansas State University, Manhattan

SUMMARY

Firms have been reluctant to develop mobile market-ing initiatives because of the time and effort needed, aswell as not fully understanding its value to their custom-ers. As firms have begun to advertise on the mobilechannel, there is still reluctance to become involved inmobile commerce. This study seeks to understand thebenefits of the adoption of a mobile commerce strategy tothe firm.

Several advantages exist for consumers, as well asmarketers, in using mobile devices for commerce. Cus-tomers can be targeted based on location, allowing mar-keters to target those within a geographic area. Since mostmobile devices have a single owner, personalization canbe offered along with location flexibility, allowing busi-ness to be transacted anywhere, anytime, to a specificcustomer. However, there are also a number of disadvan-tages of m-commerce. Usability issues are of concern,including the smaller screen, and the fact that most mobiledevices are not as fast as personal computers. As of now,there are different technology standards by different mo-bile companies while each is moving toward the 4Gtechnology. Many organizations, however, have not beeneager to develop strategies to take advantage of mobiledevices for marketing purposes. Because of the wide-spread use of mobile devices by consumers, the questionbecomes what are the benefits to using mobile commerce.

In order to determine the efficacy of m-commerceadoption, firms must also understand the benefits of amobile strategy, even with the pressure to adopt. Per-ceived benefits refer to the anticipated advantages that canbe gained by an organization when m-commerce has beenadopted (Chwelos et al. 2001).

Mobile commerce would allow firms to improvecustomer service by offering products and services con-

veniently, offer faster response to customer concernsthrough mobile access, as well as the ability to offer higherservice quality through mobile communication. Externalfactors include those benefits that enhance the ability offirms to compete. Benefits that allow a firm to keep andexpand the customer base would certainly add to a firm’scompetitive advantage. Internal factors are those within afirm that allow for greater performance; these can be interms of cost savings, profitability and operational effec-tiveness. Influencing the decision to enter an online space,in many cases, can be spurred on by relationships withothers through industry, business and trade associations.

A total of 316 managers from various industriescompleted the survey. Respondents found that the per-ceived consumer benefits (PCB) of using m-commerceare positively associated with the attitude toward usingm-commerce (ATT) as a marketing strategy. However,respondents did not have a positive attitude toward exter-nal benefits (PEB) or internal benefits (PIB) of m-com-merce. There was a positive association between attitudeand the intention to use m-commerce (INT) as a marketingstrategy. The results support the belief that firms do findthere are perceived benefits of mobile commerce.

This research has shown that firms are more likely todevelop a mobile commerce strategy because of the ad-vantage it would give over the competition. However,firms are unlikely to choose a mobile commerce strategyif it aids in the relationship with customers, or with theirrelationships with other businesses. The framework testedin this paper will be a useful first step in the determiningthe underlying antecedents to a firm’s decision to usemobile commerce. The importance of mobile commercehas its underpinnings as a useful tool that companies arebeginning to become involved in as the potential benefitsare understood, as found in this investigation. Referencesare available upon request.

For further information contact:Esther Swilley

College of BusinessKansas State University

201A Calvin HallManhattan, KS 66506–0506

Phone: 785.532.6135Fax: 785.532.3709



THE MODERATING EFFECT OF CREATIVITY ON CONSENSUSAND PERFORMANCE: A MODERATED POLYNOMIAL

REGRESSION MODEL

Linda M. Orr, University of AkronFrederik Beuk, University of Akron

Hyeong-Gyu Choi, The University of Memphis

SUMMARY

This study investigates the impact of marketing con-sensus on firm performance. We particularly focus on theeffect that decision making consensus regarding the firm’smarketing strategy has on firm-level performance. Animportant moderator of this relationship is the creativityor innovativeness of the firm’s products or services. Wetest three hypotheses:

H1: Consensus is related to firm performance in a curvi-linear manner (inverse U-shaped relationship).

H2: Marketing creativity moderates the effect betweenconsensus and firm performance.

H3: Marketing creativity directly increases firm perfor-mance.

We use both subjective and objective data gatheredfrom 174 senior executives of North American companiesin various industries. Consensus was measured with itemsadapted from Hult, Ketchen, and Slater (2004), Moorman(1995), Menon and colleagues (1999), and Johnson, Sohi,and Grewal (2004). The scale used in this study to mea-sure marketing creativity was adapted from Im and Work-man (2004). Objective financial data were obtained frommultiple secondary data sources (Compustat, Hoover’s,and DandB’s Million-Dollar Database) for the year im-mediately following the survey.

We use a curvilinear model that explains 37% of thevariance and shows substantially different results fromthe previously frequently used linear models. We suggestthat one reason for the often conflicting empirical findingsin the consensus – performance literature is because ofthis lack of appreciation for a non-linear relationship. Inaddition, we demonstrate how marketing creativity is akey moderator of this relationship.

Our study provides important insights for managers.It shows that managers across all ends of the continuum ofconsensus styles can be effective: from the completeautocratic manager, to the extreme consensus focusedmanager. In short, neither extreme is right. Results of ourstudy show that managers should indeed attempt to com-municate with their employees, attempt to incorporateemployees’ input, encourage the challenging of othersopinions, seek to build consensus during strategy making,and ensure that there is a great deal of buy-in on decisions.However, managers should do all of this only up until apoint. At the end of the day, managers also need to makesome decisions with little or no staff input, enforce poli-cies and decisions, and put plans into action withoutcomplete agreement. Going against the grain and takingaction without complete buy-in may in fact be a goodbusiness practice, particularly in markets that are charac-terized by more intense competition. The real managerialskill in this regard is to recognize when more consensusadds value to the firm, and when it does not.

Furthermore, our study shows that the level at whichconsensus gets “too high” depends on the level of creativ-ity within the organization. In highly creative organiza-tions, the optimal level of consensus is much lower than inless creative organizations. In other words, if a companyis constantly coming up with new products and ideas andimplementing these rapidly, there is hardly time for con-sensus building. Whereas, firms with low levels of cre-ativity, and perhaps very stable, not rapidly changingproduct lines, can afford the time and energy to buildconsensus. This finding should also guide academics infuture research looking to examine how creativity affectsorganizations’ decision-making processes. For example,new product development processes may be impacted orhindered by consensus making or other decision-makingprocesses. References are available upon request.


For further information contact:Linda M. Orr

Marketing DepartmentCollege of Business Administration

University of Akron259 South Broadway, Room 307

Akron, OH 44325Phone: 330.972.5447



DOES CEO TENURE REALLY MATTER? THE MEDIATING ROLE OFEMPLOYEE AND CUSTOMER RELATIONS

Xueming Luo, The University of Texas at Arlington,Vamsi K. Kauri, University of Missouri, Columbia

Michelle Andrews, The University of Texas at Arlington,

SUMMARY

Does CEO tenure really matter? Over the decades,this question has garnered immense attention from bothmanagers and researchers, spawning many theories in thestrategy literature. CEO career stage theory (Hall 1976)seems to have won the widest acclaim. This theorysuggests that CEO tenure is characterized by multipleseasons. Specifically, Hambrick and Fukutomi (1991,p. 719) note that “there are discernible phases, orseasons, within an executive’s tenure in a position, and[those] seasons give rise to distinct patterns of executiveattention, behavior, and ultimately, organizational perfor-mance.” CEOs learn rapidly during their initial years incharge, but then grow stale in the saddle as they lose touchwith the external environment (Henderson, Miller, andHambrick 2006). Thus, extant literature acknowledgesthat the relationship between CEO tenure and firm perfor-mance is an “inverted-U.” That is, firm performanceincreases until CEO tenure reaches an optimal point andgradually declines thereafter (Miller and Shamsie 2001).

However, the performance relevance of CEO tenureis a complex phenomenon that scholars believe goesbeyond the simple, direct effects (Hambrick and Fukutomi1991). Therefore, to get a holistic view of the causallinkages between CEO tenure and firm value, it is impor-tant to account for the underlying mechanisms that chan-nel CEO tenure’s impact (Simsek 2007). Nevertheless,even after several calls (e.g., Souder, Simsek, and Johnson2011), our knowledge of the intermediate factors thatchannel the impact of CEO tenure on organizationalperformance is surprisingly limited.

In bridging this crucial knowledge gap, we proposetwo novel routes through which CEO tenure influencesfirm value creation. The first channel stems from a firm’srelationship with one of its most important internal stake-holders – employees. As CEO tenure increases, CEOsgain more influence, power, and credibility (Hambrickand Fukutomi 1991; Wu, Levitas, and Priem 2005).However, longer CEO tenure rarely translates into supe-

rior performance automatically without the support offirm human capital. CEOs with longer tenure can leveragetheir reservoir of influence and credibility to create “unityof purpose” and foster trust between the firm and itsemployees (Lang, Ronit, and Schneider 2008; Simsek etal. 2005; Souder et al. 2011), which positively affect firmperformance (Wang, He, and Mahoney 2009). Therefore,the potential rent-generating role of CEO tenure is path-dependent and can partially be accounted for by the firm-employee relationship strength.

The second channel that we propose is rooted in thefirm’s relationship with its key external stakeholders –customers. As their tenure and market knowledge in-creases, CEOs such as Bill Gates and late Steve Jobs maydevelop a charisma that invigorates and draws customerscloser to the firm (Canella and Hambrick 1993). Duringtheir career seasons, CEOs’ strategies to tackle emergingproduct-market needs and evolving customer demandsaffect the strength of firm-customer relationships(Hambrick 2007; Musteen et al. 2006), which subse-quently influence firm value (Bantel and Jackson 1989).Indeed, without attracting strong customer segments thatare loyal to the firm’s product offerings, even experiencedCEOs cannot sustain competitive advantages in the mar-ketplace. Thus, the potential performance impact of CEOtenure can be channeled by the extent to which CEOtenure affects firm-customer relationship strength.

Our findings support these mechanisms for the im-pact of CEO tenure on firm value creation, i.e., via thestrength of firm relations with two of the most importantstakeholders, employees and customers. We also find thatCEO tenure has a linear impact on firm-employee rela-tionships but an inverted-U shaped impact on firm-cus-tomer relationships, thus demonstrating the complexityregarding the internal versus external focus of CEOsacross their career stages. These results also help disen-tangle the mixed findings pertaining to the ultimate per-formance impact of CEO tenure. References are availableupon request.



The University of Texas at Arlington,Arlington, TX 76019Phone: 817.272.2279



WHEN THE FOG DISSIPATES: THE CHOICE OF STRATEGICEMPHASIS IN A PARTNER

Todd Morgan, Kent State UniversitySergey Anokhin, Kent State University

SUMMARY

For many established firms the ability to innovate iscritical when it comes to building a foundation on whichtheir competitive advantage would rest (Varadarajan andCunningham 1995). Increasingly, this requires organiza-tions to look not only inwardly but also outwardly in theirsearch for innovative ideas (Rice et al. 2000). The grow-ing prominence of interorganizational relationships (Mor-gan and Hunt 1994) is a testament to the importance ofexternal sources of innovative ideas (Chesbrough andTucci 2004). A growing trend among large public compa-nies interested in innovation is to enter equity-basedpartnerships with smaller privately owned, growth-oriented firms that have the technological insight butmay lack innovative breadth and the capabilities to bringtheir technological prowess to fruition (Swaminathanet al. 2008). Yet, as established companies differ in theirstrategic emphasis with respect to allocating resourcesbetween value creation and value appropriation (Day1994), so does their attractiveness for smaller technologi-cally oriented players.

When corporations emphasize value creation, thesuccess of the company relies on innovation and outper-forming competitors with respect to the technologiesemployed (Mizik and Jacobson 2003) When value appro-priation is emphasized firms rely on heavy advertising(Mizik and Jacobson 2003). Potentially, both strategicemphases may be attractive to small firms. Value appro-priation-focused large companies tend to defend theirposition in the market against competition by erectingbarriers to imitation and through the use of isolatingmechanisms such as brand-based advertising (Mizik andJacobson 2003). Value creation-focused corporations maybetter understand the small firm’s technology and couldassist in its development and commercialization(Dushnitsky and Shaver 2009) thus arguably providing aneven more attractive deal to small firms.

However, the mutual attraction between large com-panies and smaller firms is not without challenges.Because public companies openly pursue the innova-tive agenda, they may try to misappropriate the tech-nologies developed by small firms. Unless the largecorporation has established a strong track record thatclearly suggests that it does not mistreat its smaller coun-terparts, the small firm is wise to exercise caution when

considering joining forces with an otherwise attractiveequity partner. Like the fog, information asymmetry blursthe vision for the smaller firm (Akerlof 1970), and thesmart strategy may be to steer clear of all partnershipsbecause some of them may hurt the firm.

When the ultimate intentions of the large corporationare foggy, the small firm may shield itself against misap-propriation by choosing to partner with companies thatemphasize advertising over R&D and thus may not havesufficient absorptive capacity (Cohen and Levinthal 1990)to decode and internalize the small firm’s knowledgebase. Even though working with a technologically cap-able partner should be the first choice for a small firm(Dushnitsky and Shaver 2009), when the fog of informa-tion asymmetry is thick it is perhaps best to seek a “safe”partner who may ease the market entry, provide legiti-macy and access to distribution channels (Chesbrough2002) rather than flirt with a technological powerhousethat may be of great help but is as likely to snatch the keyideas from the small firm to later claim them as its own.

On the contrary, when the information asymmetry isreduced and small firms can make an accurate assessmentof the true intentions of the large corporation, they willfavor R&D-centered companies that focus on value cre-ation. If the corporate partner with years of technologicalexpertise accumulated in its R&D department has held itspart of the deal with other small firms in the past, workingwith such a partner may be of tremendous value and couldbe instrumental in perfecting the technology for the ulti-mate success in the marketplace. In other words, thisimpact of strategic emphasis on attractiveness may becontingent on information asymmetry between the largecompany and its smaller counterpart. The fog of informa-tion asymmetry could make them choose much safer(although arguably less valuable) value appropriation-focused partners that emphasize advertising. As the fogdissipates, however, value creation-centered companies’attractiveness to small firms may increase.

Our dataset covers 233 instances of equity partner-ship between large public companies and 1445 smallprivate firms, with an average large company supportingover six small firms in a given year. Because the numberof smaller firms attracted by the established corporationsis a count variable, we utilized a population-averagednegative binomial estimation technique with equal-


correlation structure (McCullagh and Nelder 1989). Aset of hierarchical models was fitted by use of the gener-alized estimation equation approach (Liang and Zeger1986); all models proved significant. Information asym-metry was a significant predictor of the establishedcompany’s attractiveness to small firms but strategicemphasis was not. As expected, information asymmetrywas negatively related to the attractiveness of an estab-lished firm to smaller firms (β = -.33, p < .05). The dataalso suggests that all else being equal, strategic emphasisdoes not seem to matter for small firms seeking a largerpartner thus leaving preference for strategic emphasiswithout statistical support. The interaction of the informa-tion asymmetry and strategic emphasis is positive andsignificant (β = .59, p < .05) thus providing support to themoderating relationship. Interestingly, in the presence ofthe interaction term, strategic emphasis variable gainsmarginal significance (β = -.26, p < .10) thus suggestingthat small firms prefer to partner with large corporationsthat emphasize R&D to the exclusion of advertising.

The results of our research show that informationasymmetry may play a crucial role in the development ofan equity partnership between large public corporations

and small privately owned growth-oriented firms. Notonly is information asymmetry significant in its own rightas a predictor of the large corporation attractiveness as anequity partner to smaller firms, but also it attenuates theeffect that strategic emphasis adopted by the corporationhas on such attractiveness. Indeed, the fog of informationasymmetry, makes small firms wary of entering equity-based partnerships and also makes them unwilling topartner up with corporations that potentially have themost to offer them in terms of technological expertise.

When the information asymmetry fog is at its thick-est, small firms choose to work with corporations thatinvest in advertising at the expense of R&D. Once the fogdissipates, the picture changes dramatically. Establishedcorporations that pursue value creation strategic emphasisgain attractiveness in the eyes of the small firms whilevalue appropriation-focused corporations lose it at a simi-lar rate. Surprisingly, by itself strategic emphasis does notappear to matter much when considered without thepresence of the interaction effect mentioned above. Per-haps, the benefits and drawbacks associated with the bothends of the strategic emphasis continuum are balancedrather evenly. References are available upon request.

For further information contact:Todd Morgan

Kent State University475 Terrace DriveKent, OH 44242

Phone: 216.469.4663Fax: 330.672.5006



WHY QUALITY MAY NOT ALWAYS WIN: THE IMPACT OF PRODUCTGENERATION LIFE-CYCLES ON QUALITY AND NETWORK

EFFECTS IN HIGH-TECH MARKETS

Richard T. Gretz, Bradley University, PeoriaSuman Basuroy, The University of Oklahoma, Norman

SUMMARY

Many high-tech industries are characterized by posi-tive indirect network effects associated with the hard-ware/software relationship – consumers receive greaterbenefit from hardware with a larger selection of software,and software firms find it more profitable to provide forhardware with a larger installed base of consumers.Recent marketing literature has witnessed a majordebate about the critical drivers of success – qualityversus network effect, as well as efficiency in these typesof markets (Ratchford 2009; Reibstein 2009; Rossi 2009;Tellis, Yin, and Niraj 2009a, 2009b). Tellis, Yin, andNiraj (2009a) demonstrate that both the quality effect andnetwork effect are significant factors determining marketshares in these markets, but that quality effect is moreimportant than the network effect, and that such marketsare efficient.

Our paper contributes to this current literature and theongoing debate on quality and network effects in high-tech markets and extends the results of Tellis, Yin, andNiraj (2009a) in several ways. Using a data set from thehome video game industry (1995–2007) we examinequality and network size impacts on market share overdifferent phases of the product life-cycle. First, we repli-cate prior research and show that indeed both quality andnetwork effects are significant factors in determiningmarket shares in high-tech markets. Second, contrary tothe relatively stable impact of quality and network effectsfound in Tellis, Yin, and Niraj (2009a), we show that thequality and network effects vary over the productgeneration’s life-cycle. Third, we show that there aretimes when the network effect can indeed dominate thequality effect. Specifically we find that the network effectmay dominate the quality effect during the Growth andMaturity phases of a product generation’s life-cycle.Fourth, our results show that there is a distinct possibilitythat these high-tech markets may not be efficient in thatthe highest quality product may not be the market leader.In addition, in terms of estimations, we jointly estimateboth the demand and the supply equations allowing forbackward compatibility of the products (Ratchford 2009)and we control for the endogeneity issues raised byReibsten (2009) and Rossi (2009) by using several rea-sonable instruments with instrumental variables regres-

sion techniques via generalized method of moments(GMM) estimations.

The video game industry is ideal for study for severalreasons. First, it is a classic example of an industry whereindirect network effects play a major role in determiningmarket share outcomes: consoles become more attractiveas the number of available games for a console increases;the number of available games increases as the installedbase for the console increases. Next, quality increasesregularly through the introduction of more technicallysophisticated consoles that can play more complex andgraphically superior games. In addition to incrementalimprovements within a product generation, we observefour separate product generations over the life of the dataset; each generation delineated by a significant jump in thetechnological capability of consoles and games. Finally,each generation goes through four phases of the productlife-cycle similar to those defined by Golder and Tellis(2004): Introduction, Growth, Maturity, and Decline.

The results of our study generate several implicationsfor managers. First, in the economics literature, the over-all strong impact of network effects often imply a lock-inand a dominance of an existing product even when itmight have an inferior quality (see, among others, Farrelland Saloner 1986; Clements 2005; David 2007). How-ever, contrary to some of these results, Tellis, Yin, andNiraj (2009a, p. 147) argue that a network need not be areliable shield to protect incumbents. Constant qualityenhancement is an effective way for existing leaders todefend their current positions. Our results confirm theTellis, Yin, and Niraj (2009a) result on average. How-ever, our results also show that during specific phases ofthe generation life-cycle – Growth and Maturity phases,the network effect can be a reliable shield to thwart entry.In such situations, quality enhancements may not be aseffective as in other phases.

Second, it may be possible for network effects to slowthe introduction of a new product generation. Specifi-cally, the network effect is relatively strong during theMaturity phase of the generation life-cycle and weakerduring the Decline phase. A potential pioneer of a newproduct generation may have to strategically delay entryuntil the incumbent has entered the Decline phase.


Third, Tellis, Yin, and Niraj (2009a, pp. 147–148)argue that contrary to the lock-in predictions of econom-ics literature, under certain circumstances, networkeffects can make the market more efficient. “. . . A strongnetwork enhances the impact of quality.” Our results showthat there are some boundary conditions to these findings.Specifically, we see that the network effect is the smallestin the Introduction phase of a generation life-cycle, andhence a potentially inferior quality early mover can enterand position itself well in the marketplace. Therefore, wefind that the network effects may minimally enhance thequality effect, if at all.

Finally, recent studies have found that higher qualitycan attract larger networks thereby reinforcing the effectof superior quality. For example, Gretz (2010) found thathigher quality consoles attracted greater provision games.Our results show that higher quality may attract a largernetwork; however this occurs when the effect of a largernetwork is relatively small. That is, the reinforcement ofthe quality effect may not occur when the quality effect is

most in danger of being dominated by the network effect –during the Growth and Maturity phases.

Our research can be extended and improved in sev-eral ways. For example, we are currently unable toaccount for advertising or marketing spending as this datais not readily available. However, Tellis and Fornell(1988, p. 68) find that the effect of advertising on qualityis positive and mildly significant. However, quality af-fects advertising strongly positively in the later phases ofthe product life-cycle. Thus we might expect both a directand an indirect (via quality) impact of advertising onmarket share. Also, we are unable to account for bundlingof new products that may enable one firm to promoteadoption by including it with its console. For example,Microsoft bundled its Xbox 360 with the very populargame, or “killer application,” Halo 3 in September of2007. It is possible some “killer applications” may signifi-cantly affect market share throughout the generation life-cycle (see Gretz and Highfill 2010) in a way different thanwhat we present here. Future research should address thisaspect. References are available upon request.

For further information contact:Richard T. Gretz

EconomicsBradley University

1501 W. Bradley Ave.Peoria, IL 61625

Phone: 309.677.2298Fax: 309.677.4174



FOREIGN DIRECT INVESTMENT IMPACT ON RETAIL STRUCTURE

Boryana V. Dimitrova, Drexel University, PhiladelphiaBert Rosenbloom, Drexel University, Philadelphia

SUMMARY

Retail structure refers to the number, scale, and typeof retailers within a given geographic area such as a town,a region within a country, or even an entire country. Forexample, the retail structure in developed economies suchas the United States is dominated by large-scale, modernretailers such as hypermarkets, supermarkets, discount-ers, and shopping malls while the retail structure indeveloping economies such as India and Poland consistsmainly of small-scale, “mom-and-pop” stores (Reinartz,Dellaert, Krafft, Kumar, and Varadarajan 2011; Samiee1993). Studying retail structure is important for thefollowing reasons. First, the retail sector providesemployment to a significant portion of a nation’sworkforce, can help improve consumer welfare as well aseconomic growth in town centers, and often accounts forbetween 20 percent and 40 percent of Gross DomesticProduct (GDP) in both developed and developing coun-tries (Boylaud and Nicoletti 2001; Reynolds, Howard,Dragun, Rosewell, and Ormerod 2005; Smith and Sparks2000). Second, since numerous factors influence the retailstructure existing in a country, global retail chains need togain a better understanding of retail structure determi-nants in order to devise successful retail strategies for eachhost market. Specifically, retailers need to establish physi-cal presence in a given country in order to grow. However,while doing that, these firms directly interact with localconsumers, suppliers, and government decision makers.So, unlike manufacturers, retailers cannot hire an outsidecompany to distribute their products and services inforeign markets (Coe and Lee 2006; Goldman,Ramaswami, and Krider 2002; Reinartz, Dellaert, Krafft,Kumar, and Varadarajan 2011; Samiee, Yip, and Luk2004). The issue of understanding local retail markets hasbecome even more important in recent years as globalretail chains have continued to expand their operationsbeyond their national borders and grow their power.According to the 2011 Global Powers of Retailing Reportpublished annually by Deloitte Touche Tohmatsu Lim-ited, the top 250 global retailers have aggregate sales ofover $3.7 trillion and their average size, as measured bysales volume, is $15.05 billion. Moreover, each of the topfive global retailers has operations in at least thirteencountries in different regions of the world.

The primary focus of retail structure research hasbeen on investigating the relationship between demo-

graphic factors and retail structure (Bucklin 1972;D’Andrea 2010; Ford 1936; Hall, Knapp, and Winsten1961; Ingene and Brown 1987; Ingene and Lusch1981; Rosenbloom 1975; Takeuchi and Bucklin 1977).Although this demographic approach to studying retailstructure is of importance in explaining and predictingretail structure in geographical units ranging in sizefrom small local areas to entire countries, it ignores otherenvironmental factors. In particular, retail sector regula-tions such as foreign direct investment (FDI) restrictionsin the retail sector as well as property regulations andopening hours regulations are designed to restrict theexpansion of large-scale foreign and domestic retail chainsand, thus, protect small stores from intense competition(Boylaud and Nicoletti 2001; Czinkota 1985; Mishra2008; Pilat 1997). So, these factors are also importantretail structure determinants.

The purpose of the current study is to investigate theimpact of one set of retail sector regulations – FDI restric-tions, on two retail structure components: (1) retailerintensity (number of retail stores per 1000 people) and (2)retailer scale (sales per store). FDI restrictions are regula-tions that completely restrict or inhibit foreign retailersfrom entering a country’s retail sector or put limits onforeign equity ownership (Lapoule 2010; Mishra 2008;Reinartz, Dellaert, Krafft, Kumar, and Varadarajan 2011;Uncles and Kwok 2009). FDI restrictions are among themost widespread regulations in the retail sector and arepresent in both developing and developed countries, butare most prevalent in developing nations. The impact ofFDI restrictions on the domestic retail sector has been amajor topic of interest for both policy makers and globalretail powers as some developing countries are imple-menting strict FDI regulations, thus inhibiting entry offoreign retailers into the domestic market (Halepete, Iyer,and Park 2008; Lapoule 2010; Majumdar 2008; Mishra2008). Therefore, investigating the relationship betweenFDI restrictions and retail structure can offer importantmanagerial and public policy implications.

We find that FDI restrictions are negatively related toretailer intensity, but positively related to retailer scale.Our results also show that whether a country is a formercommunist nation, also referred to as a transition economy(TE), as well as a country’s level of economic develop-ment are moderators of the relationships examined here.References are available upon request.


For further information contact:Boryana V. Dimitrova

LeBow College of BusinessDrexel University

3141 Chestnut StreetPhiladelphia, PA 19104Phone: 201.238.9237



THE DIFFERENTIAL EFFECT OF INTERNET TRUSTMARK SOURCE:AN EXPLORATORY STUDY OF KOREAN AND U.S. CONSUMERS

Sohyoun Shin, Eastern Washington University, SpokaneK. Damon Aiken, Eastern Washington University, SpokaneVincent J. Pascal, Eastern Washington University, Spokane

SUMMARY

Based on trust in e-commerce and third-party en-dorsements as theoretical backgrounds, an exploratorystudy is undertaken to investigate the source influence ofInternet trustmarks for both Korean and U.S. consumers.In a between subjects experiment an Internet trustmarksource is manipulated (e.g., governmental affiliated orga-nization, industry experts, and typical consumer review-ers) in order to determine whether or not the trustmarksource can influence perceptions of three important con-structs (i.e., firm-specific trust, the product being adver-tised, and purchase intentions) of interest to Internetbusiness in regards to Korean and US consumers.

Results of empirical studies with the data of 394Korean and 358 U.S. consumers show that the presence ofa trustmark has a positive influence on firm-specific trust(FST) for both Korean and US consumers. Further, forKorean consumers, an industry expert source trustmarkappears to have the most positive influence on the out-come measures, whereas a government affiliated or con-sumer reviewer source trustmark appears to have the mostpositive influence with U.S. consumers. Moreover, as oneof the outcome measures, product evaluation (PE) andpurchase intention (PI) appear more positively assessed

when trustmarks are presented to Korean consumerswhereas in the U.S. sample, having a trustmark on websitedoes not appear to increase neither PE nor PI, implyingthat different routines of customers’ information process-ing are affecting purchase decisions in the Korean andU.S. contexts.

In addition, generalized Internet commerce trust(GICT) is found to moderate the influence of trustmarksource for both Korean and U.S. consumers. However, themoderating effect differs by trustmark condition. In par-ticular, for the governmental trustmark group in the U.S.sample, FST, PE, and PI are all moderated by GICT. Forthe industry experts’ trustmark, FST, PE, and PI aremoderated by GICT level for the Korean sample whereasGICT fails to moderate any of the relationships in the U.S.sample.

As an initial comparison study, the research findingssuggest managerial implications in regards to using tar-geting strategy based on a customer variable (i.e., generaltrust level on Internet transaction) and for internationalmarketing concerning the selection of appropriate typesof trustmark based upon consumer nationality. Limita-tions of the study are presented along with directions forfurther research. References are available upon request.

For further information contact:Sohyoun Shin


Spokane, WA 99202–1677Phone: 509.828.1243



MARKET ORIENTATION AND BUSINESS PERFORMANCE IN MNCFOREIGN SUBSIDIARIES: THE MODERATING EFFECTS OF

INTEGRATION AND RESPONSIVENESS

Riliang Qu, Renmin University of China

SUMMARY

Although there is a growing body of empirical evi-dence to attest for market orientation’s positive impact onbusiness performance, several researchers have reportednonsignificant or even negative effects for this associa-tion. On the other hand, little research has examined theimpact of market orientation on the business perfor-mance of multinational corporation’s foreign subsid-iaries despite the multinational companies’ dominancein many industrial sectors in many countries. In the fieldof multinational subsidiary management, there is also aneed to examine the impact of integration and responsive-ness on MNC subsidiaries’ strategies.

To address the above knowledge voids, we investi-gate the market orientation and business performancerelationships in the context of multinational companies’subsidiaries. In specific, we examine whether the relation-ships are moderated by the degrees of integration and

responsiveness pressures under which a subsidiary isoperating.

To test the hypotheses relating to the moderatingeffects, we carry out a split-group analysis on a sample of252 foreign subsidiaries in the United Kingdom collectedvia a mail survey. The results reveal that marketorientation’s impact on business performance is largelymoderated by the local responsiveness pressures but onlyto a limited extent by the integration pressures.

Our findings contribute to ongoing debate amongacademics on whether market orientation’s effects onbusiness performance are universal or whether they arecontingency-based by providing new empirical evidenceto suggest the latter. Our research also contributes to theinternational business literature by demonstrating thebenefits of separately examining the effects of integrationand responsiveness instead of using the established I-Rframework. References are available upon request.

For further information contact:Riliang Qu

School of BusinessRenmin University of China

59 Zhongguancun Dajie StreetBeijing, 100872

ChinaPhone: +86.10.82500434

Fax: +86.10.82509172E-Mail: [email protected]


EMERGENCE OF ONLINE SHOPPING IN INDIA:SHOPPING ORIENTATION SEGMENTS

Kenneth C. Gehrt, San Jose State UniversityMahesh N. Rajan, San Jose State University

G. Shainesh, Indian Institute of Management, IndiaDavid Czerwinski, San Jose State UniversityMatthew O’Brien, Bradley University, Peoria

SUMMARY

Although Internet retailing in India is on the verge ofrapid growth, relatively little is currently known aboutIndian non-store shopping behavior in general and Indianonline shopping in particular. The purpose of this study isto explore Indian shopping orientations and to examinehow they relate to online shopping by identifying shop-ping orientation-defined segments and profiling them interms of website characteristic importance, online shop-ping behavior, and demographics. By examining theseissues, the research will provide a foundation of under-standing of Indian shopping in general and online shop-ping in particular.

Since online shopping is not yet widespread in India(Keniston 2004), a data collection procedure that tends toover-sample online shoppers and potential shoppers wasselected (cf., Gehrt et al. 2007); thus, an online surveymethod was chosen. A total of 2500 subjects were ran-domly contacted from a reputable research company’svalidated, opt-in panel of online respondents. Eight hun-dred forty responded yielding 536 fully completed re-sponses for a 21 percent response.

The analysis involved several procedures. Factoranalysis was used to identify underlying shopping orien-tation themes. Cluster analysis was then used to identifysegments of respondents who shared similar profilesacross the shopping orientations. Factor analysis was alsoused to identify key website dimension factors. Finally,appropriate statistical analyses were used to characterizethe clusters in terms of website attribute importance,online shopping behavior, and demographics. Factor analy-sis of the 39 shopping orientation statements yielded fourfactors/shopping orientations with an eigenvalue > 1.00on which three or more statements loaded at > .50.Twenty-three of the statements came into play. The pro-cedure yielded factors with Cronbach coefficientreliabilities ranging from .81 to .86. The shopping orien-tations included value, quality with convenience, recre-ational, and reputation with convenience. Four criteriawere used to identify the optimal cluster solution whichyielded a three-cluster solution including value singular-ity, quality at any price, and reputation/recreation clusters.

Factor analysis was also used to summarize 20 websitestatements, yielding only two factors/website dimensionswith an eigenvalue > 1.00 on which three or more state-ments loaded at > .50. Fifteen of the 20 statements cameinto play for the two resultant website dimensions: (1)website responsiveness and security and (2) websitedesign and product assortment. In the case of theanalysis of both the shopping orientations and websitedimensions, several factors emerged that had two distinctthemes. Despite two themes emerging for two of the fourshopping orientations and two of two website dimension,Cronbach alphas were very high.

The analyses suggest that there are two segmentsready to be tapped in the Indian online shopping market;the Quality at any Price Segment and the Reputation/Recreation Segment. The segments’ Internet usage pat-terns and online shopping patterns are very similar. Wherethey differ is in the shopping orientation defined motiva-tions and in some of the products that they are inclined topurchase. For the Quality at any Price Segment, priceappears to be a relatively unimportant deterrent to acquir-ing quality. These young professionals tend to purchasetravel, utilities, electronic media, and consumer electron-ics. The Reputation/Recreation Segment is interested inacquiring name brands, conveniently, and derives enjoy-ment from the act of shopping. This young, blue collar/clerical/service industry segment is well educated. Theytend to purchase a wide array of computer-related items,clothing and accessories, as well as the categories that cutacross the three segments, travel, utilities, and electronicmedia. These two segments represent India’s pioneeronline shoppers and contrast with U.S. pioneer onlineshoppers who were substantially motivated by priceissues (Maguire 2005). Although the Value Singular-ity Segment’s online buying profile is low at this point,they do show some interest. Online sellers may rely on thepassage of time and the diffusion of innovation process tobring these consumers on board. The results show that theonly motivational appeal that might be used to speedalong their adoption process is the value appeal.

The dimensionality of the shopping orientation factoranalysis procedure for this sample of Indian consumers, aswell as the website dimension procedure, appear to berelatively simpler than similar analyses in highly devel-


oped economies. Gehrt et al. (2007), for instance, used analmost identical scale for an analysis of Japanese consum-ers and arrived at a solution of seven factors, each withvery singular themes. Thus, the Indian consumer’sconceptualization of the marketplace, immersed in anemerging economy, may not yet fully expanded withrespect to their marketplace experience and their percep-

tions have not yet become as fully elaborated as they arelikely to once their marketplace experience deepens andmarketplace choices expand (cf., Mahi and Eckhardt2007). Not only did fewer factors emerge but the Indianshopping orientation factors did not have the singularityof theme seen in research conducted in more developedeconomies.

For further information contact:Kenneth C. Gehrt

Marketing and Decision SciencesCollege of Business

San Jose State UniversitySan Jose, CA 95192–0069

Phone: 408.924.3534Fax: 408.924.3445



IT’S NICE TO BE IMPORTANT, BUT IT’S MORE IMPORTANT TO BENICE: COUNTRY-OF-ORIGIN EFFECTS IN PRODUCT FAILURES

Huimin Xu, The Sage Colleges, AlbanyAda Leung, Penn State University, Reading

Terry Yan, Colorado State University, Fort Collins

SUMMARY

Extant COO literature largely focuses on how anorigin country’s competence across board or in a specificindustry shapes perceptions of its products (e.g., Hongand Wyer 1989). Country competence refers to degrees ofeconomic and technological development and the result-ant power and influence in the global society. This paral-lels Fiske’s definition of perceived people competency:“Out-groups are perceived as more competent to theextent that they are perceived as powerful and high status”(Fiske et al. 2002). German luxury cars, Japanese gadgets,and Swiss watches all rub off the origin country’s reputa-tion of being generally competent and advanced. Relativeto the competence dimension, the warmth perception ofan origin country is seldom examined. Similar to howFiske et al. (2002) define people warmth, we definecountry warmth perception as how much an individualsees a foreign country as well-intentioned, friendly, warm,and cooperative. This perception may derive from one’sunderstanding of past and current political and economicrelationship between the foreign country and one’s owncountry. Even though the contents of country imagesseem diverse and arbitrary, Fiske et al. (2002) contend thatthe images of out-groups can be reduced to two dimen-sions: competence and warmth. When people meet otherindividuals, they want to know: (1) what the other’s goalswill be in relation to the self or in-group and (2) howeffectively the other will pursue those goals. These twodimensions can be orthogonal to each other. In the contextof origin country perceptions, four quadrants result fromthese two dimensions. Some countries enjoy an overallimage of being both competent and warm. Some areperceived as strong and competent, but not particularlyfriendly or warm. Some countries come across as warm,though not particularly competent. Yet some are seen aslow on both.

Different combinations of perceived warmth andcompetence result in unique intergroup emotions. In par-ticular, stereotypes of low warmth justify taking actionagainst envied groups by casting the groups as beingconcerned only with furthering their own goals. Webelieve these premises also pertain to consumers’ reactionto foreign businesses. Businesses from “cold” countrieswould receive a harsher treatment relative to those from

“warm” countries, especially when a punishment is ap-parently justified.

The current research is interested in products withwhich safety can readily become an issue. Whether aproduct is safe and sound usually is not obvious toconsumers at the point of purchase. COO’s role is morepronounced when products pose such risks. When anorigin country is perceived as warm and well-intentioned,consumers are more likely to trust its products to be safeand harmless.

H1: A perception of the origin country being warm ispositively related with purchase intention.

Many products are outcomes of collaborationbetween businesses from different countries. Due tothe complexity in supply chain, if products fail, thereis often a period of time when consumers do not knowwhich business is (most) culpable. With culpability in-creasingly ambiguous, it would be useful to companies tounderstand if observers are likely to give a company thebenefit of the doubt. Even if a company is later exoner-ated, its brand can still suffer damage if consumers blamethe company for the product failure early on. Consumerswith warm feelings toward the origin country selectivelyseek and process confirmatory information, thereby less-ening the negative impact of the failure on perceptions ofthe company.

H2: In product failure, a perception of the origin countrybeing warm enhances the perception of the company’ssufficient motivation/effort to produce safe goods.

After it is ascertained that a foreign company is atfault for the product failure, another question arises forconsumers: is it because the company does not careenough to enforce quality standards, or is it because thecompany does not have required expertise? Althoughneither scenario provides peace of mind, the former ismore worrisome, because motivation and intention tendto be seen by lay people as deeper-seated and less likelyto change (Dweck 1999). As Fiske et al. (2002) pointedout, the warm but not competent out-groups can receivepity and sympathy, whereas the competent but not warmout-groups receive harsher treatments: envy and even


exclusion. We posit that consumers would be willing togive the stumbled company a second chance if the com-pany is seen as well-intentioned more than if it is seen ascapable.

H3a: In the wake of product failure, a perception of theorigin country being warm, more than that of itbeing competent, leads to consumers’ repatronageintention after the troubled company issues a recall

H3b: In the wake of product failure, a perception of theorigin country being warm, more than that of itbeing competent, leads to purchase intention ofimproved products

H3c: In the wake of product failure, a perception of theorigin country being warm, more than that of itbeing competent, leads to perceived excessivenessof a specific amount of government-imposed fine.

We run two studies using different choice contexts(i.e., fruit juice and batteries). Warmth and competenceperceptions were measured. Regression analyses largelysupported the hypotheses. Coming from a powerful, com-petent country is good news to a foreign business, becausethis translates to greater purchase intention in the absenceof product failure. However, once product failure occurs,perceived competence ceases to help the business inquestion. This is shown from all the non-significant coef-ficients in predicting post-failure attitudes. Instead, abackground in a friendly country turns out to be morebeneficial. Further, perceived motivation mediates therelation between warmth and purchase intention forimproved products. In sum, the current research high-lights an orthogonal but largely neglected dimension, i.e.,perceived warmth, as a factor no less important thanperceived competence in shaping consumer attitudes.References are available upon request.

For further information contact:Huimin Xu

The Sage Colleges140 New Scotland Avenue

Albany, NY 12208Phone: 518.292.1700



THE EFFECTS OF PRODUCT DIVERSIFICATION ANDGLOBALIZATION ON THE PERFORMANCE OF

LARGE INTERNATIONAL FIRMS

Tianjiao Qiu, California State University, Long Beach

SUMMARY

Product diversification has been a widely adoptedfirm strategy for growth in the global marketplace(Ravichandran et al. 2009). However, the research on theeffects of product diversification on firm performanceacross various disciplines, such as marketing, strategicmanagement, and finance, produces mixed results (e.g.,Graham, Lemmon, and Wolf 2002; Soni, Lilien, andWilson 1993). One stream of research that focuses oncomparing financial performance of diversified Ameri-can firms with that of specialized American firms demon-strates that product diversification leads to value discount,and diversified firms have significantly lower marketvalue than single-segment firms (e.g., Berger and Ofek1995; Lang and Stulz 1994). In contrast, another stream ofresearch suggests that diversification contributes to firmsurvival and long-term performance (e.g., Chiang 2010).Furthermore, studies of the product diversification-per-formance link have been centered on American firms, andfew studies have explored the impact of product diversi-fication on firm performance with more diverse interna-tional backgrounds (Lins and Servaes 1999). The advanceof globalization makes product diversification an evenmore prominent marketing strategic initiative across firmsworldwide. How product diversification impacts firmperformance under various social and economic global-ization forces is an intriguing question for not only thepractitioners but also the academicians (Wiersema andBowen 2008).

The purpose of this study is two-fold. First, the studyintends to examine how product diversification impactslarge international firm performance. Product diversifica-tion refers to the extent to which a firm manages its market

segments based on its multiple and disparate products(Hitt, Hoskisson, and Kim 1997). Second, the studyintends to investigate how social and economic globaliza-tion impacts the relationship between product diversifica-tion and the performance of large international firms.Despite the recent globalization of economic and socialactivities (Dreher 2006), empirical evidence on the inter-action between the degree of globalization and the successor failure of firm product diversification strategic initia-tives is rather scarce. The study attempts to fill this gap byarguing that examining performance implications of prod-uct diversification needs to take into consideration eco-nomic and social globalization in which all firms areembedded.

We build our sample through Fortune global mostadmired companies, spanning the years 2006–2009, anddevelop a data set from multiple secondary data sources.We then empirically test a conceptual model of the effectsof product diversification and globalization on the perfor-mance of large international firms with hierarchical linearmodeling techniques. Our results demonstrate that diver-sification in multiple product segments leads to a higherfinancial performance of large international firms. Fur-thermore, economic and social globalization, indicated byactual flows, personal contacts, and information flows,moderates the above positive relationship. For large inter-national firms whose home countries promote interna-tional trade and capital investment, encourage cross-border inter-personal communication, and provide con-venient social media for interactions; the adoption ofproduct diversification strategy can significantly enhancetheir competitive positions in the global marketplace.References are available upon request.

For further information contact:Tianjiao Qiu

California State University, Long Beach1250 Bellflower Boulevard

Long Beach, CA 90840Phone: 562.985.4821



WHAT HAPPENS TO “BRAND JAPAN” WHEN TOYOTA SUDDENLYACCELERATES? THE SPILLOVER EFFECTS OF BRAND

TRANSGRESSIONS ON COUNTRY IMAGEAND RELATED BRANDS

Peter Magnusson, Florida International University, MiamiVijaykumar Krishnan, Northern Illinois University, DeKalb

Srdan Zdravkovic, Bryant University, SmithfieldStanford A. Westjohn, University of Toledo

SUMMARY

Country of origin (COO) continues to be one of themost important phenomenon in the international market-ing literature (e.g., Balabanis and Diamantopoulos 2011;Magnusson, Westjohn, and Zdravkovic 2011; Sharma2011). Extant COO research has largely viewed the cause-and-effect relationship between country and brands as thecountry is the “cause” and the brand is the “effect.” Forexample, Pharr’s (2005) major review of the COO litera-ture graphically depicts the causal relationship as country-specific beliefs à country-of-origin related thoughts àcountry-of-origin brand evaluations à purchase inten-tions.

The objective of this study is to investigate this causalassumption and examine whether the brand can be viewedas the cause, which affects country image and relatedbrands. We examine the extent to which a highly publi-cized negative event involving an iconic brand influencesattitudes toward its associated country image, its industry,and related brands. In effect, this study examines theexplanatory power of a causal sequence in which brand“behavior” affects country image and other related brands.

Toyota is the perfect example of an iconic brand thathas recently experienced major negative publicity. Afternumerous consumer complaints and alleged car accidentswith fatal consequences, Toyota was recently forced torecall over six million vehicles from the U.S. market andmore than eleven million worldwide (Maynard 2010).The incidents received heavy media attention and led tomultiple congressional hearings. One letter issued by theUnited States Congress to Toyota executives concludedthat “Toyota resisted the possibility that electronic defectscould cause safety concerns, relied on a flawed engineer-ing report, and made misleading public statements con-cerning the adequacy of recent recalls to address therisk of sudden unintended acceleration” (Ingram 2010).Although subsequent investigations have largely exoner-ated Toyota (Wald 2011), Toyota’s brand value has

decreased by more than $5 Billion (16%) (Interbrand2010) and Consumer Reports found a 46 percentage-pointdecline in consumer satisfaction.

Although the consequences have been severe forToyota, we are concerned with examining the broadereffects of this event on Japan’s country image and otherJapanese brands. The Japanese government acknowl-edged a fear that Toyota’s massive recall could have aspillover effect on other Japanese exports (Mochizuki2010), but nobody has been able to assess changes inconsumer attitudes before and after Toyota’s recall. Withthe help of unique longitudinal data, we are able toexamine attitudes toward Japan’s country image andJapanese brands pre- and post-crisis, which suggests asignificant negative spillover.

The longitudinal study observes a very interestingphenomenon, however, due to the longitudinal and real-world nature of this part of the study, we are unable tocontrol for potentially confounding factors. Therefore,subsequently, we develop a conceptual framework toexplicate the causal mechanisms involved. We empiri-cally examine the conceptual framework in four experi-mental studies that mimics an extensive recall similar tothe one experienced by Toyota.

In the experiments, we replicate a negative within-country spillover effects in three separate experiments(Mercedes-Benz-Germany, Hyundai-South Korea, andBeck’s-Germany). In advanced economies, the negativespillover on related brands is mediated by micro countryimage. In emerging markets, spillover is mediated by bothmicro and macro (Pappu and Quester 2007) countryimage. For low-involvement products (Beck’s beer),spillover is limited to brands within the product category,but not to brands outside the immediate category. Finally,the last experiment suggest that foreign prototype brandtransgression leads to a more positive attitude towarddomestic brands, which is mediated by consumer ethno-centrism.


We believe our study makes the following contribu-tions to the literature. First, we shed light on a causalrelationship that heretofore has largely been assumed tomove in the direction from country-level beliefs to brand-level evaluations (Bilkey and Nes 1982; Lee and Ganesh1999; Pharr 2005). Instead, this study investigates thepotential for the effects to move in the opposite direction,i.e., from brand-level to country-level beliefs, sheddingnew light on one of the core issues in the internationalmarketing literature (e.g., Anholt 2010; Magnusson et al.2011; Sharma 2011). Second, the theoretical frameworkand resulting hypotheses are grounded in prototype theory(Rosch 1978). Thus, we provide a new theoretical lens to

help us understand the relationship between brand imageand country image. Third, we identify the generalizabilityand boundaries of the theoretical framework by examin-ing multiple experimental contexts: advanced economiesvs. emerging economy brands, high vs. low-involvementproducts. Fourth, we examine whether a foreign brandcrisis affects consumer ethnocentrism and pro-domesticbrand attitudes. Prior research has primarily viewed con-sumer ethnocentrism as a chronic trait, yet, we examinewhether consumer ethnocentrism may also be a state ofmind that responds to arousal. References are availableupon request.

For further information contact:Peter Magnusson

Florida International UniversityUniversity Park, RB307

Miami, FL 33199Phone: 305.348.2571



THE REAL MCCOY: PRODUCT ETHNICITY, AUTHENTICITY ANDCOSMOPOLITANISM IN EVALUATIONS OF SEARCH AND

EXPERIENCE GOODS

Michael Wachter, Cleveland State UniversityJieun Park, Cleveland State University

SUMMARY

Real. True. Genuine. Consumers seek out and selectbrands that demonstrate high levels of authenticity toderive the symbolic benefits associated with these brandsand to use the brand as part of the construction of theirindividual identity. Successful brands engender percep-tions of being real, true, and genuine by rooting them-selves in authenticity. In maintaining a delicate balancebetween authenticity and social significance (Keller 2003),global brands like Apple, IKEA, and Virgin promptverisimilitude in the product without sacrificing theirrelevance to consumers. It is these symbolic benefitsconferred by a brand’s authenticity that helps explainhow product ethnicity, or the stereotypical associa-tions between a product category and its country of origin(Usunier and Cestre 2004), and product evaluations, interms of perceived image and purchase intention. Thisstudy proposes that perceptions of authenticity mediatethe relationship between cosmopolitanism and productevaluation. Additionally, congruity of a product with theexpected product ethnicity influences the effect of au-thenticity on consumer evaluations, particularly in thecase of experience products, relative to search products.

As “citizens of the world,” cosmopolitans possess agreater desire to consume products, brands, and ideas ofhigh quality. Additionally, cosmopolitans seek out prod-ucts which are authentic to a particular culture. Cannonand Yaprak (2002) suggest cosmopolitan individuals mayforego the functional benefits afforded by a product inorder to gain the symbolic benefits of a more authenticproduct or experience. A classic example of the impor-tance of symbolic benefits describes Westerners who eatAsian food using chopsticks despite their heighteneddexterity with knife and fork. Chopsticks make the expe-rience more real – more authentic. Cosmopolitans aremore likely to seek out and consumer products for thesesymbolic benefits. From this, it can be posited that higherlevels of cosmopolitanism are associated with productsrated more authentic by the consumer.

Among other dimensions, consumers judge authen-ticity based upon an object’s heritage and pedigree, stylis-tic consistency, quality commitments, and, most impor-tantly for the purposes of this study, by the object’srelationship to place (Beverland 2005; Grayson and

Martinec 2004). As a result, spatial origin for a productconfer a perception of authenticity to the brand, increasingthe importance of country of origin (COO) to productswith a strong product ethnicity. The most widely writtenstream in international marketing literature is the effectsof country of origin on product evaluations and purchaseintentions. A product’s COO acts as both an implicit andexplicit cue which influences consumers’ judgments andbehavior intentions through halo and summary effects.

Research indicates that congruity between the prod-uct category and product ethnicity influences consumerperceptions and purchase intentions for the brand orproduct (Usunier and Cestre 2007). Furthermore, con-sumers’ misclassification of the product with the correctCOO and the inability to associate the brand with anyCOO results in lower evaluations of brand image whencompared to accurate brand-country classifications(Balabanis and Diamantopoulos 2011). However,researchers debate exists the importance of correctlyassociating a product with its COO. Samiee, Shimp, andSharma (2005) challenge the diagnosticity of COO effectsin consumer judgments and purchase intentions. In theirstudy of American students, brand origin recognitionaccuracy was successful for about one-third of the 84brands tested, suggest low diagnosticity for this attribute.

Past research on product ethnicity and authenticity inthe international setting do not differentiate between howthese constructs affect experience and search goods. Nelson(1970, 1974) has been credited with defining the distinc-tion between experience and search goods based upon theavailability of diagnostic information on a product’s at-tributes. As opposed to search goods, experience goodsare dominated with attributes which cannot be assessedprior to purchase and consumption of the product. Infor-mation gathering becomes a vital task undertaken byconsumers to overcome the uncertainty of product qualityinherent in experience goods (Huang, Lurie, and Mitra2009; Neelamegham and Jain 1999; Weathers, Sharma,and Wood 2007). Given the importance of explicit andimplicit information cues to overcoming information asym-metry associated with experience goods, the relationshipof authenticity and product ethnicity on brand image andpurchase intentions should be more significant than forsearch products. Therefore, it can be posited that a product’sauthenticity and its congruity with the stereotypical prod-


uct ethnicity of its category interact with experience goodsto positively influence product image and purchase inten-tion.

Respondents were recruited to take an online surveythrough an invitation broadcast using online social net-works (i.e., Facebook and Twitter). A snowball techniquerequested them to pass along the survey to others. A totalof 126 surveys were started; 6 were deleted for missingvalues, yielding a total of 120 responses. The survey firstexposed the respondent to a two-sentence product con-cept (search or experience good) with a defined COO(matched or mismatched). After reading the concept, therespondents rated the concept in terms of its authenticityas well as personal evaluations of the product image, andpurchase intention. The respondents then answered ques-tions related to cosmopolitanism and key demographics(gender, age, income, education, birth country).

Multiple regression analyses tested the hypotheses inthe conceptual model. A regression of cosmopolitanism(COS) on Authenticity (AUTH) shows a statisticallysignificant relationship. The mediating role of AUTH inthe relationship between COS and brand evaluations(IMAGE and PURCH). Using the tests outlined by Baronand Kenney (1986), regressions were run for COS onIMAGE and PURCH and the full model with both COSand AUTH as independent variables. The relationshipsbetween COS with IMAGE hold a weak significance,while a significant relationship shown between COS andPURCH. The results show strong significance in therelationship between AUTH with IMAGE and PURCH.

Full models with both COS and AUTH as independentvariables show significant relationships with IMAGE andPURCH. However, COS is not significant in these mod-els, suggesting full mediation by AUTH on IMAGE andPURCH.

The study posits AUTH interacts with productethnicity (MATCH) and experience or search goods(TYPE) to moderate brand evaluations (IMAGE andPURCH, respectively). The three-way interaction ofAUTH*MATCH*TYPE offers no significance on IM-AGE, despite the significance of the model and AUTH.When analyzing the model for PURCH, not only is themodel significant, but both AUTH and the three-wayinteraction are as well. This suggests that purchase inten-tion differs based on whether product is an experience orsearch good, the product ethnicity matches stereotypicalassociations, and the level of authenticity

This paper makes three important contributions to thestudy of COO effects. First, the author suggests cosmo-politanism as an antecedent to perceived product au-thenticity. Second, it extends research on product ethnicityand its influence on brand evaluations by explicating themediating role of authenticity. Finally, the article extendsliterature on COO effects by discussing the importance ofproduct ethnicity in the context of experience goodsrelative to search goods. Product ethnicity, as an explicitcue, interacts with iconic authenticity of brands possess-ing a stereotypical product-country association which inturn activates the symbolic benefits of the brands. Refer-ences are available upon request.

For further information contact:Michael Wachter

Cleveland State University2121 Euclid Avenue, BU 464

Cleveland, OH 44115Phone: 216.687.4784



THE HOFSTEDE PARADOX: FACT OR FICTION?

Terry Clark, Southern Illinois University at CarbondaleSongpol Kulviwat, Hofstra University, Hempstead

Juan (Gloria) Meng, Minnesota State University, Mankato

SUMMARY

The marketing and business literatures seem to begenerally beset by two opposing errors in regards to theirconceptualizations and handling of the culture concept.On the one hand, it is treated anecdotally. On the otherhand, culture is represented in terms of a fixed number of“dimensions.” These dimensions are usually theory-basedmeasures of indices for which methodological rigor isstressed. Each culture characterized by such dimensionsis represented by a unique configuration of the indices.These configurations are then interpreted as having impli-cations for strategic and/or consumer behavior. The mostwidely used characterization of culture of this type isHofstede’s familiar five dimensions.

At first glance, the anecdotal approach to cultureseems weaker and unsupportable, while the dimensionalapproach appears stronger. In fact, both approaches maybe equally inadequate – the anecdotal because it tends toproduce a subjective and unrepresentative picture of theculture in question, the dimensional approach because it

is theoretically thin and overly abstract. Moreover, thedimensional approach can be misleading insofar as itinsinuates confidence, not on the basis of theoreticalvalidity, but methodological rigor—that is to say becauseit produces numbers.

The paper aims to provide a state-of-the-art review ofthe culture concept as characterized in the anthropologyliterature in the hope of stimulating discussion and bal-ance with regards to our own handling of the concept. Thepaper highlights the anthropologist’s conceptualizationsof culture from different schools of thought by introduc-ing and illustrating some core concepts widely used inanthropology but which have not yet been integrated intothe theoretical language of marketing discipline. It furtherdiscusses the methodological issues and problems, spe-cifically the progress in understanding cultural differ-ences will be hampered if the field continues to relydominantly on Hofstede’s (1980; 2001) benchmark dataor anecdotal approach. References are available uponrequest.

For further information contact:Songpol Kulviwat

Department of Marketing and International BusinessFrank G. Zarb School of Business

Hofstra University222 Weller Hall

Hempstead, NY 11549–1340Phone: 516.463.5519



ETHNOCENTRISM, CONSUMER ETHNOCENTRISM, RELIGION ANDRELIGIOSITY: A CONCEPTUAL REFORMATION

Daniel Friesen, Wayne State University, DetroitAttila Yaprak, Wayne State University, Detroit

SUMMARY

In the context of ethnocentrism and consumer ethno-centrism (CET), we reflect in this paper on the paucity ofattention given to the religiosity of consumers in market-ing literature, considering the extent to which large popu-lations of the world consider religion an important part ofevery day life. To the extent that consumers in variouscountries see life through various religious lenses, animportant part of consumer decision-making may bemissing from our body of knowledge as marketers.

We review the history of ethnocentrism in general, asa subject of mild interest to scholars until the periodimmediately following the Second World War and theHolocaust, when the construct went through something ofa Cambrian explosion, to the point where nearly 2000articles appeared in journals in the last decade of the 20th

century.

We then discuss CET, a construct introduced in the1980s to describe a sense among American consumersthat there was something inherently immoral or unpatri-otic about purchasing “foreign” products when Americanproducts were available (Shimp & Sharma 1987). Thoughin their seminal article Shimp & Sharma distinguishedCET from ethnocentrism, several times in the same articlethey referred to the ethnocentrism of consumers. Weconsidered it necessary, therefore, to make a clear distinc-tion in our paper between the two concepts.

In 2006 Shankarmahesh published a wide-rangingliterature review of CET’s antecedents and consequences.Shankarmahesh refers to religious intolerance as an ex-treme form of conservatism (per Sharma, Shimp & Shin1995). His review reveals, almost by omission, that scantattention has been given to the religiosity of consumerswith regard to CET. Generally, religiosity gets mention inthe context of prejudice; it does not appear in neutral orpositive terms. It may reflect something of an academicethnocentrism that people around the world who take theirreligious perspectives seriously in forming their

worldviews do not get serious consideration as consum-ers, except in denigration.

We propose that ethnocentrism and CET are separateconstructs, and that religiosity, but not a person’s specificreligion (Buddhist, Christian, Hindu, Muslim, etc.), willinfluence CET and buying behavior. For this opening ofour research stream we have restricted our proposedinvestigation to Christian and Muslim populations, andhope to extend our studies further to include Hindu andother theistic and non-theistic worldviews.

We expect to find two inverted U-shaped curves,where CET and willingness to purchase foreign productsincrease as a population moves from low to moderatelevels of religiosity, and then decline at higher levels ofreligiosity. We further expect that the effect of religiosityon willingness to purchase foreign products to be partlymediated by consumer ethnocentrism.

We have proposed four studies to investigate ourproposals. The first surveys students in an Americanuniversity, where the population is quite diverse, usingShimp & Sharma’s (1987) CETSCALE and the ReligiousCommitment Inventory-10 developed by Worthingtonand his colleagues (1988, 2003).

As the first study is designed to demonstrate correla-tive relationships, we have designed an experiment withUS university students where religiosity and CET areprimed for one group of students and not for another. Athird study replicates and extends the first, using an onlinepanel survey of consumers in the US, rather than focusingon students. A second replication using university stu-dents in other countries extends the investigation interna-tionally.

We expect to follow up our study by examiningreligiosity, CET and purchase intent among consumers incountries where different religions, e.g., Buddhism, Hin-duism, Judaism, are dominant, where religions compete inthe public square, and where the population is predomi-nantly secular. References are available upon request.


For further information contact:Daniel Friesen

Wayne State University300 Prentis Building, 5201 Cass Avenue

Detroit, MI 48202Phone: 313.577.4406



CONSUMERS’ GLOBAL IDENTITY AND NATIONAL IDENTITY:AN EMPIRICAL STUDY

Joyce X. Zhou, Emporia State University, EmporiaNitish Singh, Saint Louis University, Saint Louis

Jun Yu, Emporia State University, Emporia

SUMMARY

With the advent of globalization we are seeing moreinterconnectedness of economies, interdependencies ofpolitical and economic systems, and diffusion and knowl-edge of world cultures (Cleveland, Laroche, andPapadopoulos 2009). This interconnected world has ledto the process of cultural homogenization and culturalheterogenization creating a system of global interactionsthat are triumphantly universal and resiliently particular.In other words we are seeing not only the emergence ofglobal consumer culture but also a resurgence of interestand identification with national cultures (Arnould et al.2004; Keillor et al. 2001). The goal of this study is toexplore how this heightened awareness of global cul-ture and national culture impacts consumer’s globaland national identity and its impact on global consumersegmentation. In this paper, we discuss four segmentsbased on interaction of global and national identity.

Low Global and High National Identity (LGHN): Atthe particularistic end of Robertson’s (1997) global pro-cesses framework are individuals belonging to “nationalsocieties.” They have high awareness of their inherentcultural particularism clearly safe and thriving in the formof national societies or national cultures.

Low Global and Low National Identity (LGLN):Subjects low on both, global and national identity can beclassified under the ‘individuals’ element of Robertson’s(1997) framework, which refers to a condition of extremeparticularism. Subjects in this segment have no clearaffinity to any particular cultural values or concepts.

High Global and High National Identity (HGHN):Based on Robertson’s (1997) global processes frame-work, individuals high on both global and national iden-tity can be classified under the element “world system ofsocieties.” Such individuals not only incline toward uni-versalism but also acknowledge the world being diversi-

fied in cultural values and feel belongingness to bothorientations of global and national cultural identity. Thus,while having a vision and acceptance of universalism,they also acknowledge and adhere to the particularisminherent at the national cultural level.

High Global and Low National Identity (HGLN):According to Robertson’s (1997) framework the “human-kind” element refers to a universalistic vision of the worldand global human values. Consumers in this segment bestrepresent global consumer culture which in essence isakin to “universalism,” which captures the concept ofglobal consciousness or the awareness of the compressionof the world and the intensification of global interdepen-dence (Friedman 1994; Robertson 1992).

Consumer samples from the United States and Chinaare collected to assess the presence of the four segments.Global and national identities are measured by 7-pointscales from Der-Karabetian and Ruiz (1997). We use acluster analysis to see if respondents fall into differentgroups based on their global and national identity scores.Three of the four segments – HGHN, LGHN, and LGLNare clearly present. The global and national identities ofthe remain cluster, however, are in the mid-range of the 7-point scale, and are thus not conforming to the patternspecified earlier for a HGLN segment.

The findings of this study, collectively, provide strongevidence that consumers in today’s world possess com-plex identity orientations. The first finding, that globalidentity and national identity are not negatively corre-lated, both in the combined sample and the three indi-vidual samples, indicates that we should not view today’sconsumers through simplified lenses. In other words,being high on the former does not lead to being low on thelatter. Global identity and national identity do not neces-sarily move in different directions. References are avail-able upon request.

For further information contact:Jun Yu

Emporia State University1200 Commercial St.Emporia, KS 66801

Phone: 620.341.5784 ♦ Fax: 620.341.6345E-Mail: [email protected]


THE EFFECT OF BLOG INTERACTIVITY AND PERCEIVED TRUST ONVISITOR RESPONSE: THE MODERATING ROLE OF BLOGGER

EXPERTISE AND CONSUMER INVOLVEMENT

Yueming Zou, Old Dominion University, NorfolkKira Karande, Old Dominion University, Norfolk

SUMMARY

As the Internet has exploded, the web has createdopportunities for electronic word-of-mouth (eWOM) com-munication through electronic media, such as online dis-cussion forums, bulletin board systems, blogs, and socialnetworking sites (Goldsmith and Wu 2006). According tothe “state of blogosphere 2011” report, bloggers areincreasingly having influence over readers and otherbloggers (emarketer.com 2011). These statistics indicatethat the manner in which the blogger communicatesdetermines the extent to which their point of view andrecommendations are accepted. Thus, blog interactivitypotentially impacts the influence of blogger opinions andrecommendations on the consumers’ decision process.Our study focuses on the effect of blog interactivity onconsumer’s response in terms of attitude toward the blog(Att

blog), attitude toward the brand (Att

brand), and purchase

intention (PI). It identifies the process by which bloginteractivity impacts consumer response by identifyingtrust in the blog as the mediating variable. Moreover, italso proposes that the effects of blog interactivity on trust,and trust on consumer response both depend upon bloggerand consumer characteristics, specifically blogger exper-tise and consumer involvement.

A highly interactive online experience respondsquickly to visitors’ actions and requests, treats visitors asactive participants in the communication, and ensures thattheir questions are answered in a timely manner. Thisexperience reduces the frustration associated with waitingand feeling ignored and manipulated by the blogger, andpotentially results in a more satisfactory communicationexperience. Trust reduces uncertainty by ruling out unde-sirable future actions of other parties. In the online envi-ronment, trust is an important mechanism for reducingcustomers’ uncertainty, and, therefore, influences theirtransaction intentions. Thus, we hypothesize that bloginteractivity has positive impact on consumers’ response.Also, consumers’ trust in the blog mediates the relation-ship between interactivity and consumers’ responses.

Bloggers who are seen as having more expertise as aresult of their education, background, and experience arelikely to be seen as more trustworthy than those who donot have the relevant background. The beneficialeffect of blog interactivity on perceived trust is, there-

fore, enhanced by the expertise of the blogger. Highlyinvolved consumers tend to process product related infor-mation a lot more than less involved consumers. As aresult, they like to consider many sources of informationand different types of information in their decision mak-ing process. Thus, the involved consumers are likely to bemore demanding, and find the blogger to be trustworthywhen the blog is interactive. Thus, we hypothesize thatblogger expertise and consumers’ involvement will mod-erate the relationship between blog interactivity and con-sumers’ trust, and also moderate the relationship betweenconsumers’ trust and consumers’ response.

To test these hypotheses, we gathered data using ascenario based experimental design. A 2 (interactivity:high vs. low) × 2 (expertise: expert vs. non-expert) × 2(involvement: high vs. low) between subjects factorialexperiment was used, with interactivity and perceivedblogger expertise being manipulated and consumerinvolvement being measured. In the scenario, the respon-dents were asked to assume that they want to buy a newmusic player and that they are searching online for infor-mation and others’ suggestions. They visit several blogsto find information that will help them select the rightbrand. One of the blogs they visit is Sarah Johnson’s,which they are familiar with.

The mediating role played by trust in the effect ofinteractivity on PI, Att

brand, Att

blog was tested using proce-

dures recommended by Baron and Kenny (1986). Theeffects of blog interactivity on PI and Att

brand are both

completely mediated by consumers’ trust. The effect ofblog interactivity on Att

blog is partially mediated by con-

sumers’ trust.

We used ANOVA to test the main effect of bloginteractivity and the moderating effects of blogger exper-tise and involvement. In evaluating the moderating effectsof blogger expertise, we find that the effect of the interac-tion between blog interactivity and expertise on consum-ers’ trust is significant. Even though the effect of theinteraction between consumers’ trust and expertise on PIand consumers’ Att

blog are not significant, consumers’

Attbrand

is influenced by consumers’ trust to a greater extentfor experts than for non-experts. With regard to themoderating effect of involvement, the interactionbetween blog interactivity and consumers’ involvementon their trust is significant. Neither the effects of interac-


tion between consumers’ trust and involvement on PI andAtt

brand are significant. Only consumers’ trust interacts

with their involvement significantly to influence consum-ers’ Att

blog.

The present research provides some interesting theo-retical findings. First, trust in the blogger is central to theeffectiveness of a blog. Second, the perceived bloggerexpertise and consumer involvement are factors that mod-erate this process. The findings of this research indicate

that making a blog interactive increases the trust thatvisitors will have in the blogger, which improves theattitude in the brand being recommended by the blogger.Compared to blogs by experts, this effect is stronger whenthe blogger is perceived to have less expertise. Also,involved consumers reward interactivity more than lessinvolved consumers. As involved consumers likely use avariety of information sources, it does not change theirAtt

brand and PI; however, it improves the Att

blog, which is

likely to result in repeat visits.

For further information contact:Yueming Zou

College of Business and Public AdministrationOld Dominion University

Norfolk, VA 23508Phone: 530.304.3257



HOW QUALITY OF LIFE AFFECTS SOCIAL NETWORKING SITE USEINTENTION: ROLE OF PERCEIVED SELF DISCLOSURE

AND SOCIAL TIE QUALITY

Chung-Chi Shen, National Chiayi University, TaiwanJyh-Shen Chiou, National Chengchi University, Taiwan

Chun-Ming Yang, Ming Chuan University, TaiwanSzu-Yu Chou, National Chengchi University, Taiwan

SUMMARY

Online social interaction has become the primary useof home computers upon searching in terms of using time(Hamburger and Artzi 2003; Morahan-Martin andSchumacher 2003). In the midst of all this social activity,people are managing relationships on the Internet withthose whom they originally met in real life. Internet-basedcommunity has become important medium changingpeople’s everyday life. In 2008, Facebook started itsservice in Taiwan. In December 2009, just one year afterFacebook debuted in Taiwan, its registered users in Tai-wan have surpassed Taiwan’s leading social networkingsite Wretch (insightxplorer.com, 2009), with 9,932,740Facebook users on June 30, 2011, penetration is about43.1 percent (Internet World Stats 2011).

Our research investigates what factors affecting con-sumers’ acceptance of social networking site (SNS) andthe impact of perceived self disclosure and social ties ontheir attitude and intention toward using the social net-working service. Our research model integrates con-structs from social psychology, technology acceptancemodel and social capital theory. Tie strength is presumedhas moderating effect on user’s attitude toward usingonline social networking site. The scales were eitherbased on previous research, or developed according toChurchill’s (1979) recommendations. Seven-point scales,from “strongly disagree” to “strongly agree,” were used

throughout the questionnaire, except life dissatisfac-tion will be measured by using scales from terrible todelighted. To ensure the weblog and community thatrespondents’ thinking are appropriate for the study con-text while answering each group of items, we solicitrespondents to recall the microblogging site that they mostuse in the beginning of the questionnaire. The model wastested using the two-step structural equation procedure(Anderson and Gerbing 1988). First, a ConfirmatoryFactor Analysis (CFA) was employed to evaluate con-struct validity in the measurement model. A full SEMmodel was performed to test the hypothesized relation-ships after CFA obtains an appropriate model fit thatindicating the model has good validity. Respondents weredivided into either strong or weak group by the scores ofsocial tie quality scales.

Results show that perceived self disclosure and pres-sure reducing play a mediating role in the relationshipfrom loneliness and life dissatisfaction to attitude towardusing SNS, which in turn affects intention to continuallyuse SNS. Social tie strength has moderating role in deter-mining whether users use SNS when they are willing totell other about their mind via SNS. Finally, the articleconcludes with implications for web-based service pro-viders. Future research suggestions relevant to the onlinesocial capital and consumer loyalty are also provided.References are available upon request.

For further information contact:Chung-Chi Shen

Graduate Institute of Marketing and LogisticsNational Chiayi University

580, Sinmin Rd.Chiayi City 60054

TaiwanPhone: 886.9.31749150



FACTORS AFFECTING ONLINE TRUST IN ONLINE SHOPPING: THEROLE OF NETWORK EXTERNALITY AND INTERNET SKILL

Kazuhiro Kishiya, Kansai University, JapanNao Yamamoto, Nagoya City University, Japan

ABSTRACT

Recent research has focused on online trust. Still, fewresearchers have applied the “network effect” that existsin online shopping. Furthermore, the perception of onlinetrust and network externality depends on Internet ability.The authors show the relationship between Internet skill,network externality and online trust in online shopping.

INTRODUCTION

In this paper, we examine the relationship betweennetwork externality, online trust and Internet skill inonline shopping. First, we focus on online trust in shop-ping sites, which influences the likelihood of purchase. Inonline shopping, the consumer perceives the various risksfairly. The perception of these risks is likely to decreasethrough online trust on a shopping site. However, unlikeoffline trust, online trust itself is less common and not asobvious. We propose that trustworthiness online is basedon perception of network externality.

Second, we extend network externality to researchthe field of online shopping. Little research has beenconducted on the relationship between network external-ity and online trust. Due to uncertain information onshopping sites, the source of “word of mouth” recom-mendations is uncertain and often lacking in credibil-ity. Instead of the credibility of each recommendation,we indicate that network externality, which is caused byan effective and broad range of information from a largenumber of participants, affects online trust, and accord-ingly, purchase intention.

Third, Internet skill also influences the perception ofnetwork externality and online trust, playing an importantrole in users’ attitudes. There is a large amount of uncer-tain information on the web. Without Internet skill, con-sumers cannot search for the information they require andconfirm the authenticity – necessary to maintain their trustin shopping sites – or perceive that the variety and numberof comments shown are useful as indicative of networkeffects.

LITERATURE REVIEW

Online Shopping and Online Trust

Online shopping is likely to be risky, compared withtraditional store shopping (Forsythe et al. 2006).

First, greater information asymmetry between sellerand buyer through spatial and temporal separation leads tothe perception of higher risk in consumers (Pavlou 2003;Zhou, Dai, and Zhang 2007). The lack of deeper productinformation in purchases and the possibility of the productnot being delivered causes the buyer to perceive productand financial risk.

Second, problems associated with security and pri-vacy are also likely to increase the perception of risk(Belanger, Hiller, and Smith 2002; Yoon 2002). Anxietyabout exposure and misuse of personal information inhib-its consumers from making online shopping purchases.The open nature of online transactions globally createsuncertainty for consumers (Pavlou 2003). Therefore therole of trust, which reduces the perception of risk inpurchases, is a crucial factor in e-commerce. Thus, recentstudies have been investigating the issue of online trust(Hoffman, Novak, and Perlta 1999; Yoon 2002; Belanger,Hiller, and Smith 2002; Anderson and Srinivasan 2003;Bart et al. 2005; Ling, Chai, and Piew 2010).

As for trust itself, Morgan and Hunt (1994) definedtrust as confidence in the exchange partner’s reliabilityand integrity. Doney and Cannon (1997) defined trust asthe perceived credibility and benevolence of a target.Requiring integrity and benevolence, there is vulnerabil-ity and uncertainty in exchange and transactions. Trustcan decrease the various risks regarding transactions(Coleman 1990; Moorman, Deshpandé, and Zaltman1993). More so than offline transactions, e-commercetransactions are accompanied by uncertainty and vulner-ability due to the spatial and temporal separation andopenness that are inherent characteristics of e-commerce.Online trust reduces consumer uncertainty significantly(Yoon 2002; Pavlou 2003; Chen, Griffith, and Shen 2005;Ling, Chai, and Piew 2010).

In addition, online trust is necessary to reduce theeffort to find additional information. If consumers con-sider a shopping site to be trustworthy, they may stick tothat site. Thus, online trust reduces the demand for newinformation (Yoon 2002).

Network Externality As a Source of Trust

Network externality is present when the value of anetwork increases as the number of users of the networkincreases. For example, the value of a telephone network


depends on the number of people with access to thatnetwork (Katz and Shapiro 1985; Shanker and Bayus2004; Song, Parry, and Kawakami 2009). Examples ofproduct success are Microsoft’s MS-DOS operating sys-tem and the QWERTY keyboard. Network externalitytheory has been used to explain technology adoptiondecisions (Song and Walden 2007). That concept hasbeen applied to research on digital products, includingVCRs (e.g., Redmond 1991), home video games (e.g.,Shankar and Bayus 2002), and MD players (e.g., Song,Parry, and Kawakami 2009). Also, network externality isapplied to research on information services (e.g., Songand Walden 2007). We will examine the importance ofnetwork externality vis-à-vis online shopping sites. Thereasoning is as follows:

First, network externality works with social interac-tion, including WOM and observational learning (Chen,Wang, and Xie 2011). Through WOM recommendations,people acquire the necessary knowledge for decisionmaking (Redmond 1991). In addition, network external-ity is effective in virtual communities through electronicWOM (Balasubramanian and Mahajan 2001; Shankerand Bayus 2004). On online shopping sites, there areusually a huge number of user comments, which mightinfluence other consumers’ purchase choices. Consumersmight assign importance to the variety and range of usercomments, because that enables them to acquire variousand detailed information.

Moreover, online shopping sites report data on thepurchase actions of others, such as sales rankings, tofacilitate consumer observational learning. Amazon.comrecommends products bought by other customers withsimilar purchase records. Online shopping sites thus helpconsumers make purchases by displaying a variety ofconsumer data in a timely manner. If the data is based ona large number of participants, the data is perceived to bemore accurate and user trust is increased.

Second, network externality includes a psychologi-cal effect. Consumers speculate about product and servicequality from the number of adopters. Consumers infer thatproducts and services that are bought by a large number ofconsumers are of reliable quality. As a result, they are ableto reduce some risk in making the purchase (Padmanabhan,Rajiv, and Srinivansan 1997; Song, Parry, and Kawakami2009). In online shopping, in which the consumer hasdifficulty obtaining exact and certain information com-pared with store shopping, network externality might bea powerful tool with which to increase trustworthiness.We will examine the relationship between network exter-nality and online trust.

Internet Ability

Ability is one of the factors that determines the degreeto which individuals process information (MacInnis and

Jaworski 1989; Gruen, Osmonbekov, and Czaplewski2006). In particular, the Internet leads to informationasymmetry between individuals. On the web, it is neces-sary for consumers to have some skills such as being ableto search and make decisions on whether the informationthey find is correct. Without such skills, an individualmight not acquire the information they require, despitehaving searched for it.

In addition, Internet ability might influence the per-ception of the user (Hoffman and Novak 1996; Novak,Hoffman, and Yung 2000). Zhou and Bao (2002) indicatethat people with high level skill extract important infor-mation from web advertising. The perception of users ininterpreting web pages varies even when exposed to thesame page, and depends on their Internet ability. Thus,Internet ability facilitates the finding of information andthe acquisition of important and appropriate information.We will examine Internet ability and how it relates toonline trust and the perception of network externality.

THEORETICAL MODEL AND HYPOTHESES

In this section, we describe our research hypotheses,which concern the relationships between online trust,network externality and Internet ability.

As for online shopping sites, user comments whichshow their trial and usage of products and services helpconsumers to make purchases. In other words, consumerscan reduce their uncertainty by referring to eWOM infor-mation (Park and Lee 2009). In an online community,those with an interest in specific brands and productcategories share their knowledge regarding productadvice and usage (Kozinet 1999; Brown, Broderick,and Lee 2007). In particular, when consumer judgmentalcriteria are ambiguous, WOM affects consumer choice(Bone 1995). As opposed to conventional word of mouth,eWOM lacks vivid, deeper information and the opportu-nity for interaction. On the other hand, there is variousinformation in eWOM. In addition, online consumers canfreely select eWOM based on their needs (Chu and Kim2011). Therefore, they seek shopping sites featuring hugeamounts of, and diverse, eWOM, from among which theyare able to find eWOM comments to match their needs.

Moreover, a greater number of participants result ina more accurate database for promoting observationallearning. A consumer also might seek product informa-tion and make a purchase referring to the sales ranking andrecommendations of the online shopping site, especiallyto reduce search effort. The accuracy or degree of perfec-tion, which might influence consumer trust, depends onthe number of participants. Thus, online trust might beinfluenced by the quantity of information gathered duringthe online experience (Smith, Menon, and Sivakumar2005). The perception of the network externality effect –


which is led by the existence of a large amount of recom-mendable eWOM and observational learning – mightincrease trust in the shopping site itself. In addition, if auser observes a large number of participants, the user’sperception of the trustworthiness of the site is likely toincrease, thereby relieving the user’s anxiety, and increas-ing the likelihood of purchase based on trust.

H1: Perception of network externality is positively asso-ciated with online trust.

H2: Perception of network externality is positively asso-ciated with purchase intention.

H3: Perception of trust is positively associated with onlinepurchase intention.

In the e-commerce context, consumers need to makethe decision to buy a product or service from a hugeamount of information and comments, in a state of uncer-tainty. Those with high level skill are probably familiarwith the Internet. They are able to find out specific andrequired information, confirm the authenticity and refer tothe online shopping data effectively, while those with lowlevel skill may be unfamiliar with the Internet and as aresult may be unable to use the Internet effectively (Gruen,Osmonbekov, and Czaplewski 2006). Furthermore, whenusers conduct beyond their skill, they are in a state ofanxiety (Mathwick and Rigdon 2004), led to losing onlinetrust. Thus, their trust in online shopping is not main-tained.

Moreover, the perception of network externality mightbe influenced by Internet skill. Through Internet skill,

FIGURE 1Overall Model

consumers can recognize and infer diverse and broadranges of participants in online shopping. They recognizethe benefits deriving from the network effect. On the otherhand, those with low level skill cannot find out therequired information and grasp the exact meaning fromeWOM, and misuse marketing tools such as observationallearning. Therefore, they do not recognize the networkeffect.

H4: Internet Skill is positively associated with the percep-tion of online trust.

H5: Internet Skill is positively associated with the percep-tion of network externality.

Scale Development and Data Collection

To test our hypotheses, we asked which shoppingsites respondents used the most frequently. Regardingthe sites they selected, respondents were asked to giveanswers regarding Network Externality, Online Trustand their Purchase Intentions. The concept and measuresof this study are reported in the Appendix. A translation/back-translation process was used to ensure equivalencein the meaning of all parts of the Japanese and Englishversions of the questions (Douglas and Craig 1983).

Questions and five-point rating scales (Strongly Dis-agree (1)-Disagree-Neutral-Agree-Strongly Agree (5))were selected, as in previous studies. To measure onlinetrust, four items were adopted from Anderson andSrinivasan (2003). To measure perceived network exter-nality, three items were adopted from Sony and Walden

OnlineTrust

InternetSkills

PurchaseIntention

NetworkExternality

+

+

+

+

+

Age Gender


(2007). To measure purchase intention, three items wereadopted from Hausman and Siekpe (2008). And finally, tomeasure Internet skill, four items were adopted fromNovak, Hoffman, and Yung (2000). Measurements ofperceived network externality were adjusted in the con-text of online shopping.

We included the control variables of Age and Genderfor purchase intention in our analysis. Gender was used asa dummy variable in which “male” was assigned the valueof 1 and “female” the value of 2. Only gender waspositively significantly related to purchase intention. Theonline questionnaire survey was conducted in March2008 with the help of a marketing research company. Theage and gender of the participants are summarized inTable 1. The size of the sample was 1191, comprising 599males and 592 females. The daily average viewing time ofparticipants in the week prior to the survey was 4.7 hours.The number of online purchases per month was 5.44.

Measurement Model

We evaluated our measurement model following thetwo-step approach recommended by Anderson and

Gerbing (1988). The result of a confirmatory factor analy-sis yielded a satisfactory result (Chi-square = 427.829,d.f. = 71 p > 0.001, CFI = 0.96 NFI = 0.96, IFI = 0.96,ROSEA = 0.06). All loading items were found to be morethan 0.6. Composite reliabilities for the items of eachconstruct ranged from 0.80 to 0.95. We also reported thesquare root of the Average Variance Explained (AVE). Toevaluate discriminant validity, the AVE of each constructis required to be greater than 0.50 and the construct’slargest correlation with other constructs (Fornell andLarcker 1981). The AVE satisfies those criteria. Table 2shows the correlation, composite reliabilities and theAVE.

HYPOTHESES TESTING

We used structural equation modeling to test ourhypotheses (Anderson and Gerbin 1988). The results aresummarized in Table 3. The fit statistics indicated that ourmodel fit the data well (Chi-square = 608.973, d.f. = 99p > 0.001, CFI = 0.95, IFI = 0.955 ROSEA = 0.06). Withregard to our hypotheses, the perception of NetworkExternality was positively and significantly related toOnline Trust (β = 0.57 p > 0.001) and Purchase Intention(β = 0.22 p > 0.001). Online Trust was positively andsignificantly related to Purchase Intention (β = 0.58 p >0.001). H1, H2, H3, are thus supported.

With regard to the Internet Skill variable hypotheses,Internet Skill was positively and significantly related toperception of Network Externality (β = 0.34 p > 0.001)and Online Trust (β = 0.09 p > 0.001). These resultsprovide support for H4.

DISCUSSION

We examined the impact of network externality ononline trust and purchase intention, and Internet skill on

TABLE 2Correlation, Cronbach’s Alpha, Composite Reliabilities, and the Average Variance Explained

Mean S.D. a b c d ααααα CR AVE

Perceived N. E. (a) 3.81 0.67 0.75 0.82 0.74

Online Trust(b) 3.95 0.65 0.20* 0.91 0.92 0.86

Purchase Intention(c) 4.02 0.73 0.20* 0.25* 0.91 0.94 0.89

Internet Skill(d) 3.61 0.76 0.09* 0.08* 0.07* 0.84 0.87 0.79 ***

p > 0.001

TABLE 1Sample of Age and Gender

Male Female

10s 94 9520s 100 10130s 100 10140s 103 10250s 100 9860s 102 95


the perception of network externality and online trust. Thefindings extend the research area of online trust andnetwork externality in several ways.

First, online trust is influenced by network external-ity. So far, studies about online trust have simply focusedon relationships regarding the various risks and experi-ences in online shopping (Zhou, Dai, and Zhang 2007).However, online trust works with network externalities.The results are determined by the characteristics of eWOM,which is shallow and less certain due to the lack ofobservable and contextual information. Instead, a con-sumer might make a decision from a broad range ofinformation and an inference that there are a large numberof participants. This is supported by the example of thelong tail phenomenon. Sites with large numbers of partici-pants are successful at competing.

Second, perceptions of network externality and onlinetrust are mediated by Internet skill. That fact causesresearchers to examine context of trust in their studies. Sofar studies conclude that integrity and benevolence arerequisite for maintaining trust. That presumes the charac-teristics of the relationship are stable and the possibility ofmonitoring partner’s intention and behavior exist. Incontrast, a relationship on the web is unstable and infor-mation brought from that relationship is shallow anduncertain. Therefore, Internet skill that includes the abil-ity to search and make accurate evaluations is necessary.

LIMITATIONS AND DIRECTION OFFUTURE RESEARCH

Little is known about the relationship between net-work externality and online trust. We examined thisrelationship in our research; however, our model is beingdeveloped and should be tested in future research.

First, we posit that the eWOM effect and observa-tional learning mediate the network externality variableand online trust variable, followed by purchase intention.We did not measure these variables. The eWOM andobservational learning should be identified in future re-search. We will examine the relationships between theeWOM, observational learning and network externalityvariables.

Second, online trust will be examined theoreticallyand empirically. For instance, strength of the relationshipof Internet skill and online trust might be dependent on thediffusion of Internet usage and Internet shopping.

Third, as we collect data in Japan, we can consider therelationship between the perceptions of network external-ity and online trust as reflecting the uniqueness of Japa-nese culture. Japan is regarded as a collective culture(Triandis 1995), And thus cultural variables might influ-ence this relationship. In the future, to extend that model,the analysis of national comparisons should be researched.

TABLE 3Results of Structural Equation Model

Hypotheses From To Standardized Estimate

H1 Perceived N.E. Online Trust 0.57***

H2 Perceived N.E. Purchase Intention 0.22***

H3 Online Trust Purchase Intention 0.58***

H4 Internet Skill Network Externality 0.34***

H5 Internet Skill Online Trust 0.09

Chi-square = 493.480, d.f. = 85 p > 0.001, CFI = 0.96, IFI = 0.96, RESEA = 0.06***p > 0.001


ENDNOTE

The authors acknowledge financial support by MEXTKAKENHI Grant-in-aid for Scientific Research (B)#21730351 (2011–2014) of Japan Ministry of Education,Culture, Sports, Science, and Technology, Strategic Projectto Support the Formation of Research Bases at PrivateUniversities Matching Fund Subsidy from MEXT (2009–2013) and JOHOKU SENKOU CO., LTD.

REFERENCES

Anderson, James C. and David W. Gerbing (1988), “Struc-tural Equation Modeling in Practice: A Review andRecommended 2-Step Approach,” PsychologicalBulletin, 103 (3), 411–23.

Anderson, Rolph E. and Srini S. Srinivasan (2003), “E-Satisfaction and E-Loyalty: A Contingency Frame-work,” Psychology & Marketing, 20 (2), 123–38.

Balasubramanian, Sridhar and Vijay Mahajan (2001),“The Economic Leverage of the Virtual Commu-nity,” International Journal of Electronic Commerce,5 (3), 103–38.

Bart, Yakov, Venkatesh Shankar, Fareena Sultan, andGlen L. Urban (2005), “Are the Drivers and Role ofOnline Trust the Same for All Web Sites and Con-sumers? A Large-Scale Exploratory Empirical Study,”Journal of Marketing, 69 (4), 133–52.

Belanger, France, Janine S. Hiller, and Wanda J. Smith(2002), “Trustworthiness in Electronic Commerce:The Role of Privacy, Security, and Site Attributes,”Journal of Strategic Information Systems, 11 (3/4),245–70.

Bone, Paula F. (1995), “Word-of-Mouth Effects on Short-Term and Long-Term Product Judgement,” Journalof Business Research, 32 (3), 213–23.

Brown, Jo, Amanda J. Broderick, and Nick Lee (2007),“Word of Mouth Communication Within OnlineCommunities: Conceptualizing the Online SocialNetwork,” Journal of Interactive Marketing, 21 (3),2–20.

Chen, Yubo, Qi Wang, and Jinhong Xie (2011), “OnlineSocial Interactions: A Natural Experiment on Wordof Mouth Versus Observational Learning,” Journalof Marketing Research, 48 (2), 238–54.

Chen, Qimei, David A. Griffith, and Fuyuan Shen (2005),“The Effects of Interactivity on Cross-Channel Com-munication Effectiveness,” Journal of InteractiveAdvertising, 5 (2), 19–28.

Chu, Shu-Chuan and Yoojung Kim (2011), “Determi-nants of Consumer Engagement in Electronic Word-of-mouth (EWOM) in Social Networking Sites,”International Journal of Advertising, 30 (1), 47–75.

Coleman, James (1990), Foundation of Social Theory.Cambridge, MA: Belknap Press of Harvard Univer-sity Press.

Doney, Patricia M. and Joseph P. Cannon (1997), “AnExamination of the Nature of Trust in Buyer-SellerRelationships,” Journal of Marketing, 61 (2), 35–51.

Douglas, Susan P. and C. Samuel Craig (1983), Interna-tional Marketing Research. Englewood Cliffs, NJ:Prentice-Hall, Inc.

Fornell, Claes and David F. Larcker (1981), “EvaluatingStructural Equation Models with Unobservable Vari-ables and Measurement Error,” Journal of MarketingResearch, 18 (1), 39–50.

Forsythe, Sandra, Chuanlan Liu, David Shannon, and LiuC. Gardner (2006), “Development of a Scale toMeasure the Perceived Benefit and Risks of OnlineShopping,” Journal of Interactive Marketing, 20 (2),55–75.

Gruen, Thomas W., Talai Osmonbekov, and Andrew J.Czaplewski (2006), “eWOM: The Impact of Cus-tomer-to-Customer Online Know-How Exchange onCustomer Value and Loyalty,” Journal of BusinessResearch, 59 (4), 449–56.

Hausman, Angela V. and Jeffrey S. Siekpe (2009), “TheEffect of Web Interface Features on Consumer OnlinePurchase Intentions,” Journal of Business Research,62 (1), 5–13.

Hoffman, Donna L. and Thomas P. Novak (1996), “Mar-keting in Hypermedia Computer-Mediated Environ-ments: Conceptual Foundations,” Journal of Market-ing, 60 (3), 50–68.

____________, ____________, and Marcos Peralta(1999), “Building Consumer Trust Online,” Commu-nications of the ACM, 42 (4), 80–85.

Katz, Michael L. and Carl Shapiro (1985), “NetworkExternalities, Competition, and Compatibility,”American Economic Review, 75 (3), 424–40.

Kozinets, Rovert V. (1999), “E-Tribalized Marketing?The Strategic Implications of Virtual Communitiesof Consumption,” European Management Journal, 7(3), 252–64.

Ling, Kwek C., Lau T. Chai, and Tan H. Piew (2010),“The Effects of Shopping Orientations, Trust andPrior Online Purchase Experience Toward Custom-ers’ Online Purchase Intention,” International Busi-ness Research, 3 (3), 63–76.

Macinnis, Deborah J. and Benard J. Jaworski (1989),“Infromation-Processing From Advertisement – To-ward an Intergrative Framework,” Journal of Mar-keting, 53 (4), 1–23.

Mathwick, Charla and Edward Rigdon (2004), “Play,Flow, and the Online Search Experience,” Journal ofConsumer Research, 31 (2), 324–32.

Moorman, Chistinem, Rohit Deshpande, and GeraldZaltman (1993), “Factors Affecting Trust in Market-Research Relationship,” Journal of Marketing, 57(1), 81–101.

Morgan, Robert M. and Shelby D. Hunt (1994), “TheCommitment-Trust Theory of Relationship Market-ing,” Journal of Marketing, 58 (3), 20–38.


Novak, Thomas P., Donna L. Hoffman, and Yiu F. Yung(2000), “Measuring the Customer Experience inOnline Environments: A Structural Modeling Ap-proach,” Marketing Science, 19 (1), 22–42.

Padmanabhan, V., Surendra Rajiv, and Kannan Srinivasan(1997), “New Products, Upgrades, and New Re-leases: A Rationale for Sequential Product Introduc-tion,” Journal of Marketing Research, 34 (4), 456–72.

Park, Cheol and Thae M. Lee (2009), “Information Direc-tion, Website Reputation and eWOM Effect: A Mod-erating Role of Product Type,” Journal of BusinessResearch, 62 (1), 61–67.

Pavlou, Paul A. (2003), “Consumer Acceptance of Elec-tronic Commerce: Integrating Trust and Risk with theTechnology Acceptance Model,” International Jour-nal of Electronic Commerce, 7 (3), 101–34.

Redmond, William H. (1991), “When Technolgies Com-pete: The Role of Externalities in Nonlinear MarketResponse,” Journal of Product Innovation Manage-ment, 8 (3), 170–83.

Shankar, Venkatesh and Barry L. Bayus (2003), “Net-work Effects and Competition: An Empirical Analy-sis of the Home Video Game Industry,” StrategicManagement Journal, 24 (4), 375–84.

Song, Jaeki and Eric Walden (2007), “How Consumer

Perceptions of Network Size and Social InteractionsInfluence the Intention to Adopt Peer-to-Peer Tech-nologies,” International Journal E-Business Re-search, 3 (4), 49–66.

Song, Micheal, Mark E. Parry, and Tomoko Kawakami(2009), “Incorporating Network Externalities intothe Technology Acceptance Model,” Journal of Prod-uct Innovation Management, 26 (3), 291–307.

Smith, Donnaieve, Satya Menon, and K. Sivakumar (2005),“Online Peer and Editorial Recommendations, Trust,and Choice in Virtual Markets,” Journal of Interac-tive Marketing, 19 (3), 15–37.

Triandis, Harry C. (1995), Individualism & Collectivism.Boulder, CO: Westview Press.

Yoon, S. Joon (2002), “The Antecedents and Conse-quences of Trust in Online-Purchase Decision,” Jour-nal of Interactive Marketing, 16 (2), 47–63.

Zhou, Lina, Liwei Dai, and Dongsong Zhang (2007),“Online Shopping Acceptance Model: A CriticalSurvey of Consumer Factors in Online Shopping,”Journal of Electronic Commerce Research, 8 (1),41–62.

Zhou, Zheng and Yeqing Q. Bao (2002), “Users’ Atti-tudes Toward Web Advertising: Effects of InternetMotivation and Internet Ability,” Advances in Con-sumer Research, Xxix, 29, 71–78.

APPENDIX

Construct Measure Source

Perceived N.E. I believe that I could rely on this online shopping Song and Waldensite to reduce time and effort for information (2007)because of many usersI believe that this online shopping site will bedominant among all the shopping sitesI believe that I could benefit by this onlineshopping site

Online Trust The performance of this shopping site meets my Anderson andexpectations Srinivasan (2003)This online shopping site can be counted on tosuccessfully complete the transactionI can trust the performance of this online shoppingsite to be goodThis online shopping site is reliable for onlineshopping

Purchase Intention I will definitely buy products from this online Hausman and Siekpeshopping site in the near future (2008)I intend to purchase through this onlineshopping site in the near futureIt is likely that I will purchase through thisonline shopping site in the near future


APPENDIX (CONTINUED)

Construct Measure Source

Internet Skill I am extremely skilled at using the Web Novak, Hoffman andI consider myself knowledgeable about good Yung (2000)search techniques on the WebI know somewhat less than most users aboutusing the Web ( R )I know how to find what I am looking for onthe web

For further information contact:Kazuhiro Kishiya

Faculty of CommerceKansai University

Suita Osaka 564–8680Japan

Phone: (06)63680659Fax: (06)63680659



EFFECTS OF PERSONALIZED E-MAIL MESSAGES ON PERCEIVEDRISK: MODERATING ROLES OF CONTROL AND INTIMACY

Sung-Won Lee, University of Seoul, South KoreaJi Hee Song, University of Seoul, South Korea

Hye Young Kim, University of Seoul, South Korea

SUMMARY

Technological advances have dramatically trans-formed the way people communicate. For example, theInternet and mobile devices have become an increasinglyimportant part of communication between firms and theircustomers. New communication platforms have enabledfirms to provide their customers with more personalizedservices by allowing for the collection of personal data ontheir customers. However, some researchers have notedthat such personalized services may entail an invasion ofprivacy. Here, we examined the effects of personalizationin computer-mediated communication on risk percep-tions. In particular, we investigated the moderating rolesof control and message intimacy in the relationshipbetween personalization and risk perceptions.

Individuals’ perceived risk varies according to thelevel of personalization. An increase in the level of per-sonalization may reduce consumers’ risk perceptions, andwhen the level of personalization exceeds the “optimum”level, consumers may be uncertain about personalizedservices (i.e., an increase in the level of perceived risk).The present study considers three levels of personaliza-tion (low, moderate, and high) and assumes that consum-ers are most certain when they receive moderately person-alized services. In this regard, we propose the followinghypothesis:

H1: The relationship between the level of personalization(from a low to a high level) and perceived risk isquadratic, that is, a U-shaped relationship.

According to cognitive control theory, under stress-ful situations, consumers may perceive their circum-stances to be more controllable when they receive moreinformation on their situation. That is, consumers reevalu-ate stressful situations when they have more information.Therefore, when consumers are exposed to some riskthrough highly personalized messages, their level of riskmay be reduced if they have more information on (i.e.,more control over) the use of their personal information.In this regard, we propose the following hypothesis:

H2: The effect of a personalized e-mail on perceived riskis moderated by control over personal information.

That is, when consumers have the ability to controlpersonal information on their e-mails, there is nodifference in their risk perception between e-mailmessages reflecting a low level of personalizationand those reflecting a high level of personalization.

Another moderating variable that can reduce con-sumers’ risk perception concerning highly personalizedmessages is the use of intimacy in personalized services.A highly personalized message reflecting a high level ofintimacy (e.g., the use of a natural language, the identifi-cation of the sender’s personal information) may reduceconsumers’ perceived risk concerning the message. Bycontrast, when consumers receive a message reflecting alow level of personalization, intimacy may not have asignificant effect on their perceived risk. In this regard, wepropose the following hypothesis:

H3: The effect of a personalized e-mail message on per-ceived risk is moderated by the message’s level ofintimacy. That is, when consumers get e-mail mes-sages reflecting a high level of intimacy, there is nodifference in their risk perception between emailmessages reflecting a low level of personalizationand those reflecting a high level of personalization.

To test the proposed relationships, we consider an e-mail marketing campaign in which a bank sends e-mailsto their target customers to promote their financial prod-ucts. Three lab experiments were conducted. In the firstexperiment, we found that there was no significant differ-ence in perceived risk between minimally personalizede-mail messages and moderately personalized ones, butperceived risk increase from moderately personalizede-mail messages to highly personalized ones. Thus, theseresults provide support for H1 between moderate and highlevels of personalization but not between low and mediumlevels of personalization. The results of the second experi-ment indicate that control over personal informationreduced the negative effects of a high level of person-alization on perceived risk, supporting for H2. Third,intimacy in e-mail messages regulated the effects ofpersonalization on perceived risk, supporting for H3.

The results have important theoretical implications.We considered three levels of personalization, whichshould be useful for manipulating personalization. In


addition, we verified that optimal stimulation level theory,cognitive control theory, and social presence theory areuseful for examining computer-mediated communicationphenomena. The results have important practical implica-tions. If customers are given some control over their

personal information maintained by service providersand/or if personalized services are delivered in a moreintimate manner, then firms can maximize the effectsof personalized services. References are available uponrequest.

For further information contact:Ji Hee Song

University of SeoulSiripdae-gil 13, Dongdaemun-gu

Seoul, South Korea 130–743Phone: 82.2.2210.5741

Fax: 82.2.2246.0570E-Mail: [email protected]


LINKING THE VIRTUAL WORLD WITH THE REAL WORLD: HOWTO BUILD ONLINE RELATIONSHIPS THAT LEAD

TO OFFLINE INTERACTIONS

Sukanya Seshadria, University of Massachusetts, BostonWerner H. Kunz, University of Massachusetts, Boston

SUMMARY

Prior research has investigated the relationships be-tween users of networking platforms in the online world.This paper focuses on using online communities to linkthe virtual world to the real world and how online commu-nities serve the purpose to develop an offline relationshipto a stranger. We developed a framework based on cueutilization theory, social balance theory, and uncertainty

reduction theory and tested it in a field experiment withmembers of a travel community website. We show that thecommunity reputation, the online communication, andthe perceived similarity between the counters parts play asignificant role for a potential offline relationship. Fur-ther, trust and sympathy are important mediators for thisdecision process. Theoretical and managerial implica-tions are also discussed in the paper.

For further information contact:Werner Kunz

College of ManagementUniversity of Massachusetts, 100

Morrissey BoulevardBoston, MA 02125

Phone: +1.617.291.8736Fax: +1.617.287.7877



101 PEOPLE LIKE THIS: EVALUATING THE FACEBOOK MESSAGESTRATEGY EFFECTIVENESS OF FORTUNE 500 COMPANIES

Kunal Swani, University of Massachusetts, AmherstGeorge R. Milne, University of Massachusetts, Amherst

Brian P. Brown, Virginia Commonwealth University, Richmond

SUMMARY

Marketers allocate a substantial amount of theirresources to word-of-mouth (WOM) communication.A sizeable amount of this spending is being diverted tosocial networks and online communities due to theirincreasing popularity, reach and potential impact on mar-keting outcomes. WOM marketing is considered to be apowerful and effective marketing strategy that is replac-ing and augmenting traditional media communications.

Social media sites have accelerated WOM communi-cation via various social plugins (e.g., Likes, YouTube,Tweet). The Facebook Likes plugin is the most adoptedsocial plugin in the social media space. While researchershave considered the hardware (devices) and software(social plugins) technology advances that have acceler-ated the use of WOM on social sites, less attention hasbeen given to understanding the message strategies thatactually influence online WOM effectiveness.

Building on the network coproduction model, thisresearch attempts to address such gaps by exploring therelationship between message strategies and the “liking”of company Facebook communications (Facebookwallposts) so that marketers can determine how to com-municate and advertise effectively on social media sites.We refer to the act of liking content and subsequentlyspreading likes through customer networks as word oflikes (WOL). The network coproduction model takes intoaccount the new dynamics of online WOM marketing byrecognizing that marketers use tactics and metrics thattarget and influence consumers and opinion leaders. Fur-ther, it suggests that consumers are active co-producers ofvalue where communications and meanings are exchangedamong members of the consumer network.

In this research, we attempt to determine the messagestrategies most likely to promote liking of messages onFacebook and how these message strategies vary based onmarket and offering characteristics. Various markets (B2B/B2C) and offerings (product/service) use different brand-ing strategies and message appeals in their marketingcommunications. Accordingly, we classify Facebookaccounts as B2B or B2C accounts, and classify com-pany offerings as either product or service offerings.Three types of message strategies used in message posts

are evaluated: (1) the use of corporate brand names, (2) theuse of emotional appeals, and (3) the use of direct calls topurchase. Corporate brand names are present when asocial media message has a corporate brand name men-tioned in the message. Emotional appeals attempt to stirup either negative or positive emotions. Direct calls topurchase message strategies involve explicit statementsor cues that encourage immediate purchase (e.g., “sale,”“buy now”). We examine the effectiveness of these mes-sage strategies by analyzing Facebook messages postedby Fortune 500 Facebook accounts.

We hypothesize that the effectiveness of these mes-sage strategies differs by the market context (i.e., B2B andB2C) and whether a product or service offering is beingpromoted. Use of emotional appeals and direct calls topurchase in social media messages is likely to generatemore WOL for B2C Facebook accounts than for B2Baccounts. Furthermore, using of corporate brand names inmessages is likely to generate more WOL for B2B ac-counts than for B2C accounts. Service Facebook accountswhen using corporate brand names and emotional appealsin messages are likely to generate more WOL than prod-uct accounts.

Method and Results

We conducted a content analysis of 1,146 wallpostsfrom 195 Facebook accounts. To test our model, we useda random coefficient level 2 HLM model with messages(Level 1) nested in Facebook accounts (Level 2). Thedependent variable for our analysis was the number oflikes for a message. The total number of fans for theFacebook company page and message time, time whenthe message was sent out to the time when it was archived,was added as control variables. Four of our six hypotheseswere supported. More specifically, our findings suggestthat B2B Facebook accounts should include their corpo-rate brand names and avoid direct calls to purchasemessage strategies. Furthermore, results suggest that ser-vice Facebook accounts should use emotional appeals intheir messages.

General Discussion

Depending on whether an account is B2B/B2C orproduct/service, the effectiveness of Facebook message


strategies (in terms of increasing WOL) differ. Our analy-sis reveals that the use of a corporate brand name strategyeffectively promotes WOL in B2B Facebook companyaccounts compared with B2C company accounts. Fur-thermore, the use of direct calls to purchase in a messageis less effective in B2B Facebook company accounts thanin B2C company accounts. The use of emotional appealsin a message promoted more likes in service Facebookcompany accounts than in product company accounts.

This study contributes to the marketing communica-tions and WOM/online WOM literature in the followingways. First, it suggests factors that influence WOL strat-egies in the B2B and B2C environments and when mar-keting product and service offerings. Second, it exploresthe influence of new technology in the form of socialmedia sharing as it relates to online WOM and thereforeadvances our knowledge of online WOM. Third, it pre-sents a novel technique for observing online WOM activ-ity. References are available upon request.

For further information contact:Kunal Swani

Eugene M. Isenberg School of ManagementUniversity of Massachusetts Amherst

121 Presidents DriveAmherst, MA 01003Phone: 216.212.1851



DOES SOCIAL MEDIA MATTER FOR MARKETING? THE EFFECTSOF SOCIAL MEDIA ENGAGEMENT ON THE

CONSUMER-BRAND RELATIONSHIP

Benedikt Jahn, Ludwig-Maximilians–University MunichWerner H. Kunz, University of Massachusetts, Boston

SUMMARY

Social media has radically changed consumer behav-ior over the last few years. Brand pages on channels likeFacebook, Twitter, or YouTube are getting more andmore popular. Customer engagement on such channelsalso changes the idea of relationship marketing. Today,businesses are trying to get in contact and engage withexisting and potential customers by means of brand pageson social media platforms. Social Media Engagement hasbecome an important goal for many companies. But howrelevant is consumer engagement on social media forsuccessful branding? Many managers are still asking: Is itworthwhile putting more effort and investment into thesocial media phenomenon, and does it pay off? Despitethe importance of this question, a conceptual framing andunderstanding of social media engagement and its effectsis still missing.

Therefore this study develops a framework of socialmedia engagement and analyzes how consumer engage-ment on social media sites affects the consumer-brandrelationship. The proposed framework is based on classi-cal concepts of customer engagement and brand attach-ment theory. The basic idea of the framework is thatpassive and active brand page participation should lead tohigher brand attachment, which should in turn lead tobrand word-of-mouth and loyalty. To test our framework,we applied a multi-step approach incorporating a series ofqualitative and quantitative analyses. First, we executedfocus groups and a diary study. Next, we executed asurvey on Facebook (n=910) and tested the proposedframework using a structural equation model. Finally we

executed a longitudinal field experiment with Facebookusers (n = 100) in a pre-post within-subject design. Indifferent analyses we can show a significant influencefrom passive and active brand page participation on word-of-mouth and loyalty, mediated by brand attachment.

This study contributes in several ways to the researchand management of social media platforms. First, weshow a reliable effect of social media engagement onbrand pages for the customer-relationship management ofcompanies. Second, we don’t just show an effect, butdetect with passive and active participation two driverswith different relevance for the success of brand pages.Third, we show the exposure pathway of social mediaengagement over the customer relationship and prove thecausal direction of the effect in a longitudinal field experi-ment.

Based on the results, we can infer multiple implica-tions for the management of social media activities ingeneral and especially for brand pages on social network-ing sites. In general we can conclude that brand pages arean excellent tool for brand management. The results showsocial media engagement as a successful strategy forretaining and maintaining relationships, but also forattracting new fans. Ideally, fans would see brands asreal “friends” in their social networks, which play animportant part in their everyday lives. In this case, brandcommunication is no longer automatically perceived asdisturbing advertising but as interesting and reasonableconversation among friends. References are availableupon request.

For further information contact:Werner Kunz

College of ManagementUniversity of Massachusetts, 100

Morrissey BoulevardBoston, MA 02125

Phone: +1.617.291.8736Fax: +1.617.287.7877



EVALUATION AND AUTHENTICATION OF MUSIC SHARED THROUGHSOCIAL NETWORKING: eWOM OF CULTURAL PRODUCTS

Paul G. Barretta, The University of Texas – Pan AmericanMichael S. Minor, The University of Texas – Pan American

SUMMARY

People with prior knowledge about an artist mightemploy a different evaluation process than those unfamil-iar with the artist. These differences can lead to variancesin how an electronic word-of-mouth (eWOM) messagemight be communicated. How senders and receivers ofeWOM evaluate subject matter is an important element inunderstanding the ways in which messages and contentspread throughout a social network (Kozinets et al. 2010).When evaluating music content there is a process ofauthentication, which includes three dimensions – sincer-ity, expertise, and dependability (Barretta 2012). Examin-ing familiarity involves two important areas of research:perceived authenticity and word-of-mouth.

Music, regardless of genre, has been found to be away of expressing one’s identity (Bennett 1999; Cateforisand Humphreys 1997; Gardikiotis and Baltzis 2011;Hesmondhalgh 2008; Jackson 2002; Miklas and Arnold1999; Nuttall 2008b), and a means to developing interper-sonal relationships (Arnett 1993; Boer and Fischer 2011;Larsen et al. 2009; Nuttall 2008a; Nuttall and Tinson2008; Packer and Ballantyne 2010; Roy and Dowd 2010).It is not surprising that social networking sites likeFacebook have exploded as a platform for musiciansand fans to share music (Peoples 2011; Sargent 2009).Based on the notion that eWOM in contemporarysociety involves coproduced (Kozinets et al. 2010), cu-mulatively constructed and shared messages (Trusov et al.2009), it is important to both theory and practice that weexplore how consumers assess shared music. By develop-ing our understanding of how music is assessed on a socialnetworking platform, marketers can aim for strategies andcontent that will be most effective in reaching the desiredtarget market.

Prior literature on eWOM and exploratory qualita-tive research led to the following hypotheses:

H1: Receivers of shared music will evaluate the contentbased on three dimensions of perceived authentic-ity: Sincerity, Expertise, and Dependability.

H2a: Consumers unfamiliar with the source (musician/band) will rely upon perceptions of skill level(expertise), but not on pre-existing perceptions ofthe artist (sincerity).

H2b: Consumers with high familiarity will rely moreheavily on pre-existing perceptions of the artist(Sincerity) than on Skill level (Expertise).

Three factors mentioned in H1 (alpha = .91/.88/.83)explained dependent variable “overall evaluation” withan effect size (R2) of .568, explaining 57% of the variancein overall evaluation (n = 310). This result supportsHypothesis 1.

To test hypotheses 2a and 2b, the same independentand dependent variables were used to perform multiplelinear regression analysis for each group of reportedfamiliarity with the band performing the music:

• For participants who had never heard of the bandbefore, Expertise and Reliability were significantpredictors of evaluation (p<.001), but Sincerity wasnot a significant predictor. The independent variablesexplained 63% of the variance in overall evaluation(n = 120).

• For participants who were vaguely familiar with theband, Expertise and Sincerity were significant pre-dictors of evaluation (p < .001), but Reliability wasnot a significant predictor. The independent variablesexplained 42 percent of the variance in overall evalu-ation (n = 101).

• For participants who were very familiar with theband, only Sincerity was a significant predictor ofevaluation (p < .001), neither Expertise nor Reliabil-ity was a significant predictor. The sole independentvariable explained 51% of the variance in overallevaluation (n = 89).

These results indicate support for hypotheses 2a and2b. When consumers evaluated the music performanceclip of a band with whom they were not familiar, theyrelied upon their perceptions of the expertise and depend-ability of the band, which reflect the skills of the band.However, when pre-existing knowledge of a band ispresent, either good or bad, the evaluation was based onlyon how sincere they felt the band to be, not on expertise ordependability.

Our findings contribute to knowledge of how pre-existing familiarity influences two areas of research,


perceived music authenticity and eWOM. In the net-worked approach to eWOM, these two areas are related,so it is important to discuss the findings as they relate toeach area, as well as in combination.

As the results relate to the authentication process,marketers can determine the best content for a targetmarket based on that segment’s level of familiarity. Forexample, when marketing music by a band with the goalof expanding a fan base by reaching consumers notalready familiar with them, content should emphasize themusicians’ skills. However, if the objective is to buildsupport for an artist likely to be familiar to a targetaudience, either because it is an established artist expand-ing its fan base or building support with existing fans, thenelements of sincerity should be reflected in the content. Interms of eWOM, level of familiarity with a band couldinfluence what message is communicated to a person’s

social network. Therefore, content that may be seededshould match the objectives of one’s target audience.

There is an unavoidable element of snowball recruit-ment that like-minded individuals will participate. On onehand, this is a positive aspect of the design because itmimics how eWOM content is spread. On the other hand,the content is primarily a survey, not a music file, andtherefore may not follow the same path as a music clip thatcarries coproduced meanings as it spreads through thenetwork. Of interest for future research is replication ofthe SED authentication process using the dimensionsSincerity, Expertise and Dependability. We are encour-aged by the extension of scales from advertising literaturethat were used by Ohanian (1990) to define and measureperceived expertise, trustworthiness, and attractivenessof celebrity endorsers. References are available uponrequest.

For further information contact:Paul G. Barretta

The University of Texas – Pan American1201 W. University DriveEdinburg, TX 78541–2999

Phone: 570.760.3289Fax: 956.665.2242



SOCIAL NETWORKING SITES AND PLANNED BEHAVIOR

John T. Gironda, Florida Atlantic University, Boca RatonPradeep K. Korgaonkar, Florida Atlantic University, Boca Raton

SUMMARY

The rise of social media has not only changed theinternet as we know it, but has dramatically changed theway people communicate and interact. Under the largerumbrella of social media, social networking sites (SNSs)such as Facebook, LinkedIn, Twitter and MySpace havehad a particularly substantial impact. The ascension tomainstream popularity of SNSs has been nothing short ofincredible considering that only ten years ago thesewebsites did not even exist and now the estimate of thenumber of SNS users is over 750 million people (Facebook2012; Shih 2010; Twitter 2012). Needless to say thedramatic rise in popularity of this new medium hascreated huge opportunities for marketers. If marketerscan develop a more comprehensive awareness ofconsumer’s social networking site intentions and behav-ior they may be able to use these insights in the develop-ment and implementation of more effective SNS adver-tisements, business SNS pages and other marketing toolsin order to better serve, attract and retain customers.

This research works toward developing a more thor-ough understanding of consumer social networking siteusage by empirically exploring the antecedents of con-sumer intention and behavior in regards to three activitiesrelated to social networking sites: (1) general social net-working site usage, (2) the joining of a business’s socialnetworking site page and (3) clicking on an advertisementon a social networking site. These three types of activitiesrepresent some of the most common usage of SNS sitesfrom a marketer’s perspective. To assist with this explo-ration we drew upon the decomposed theory of plannedbehavior (DTPB) as a theoretical lens.

The DTPB represents an extension of the theory ofplanned behavior (TPB) (Ajzen 1991). The TPB positsthat intention is positively related to behavior, and thereare three constructs that can be used to predict intention(attitude, subjective norms, and perceived behavioralcontrol). In addition to predicting intention, perceivedbehavioral control is also said to predict behavior as well.The DTPB extends this line of thought by proposingantecedents to attitude (relative advantage, compatibility,and complexity), subjective norms (normative influences),and perceived behavioral control (self-efficacy and facili-tating conditions) respectively. For an in depth review ofthe DTPB see Taylor and Todd (1995).

Data for this research was collected via a self-administered online survey from 467 subjects at alarge urban university in the southeast United States.The measures utilized were all reflective indicators de-veloped from Taylor and Todd (1995) as well as based onrecommendations by Ajzen (2011) and adapted to fit thecontext of this study. Data analysis was carried out withLISREL 8.8 (Jöreskog and Sörbom 1996) using structuralequation modeling (SEM) with maximum likelihood (ML)estimation analyzing a covariance matrix. In accordancewith the two-step model building approach (Andersonand Gerbing 1988), a confirmatory factor analysis (CFA)measurement model was generated and analyzed for eachone of the three distinct SNS behaviors prior to testingrelationships between latent constructs. Following theCFA, structural path models with respect to each of thethree SNS behaviors were also generated and analyzed.Each of the three measurement models as well as each ofthe three structural models demonstrated factor loadings,goodness-of-fit statistics and measurement scalereliabilities that were consistent with generally acceptedthresholds.

Results of the structural model path analysis indi-cated that intention was significant and positively relatedto behavior for each of the three SNS activities examined.In addition, attitude exhibited the strongest relationshipwith intention, as a significant and positive relationshipwas found between attitude and intention for each of thethree structural models tested. In terms of the antecedentsto attitude, relative advantage and compatibility bothrespectively demonstrated a significant and positive rela-tionship with attitude for each of the three SNS activitiesexamined. Complexity however only demonstrated a sig-nificant (and as expected negative) relationship withattitude for the SNS behavior of joining a business’s socialnetworking site page.

Normative influences showed a significant and posi-tive relationship with subjective norms for each of thethree SNS behaviors; however subjective norms onlydemonstrated a significant relationship with intention forthe activity of clicking on an advertisement on a socialnetworking site.

The construct of perceived behavioral control dem-onstrated a positive and significant relationship with bothintention and behavior in terms of general social network-


ing site usage. However, these same relationships werenot significant for the behavior of joining a business’sSNS site and were significant but negative for the behav-ior of clicking on an ad on a social networking site. For theexpected antecedents to perceived behavioral control,facilitating conditions displayed a significant and positiverelationship with perceived behavioral control for each ofthe three SNS activities. Furthermore, self-efficacy dem-onstrated a significant and positive relationship withperceived behavioral control for two out of the threeactivities; general SNS usage and joining a business’s

social networking site page respectively. However, thisrelationship was significant but unexpectedly negative forthe activity of clicking on an SNS ad.

In summary, this study contributes to a more thor-ough understanding of consumer social networking siteusage and in the process provides several insights to bothpractitioners and researchers alike by discovering what is(and is not) related to consumers’ intention and behaviorregarding three distinct social networking site activities.References are available upon request.

For further information contact:John T. Gironda

Department of MarketingFlorida Atlantic University

201 Fleming Hall777 Glades Road

Boca Raton, FL 33410Phone: 561.297.3036



WHAT CONSUMERS REALLY THINK OF TARGETED ONLINEADVERTISING: A SEGMENTED APPROACH1

Bettina West, Ryerson University, TorontoAvner Levin, Ryerson University, TorontoMary Foster, Ryerson University, Toronto

SUMMARY

The Internet has changed how we shop, communicateand gather information. This is particularly true for 18 to34 year olds, who spend more time online than any otherage group (IAB Canada 2009). This study examinesyoung consumers’ attitudes and responses toward tar-geted online advertising, and the impact of different levelsof social media use on their receptiveness to such mes-sages. Such information can assist marketers to gain aclearer understanding of consumers’ attitudes andresponses to online media and allow them to makebetter informed decisions about allocating promotionbudgets and targeting consumers with tailored messagesin the midst of a rapidly changing technological climate.

We began the study by running two focus groups withconsumers between the ages of 18 and 30 to gain insightinto social media behaviors and attitudes toward targetedonline advertising. Each focus group was then followedby a “town hall” with an additional 60 individuals who hadobserved the focus groups and provided additional com-mentary to enhance our understanding of consumers’interactions with digital media. We then designed andconducted a quantitative survey incorporating insightsgathered from our qualitative stage as well as items fromprevious studies by Foster et al. (2010) and McDonald andCranor (2010).

In total, 1037 business students enrolled in a largeurban Canadian university completed the online survey.We employed SPSS (Version 19) for our analysis proce-dures which included exploratory factor analysis (EFA),cluster analysis and ANOVA. Confirmatory factor analy-sis (CFA) was performed using AMOS (Version 19).

Results of the CFA and cluster analysis revealed fourSMUG segments (identified as Minimally Involveds,Socializers, Info Seekers and Social Media Technology(SMT) Mavens) which were then used as the independentvariable for further analysis into consumer attitudes andbehaviors about targeted online advertising.

In general, targeted online ads tended to be perceivedas ineffective, both because they are not viewed as par-ticularly interesting or relevant, and because they rarelylead to click-throughs and product purchases. However,

closer examination of the four segments revealed subtledistinctions that offer potential targeting opportunities formarketers.

While all respondents displayed some degree ofinterest in personalized information (receiving personal-ized online materials such as ads, discount coupons andnews items), the SMT Maven group was more interestedthan others in ads and news items. They were also signifi-cantly less disinterested in paying a small monthly feeto either receive or block targeted online ads, perhapsbecause they spend more time online than other seg-ments and are therefore more particular about being ableto customize their online experience.

Overall, respondents displayed mild privacy-relatedconcerns by disagreeing with the statement “I don’t careif advertisers collect data about my search items” andagreeing that targeted online advertising is “creepy,”suggesting that the concept of online privacy is not well-developed among most users. Interestingly, the SMTMavens were the most likely to be unconcerned aboutonline privacy issues, while Socializers reported the stron-gest concerns, perhaps because Socializers tend to spendmost of their time on sites such as Facebook which theyperceive to be for private use. However, it appears thatmost respondents do not really understand the nature andimplications of targeting, and therefore do not view it asa potential invasion of privacy. As targeting capabilitiesbecome more sophisticated and effective than respon-dents understand them to be, disinterest in protectiveaction suggests either a worrisome (to privacy advocates)acceptance of the commercial use of information, orperhaps an expectation for government policy interven-tion, rather than individual action.

Info Seekers appeared to be the most accepting of theinevitability of online advertising, as they were signifi-cantly more likely to believe that online advertising isnecessary for the Internet and putting up with it is the pricethey pay for having access to free sites. This is consistentwith their use of the online environment as a research toolfor information.

SMT Mavens were the most likely to agree thattargeted ads feature products or services of interest tothem, and that they actively seek out relevant advertising.


This group was also significantly more likely to believethat including an interactive component in online adswould be more attention-getting. This is consistent withtheir heavier involvement in the online environment andthe diversity of activities in which they participate. InfoSeekers, on the other hand, were the most likely to agreethat random ads are more annoying than targeted ads,likely because random distractions would be more disrup-tive to their specific online search behaviors.

SMT Mavens were significantly more likely to ‘takeaction’ by clicking on links in online ads that come up nextto search results or on Facebook, making an online pur-chase as a result of an e-mail advertisement or an onlinead, and clicking on a link in an ad that comes up after anonline purchase. This is not surprising given that thisgroup is the most sophisticated in terms of range of onlineactivities and frequency of use. They were also signifi-cantly more likely to use the available tools to control theironline environment, such as clicking on “like” to accessmore information, clicking on “offensive” to stop targetedonline advertising and changing their Facebook profile toavoid online advertising.

Currently, advertisers attempt to target users basedon the profile information provided through social mediasites and their history of websites visited. From a manage-rial perspective, the existence of distinct segments ofonline users both in terms of behavioral patterns andmotivational triggers is significant for the development ofonline advertising strategies because it underlines theimportance of customization to meet the needs of differentsegments. By creating a SMUG-based customized onlineexperience, advertisers may be able to capture moreattention from each of these consumer segments as well aspossibly gain more positive response behaviors. Thechallenge moving forward is to incorporate design fea-tures and formats that address the unique needs of eachsegment without having a cluttered result that is confusingto all segments. References are available upon request.

ENDNOTE

1 The research on which this paper is based was funded bythe Office of the Privacy Commissioner of Canada.We also thank our Research Assistants, Colin Rogersand Roman Cezar for their help.

For further information contact:Bettina West


Toronto, ON M5B2K3Phone: 416.979.5000



ENGAGING THE FACEBOOK USER TO FOSTER SALES – DRIVERS OFADVERTISING EFFECTIVENESS IN SOCIAL NETWORKS,

INCORPORATING SYNERGIES AND TIME LAGS

Jens-Christian Reich, RWTH Aachen University, GermanyMalte Brettel, RWTH Aachen University, Germany

SUMMARY

With the growth of online advertisements (ads),numerous new, interactive ad channels have emergedattracting continuous attention of both researchers andpractitioners. Among these new ad channels, social net-works play an increasingly important role in today’smarketing mix (Forrester 2009). Due to their highlyinteractive character, these networks open up unprec-edented possibilities and challenges for managers andresearchers that can be used to market a product or service(Trusov et al. 2009). First research in this field based onsurveys and experiments in small networks has indicatedthat engaging the customer in a social network can buildcustomer loyalty (Bagozzi and Dholakia 2006). A cus-tomer can easily broadcast his or her opinion to a multi-tude of people with just one click and therefore functionas a brand advocate to acquire new customers (Trusovet al. 2009). Acquisition is facilitated by the fact that thesepeople often have similar needs and interests (Yang andLin 2006). Combining social networks with other adchannels can further increase their effectiveness, althoughinitial studies demonstrate a comparably low ad effective-ness (Mabry and Porter 2010).

However, unlike more traditional ad channels likeTV, print or search ads (Naik and Peters 2009), these newad channels still lack a profound proof of their salesimpact based on real field data. In particular, little isknown about what are the drivers of advertising success ina big social network like Facebook. We thus address fourmain research questions: (1) Which ad-related activitieswithin a social network drive short-term sales (Winer2009)? (2) How long do the ad effects triggered by thoseactivities last (Osinga et al. 2010)? (3) Which activity hasthe strongest long-term sales effect (Wang et al. 2009)?(4) Are there synergies between the different activities(Rust et al. 2004)?

To analyze advertising effectiveness, advanced mod-els have been developed that demonstrate that an ad effectcan carry-over to the next day and beyond (time lags; cf.,Naik and Raman 2003, who calculate a carry-over of 93percent for TV and 37% for print ads) and that simulta-neous use of different ads can increase their effectiveness(synergies; cf., Naik and Peters 2009, who find synergiesfor TV, print and search ads). We apply an integrated

direct aggregation approach to account for time lags(Srinivasan and Weir 1988); Herrington and Dempsey2005) and calculate interaction terms to model synergies(Green 1973; Völckner and Sattler 2006). For our analy-sis, we use aggregate-level, daily field data obtained fromone of Germany’s top ten e-commerce retailers. Thisunique sample spans a period of 365 days from October2010 to September 2011 and includes four ad-relatedactivities on Facebook. Based on the established frame-work of Vakratsas and Ambler (1999), we derive ourhypothesis on how advertising works:

• Short-term, we expect a positive sales impact whichis comparably weak for stream impressions (H1a);weak to moderate for page views (H1b); moderate tostrong for likes (H1c); and strong for contributions(H1d).

We expect the carry-over to be comparably short forstream impressions (H2a); short to mode-rate for pageviews (H2b); moderate to high for likes (H2c); and highfor contributions (H2d).

• Long-term, we expect positive sales effects that arecomparably weak for stream impressions (H3a); weakto moderate for page views (H3b ); moderate to highfor likes (H3c); and high for contributions (H3d).

We infer positive synergies and their sales impact tobe comparably weak for stream impressions and pageviews (H4a); moderate for stream impressions and likes(H4b); moderate for stream impressions and contribu-tions (H4c); moderate for page views and likes (H4d);moderate for page views and contributions (H4e); high forlikes and contributions (H4f).

We find that all ad-related activities in our analysis,except for stream impressions (H1a rejected), positivelyaffect sales. A like of a group’s content has the strongestshort-term sales effect, followed by a contribution (par-tially confirming H1c-d) and a page view (confirmingH1b). Active contributions followed by likes show a high,page views a medium carryover effect, as expected, veri-fying H2b-d. Long-term, active contributions have thestrongest sales impact followed by likes and a rather weakeffect for page views (hypotheses H3b-d supported). Outof our six hypotheses on synergies, three are fully sup-


ported (H4b, H4c, and H4f). Likes and contributionsestablish a strong synergistic effect and also show amoderate synergy with stream impressions as expected.Besides these effects, we do not find evidence for signifi-cant synergies among the other activities in our focus,leading to a rejection of H4a, H4d, and H4e.

For future research, our results underscore the impor-tance of incorporating different time lags as well assynergies when modeling advertising effectiveness. Prac-titioners can use this approach to optimize their advertis-ing as it is based on data that is readily available onlinethrough Facebook at no additional charges. Generalizing

our empirical results, a marketer should use social net-works for advertising to drive sales. In doing so, it isimportant to engage the user and generate both likes andcontributions which mutually increase their strength tocreate a strong short- and especially long-term saleseffect, the strength of Facebook as an ad channel. More-over, stream impressions should be used as a trigger tomotivate engagement with the group and consequentlyimpact sales. We encourage more research in this field togain a solid understanding of how to further increase adeffectiveness in a social network (e.g., through an analysisof different design elements like ad message content).References are available upon request.

For further information contact:Jens-Christian Reich

RWTH Aachen UniversityKackertstr. 7, 52072, Aachen




SOCIAL MEDIA MARKETING INTENSITY AND ITS IMPACT ON THERELATION BETWEEN DYNAMIC CAPABILITIES

AND BUSINESS PERFORMANCE

Felipe Uribe, Barcelona Autonomous University, SpainJosep Rialp, Barcelona Autonomous University, SpainJoan Llonch, Barcelona Autonomous University, Spain

Henry Robben, Nyenrode Business Universiteit, The Netherlands

SUMMARY

The rise of Online Social Networks (OSN) such asFacebook, Twitter, YouTube or LinkedIn, has changedthe way that people communicate through the Internet.Firms have realized that their clients are part of the OSN,increasing the interest of marketers to explore it as a newmarketing tool. However, its strategic importance stillseems not clear given the novelty and the difficulty ofmeasuring their impact on business performance. Our aimis to contribute with the crescent but still limited study ofthe use of Social Media Marketing (SMM), taking theintensity of use of SMM as a moderating variable betweenthe dynamic capabilities of market orientation (MO),entrepreneurial orientation (EO), and performance (PERF).

Dynamic capabilities such as MO and EO allow firmsto integrate, build, and reconfigure competencies andalign them to market changes to obtain a better perfor-mance (Teece, Pisano, and Shuen 1997). A strong MOcharacterizes an organization´s disposition to deliversuperior value to its customers (Slater and Narver1994); and, an EO determines how firms identify newopportunities and respond to changes in the environmentto achieve successful results. Several studies have consid-ered the relation MO-PERF, and EO-PERF. Also, someauthors argue that the interaction MO-EO is synergistic,as they have complementary approaches. The relationshipMO-PERF, and EO-PERF as well as being direct, may bemediated by many control variables (Jaworski andKohli 1993). A conceptual model is proposed suggest-ing that the intensity of use of SMM moderate therelation between MO, EO, and PERF.

Through LinkedIn, we invited Spanish marketingchief officers, digital marketing directors and communitymanagers to be part of a network, between February andOctober 2011. To be eligible, firms must have used at leastone OSN as a marketing tool. Seven hundred eighty-fourcontacts were linked, making possible to communicatedirectly, avoiding non appropriate responders, achieving191 complete responses (24.3%) for the analysis. Formeasuring MO we used an adaption of the MKATOR

scale by Narver and Slater (1990), which consists of threesubscales reflecting the behavioral components of cus-tomer and competitor orientation, and inter-functionalcoordination. For EO we used the Covin and Slevin(1989) scale, measuring firm innovativeness, proactivenessand risk taking. To measure PERF, we adopted the con-struct proposed by Hooley, Greenley, Cadogan, and Fahy(2005), which consists of subscales for customer, marketand financial results.

SMM intensity was calculated using informationabout: number of Facebook fans and Twitter followers;number of YouTube videos; frequency of SMM activi-ties; SMM team size; among others. Some variables werere-coded using quartiles, to generate five groups of clas-sification in order to use it all in a Multi CorrespondenceAnalysis. Using the most discriminative ones, three cat-egories were created for SMM intensity. This new vari-able allows doing a multi group analysis using StructuralEquations Modeling (SEM).

To refine the measures and assesses the reliabilityand validity of constructs we used an exploratory factoranalysis. Cronbach´s alpha was calculated, obtaining sat-isfactory results for all construct internal consistency. AConfirmatory Factor Analysis (CFA) using AMOS v.18was conducted and confirms these results. Using SEM,we estimated a model including MO, EO and PERF. Theoverall fitness indices of the model suggest a good fit forthe construct model. χ2 = 417.370 (p < .001); χ2/df =1.355. In our case 1.355, shows a good model fit. The CFIindex of 0.956 indicates again a good model fit, becausewhen these values are close to 1 data validate the theoreti-cal model. RMSEA is 0.043, below the acceptable level of0.05, which indicates a close fit of the model in relation tothe degrees of freedom.

The results shows: (i) the positive and direct effect ofMO on PERF, supporting the notion about this relation;(ii) the positive and direct effect of MO on EO, accordingprevious studies; and iii) there is no evidence for support-ing the direct effect of EO on PERF, which is indirectthrough other variables. One of these variables could be


the MO itself, because firms that combine EO and MOshow higher performance than firms that only developone orientation. To test our hypotheses we applied amultiple group technique. Data were divided consideringthe “Intensity” variable. SEM estimates were calculatedfor each group. To check if there are differences betweenthe groups we used a tool developed by Gaskin (2011),which evaluates the differences by the critical ratiosmethod. Results show that H1 is confirmed: the intensityof use of SMM moderates the relationship MO-PERF.There is a significant difference between Medium andHigh intensity. However, there is no significant differencebetween Low and Medium. H2 is also confirmed: theintensity moderates the relation MO-EO, but only com-paring Medium and High. No significant differences werefound for Low and Medium. Finally, H3 is not confirmed:the intensity does not moderate the relation EO-PERF.

Several findings emerged from our study. First, theresults confirm for Spanish firms that use SMM theprevious findings about the direct and positive relationMO-PERF; and MO-EO. Our results agree with previousstudies that did not find evidence for supporting the directeffect of EO on PERF. Second, the results show that firms

must use OSN in an intensive way to potentiate therelation MO-PERF. Just if firms participate actively inOSN, really develop a SMM strategy, invest in SMM andhave a community manager or a digital marketing agency;they could take profit of the potential of OSN as amarketing tool. Results show that there is not enough withhaving some OSN profiles and do some basic SMMactions. A dedicated and strong effort in SMM is neededto observe a real impact in performance. Third, the resultsprove that firms must use OSN in an intensive way tomoderate the relation between MO and EO. Fourthly,there is no evidence to support that a higher intensity in theuse of the SMM moderates the EO–PERF relation. Prob-ably it is not enough of being innovative, proactive and totake risks with SMM to impact on business performance.

Summarizing, only a high intensity in the use ofSMM could help to moderate the MO effect over businessperformance, and the MO effect over EO. It is not enoughto be a SMM amateur; firms must be professionals to takeprofit of this emerging and growing marketing tool. Thisstudy is limited by several factors that should be addressedin future research. Implications for managers, policymakers and academia are discussed.

For further information contact:Felipe Uribe

Barcelona Autonomous UniversityDepartament d’Economia de l’Empresa

UAB, Campus Bellaterra, Edifici BCerdanyola, Barcelona 08193

SpainPhone: +34.93581.1209



NETNOGRAPHY AND METAPHYSICAL BRANDING

Steve Oakes, University of Liverpool, United KingdomNoel Dennis, Glasgow Caledonian University, United Kingdom

Helen Oakes, Keele University, United Kingdom

SUMMARY

The current study examines online forum evaluationand introduces the concept of metaphysical branding,considering how it contributes to the success of the best-selling album in jazz history (Kind of Blue by MilesDavis). Analysis of Kind of Blue postings draws on amusic rating website publishing authors’ personal thoughtsand opinions in a diary-like format. Such online word-of-mouth communication involves a networked co-produc-tion of narratives with the potential to alter marketingmeanings (Kozinets et al. 2010). While online communi-cation epitomizes the modernity of the digital age, thecurrent paper considers how such communication mayparadoxically reflect an innate metaphysical yearning thatrejects technology in favor of mythology and spirituality.Text on the Kind of Blue CD cover communicates themetaphysical by depicting a spiritually uplifting experi-ence in listening to the album. For example, the liner notesstate: “If you’re going to heaven, you might as well go firstclass all the way.” While the concept of “heaven on earth”pulses through many marketing campaigns such as thosefor Hawaiian holidays (Borgerson and Schroeder 2002),the concept of metaphysical branding is broader than this.It identifies how purchase and consumption of the brandedproduct potentially provide an access point for consumersseeking to move along a pathway of self-improvement.Online word-of-mouth communication is an importantpart of this process since any critical comments regardingthe sacred product may be infallibly regarded by theonline community of devotees as sacrilegious or heretical.However, metaphysical branding also underlines the cog-nitive effort required of the uninitiated consumer in con-tributing toward self-improvement.

Netnography “adapts ethnographic research tech-niques to the study of cultures and communities emergingthrough computer-mediated communications and usesinformation publicly available in online forums” (Kozinets2006, p. 130). The current study uses a netnographicapproach based on analysis of online textual discourseposted on a music rating website involving posting ofcomments regarding music that allows others to reply.Kozinets identifies boards as valuable forums for commu-nities of consumption that ‘offer one of the best places tofind consumption-related topics’ (2006, p. 131), provid-ing a counterbalance to corporate communication. Inconsumers’ stories, brands are frequently related to reli-

gious devotion and provide netnographic understandingthat is useful in terms of branding strategy and positioning(Kozinets 2006). Netnography is used to gather onlinedata from consumers who have purchased Kind of Blue,and is particularly appropriate in this context since it canprovide information on the metaphorical meanings, sym-bolism and consumption patterns of online music con-sumer groups (Kozinets 2002). One hundred seventy-fivepages of Kind of Blue postings are identified and analyzedfrom the music rating website: http://rateyourmusic.com/release/album/miles_davis/kind_of_blue/

Postings commenced in January 2002 and were ac-cessed by the researchers on 19th July 2011. After eachresearcher had read all postings, agreement was reachedon a preliminary list of metaphysically themed categories:Rite of passage, sacred and profane consumption, andself-improvement.

Rite of passage: Numerous users indicate this wastheir first jazz album, suggesting it is a rite of passage entrylevel album for novices seeking access to the art form. Itmarks an aspirational pathway toward spiritual and aes-thetic enlightenment. The album symbolically gives birthto appreciation of the genre. The current data frequentlyrefers to a metaphysical doorway for the novice listener:

Hearing this album, my first jazz album, was an opendoor into this hallway of the universe (October 20,2009).

Sacred and profane consumption: The concept ofsacred consumption and references to a heavenly listen-ing experience occur regularly throughout the data. How-ever, frustrated listeners unable to enjoy the music mayrespond with blasphemy:

I’m blaspheming here, since Miles Davis is a legendand this album in particular is hailed as such anamazing classic, but I honestly don’t get it (August 2,2009).

Questioning the unquestionable quality of the albumis acknowledged as heretical:

I’m a heretic. Why? Because, simply, I can name adozen jazz albums I prefer to this one (December 10,2006).


The metaphor of music as heavenly reward is statedby numerous users and scores of postings convey evangeliconline recommendations of the music:

A holy relic that I also happen to love (May 2, 2010).

Myths surround the sacred recording to document itsstatus:

A hallowed monument in the history of jazz (Decem-ber 1, 2008).

The evidence highlights consumers’ deep-seated needfor the mystical:

One of the most influential jazz albums ever made-theHoly Grail (February 19, 2005).

Self-improvement: Diligent attempts at aesthetic self-improvement are reported in the postings, some of whichlead to success, others to failure. A quasi-spiritual searchfor faith and enlightenment is evident. For some listeners,patience and perseverance eventually lead to revelation:

After its umpteenth-millionth repeated listen – Ifinally get Kind of Blue (December 30, 2010).

Some comments convey the dedication of religiousasceticism:

You might feel like just giving up and walking away.Still, if you keep on trying, and you keep at it, thenyou’ll come out feeling renewed (April 14, 2007).

Metaphysical branding is identified as an overarchingconcept binding together discrete themes in the data andis likely to be of value in marketing various creative arts.While spiritual discourse in Kind of Blue may have amarketing origin, the metaphysical mythology is eagerlyperpetuated by users who incorporate extensive religiousmetaphor into their comments. Postings indicate that theconsumers are not just passively responding to marketingcommunications, but are actively sacralizing Miles Davisand Kind of Blue, thus reverencing the metaphysical byco-creating transcendent meaning in their lives. Sacral-ization of the secular enables them to experience the kindof extraordinary meaning previously attained primarilythrough religion (Belk et al., 1989). The current researchreveals how dissemination of brand related religiousdiscourse is particularly suited to confessional onlineforums. Due to their anonymity, such forums avoid thefear of embarrassment of discussing metaphysical con-cepts using traditional word-of-mouth communications.Web-based forums therefore create unique marketingopportunities to harness and disseminate metaphysicalbrand linkages. An important managerial implication ofthe current study is that religious and mythical motifs haveinvested Kind of Blue with self-perpetuating, life cycleextending meanings. The concept of metaphysical brand-ing could be developed by examining its application in themarketing of successful CD’s and performers in variousmusical genres including the charismatic deities of popand rock music. Metaphysical branding research is alsovaluable when considering the global marketing of sportsteams which typically engenders quasi-religious, wor-shipful devotion and unquestioning, lifelong brand loy-alty in its fan base. References are available upon request.

For further information contact:Steve Oakes

University of Liverpool Management SchoolChatham Street,

Liverpool L69 7ZHUnited Kingdom

Phone: +44(0)151.795.3010Fax: +44(0)151.795.3001



INTEGRATION VS. REGULATION: WHAT REALLY DRIVES USER-GENERATED CONTENT IN SOCIAL MEDIA CHANNELS?

Welf Weiger, Georg August University Goettingen, GermanyHauke Wetzel, Georg August University Goettingen, Germany

Maik Hammerschmidt, Georg August University Goettingen, Germany

SUMMARY

Brand related user-generated content (UGC) plays avital role in social media channels. Seminal studies showthat UGC provides valuable opportunities to influencebuying behavior and to enhance sales revenue (Dhar andChang 2009; Zhu and Zhang 2010), brand equity(Christodoulides, Jevons, and Bonhomme 2012) and stockmarket performance (Tirunillai and Tellis 2012). How-ever, UGC also represents a potential threat for brandsbecause proactive consumers cannot only advocate abrand but they can also easily inflict damage on it. Thisconflict is fueled by the tremendous rise in UGC creationbecause consumers are wrenching significant controlover brand communications from brand managers(Willems 2011).

Consequently, managers are getting increasinglyinterested in identifying consumers’ motives forengaging in UGC. However, research that picks up onthis issue is in its infancy. It has been shown that consum-ers’ integration motives, such as the desire to participateand collaborate in a community, drive UGC creation(Christodoulides et al. 2012; Sheldon, Abad, and Hinsch2011). At the same time, regulation motives, such asseeking for economic rewards, may motivate consumersto produce UGC (Hennig-Thurau et al. 2004; Ghose andSang Pil 2011). However, these findings focus on theisolated effects of integration and regulation. But how isUGC creation affected once external regulation and inte-gration motives coincide? On the one hand, preliminaryfindings in UGC research suggest that integration andregulation motives may complement each other in theireffect on UGC creation (Hennig-Thurau et al. 2004). Onthe other hand, literature on behavior control suggests thatthe outcomes of internal motives may be contingent onexternal reward and governance structures (Hernandez2008). Specifically, external regulation might underminethe impact of integration motives on UGC creation due toa motivation conflict. Thus, we examine whether and howintegration and regulation interact and what consequencessuch a potential interaction has for UGC creation andUGC management.

In answering these questions, we make several con-tributions. First, consistent with prior findings, we showthat both integration and regulation have a direct and

positive effect on UGC creation. By considering these twomotivational extremes, we present a framework that com-prises the range of motivations that should be consideredby managers who desire to enhance UGC.

Further, we contribute by considering the interplaybetween integration and regulation. Self-determinationtheory postulates that motivational conflicts affect taskperformance (Ryan and Deci 2000). Thus, we proposethat regulations make users feel forced into action whichblocks integration motives and ultimately results in adecrease of UGC creation. Critical for managers, ourresults show that regulation has a negative moderatingeffect on the impact of integration on UGC creation.Because managers address regulation motives much easierthan integration motives, this finding leads to a controldilemma for brand managers.

Finally, we examine the role user’s consumptionlevel of firm-generated content (FGC) for the aforemen-tioned dilemma. As a high consumption of FGC fostersfeelings of relatedness to a brand, a user internalizes theexternal goal. Thus, the activity is rather perceived as self-determined than other-determined. In line with this argu-mentation, our findings indicate that besides having apositive impact on UGC, FGC consumption alleviates theaforementioned dilemma by mitigating the underminingeffect of regulations on the impact of integration motives.This result does not only demonstrate that managers canenhance UGC creation by providing FGC. It also impliesthat FGC is a necessary prerequisite for regulationmotives (i.e., providing incentives for UGC) to comple-ment rather than cannibalize the role of integrationmotives for enhancing UGC.

In sum, this study improves the understanding ofUGC drivers. Using blogger data, we demonstrate thatboth integration and regulation are important motivationsof UGC creation. Users create content because they desiresocial interaction as well as economic incentives. How-ever, the results indicate that the regulation motive, whichis addressed by incentive strategies, undermines the posi-tive relationship between the integration motive and UGC.Moreover, this undermining effect is alleviated for userswho show high consumption of FGC. Thus, firms shouldtarget incentives on heavy consumers of FGC in order toavoid undesired results. References are available uponrequest.


For further information contact:Welf Weiger

Marketing and Innovation ManagementGeorg August University Goettingen





IS THERE A DARK SIDE TO CUSTOMER CO-CREATION? EXPLORINGCONSEQUENCES OF FAILED CO-CREATED SERVICES

Matthias Handrich, EBS Business School, Germany Sven HeidenreicH, EBS Business School, Germany

SUMMARY

Recently innovative, technology-based services likemobile apps are on the rise (Hoyer et al. 2010; Ostromet al. 2010). Such technology-based services (TBS)require the customer to co-create the service by ac-tively engaging in the service provision and consumption(Grönroos 2011, forthcoming; Witell et al. 2011). There-fore, it is no surprise that, customer co-creation hasbecome one of the dominant research topics in recentyears. However, despite the theoretical and practical rel-evance, empirical results on customer co-creation, espe-cially for TBS are scarce ( Bolton and Saxena-Iyer 2009;Dong et al. 2008; Meuter et al. 2005). Furthermore, noone has yet examined the downsides of customerco-creation. More specifically, no one has examinedthe ramifications of failed co-created TBS. Thus, thepresent paper examines how the degree of co-creation(high vs. low) will affect customer satisfaction in case ofsuccessful and failed TBS. Additionally, we test theeffectiveness of proactive and reactive service recovery tocounteract the possible negative consequences of servicefailures for TBS.

Drawing on equity theory we argue that the fairnessof every transaction is based on a customer’s overallevaluation of investment (provided input) and reward(received output; Adams 1963). If a customer’s input/output ratio is smaller than the input/output ratio of theservice provider, e.g., because a service failure occurred,the customer feels unfairly treated (de Ruyter and Wetzels2000; Homans 1968). Stated differently, if the customerhas provided his/her input (e.g., paid the price and in-vested time), but did not receive the expected output (thecustomer received a failed service) he/she is dissatisfied(Lapidus and Pinkerton 1995).

Based on a thorough literature review, we find thatfor highly co-created TBS customers have to invest aconsiderable amount of time, effort and share informationcompared to the same service low on co-creation (Chan etal. 2010; Etgar 2008; Hoyer et al. 2010). Consequently, inthe event of a service failure the input/output ratio is largerfor highly co-created TBS than for low co-created ones.Thus, customers who experience a service failure usinghighly co-created TBS are more dissatisfied than custom-ers using failed, low co-created TBS.

Furthermore, and also in line with equity theory,we propose that the type of service recovery dependson the co-creation level. Customers using a TBS lowon co-creation prefer a proactive service recovery asthey do not want to provide the same amount of input inthe service process as customers using high co-creationTBS (Bitner et al. 1997). Thus, these customers prefer aneffortless service recovery, in this case a proactive servicerecovery which is initiated and managed by the company(Smith et al. 1999). In contrast as customers using a highco-creation TBS have invested a considerable amount ofinputs (Etgar 2008; Hoyer et al. 2010) they are moredissatisfied when a service failure occurs. Thus, thesecustomers feel the urge to complain to the service providerand “blow of some steam” (Goodwin and Ross 1990;Grégoire et al. 2009). Consequently, they spread negativeword-of-mouth or they directly complain to the serviceprovider. Complaining or spreading NWOM representsviable customer reactions which help to restore equity notonly by changing a customer’s input/output ratio, but alsoby altering the input/output ratio of the service provider(Alexander 2002; de Ruyter and Wetzels 2000). Thus, bycomplaining or spreading NWOM a customer harms thereputation of the company (Fullerton and Punj 1997)which subjectively decreases a company’s output andfinally its input/output ratio thereby restoring equity andreducing customer dissatisfaction (de Ruyter and Wetzels2000; Lapidus and Pinkerton 1995; Szymanski and Henard2001). Knowing this, it seems reasonable that an effectiveservice recovery for a high co-creation technology-basedservice has to provide the customer with the opportunityto complain. Thus, in line with equity theory, we suspectthat for customer using high co-creation TBS a reactiveservice recovery is more effective in restoring customersatisfaction.

To test our two hypotheses we conducted a scenario-based online experiment (n = 288) describing the use oftwo different mobile apps (an app to buy a train ticket andan app to book a flight). We used analysis of variance totest our hypotheses. We chose mobile apps as researchcontext as they represent an innovative and representativetype of technology-based service which allows for varia-tion in co-creation (Ding and Ng 2011; van Beuningenet al. 2009). We especially picked ticketing mobile apps,as these apps are used worldwide by 230 million peopletoday and are expected to be used by over 750 million


people by the year 2015 (Juniper Research 2011). Basedon rankings within the commonly used app-stores (apple,android, OVI, blackberry) we selected the 10 most popu-lar ticketing apps. Subsequently, we developed two prod-uct descriptions for each app, manipulating the level ofco-creation as being low or high. We used expertjudges to rate the descriptions with respect to wording,sequence, format and layout, difficulty and effectivenessof our co-creation manipulations (Ngo and O’Cass 2009).Finally, the judges chose two apps, one to book a flightand one to buy a train ticket as the manipulations of thelevel of customer co-creation were seen as most effectivefor these apps.

Our results show that customers using TBS high onco-creation will be more satisfied than customers usingthe same TBS low on co-creation, whereas in the event ofa service failure customers with TBS low on co-creationwill be less dissatisfied than the ones using high co-creation TBS. Furthermore, we find that for customersusing failed low co-creation TBS a proactive servicerecovery is more effective in restoring customer satis-faction than a reactive service recovery. However, in caseof failed high co-creation TBS, the situation is reversed,meaning that those customers react more positively toreactive service recovery than to proactive. Thus, both ourhypotheses can be confirmed.

Our research not only extends current customerco-creation literature significantly, but also providesmanagers with valuable insights on how to deal withfailed, co-created, technology-based services. First ofall, managers have to be aware of the consequenceswhen offering high co-creation TBS. As our resultsshow, the volatility of customer satisfaction is muchstronger for high co-creation than for low co-creationTBS. Thus, customer satisfaction can be either very highin case of a failure-free service or very low in the event ofa failure. As customer satisfaction is linked to companyprofits (Anderson and Mittal 2000) managers shouldcarefully consider which level of customer co-creationthey want to allow within the service design process.

Second, according to our results managers have tochoose the type of service recovery before they decide onthe components of the recovery. The type of servicerecovery depends on the level of customer co-creation.Consequently, mangers should advice their companies tooffer reactive service recoveries for high co-creationTBS and proactive service recoveries for TBS low onco-creation. In the next step, they can decide on thecomponents of the service recovery (e.g., price dis-counts, free add-on service, apology etc.). Thus, oursecond hypothesis provides managers with crucial infor-mation to effectively and efficiently plan their servicerecovery strategies.

For further information contact:Matthias Handrich

EBS Business SchoolRheingaustraße 1

Oestrich-Winkel, Hessen 65375Germany

Phone: +49(0)170.3521869 E-Mail: [email protected]


THE ROLE OF MORAL IDENTITY IN ONLINE CONSUMERREVIEW BEHAVIOR

Nan Zhang, Southern Illinois University, CarbondaleMavis T. Adjei, Southern Illinois University, Carbondale

SUMMARY

Consumers’ online product reviews have been shownto have a profound influence on the market performanceof products being reviewed (Pan and Zhang 2011). How-ever, although a number of studies have delved into theconsequences of online reviews on review readers’ con-sumption behaviors (e.g., Chevalier and Mayzlin 2006;Zhu and Zhang 2010; Pan and Zhang 2011), crucial gapsstill remain in the marketing literature. For example,online product review research has yet to study how tomotivate consumers to provide honest and unbiased re-views. The purpose of this study is to begin to identifysocio-psychological factors (here, moral identity) that canmotivate consumers to create online reviews, and sub-sequently, enhance reviewers’ behavioral loyalty. Wedevelop and test a theory-based model of the value ofmoral identity in mobilizing online product reviews andultimately, the reviewers’ behavioral loyalty.

The conceptual model for this study was developedby integrating the social psychology (e.g., Aquino et al.2009) with the customer purchase behavior literature(e.g., De Wulf, Odeken-Schroder, and Iacabucci 2001).We propose that self-importance of moral identity (SIMI)leads to increased review intentions and review honesty,which in turn positively affects behavioral loyalty of thereviewers. Also, the influence of moral reviewing behav-ior on reviewers’ behavioral loyalty is moderated byreview valence. Further, we propose that SIMI can beprimed to increase moral review intentions and honesty.

Drawing on moral identity theory (MIT), this studyargues that, SIMI promotes review intentions and moralreviewing behaviors such as increased review intentionand review honesty. The key premise of MIT is that salientor self-important moral identity motivates consumers toengage in moral behaviors, which are perceived to beconsistent with the high moral identity (Forehand,Deshpandé, and Reed 2002). As a reviewer post morefrequently and honestly, his or her involvement with theproduct or brand being reviewed increases (Park, Lee, andHan 2007), which exerts greater commitment and behav-ioral loyalty to the product or brand (Howard and Sheth1969; Iwasaki and Havits 1998; Tyebjee 1979). The effectof moral review behavior on reviewers’ behavioral loy-alty will be stronger when review valence is positive

versus negative. This is because online product reviewersencounter internal pressure to behave consistently withthe opinions they express in reviews, whether positive ornegative (Davidow 2003). Finally, SIMI can be primed toincrease moral reviewing behaviors because people withmoral identities that are primed to be salient are likely totake moral actions (Reed, Aquino, and Levy et al. 2007).

This study focuses on the moral identity of consum-ers in an online product reviews context. This study usedtwo experimental studies to investigate the value of SIMIin increasing review intentions and review honesty, andsubsequently, behavioral loyalty. In Study 1, we recruited250 participants via Amazon’s MTurk service. This studywas used to measure the following constructs: self-impor-tance of moral identity, review intention, review honesty,and behavioral loyalty (i.e., share of wallet, frequency ofpurchase). Structural equation modeling was used toanalyze the data.

As in study 1, 250 subjects were recruited fromAmazon’s MTurk service in Study 2. The purpose ofStudy 2 is to investigate whether SIMI can be primed toinfluence review behaviors. The study contains threeparts. Participants were randomly assigned to either theprimed condition or the non-prime condition.

In summary, our results show that SIMI is a prom-ising construct for eliciting individuals’ online reviewbehaviors, which in turn, leads to behavioral loyalty.More importantly, an individual’s SIMI can be primedto increase his/her intentions to provide online reviewsfor products, in an honest manner. Contrary to our expec-tations, review honesty does not increase behavioral loy-alty, and review valence does not moderate the effect ofreview behaviors on behavioral loyalty is not supported.These results suggest that review intention lead toreviewers’ behavioral loyalty regardless of thevalence of review.

Our results imply that SIMI works as an essentialmechanism that is related with accessibility of consumers’willingness to be helpful and honest, and thus increasesconsumers review intention and review honesty. Also,when building an online review system, managers shouldrecognize the window of opportunity where they can“prime” consumers’ SIMI (e.g., use moral appeals wheninviting people to review a product online) in order to


increase moral review behaviors, which generate review-ers’ behavior loyalty. We recommend that, based on thecontinuous improvement of satisfaction, the leverage ofSIMI should be utilized to enhance the effectiveness of the

investment in building customer satisfaction, and to avoidthe unnecessary loss caused by exaggerated reviews (e.g.,exaggerated negative online reviews). References areavailable upon request.

For further information contact:Mavis T. Adjei

Department of MarketingCollege of Business and AdministrationSouthern Illinois University Carbondale

Rehn Hall 229, Mailcode 4629Carbondale, IL 62901Phone: 618.453.4341



WHO CARES ABOUT CROWDSOURCING FROM A VIRTUAL BRANDCOMMUNITY? THE CASE OF MARVEL.COM

Paul G. Barretta, The University of Texas – Pan American

SUMMARY

This research asks whether or not crowdsourcingfrom a Virtual Brand Community (VBC) has an influenceon consumer perception of a resulting innovation; ifknowledge that ideas were crowd-sourced from a VBCaffects brand attachment; and if so, is it the same formembers as non-members? Theoretically, it tests oneaspect of the VBC phenomenon -- the similarity or differ-ence in perspective between members and non-members.Pragmatically, it provides direction to managers abouthow effective their VBC is as a source of innovation.

The present study empirically examines an assump-tion that ideas resulting from a social process of a VBCmay have potential for commercialization by the brandowner (Franke and Shah 2003). Commercialization ofideas sourced from a VBC implies non-members wouldperceive the innovation the same as members. However,previous findings which show differences in brand com-mitment (Kim et al. 2008a) and word-of-mouth engage-ment (Woisetschläger et al. 2008; Yeh and Choi 2011)suggest members and non-members might have differ-ences in perception of those ideas. Therefore:

Hypothesis 1a: Satisfaction with a brand’s productwill be greater for members than for non-members.

Hypothesis 1b: Satisfaction with a brand’s productinnovation that was sourced from a Virtual BrandCommunity will be greater for members, but not fornon-members.

The notion that crowdsourcing from a VBC can be atool to increase brand perception in general (both insideand outside the community) is challenged by measuringbrand attachment, suggested to include brand promi-nence and brand connectedness by Park et al. (2010).

Hypothesis 2a: Brand Attitude will be greater formembers than for non-members.

Hypothesis 2b: When innovation is sourced from aVirtual Brand Community, Brand Attitude will begreater for members, but not for non-members.

Hypothesis 3a: Brand Prominence will be greater formembers than for non-members.

Hypothesis 3b: When innovation is sourced from aVirtual Brand Community, Brand Prominence willbe greater for members, but not for non-members.

Hypothesis 4a: Brand Connectedness will be greaterfor members than for non-members.

Hypothesis 4b: When innovation is sourced from aVirtual Brand Community, Brand Connectednesswill be greater for members, but not for non-mem-bers.

The context used to test the hypotheses was the web-based brand community Marvel.com. Product lines of thiswell-known brand are amenable to testing for perceptionsof innovation. Comic book story lines represent discourseof social commentary (Stevens 2011); and charactersexist in a world that “re-sets” to create new story lines(KustomKoolMedia 2010), such as the Civil War series.These re-sets are a form of innovation used to develop newproducts for both existing and new customers. Afterseveral points of contact with Marvel Entertainment,LLC, confirmation was obtained by telephone that thecompany is aware of the research and no attempts weremade by the company to either disrupt or participate in theeffort.

A web-based survey was distributed to fans of comicbooks and related products. Due to necessary deception,e-mail addresses were collected by holding a drawing fora gift certificate to the winner’s favorite comic retailer.After a sufficient number of responses were collected,each individual was contacted electronically to reveal thenecessary deception. The online survey employed a sce-nario-based experiment to measure satisfaction, brandattachment, and brand attitude. The first page of thesurvey enabled respondents to self-select into communitygroup membership by asking about their affiliation withthe website Marvel.com. Respondents were then random-ized using the web-based survey into three groups forsource of innovation.

The scenarios manipulated readers’ perception ofwhere the idea for Civil War was developed using neces-sary deception. In the community group, the first sentenceof the scenario indicated that the idea for the new storylines came from members of the Marvel.com community.In the company group, the first sentence indicated that the


idea came from writers and editors of Marvel Comics. Thecontrol Group scenario did not indicate any source ofinnovation. Respondents were then asked questions aboutthe Civil War series to record their evaluation of theinnovation. These questions were followed by scales forBrand Attachment and Brand Attitude adapted from Parket al. (2010). Because data was gathered using on-lineposting of the survey, a manipulation check questioncould not be included in the main survey. Instead, thesame scenarios were used in a separate survey that askedfive questions, one of which was a manipulation check forthe source of innovation treatment. Fifty-seven usablesurveys with equal numbers of respondents in each grouprevealed the manipulation was successful for 90% ofrespondents.

Evidence was found that Community Membershipcauses significant differences in all four dependent vari-ables, finding support for Hypotheses 1a, 2a, 3a and 4a.No significant differences in means were found for themain effect treatment innovation source; however, therewas a significant difference in means for the interactioneffect of innovation source and community membershipfor brand prominence and brand connectedness, the two

components of brand Attachment. Therefore, hypotheses3b and 4b were supported.

Involving VBC members in product design is effec-tive for bolstering brand attachment within the VBC,however, not necessarily for consumers who are notinvolved with the VBC. In addition, finding significantdifference in the interaction for brand attachment, but notevaluation of product innovation, implies the greaterbrand attachment may be due to crowdsourcing fromwithin the VBC, and not satisfaction with the productinnovation itself. In the present study, the source ofinnovation was specifically indicated to be sourced froma web-based community, which is different from “enthu-siastic” customers, the focus of a sample tested by Fuchsand Schreier (2011) which found opposing results. Tak-ing these two findings together implies that consumersplace a value on the input of enthusiastic consumers, butnot necessarily when the source is from a virtual brandcommunity. Therefore, perception of a web-based brandcommunity may not be as strong as physical (or combina-tion physical and virtual) brand communities, which havebeen found to include highly involved consumers(Algesheimer et al. 2005; Muniz, Jr. and O’Guinn 2001).References are available upon request.

For further information contact:Paul G. Barretta

The University of Texas – Pan American1201 W. University DriveEdinburg, TX 78541–2999

Phone: 570.760.3289Fax: 956.665.2242



THE OMNICHANNEL LUXURY RETAIL EXPERIENCE: BUILDINGMOBILE TRUST AND TECHNOLOGY ACCEPTANCE THROUGH

SYMBOLIC SELF-COMPLETION

Charles Aaron Lawry, University of Arizona, TucsonLaee Choi, University of Arizona, Tucson

SUMMARY

Luxury consumers typically are trendy, affluent andthey tend to be early adopters of technology (Lamb 2012).As such, branded mobile apps, iPad catalogs andsmartphone-optimized websites have become popularwithin the luxury goods sector. For example, LuxuryInsider details ten “must-have” luxury apps, including aCartier application that helps shoppers create custompieces of jewelry and make appointments with personalshoppers (Quek 2011). Such technologies maintain cus-tomer relationships and keep luxury devotees connectedto stores 24/7, thereby digitally expanding the merchan-dize selection and service offerings.

Custom quick response (QR) codes, therefore, maybecome important tools for integrating physical and digi-tal luxury marketing channels. A QR code is a 2-D barcode that shoppers can scan with any camera-equippedcell phone and bar code scanning software. A custom QRcode does not contain a generic black and white pattern,but a full-color brand logo is alternatively embossed intothe code. By placing custom QR codes into store windowdisplays, luxury retailers can unify the affectivity ofluxury shopping with the technical aspects of QR codescanning. Hence, the visual appeal of a window displaywith a custom QR code may increase luxury consumers’acceptance of the custom QR code and build trust inbranded mobile apps or websites.

In the current study, we specifically examine how thecognitive and affective dimensions of luxury shop-pers’ experiences with custom QR codes can relate tomobile trust and QR code scanning. Using SymbolicSelf-Completion theory (Wicklund and Gollwitzer1982), we hypothesize that consumers with hedonicluxury value perceptions (self-gratification, pleasure-seeking, and life enrichment) may use the visual appeal(experiential value) of window displays and customQR codes as props for their self-symbolizing, therebyleading to increased mobile trust and QR code scanning.At the same time, we see whether the perceived ease of use(PEOU) and perceived usefulness (PU) of custom QRcodes can enhance luxury consumers’ mobile trust andQR code scanning behaviors.

This study effectively contributes to luxury con-sumer research and the Technology Acceptance Model(Davis 1989). Luxury research has primarily been con-cerned with the symbolic, emotive and affective dimen-sions of luxury consumption (Dion and Arnould 2011;Okonkwo 2009). Previous studies have supported thataesthetic, affective and hedonic factors are central to m-commerce (Li et al. 2012; Davis 2010). Yet, as moreluxury firms opt into m-commerce, it will be valuable toalso understand how the technical and cognitive aspectsof smartphone and tablet use may compliment the affec-tive aspects of luxury shopping. Technology acceptanceresearch has often emphasized the cognitive componentsof m-commerce (i.e., Wu and Wang 2005). As luxuryfirms develop creative digital marketing strategies, it willbe equally important to know how this affectivity relatesto technology acceptance, use and/or adoption.

In order to address these goals, we recruited a na-tional sample of luxury shoppers (n = 100) from a repu-table online market research panel. Ninety-four percent ofthe respondents earned $100,000 to $200,000 per year,78 percent of the respondents had a college degree, andthere were 51 female and 49 male respondents. An onlinesurvey instrument was used to assess hedonic luxuryvalue perceptions, visual appeal (experiential value),PEOU, PU, mobile trust and BI toward scanning thecustom QR code. Measurement items for hedonic luxuryvalue perceptions were drawn from the hedonic dimen-sion of the luxury value perceptions scale (Wiedmannet al. 2009). Mobile trust was adapted from online trustmeasurement items (Bart et al. 2005). The remainingitems were drawn from established scales within theresearch literature (Mathwick et al. 2001; Davis 1989).

First, respondents were shown an image of a brandedluxury storefront window that contained a custom QRcode. Respondents were directed to use the image whenanswering the questions related to visual appeal (experi-ential value), PEOU, PU, and BI toward scanning thecustom QR code. Then, respondents were shown a corre-sponding image of the branded mobile site and were toldthat it would appear when a shopper scanned the QR code.Respondents were instructed to use the image of thebranded mobile site when answering the questions relatedto mobile trust.


Structural equation modeling (Amos 18.0) was usedfor the data analysis. In order to assess reliability andvalidity, a confirmatory factor analysis was performed onthe data. The item reliability, construct reliability, and theaverage variance extracted reached satisfactory levels(item reliability: >.50; construct reliabilities: .88 to .95;AVE: > .5). The AVE for each construct was greater thanthe corresponding squared interconstruct correlations(SIC), confirming convergent validity and discriminantvalidity (Kline, 1998). Overall, the fit of the CFA modelwas acceptable (CFI = .959; TLI = .946; RMSEA = .078;χ2 = 167.263, df = 104, p < .001).

The SEM path analyses confirmed that the χ2 value ofthe proposed model was significant (χ2 = 179.094, df =109, p < .001) and that the model had an acceptable fit(CFI = .955, TLI = .943, RMSEA = .081). We found thathedonic luxury value perceptions were positively relatedto visual appeal (β = .56, p < .01), PEOU (β = .30, p < .001)and PU (β = .41, p < .001). Findings also supported thatvisual appeal (β = .27, p < .01) and PU (β = .51, p < .01)were positively related to mobile trust. However, therewas no significant relationship between PEOU and mo-

bile trust (β = -.02, p = .897). Lastly, visual appeal (β = .19,p < .05) and PU (β = .64, p < .001) were positively relatedto BI toward scanning the custom QR code. To thecontrary, mobile trust was not significantly related to BItoward scanning the custom QR code (β = -.03, p = .758).

Perceived usefulness and visual appeal both signifi-cantly contribute to mobile trust and the intention to scancustom QR codes within a luxury retail setting. Perceivedusefulness, however, contributes more strongly to QRcode scanning and mobile trust than visual appeal. Whereasaffect and cognition appear to mutually enhance thetechnology acceptance and scanning of custom QR codes,these findings suggest that symbolic self-completion mayplay a lesser role in luxury consumers’ experiences withcustom QR codes. Similarly, perceived usefulness is a keypredictor of mobile trust beyond visual appeal. The mana-gerial implications are that luxury firms should not onlyfocus on creating visually appealing window displays andcustom QR codes. In order to build mobile trust and QRcode scanning, firms also need to ensure that luxuryshoppers understand why QR codes may be useful. Ref-erences are available upon request.

For further information contact:Charles Aaron Lawry

Division of Retailing and Consumer SciencesUniversity of Arizona

650 N. Park Ave.Tucson, AZ 85721–0078

Phone: 520.269.8049Fax: 520.621.9445



CAN YOU HEAR ME NOW? HOW PRODUCT ATTRIBUTESINFLUENCE ONLINE PHONE REVIEWS

Elliott Manzon, University of Michigan, Ann ArborRichard Gonzalez, University of Michigan, Ann ArborColleen M. Seifert, University of Michigan, Ann Arbor

SUMMARY

Prior to making a purchase, many consumers firstseek additional information to aid their decision makingprocess. One of the most trusted sources has traditionallybeen “word of mouth” information (Dierkes, Bichler, andKrishnan 2011). With the growth of the internet andonline retailers such as Amazon.com, consumers areincreasingly seeking product information through con-sumer reviews posted online (Li and Zhan 2011). Fifty-seven percent of respondents were found to consult onlinereviews prior to purchasing consumer electronics (NielsonCompany 2010). Understanding how consumers weighattributes in their product judgments and recommenda-tions is especially important in marketing because theinformation is readily available to other consumers.

We examined cell phone reviews on product websitesin two studies to determine how the attributes of phones(Ease of use, Design, Features, etc.) were differentlyweighted in judgments about overall product ratings andrecommendations to other consumers. In particular, wewere interested in whether the weighting of attributes forone’s own judgments of product quality would differ fromjudgments when recommending the products to others.Product reviews often consist of both overall ratingsrepresenting one’s personal opinion of the product and achoice of, “Would you recommend this product to afriend?” One might expect a rating and the decision torecommend a product to be similar; however, differentattribute weightings may be applied in judgments whenconsumers consider products for themselves vs. others.

Decisions for oneself vs. others have been found tovary in a variety of contexts (Kray and Gonzalez 1999;Beisswanger, Stone, Hupp, and Allgaier 2003; Zikmund-Fisher, Sarr, Fagerlin, and Ubel 2006). When makingdecisions for others, fewer attributes tend to be consid-ered, and the focus tends to be on one prominent attribute.This difference does not appear to be a matter of lowereffort, but rather a belief that another’s preferences will bedifferent than one’s own (Kray 2000). When making adecision for oneself, attributes are more evenly weightedusing a strategy of tradeoffs (Kray and Gonzalez 1999).

In two studies, we examined how the evaluation of aproduct changes depending upon whether a consumer isproviding an overall rating vs. recommending a phone for

others. Data were collected from public online reviews oncommercial websites. For each consumer review, the dataincluded an overall rating on a 1-5 scale, a choice ofwhether the reviewer would recommend the phone toa friend, and separate ratings for 5 product attributes:Design, Ease of Use, Features, Display, and BatteryLife. We hypothesized individual attributes would havediffering importance for ratings compared to recommen-dations. We predicted ratings would be most influencedby ease of use, as this attribute has been found to be amajor post-use influence on product liking (Thompson,Hamilton, and Rust 2005). We expected that recommen-dations are more likely to be influenced by easily observ-able and thus justifiable attributes, such as design andfeatures. We also expected ratings to be influenced bymore attributes, while recommendations would focusmore on one primary attribute. The framework ofBrunswik’s Lens Model (Brunswik 1955) explains howcues are utilized as predictors of different judgment states,and we employ a variant using two dependent variablesfor our analyses.

In Study 1, reviews (n = 720) from 6 different modelsof non-smart phones were recorded. An analysis wasperformed using the SUR framework to account for thebias due to the correlation between observed ratings andrecommendations. For ratings, all 5 attributes were sig-nificant predictors, p < .01. For recommendations, Fea-tures, Ease of Use, Battery, and Design were significant,p < .001, and Display was marginally significant (p =.061) as predictors. Contrary to previous findings, recom-mendations did not rest on a focal feature, and were onlymarginally less likely to weight all of the attributes.

We also analyzed these data using a conditional treealgorithm that takes into account more complicated inter-actions between predictor variables, as well as the ordinalnature of the Likert scales and the binary recommendationvariable. We used the algorithm by Hothorn, Hornik, andZeileis (2006) as implemented in the R package “PARTY.”We included all 5 predictors and used a Bonferronicorrection in order to construct a conservative conditionaltree. The algorithm finds the optimal cut points for rel-evant predictors and indicates the cut points as labels onthe branches.

For ratings, Ease of Use was a first cut variable,followed by Features and Design at the second level. The


model generated a total of 7 nodes using all 5 attributes.For recommendations, Design was the first cut predictor,followed by Ease of Use at the second level. This modelgenerated 5 nodes using 3 predictors. These results indi-cate that consumers are applying a different set of rulesand expectations depending upon whether they are ratinga product for themselves or recommending it for others.

Study 2 replicated Study 1 using the new productdomain of smart phones. One hundred and twenty reviewswere recorded for each of 6 different models of phones. Inthis study, the SUR model analysis found all of theattributes were significant (p < .01) except for Display(p = .94 and p = .21, respectively) for both ratings andrecommendations. The conditional tree analysis resultsdemonstrated that for ratings, Features was the first cutvariable, followed by Ease of Use. The model consisted ofsix nodes using 3 attributes. In the model for recommen-dations, Features was the first cut variable, followed byDesign. The model consisted of 3 nodes and 3 attributes.

These results confirm our hypothesis that a differentset of criteria was applied to recommendations for otherscompared to one’s personal rating of a product. Consistentwith our hypothesis that ease of use would be moreimportant for ratings, Ease of Use emerged as a moreimportant predictor for ratings than for recommendationsin both studies. Also consistent with our hypothesis that

recommendations would focus on easily observableattributes, Design emerged as a more important pre-dictor for recommendations compared to ratings. Themodel for recommendations was also simpler, with fewercut points and fewer attributes in Study 1. This supportsour prediction that recommendations would be based onfewer attributes. The product context (smartphones vs.cell phones) also influenced attribute importance, withFeatures more important for smartphones and Ease of Usemore important for cell phones. This difference mayreflect differing consumer demands in each product cat-egory.

Because the data were collected from a public prod-uct website, there was no attempt to control for differencesamong consumers who entered reviews for each product,though the demographics of gender and age were similar.

For marketers, there is significant value in under-standing the differing ways that attributes can influencejudgments, and how liking a product relates to recom-mending it to others. Given the increasing use of theinternet and online retailers, consumers are increasinglyreceiving product information through consumer reviewsposted online (Li and Zhan 2011). It is of practical importto understand how consumer ratings are determined, andits study also informs theories of human judgment. Refer-ences are available upon request.

For further information contact:Elliott Manzon

University of MichiganB412 East Hall

530 Church StreetAnn Arbor, MI 48109Phone: 269.598.1709



DESIGN OF REWARD SYSTEMS IN CUSTOMERREFERRAL PROGRAMS

Christoph Look, EBS Universität für Wirtschaft und Recht, Germany

ABSTRACT

Referral programs are a marketing instrument to useinterpersonal communication to win new customers. Theauthor reports the result of one experiment to analyze theinfluence of different referral reward system design ele-ments on customer referral behavior. The implications ofthe findings for designing referral rewards systems arediscussed.

INTRODUCTION

Sharing one’s own experiences with products orservices with friends is a common phenomenon and afrequent element of personal communication. (Charlettand Garland 1995) This exchange of information aboutquality, performance and price of goods also plays animportant role in consumers’ purchasing decisions (Priceand Feick 1984; Walker 2001). The key reason for theimportance of personal information sources is the fact thatit is considered as more trustworthy, reliable and lessdriven by commercial motives (Dichter 1966; East,Hammond, and Wright 2007) than other informationsources. In times when consumers are exposed to an everincreasing number of marketing messages (Derbaix andVanhamme 2003) these circumstances positively distin-guish personal communication from other sources ofinformation about products or services in the eyes ofconsumers (Bone 1995). In recent years, marketers havebecome increasingly aware of this positive influence ofpersonal communication on consumer decision processes.As a consequence, the use of personal communication ismore and more considered as an alternative marketinginstrument that requires to be managed proactively (Ryuand Feick 2007). Customer referral programs are the mostwidespread marketing instruments that are designed tomake the advantages of personal communications acces-sible for corporate marketing efforts (Bayus 1985). Still,the interest of marketing researchers in understanding thekey dependencies and relationships of these instrumentshas been fairly limited (Ryu and Feick 2007) despite thelong trace of research concerning personal communica-tion as such. Academic acknowledgment of the impor-tance of personal communication can be traced back to thebeginning of the 20th century (Gilly et al. 1998). In the1950s and 60s the work from Katz and Lazardsfeld (Katz1958) as well as Arndt (Arndt 1967) can be considered asthe root for research into the field of personal communi-cation. Over time numerous research papers have been

developed shedding light on different aspects of thephenomenon of personal communication (Yale 1989).But until the present the use of personal communication inthe context of customer referral programs has only beenaddressed by a limited number of authors (Ryu and Feick2007). One topic that has only received limited attentionin academic research endeavors is the design of the rewardsystem in referral programs. While different authors haveexplored effects of the referral incentive design, e.g.,(Biyalogorsky, Gerstner, and Libai 2001; Kornish andQiuping Li 2010; Ryu and Feick 2007; Tuk et al. 2009;Xiao, Tang, and Wirtz 2011), the influence of key ele-ments of the reward system on customer referral behaviorsuch as the characteristics of the reward itself, the rules forreceiving the reward or the requirements to be allowed tomake a referral are still largely unknown from a marketingresearch perspective. It is the objective of our research tocontribute to closing this research gap by analyzing thepossible influence of different design elements of rewardsystems on customer referral behavior and to expandexisting literature in this area. The scope of our researchis limited to consumers located in Germany and the use ofcustomer referrals for the sale of mobile telecommunica-tion services.

DESIGN ELEMENTS OF REFERRAL REWARDSSYSTEMS IN CUSOMTER REFERRAL

PROGRAMS

One of the prerequisites for analyzing the influenceof the reward system’s design on customer referral behav-ior is an understanding of its functional elements, i.e.,what design elements exist and how are they applied. Wehave used a dyadic approach to acquire this crucial infor-mation of our research. As a starting point, an extensivedesk research was under-taken to identify key designelements of reward systems currently used in existingcustomer referral programs. Through analyzing availableinformation about customer referral programs on thewebsites of 32 German telecommunication companies, afirst comprehensive list of design elements was created.This list of design elements has then been discussed in tenexpert interviews with responsible managers of seventelecommunication companies operating on the Germanmarket. Despite the limited number of expert interviews,we consider the obtained information to be a valid sourcesince the companies represent about 70% of revenue and# of customers in the German market for telecommunica-tion services. As a combined output of our desk research


exercise and the conducted expert interviews, we havebeen able to identify the following seven key designelements of reward systems.

These identified design elements of referral rewardsystems have been the basis for the derivation of ourresearch model.

THEORETICAL BACKGROUND ANDFOUNDATION

Following the concept of theoretical pluralism (Möller2007) we have constructed our research model by refer-ring to multiple theoretical frameworks. In view of theunderlying logic of referral programs – making a referralin exchange for a reward – motivational theories, socialexchange theories and utility theory were apparentlyapplicable to our research topic. Motivational theoriesexplain the action of human beings through the presenceof personal motives, rewards for actions to achieve thesemotives and the likelihood that through actually engagingin these actions one’s motives are achieved (Pinder 1998).Theories of social exchange postulate that social interac-tions are driven by an assessment of possible incurredcosts compared to achievable benefits and only if benefitsare equal or greater than costs will the interaction occur(Gatignon and Robertson 1986). When applying thesetheories to referral behavior, making a recommendationneeds to satisfy a personal motive, be initiated by areward, have a high likelihood for achieving the motive

and the benefits need to outweigh the incurred costs. Pastresearch has identified different psychological motivesfor engaging in customer referrals. Talking about a prod-uct can be a means to overcome post-purchase anxiety, acontribution to manage one’s own perception in the socialenvironment or an expression of care for others throughhelping them make better purchase decisions (Ryu andFeick 2007). Beyond these psychological motives referralprograms apparently aim at stimulating referral behaviorby appealing to financial motives via the rewards. Withreference to utility theory (O’Donoghue and RABIN1999) the reception of monetary rewards is assumed tocreate utility for the referring customer and can thereforeact as an incentive to trigger referral behavior. This line ofthought is the basis for the first hypothesis.

H1: Referral rewards positively influence customerreferral behavior.

As it is a key objective of customer referral programsto benefit from personal relationships for selling products,giving out a reward implicitly assumes that even in one’sclose personal environment, the so called strong ties(Granovetter 1983), rewards have a positive influence onsparking the desired referral behavior. Only appealing tothe aforementioned psychological motives is not consid-ered sufficient to trigger the intended referral behavior.Therefore, the key elements of motivational and socialexchange theory, i.e., the appeal to motives to initiate anaction and the equilibrium between cost and benefits for

Key Design Elements of Referral Reward Systems

Name Explanation

Reward reception The customer who makes a successful referral receives a referral reward

Reward value The reward for a successful referral has a certain monetary value for the referringcustomer

Scale of reward value The value of the reward progressively scales with the number of successfulreferrals

Reward type The reward for a successful referral can be a monetary reward (cash) or a cashequivalent in the form of a physical good

Reward recipient The recipient of the reward can be the referring customer, the referred customeror a combination of both

Reward entitlement The entitlement of the reward describes the relationship between the number ofnecessary referrals and the referral reward

Membership requirement The right to make a referral is open to all customers or is subject to a membershipin the customer referral program


a specific action, are relevant even for strong ties. Acomparable argumentation seems even more directly ap-plicable for referrals toward weak ties – that are peoplefrom one’s extended social environment incl. strangers –since in these cases the presence of psychological motivescan at least be questioned. Therefore to trigger recom-mendations to weak ties, the referral reward representsthe necessary incentive for action and compensates therequired effort. Based on these arguments, we haverefined our first hypothesis as follows.

H1a: Referral rewards positively influence customerreferral behavior toward strong ties.

H1b: Referral rewards positively influence customerreferral behavior toward weak ties.

Looking at the two identified referral reward charac-teristics, referral value and referral type, an argumentationwith similar reference to motivational, social exchangeand utility theory appears equally conclusive.

An increase of the reward value should positivelyinfluence referral behavior since a higher reward valuemore strongly appeals to financial motives for action,represents a higher compensation for incurred costs frommaking a referral and creates higher utility for the refer-ring customers.

H2: Higher referral reward values positively influencethe customer referral behavior.

As a scaling of the reward value beginning with thesecond successful reward is only a further specification ofthis aspect it should also have a positive effect on cus-tomer referral behavior.

H3: A scaling of the reward value has a positive effecton referral behavior.

To extend this argumentation further, higher referralrewards could also be a means to increase referral fre-quency as the incentive for every additional referral ishigher based on the greater utility of the reward.

H4: Higher reward values increase the intention tomake referrals more frequently.

Closely related to influence of the reward value assuch is the means by which the customer receives thereward, i.e., the reward type. Physical goods as rewardsonly exercise their motivational power if they createutility for recipient. The recipient must like or need thespecific product that is offered as a reward. With cashrewards, the recipient has the freedom to use of the rewardto purchase any desired product or can even postpone thepurchasing decision. Therefore, our utility argument leads

to the hypothesis that based on equal monetary values, adirect financial reward should be a stronger incentive forreferral behavior than a physical good.

H5: Customers prefer direct monetary rewards overcash equivalent physical goods.

So far, all discussed design elements primarily influ-ence the motive for engaging in referral behavior throughalterations of the reward characteristics. The remainingdesign elements primarily determine the likelihood toactually receive a reward and the possible risks associatedwith it. Concerning the actual reward recipient, three basicalternatives exist: giving a reward either to the referringcustomer alone, or to the referring customer and the newcustomer or only to the new customer. While the firstalternative is fundamental to all referral programs and thethird option is irrelevant based on our research, the influ-ence of giving a reward also the referred customerdeserves further attention. So far in our argumentationreferral rewards have only been associated with positiveeffects on the recommendation behavior of a customerthrough their utility and the associated motivational influ-ence. But rewards could also pose risks to the referringcustomer which in turn might weaken their function as anaction-inducing incentive. A potential risk could forexample arise in case a referred product fails to meetthe expectations of the referred customer and the purchas-ing decision is regretted. The potential reproach of havingmade a referral primarily to obtain a monetary benefitcould endanger a social relationship and thus be risk forthe referring customer (Folkes 1984). With reference tosocial exchange theory, the cost from endangering a socialrelationship could then outweigh the reward benefits.Giving a reward to all participants could therefore be ameans to reduce the perceived risk of the referring cus-tomer. In addition, the shared reward would also addutility to the purchasing decision for the referred customerso that the likelihood for a successful referral couldincrease.

H6: Customers prefer dyadic reward schemes whereboth involved parties receive a reward.

Next to the benefits from supporting product sales,referral rewards can also be an invitation for opportunisticcustomer behavior. If a customer’s purchasing decisionhas already been taken, obtaining the reward could be theonly motive for the referral. In such a case, the reward stillfulfills its function as a trigger for customer referralbehavior. But from the perspective of the company thatoffers the referral program giving out the reward does notdeliver on the originally intended objective of influencinga purchase decision. The situation is similar to whatagency theory calls a principal and agent problem (Jensenand Meckling 1976). An agent – the referring customer –acts on behalf of the principal – the business – by making


a referral. But the principal does not know if the agent actsin the intended manner or if he pursues other motives(Pratt and Zeckhauser 1985). To reduce the risk of oppor-tunistic behavior, agency theory suggests different instru-ments. The concept of signaling appears especially validin the context of referral programs. In this case, the agentprovides additional information – the signal – to theprincipal about his true intentions (Spence 1974). Appliedto the identified design elements of reward programs, acustomer’s acceptance of reward entitlement schemesgreater than one – which means more than one successfulreferral is required to be entitled to receive a reward – anda required membership could be interpreted as signals asthey both decrease the likelihood of obtaining the reward.

By changing the reward entitlement to a ratio greaterthan one, the customer is forced to make more than onesuccessful referral to receive a reward. The reward losessome of its appeal as an incentive for opportunistic actionsince the likelihood for two successful referrals can beconsidered lower that for only one successful referral.From an exchange theory perspective, the benefits forengaging in the relationship decrease while the necessaryefforts for the customer grow. On the other hand, themotivational effect of the reward for making the necessaryadditional referral should partially outweigh this effect incase of larger reward values since the higher reward valuerepresents a greater utility. Consequently, to counter thenegative effect of the referral entitlement, higher rewardvalues are required to achieve comparable motivationaleffects while reducing the risks of opportunistic behavior.

H7: With higher reward values customers are morelikely to accept reward entitlement schemes greaterthan one.

Similar to reward entitlement schemes, a requiredmembership increases the necessary effort on the side ofthe customer to obtain a reward. The likelihood to receivea reward drops to zero if becoming a member in thereferral reward program is not desired by the potentiallyreferring customer. Therefore, a required membershipcould help reduce opportunistic referral behavior. On theother hand, the additional barrier could negatively influ-ence the desired effect of rewards from a motivationaltheory point of view. A required membership raisesthe perceived cost for engaging in referral behavior inexchange for the reward. Therefore, while membershiprequirements can be a potential means to limit opportunis-tic referrals, they increase the risk of adversely affectingthe attractiveness of a referral program.

H8: The requirement of a membership negatively influ-ences the willingness to make a referral indepen-dently of the reward value.

RESEARCH METHODOLOGY

To verify the formulated research hypothesis chosean experimental design in the form of a post-test-onlycontrol group design. We manipulated the reward value(at 20• , 50• , and 100• ) as well as the requirement for amembership in the referral program (no membershiprequired, membership required) in a 3 x 2 experimentalsetup. The choice of the reward values was primarilydriven by the input from our market research combinedwith the insights from the expert interviews. The experi-ment was conducted with 375 students using a physicalquestionnaire. The questionnaire consisted of three parts,an A4 flyer describing the customer referral program of afictitious German mobile telecommunications company,a page with the instructions of the experiment followed bythe section with the actual questions. The design of theflyer was similar to typical marketing material of commu-nication companies. Both the reward value for a success-ful referral as well as the requirement for a membership inthe customer referral program were manipulated accord-ing to the intended values. Based on the instructions forthe experiment, the students were advised to answer allfollowing questions as if they were a satisfied customer ofthe fictitious mobile operator and had just received theflyer. The questions itself were based on 3-item constructseach with a 7-point Likert scale asking about the partici-pants preference for the described reward design element.The questionnaires were distributed randomly in a class-room and collected from the students directly after comple-tion.

RESEARCH RESULTS

Concerning the impact of a reward reception, partici-pants’ results indicated that for reward values of 100•compared to 20• : a significant positive effect existedregarding the referral behavior (20• : M = 4.36, SD = 1.42;100• : M = 4.83, SD = 1.44; (F(2, 369) = 3.259, p < 0.05).This result is in line with H1 and indicates that referralrewards in general have a positive impact on customers’referral behavior. On the other hand, the results concern-ing the participants’ willingness to recommend to strongties did not show any significant results; an outcomewhich is opposite to H1a. For the willingness to recom-mend to weak ties, the results showed a significant resultfor reward values of 50• compared to 100• (50• : M =4.13, SD = 1.39; 100• : M = 4.60, SD = 1.26; F(2, 369) =4.061, p < 0.05). This result supports H1b and in terms ofthe overall tendency is also in line with H1. The results forthe attractiveness of the reward value showed significanteffects in all cases; for increases of the reward value from20• to 50• or 100• and for increases from 50• to 100•(20• : M = 3.74, SD = 1.51; 50• : M = 4.95, SD = 1.48;100• : M = 5.65, SD = 1.25; F(2, 369) = 57.165, p < 0.01).This result supports H2 and the link between the


reward value and its attractiveness. The scaling ofreferral rewards resulted a positive effect in case of arequired membership (No membership: M = 4.81, SD =1.57; with membership: M = 5.16, SD = 1.34; F(1, 369) =5.221, p < 0.05) but no significant effects depending onthe manipulation of the reward value. This outcome iscontradicting H3. Concerning the referral frequency, theresults show a significant effect of the increase in rewardvalue in two cases, an increase from 20• to 50• as well asfrom 20• to 100• (20• : M = 3.90, SD = 1.56; 50• : M =4.28, SD = 1.60; 100• : M = 4.42, SD = 1.41; F(2, 369) =3.897, p < 0.05). H4 is supported by this result. For theimpact of the reward type, the results showed two effectsin favor of direct cash rewards. An increase of the rewardvalue from 20• to 100• as well as from 50• to 100•showed a significant positive effect (20• : M = 5.33, SD =1.66; 50• : M = 5.43, SD = 1.57; 100• : M = 5.82, SD =1.31; F(2, 369) = 3.649, p < 0.05). H5 is supported by theseresults as participants showed an increasing preferencefor direct cash rewards depending on the reward value.Regarding the participants preference for the rewardrecipient there is a significant effect in case of no member-ship (No membership: M = 4.74, SD = 1.58; membership:M = 4.35, SD = 1.61; F(1, 369) = 5.449, p < 0.05). Thisresult is in contradiction to H6. In addition, its theoreticalexplanation proves difficult since neither a motiva-tional theory nor an exchange theory approach providea reasonable explanation for this outcome. The result fora possible reward entitlement scheme did not show sig-nificant effects for the change in referral reward or mem-bership requirement. Therefore, H7 was not supported.Apart from the two effects concerning the scaling of thereferral rewards and the reward recipient, no signifi-cant effects could be observed regarding the require-ment for membership in the customer referral program tobe allowed to recommend. Therefore, H8 was also notsupported.

DISCUSSION AND FURTHER OUTLOOK

The results of our research provide several insightsinto how different design elements of referral rewardsystems should be selected and designed to positivelyinfluence customer referral intentions. Foremost, ourresults support the widespread use of referral rewardsas a means to positively influence customers’ referralbehavior in general (H1). Surprisingly though, a positiveeffect could only be observed for the difference betweenthe smallest (20• ) and the largest (100• ) reward value,which can be interpreted as a sign that rewards alone maynot be sufficient to spark referral behavior. This interpre-tation is further underlined by the lack of a significantdifference in customer referral intentions toward strongties (H1a). The monetary value of the referral reward doesnot seem to be an influential factor for making referrals tostrong ties. On the other hand, our results showed a

significant difference between 50• and 100• rewards forreferral intentions to weak ties (H1b). In addition, partici-pants claimed that higher reward values would increasetheir intention to recommend more frequently (H4). Takentogether, the target group for the referral attempts seem-ingly has a strong influence on whether rewards doexercise positive effects on customer referral intentions.Choosing the right reward value depends on the desiredtarget for group for referrals.

Generally though, the attractiveness of rewards growswith a higher reward value (H2) so that an increasingeffect from the reward on referral intentions can beassumed. In contrast to this result, the scaling of rewardvalues did not produce significant results for changes inreward value (H3). The potentially higher reward valuefor the second or third successful referral does not seem toplay a role in the participants’ intention about making areferral or not. This could be an indication that customersdo not anticipate more than one or two successful refer-rals. In deciding between direct cash awards or cashequivalents, the decision should favor direct cash rewards(H5).

As the design elements of reward recipient and re-ward entitlement did not show significant results in rela-tion to the change in reward value (H6 +7), their usageshould be carefully thought through. Giving a referralreward to the referring customer and the referred newcustomer seems to bear only limited risks for the rewardeffectiveness under the assumption that the customer’sreferral intention does not decrease if also the referredperson receives a reward. Our previous theoretical argu-ments are also in support if this assessment since thereward for the referred customer does not negativelyinfluence the effects of the reward for the referring cus-tomer. The utility from a reward remains unchanged andthe appeal to financial motives as well as the benefit frommaking a referral does not change.

The use of reward entitlement ratios greater than one,on the other hand, could be interpreted as a considerabledeterioration to the attractiveness of a referral programfrom the perspective of the referring customer since therequirements for receiving a reward increase substan-tially. As a consequence, reward entitlement schemespose the risk of rendering a referral program so unattrac-tive that potential gains from reducing opportunisticbehavior are negatively outweighed by the reductionin the number of acquired customers because of theoverall lower number of referrals.

The fact that the requirement for a membership in thecustomer referral program did not show significant nega-tive effects (H8) is surprising since this result is in contra-diction to overarching opinion of the expert interviews.


The majority of the interviewed experts stated that theyrefrain from have a membership for the referral programto avoid this additional barrier for making referrals. Ap-parently, for our participants the membership requirementwas not influential when assessing their referral intentionin the experimental setup.

The results of our research lead to additional researchquestions and are in parallel subject to certain designlimitations. The lack of significant results for referral

intentions to strong ties directs the research attention tothe search for other factors that are potentially moreinfluential than the reward value. In view of the resultsother referral research work (Ryu and Feick 2007), suchfactors could possibly be customer satisfaction, productinvolvement or opinion leadership. To overcome thelimitations of our research design regarding its geographi-cal focus, the focus on the telecommunication industryand the usage of a student sample, similar endeavorsshould be undertaken with an adapted research design.

REFERENCES

Arndt, J. (1967), “Word of Mouth Advertising and Infor-mal Communication,” in Risk Taking and Informa-tion Handling in Consumer Behavior, D.F. Cox, ed.Boston, 188–239.

Bayus, B.L. (1985), “Word of Mouth: The Indirect Ef-fects of Marketing Efforts,” Journal of AdvertisingResearch, 25 (3), 31–39.

Biyalogorsky, Eyal, Eitan Gerstner, and Barak Libai(2001), “Customer Referral Management: OptimalReward Programs,” Marketing Science, 20 (1), 82–95.

Bone, P.F. (1995), “Word-of-Mouth Effects on Short-term and Long-term Product Judgements,” Journalof Business Research, 32 (3), 213–23.

Charlett, Don and Ron Garland (1995), “How DamagingIs Negative Word of Mouth?” Marketing Bulletin, 6,42–50.

Derbaix, C. and J. Vanhamme (2003), “Inducing Word-of-Mouth by Eliciting Surprise – A Pilot Investiga-tion,” Journal of Economic Psychology, 24 (1), 99–116.

Dichter, E. (1966), “How Word-of-Mouth AdvertisingWorks,” Harvard Business Review, 44 (6), 147–66.

East, Robert, Kathy Hammond, and Malcolm Wright(2007), “The Relative Incidence of Positive andNegative Word of Mouth: A Multi-Category Study,”International Journal of Research in Marketing, 24(2), 175–84.

Folkes, Valerie S. (1984), “Consumer Reactions to Prod-uct Failure: An Attributional Approach,” Journal ofConsumer Research, 10 (4), 389–409.

Gatignon, Hubert and Thomas S. Robertson (1986), “AnExchange Theory Model of Interpersonal Communi-cation,” Advances in Consumer Research, 13 (1),534–38.

Gilly, Mary C., John L. Graham, Mary F. Wolfinbarger,and Laura J. Yale (1998), “A Dyadic Study of Inter-personal Information Search,” Journal of the Acad-emy of Marketing Science, 26 (2), 83–100.

Granovetter, Mark (1983), “The Strength of Weak Ties:A Network Theory Revisited,” Sociological Theory,1, 200–33.

Jensen, Michael C. and William H. Meckling (1976),“Theory of the Firm: Managerial Behavior, AgencyCosts and Ownership Structure,” Journal of Finan-cial Economics, 3 (4), 305–60.

Katz, Elihu (1958), “The Two Step Flow of Communica-tion: An Up-to-Date Report on an Hypothesis,” ThePublic Opinion Quarterly, 22, 61–78.

Kornish, Laura J. and Qiuping Li (2010), “Optimal Refer-ral Bonuses with Asymmetric Information: Firm-Offered and Interpersonal Incentives,” MarketingScience, 29 (1), 108–21.

Möller, Kristian (2007), “Marketing Research Traditions:Toward Theoretical Unification or Pluralism? Pro-ceedings of the 2006 EMAC-ANZMAC ResearchSymposium,” Australasian Marketing Journal (AMJ),15 (1), 61–66.

O’Donoghue, Ted and Matthew Rabin (1999), “Doing ItNow or Later,” American Economic Review, 89 (1),103–24.

Pinder, Craig C. (1998), Work Motivation in Organiza-tional Behavior. Upper Saddle River, NJ: PrenticeHall.

Pratt, J.W. and R.J. Zeckhauser (1985), “Principals andAgents: An Overview,” in Principals and Agents:The Structure of Business, John W. Pratt, ed. Boston,MA: Harvard Business School Press, 1–35.

Price, Linda L. and Lawrence F. Feick (1984), “The Roleof Interpersonal Sources in External Search: AnInformational Perspective,” Advances in ConsumerResearch, 11 (1), 250–55.

Ryu, Gangseog and Lawrence Feick (2007), “A Penny forYour Thoughts: Referral Reward Programs and Re-ferral Likelihood,” Journal of Marketing, 71 (1), 84–94.

Spence, A.M. (1974), Market Signaling: InformationTransfer in Hiring and Related Processes. Cam-bridge, MA: Harvard University Press.


Tuk, Mirjam A., Peeter W.J. Verlegh, Ale Smidts, andDaniel H.J. Wigboldus (2009), “Sales and Sincerity:The Role of Relational Framing in Word-of-MouthMarketing,” Journal of Consumer Psychology(Elsevier Science), 19 (1), 38–47.

Walker, L.J.H. (2001), “The Measurement of Word-of-Mouth Communications and an Investigation of Ser-vice Quality and Customer Commitment as Potential

Antecedents,” Journal of Service Research, 4 (1),60–75.

Xiao, Ping, Christopher S. Tang, and Jochen Wirtz (2011),“Optimizing Referral Reward Programs Under Im-pression Management Considerations,” EuropeanJournal of Operational Research, 215 (3), 730–39.

Yale, Laura J. (1989), Individual Differences in Interper-sonal Source Selection.

For further information contact:Christoph Look

EBS Universität für Wirtschaft und RechtStrategic Marketing

Gustav–Stresemann-Ring 365189 Wiesbaden

GermanyPhone: +4917610295210



THE POWER OF A TWEET: AN EXPLORATORY STUDY MEASURINGTHE FEMALE PERCEPTION OF CELEBRITY

ENDORSEMENTS ON TWITTER

Nicole Cunningham, Texas Christian University, Fort WorthLaura Bright, Texas Christian University, Fort Worth

ABSTRACT

Social media websites such as Twitter and Facebookprovide consumers with direct access to celebrities therebyincreasing the interactivity of the fan-celebrity relation-ship. This study expands on celebrity endorsement litera-ture in an effort to measure the effectiveness of celebrityendorsements on Twitter.

INTRODUCTION

Where celebrities go, fans will follow. This applies tosocial media as much as it does in the real world. In thepast, celebrities sought promotion through magazines andtelevision appearances. Today, celebrities are turning towebsites like Facebook and Twitter to promote theirbrands and connect with fans. Social media has revolu-tionized the way fans interact with celebrities by provid-ing direct access and removing the middleman that istraditional media. This gives celebrities the freedom topromote themselves or their sponsors directly to fans atanytime. This two-way path enables a direct relationshipbetween celebrity and fan, providing a lucrative opportu-nity for marketers and sponsors.

Celebrities like Eva Longoria have turned their act-ing careers into multi-million dollar endorsement deals.For example, Eva Longoria has endorsed Bebe Sport,L’Oreal, and Microsoft among others (“Eva Longoria,Actress – Celebrity Endorsement Ads,” 2011). Accordingto Forbes magazine, Eva Longoria earned approximately$13 million from May 2010 to May 2011, but it was herlucrative endorsement deals with L’Oreal and LG thathelped push her to the top of the list of “Highest Paid TVActresses” (Pomerantz 2011). Sometimes companies willrecruit celebrities from different industries to team up ina star-studded attempt to sell their products. Singer andactor Justin Timberlake and NFL quarterback PeytonManning appeared in a series of ads for Sony (Akhtar2011). It is believed commercials of this sort will mostlikely get viewers to take notice of Sony due to thepresence of such popular celebrities (Akhtar 2011). More-over, research shows celebrity endorsers can affect con-sumer product choice (Agrawal and Kamakura 1995) andproduct trial (Miciak and Shanklin 1994). Therefore, it islikely that marketing and advertising practitioners willcontinue to spend millions of dollars on celebrity endorse-

ments to enhance the effectiveness of persuasive commu-nication efforts.

As companies continue to make large investments incelebrity endorsers, they are also dedicating more moneyand resources to grow their online presence. A studyconducted by the University of Massachusetts DartmouthCenter for Marketing Research revealed that 62 percent ofFortune 500 companies used Twitter in 2011 and 58percent used Facebook (Barnes and Andonian 2011).Companies are investing more money into online adver-tising on social networks than ever before. It is estimatedthat nearly 12.1 percent of spending on online advertisingin the United States will be allocated to social networks in2012 (Williamson, 2011b). Research firm E-Marketerforecasts that worldwide spending on social networkadvertising will reach almost $6 billion in 2011, with $4billion of that going to Facebook and $150 million toTwitter (Williamson 2011a).

When selecting celebrities for endorsements on Twit-ter, marketers should make an extended effort to target thefemale audience. Several studies have identified a gendergap on Twitter (Heil and Piskorski 2009; Solis 2010). In2010, 57 percent of Twitter users were female (Solis2010). This trend seems to expand across other socialmedia platforms. According to the most recent Nielsenreport, “Females make up the majority of visitors to socialnetworking sites (SNSs) and blogs, and people aged 18–34 have the highest concentration of visitors among allage groups” (The Nielsen Co. 2011). Women also spendsignificantly more time on social networking sites thanmen, with women averaging 5.5 hours per month com-pared to men’s 4 hours (Abraham, Morn, and Vollman2010). This demonstrates the strong engagement thatwomen across the globe share with social sites. The focuson women college students allowed the researchers to usea convenience sample in this exploratory study and assistin the formation of new hypotheses to be used in asubsequent study focusing on the differences betweengenders in response to celebrity endorsements on Twitter.

This is an exploratory study that will extend pre-vious research on celebrity endorsements to exploretheir effectiveness in social media, specifically onTwitter. As advertisers continue to pour moreresources into advertising on social networking sites,


it is likely they will tap into celebrity endorsements,much like they have with traditional media outlets. Theaverage consumer on Twitter has approximately 300followers, while a top 500 celebrity has 300,000 or more(360i, 2010). Therefore, by activating one celebrity onTwitter, a marketer can extend his reach 1000x more thanwhat is possible with the average consumer (360i, 2010).By activating just one celebrity, the marketer would beable to achieve the same reach as he would have byencouraging 1,000 consumers to tweet about his brand(360i, 2010). Thus, this study is important for advertisingpractitioners who seek to utilize celebrities to endorseproducts on Twitter.

LITERATURE REVIEW

Social Media Adoption

In less than three years, social media has become themost popular activity on the Internet. By the end of 2008,social networking sites like Facebook, Twitter, andMySpace surpassed email in terms of reach (Ostrow2009). This is beneficial for marketing and advertisingpractitioners seeking to use social media as a means ofcommunicating with consumers. As mentioned, socialnetworking sites have been praised for providing instantaccess to celebrities. Social media gives celebrities aplatform where they can talk (or type) in their ownunedited words to directly connect to fans (Corazza 2009).

Uses and Gratifications literature implies individualsuse media to gratify their needs and wants (Katz, Blumler,and Gurevitch 1973; Rubin 1983). Gangadharbatla (2009)identified six reasons users will adopt social media: desireto belong, desire to communicate, entertainment, desire toseek information, satisfy commercial needs, and self-expression. U&G literature suggests media can servefunctions like “connecting people . . . with different kindsof others (self, family friends, etc.),” (Katz et al. 1974,p. 63). The desire to stay in touch and make new friendsis based on a basic need to belong (Gangadharbatla 2009).In addition, the desire to belong results in increasedcommunication and SNSs fill the gap by providing addi-tional access points to stay in touch with others(Gangadharbatla 2009). This is evident in how celebritiesand fans use social media. Celebrities want to stay in touchwith their fans and fans want to be a part of the celebrity’sexclusive world. Papacharissi and Rubin (2000) claimindividuals use the Internet to communicate with others asan alternative to face-to-face contact. Social media is theideal ground for this form of communication. A LadyGaga fan living in Nebraska is not likely to run into herfavorite singer, but she can “follow” one of her favoritecelebrities on a social media site and have a direct line ofcommunication to that celebrity.

One important factor that surfaces with computer-mediated communication is the interactivity feature.Interactivity is not present in other media outlets and canbe characterized by three factors: it is (1) multiway (itinvolves two or more actors), (2) immediate (responsesoccur within seconds), and (3) contingent (response ofone actor follows directly and logically from the action ofanother) (Alba et al. 1997). Interactivity in social mediaallows fans to ask questions and receive responses fromtheir favorite celebrities. This, of course, includes follow-ing their favorite celebrity to see what he or she said aboutthe latest rumor in Hollywood. Fans want this kind ofinformation and social media allows them to receive itdirectly from the celebrity. “The higher the need forinformation, the higher the likelihood they adopt SNS”(Gangadharbatla 2009, p. 11).

RQ1: Due to the interactive component of Twitter, dowomen students feel more involved with the celeb-rities they follow on Twitter?

RQ2: Because users choose who they want to follow onTwitter, do women students find celebrity endorse-ment tweets relevant?

RQ3: Do women students’ who use Twitter to connectwith celebrities have a positive attitude towardcelebrity endorsements on Twitter?

SNSs are often used to satisfy commercial needs,such as searching for product information or postingadvertisements (Gangadharbatla 2009). “If individualsperceive SNSs as ideal channels for commercial activi-ties, they are more likely to adopt and use SNSs,”(Gangadharbatla 2009, p. 11). Celebrities using socialmedia to endorse products would fall under commercialactivities. Consequently, these endorsements can also beself-promotion. One example is reality star Kim Kardashiantweeting about seeing her Kardashian Kollection on therunway during an episode of “America’s Next Top Model”(KimKardashian 2011). It is evident by the literature thatboth celebrities and fans engage in social media forvarious reasons. Moreover, both parties benefit from therelationships and interactions created through this plat-form.

Social Media Engagement

Once a user has adopted social media, he or shecan begin to engage in social media practice. The termengagement has several connotations and meaningsdepending on the circumstance. The AdvertisingResearch Foundation defined engagement as “turning ona prospect to a brand idea enhanced by the surroundingcontext” (Weigold and Pulizzi 2010, p. 5). Marketing


research company, Forrester described it as “the level ofinvolvement, interaction, intimacy, and influence that anindividual has with a brand over time” (Haven and Vittal2008, p. 3). The Forrester definition has four majorcomponents: involvement, interaction, intimacy, and in-fluence (Haven and Vittal 2008, p. 3). While celebritiescan assume the role of a brand, the relationship establishedon Twitter between celebrities and fans is not solelybased on marketing purposes. Whether they are askingfans for restaurant suggestions or answering theirquestions, celebrities use Twitter to connect and inter-act with fans on a more personal level. For the purpose ofthis study, engagement will be defined as the act ofpurposefully choosing to interact with other users througha social media website and actively participating in theonline community.

Social Media and Persuasion Knowledge

Despite diminished effectiveness, advertisementscontinue to reign as the favored method of promotion(Sliburyte 2009). One explanation for diminished effec-tiveness can be attributed to the Persuasion KnowledgeModel (PKM). PKM claims consumers learn about per-suasion through various outlets, i.e., firsthand experi-ences in social interactions with friends and family orparticipating in conversations about how consumers’thoughts and behaviors can be influenced (Friestad andWright 1994). As a consequence, the effects of actionsperformed by persuasion agents on consumers’ attitudesand behavior will also change. This is because a consumer’spersuasion knowledge shapes how they respond as “per-suasion targets,” (Friestad and Wright 1994). Accordingto Friestad and Wright (1994), “consumers’ persuasioncoping knowledge enables them to recognize, analyze,interpret, evaluate, and remember persuasion attemptsand to select and execute coping tactics believed to beeffective and appropriate” (p. 3).

Consumer Skepticism

One way consumers cope with persuasion attempts isby exhibiting consumer skepticism (Hardesty, Carlson,and Bearden 2002). One study revealed respondents indi-cated some degree of skepticism regarding celebrity en-dorsements (Bailey 2007). The respondents indicatedcelebrity endorsers would have little impact on consumersbecause they, and other consumers, do not buy productssimply because celebrities endorse them (Bailey 2007).Respondents in two other studies shared similar senti-ments and expressed doubt that celebrities used, or evenliked, the products they endorsed (Tripp, Jensen, andCarlson 1994). They also indicated that celebrities tookpart in endorsements because they were paid for them(Tripp, Jensen, and Carlson 1994). Despite participants

demonstrating consumer skepticism, there was no evi-dence it diminished consumers’ attitudes and perceptions.

The PKM will be utilized to explore whether socialmedia users are predisposed to persuasion attempts bycelebrities. Because users consciously make the decisionto “like” something on Facebook or “follow” a celebrityon Twitter, social media could be classified as highengagement. Consumers with high engagement activelysearch and follow their favorite celebrities or brands andit is possible they go into the persuasion episode knowingwhat type of information or persuasion attempts may bepresent. The PKM can also be applied to determine theeffectiveness of sponsored versus personal (testimonial)messages delivered by celebrities. It is possible consum-ers may have built up a tolerance to sponsored messages.In that case, personal messages would carry more influ-ence, even if the product is not congruent with the celeb-rity.

Celebrity Endorsement in Traditional Media

Advertisers are often faced with the challenge ofmaking advertisements noticeable and more attractive forconsumers. In the past, this has been achieved throughcelebrity endorsements. Empirical evidence indicatesapproximately 20–25 percent of advertisements involve acelebrity as an endorser (Sliburyte 2009). According toFriedman and Friedman (1979), celebrity endorsers areactors, athletes, or entertainers who are known to thepublic for their achievements. According to Atkin andBlock (1983), celebrity endorsers are considered to be“highly dynamic, with attractive and engaging personalqualities,” (p. 57). Research revealed several characteris-tics of a celebrity endorser that impact the effectiveness ofa message, including source credibility and the match-uphypothesis (celebrity-product congruence) (Kim and Na2007; Sliburyte 2009).

Source credibility is a key factor, because a crediblesource can influence opinions and consumer behaviorthrough the internalization process (Kelman 1961). Thesource credibility model suggests the effectiveness of amessage depends on the level of expertise and trustwor-thiness of an endorser (Kim and Na 2007). As defined byLafferty and Goldsmith (1999), credibility is the “extentto which the source is perceived as possessing expertiserelevant to the communication topic and can be trusted togive an objective opinion on the subject,” (p. 43). In thisdefinition, expertise is obtained through knowledge of thesubject and trustworthiness refers to the honesty andbelievability of a source (McGinnies and Ward 1980).Some authors argue perceived expertise of a celebrityendorser can affect purchase decision more than sourceattractiveness or any other factor (Ohanian 1990).


RQ4: Do women students find celebrity endorsementson Twitter credible?

Due to the conversational nature of social media, fansmay perceive endorsements as more credible and truthfulthan advertisements in traditional media. “Social tech-nologies have revved up that word-of-mouth dynamic,increasing the influence of regular people while dilutingthe value of traditional marketing,” (Li and Bernoff 2008,p. 102). If word-of-mouth through social media increasesthe influence of regular people, then celebrities shouldalso see the same effect with their endorsements. Somecelebrities have even kept their endorsement efforts strictlyto social media. For example, instead of appearing in atelevision commercial, former reality star turned designerLauren Conrad pushes her Paper Crown clothing linethrough her Twitter account (Conrad 2011). She recentlyannounced she would be launching a beauty line based onher website “The Beauty Department” (Conrad 2011a).Word-of-mouth endorsements are also more personal,which means fans may place more value on a celebrity’sendorsements. This should also be the case if the endorsedproduct falls into the celebrity’s realm of expertise. Forexample, Lauren Conrad often posts links to her websitewhich features products from other brands, like J Crewand H&M (Conrad 2011b). As a clothing designer, aconsumer can assume Conrad is well-versed in what isconsidered stylish and fashionable.

Conrad’s role as a clothing designer and style iconmatches with the products she is endorsing – clothing andbeauty products. Kamin’s (1990) match-up hypothesisproposes endorsers are more effective when there is acorresponding relationship between the endorser and theproduct. In their research, Kim and Na (2007) discoveredcredibility is important when there is a congruent relation-ship between the celebrity endorser and the endorsedproduct. On the other hand, source attractiveness provedto be more important than credibility when the fit betweenthe celebrity endorser and the endorsed product wasincongruent. Participants had more favorable attitudestoward the endorsed product when the celebrity endorserwas congruent with the endorsed product (Kim and Na2007).

Because social media eliminates traditional media asthe middleman, celebrities have the freedom to promotewhatever and whenever they want. This includes non-sponsored products and services, as well. NFL playerChad “Ochocinco” Johnson connects with his fans bygiving away several pairs of Christian Louboutin shoes tohis female followers. (Ochocinco 2011). ChristianLouboutin is a high-end luxury shoe brand known for itsred soles and appearances on Sex and the City. Therefore,Johnson does not “match up” with this particular endorse-ment.

RQ5: Do women students have a more positive attitudetoward personal or sponsored celebrity endorse-ments on Twitter?

Social media has revolutionized the way peopleinteract with celebrities. Removing the media middle-man gives celebrities the freedom to promote anything attheir own discretion. Based on the prevalence of celebrityendorsements (“Celebrity Ad Endorsements” 2011), it isclear employing celebrities for endorsements is still apopular strategy among companies. Previous researchindicated several factors critical in the effectiveness ofusing celebrity endorsements, including source credibil-ity and match-up hypothesis (celebrity-product congru-ence) (Kim and Na 2007; Sliburyte 2009). However,research to date has been limited to traditional mediaoutlets. However, the popularity and widespread use ofsocial media has made it critical for marketing and adver-tising practitioners to understand the best way to utilizesocial media. This research attempts to merge both strat-egies for the sake of creating effective persuasion commu-nication in a new media outlet.

METHOD

Design

A 109 item online questionnaire was fielded amongstudents from a major university in the Midwestern regionof the United States as well as social media websites (e.g.,Facebook, LinkedIn, and Twitter). In addition to informa-tion regarding attitude toward celebrity endorsements onTwitter, demographic data was gathered from all par-ticipants. Data was collected for a three-week period(November 15th – December 6th, 2011).

Sample

Study participants were undergraduate students whowere enrolled in a variety of courses, including digitalmedia, advertising, and communication sciences. Stu-dents were informed that participation in the study wascompletely voluntary and were offered extra credit forcompleting the survey. To supplement the student sample,a snowball technique was used to recruit respondentsfrom social media websites such as Facebook, Twitter,and LinkedIn. Participants were screened for gender andmedia usage.

The final sample (N = 113) was comprised of femalestudents who have Twitter accounts and follow celebri-ties. One hundred percent of the sample was between theages of 18–30. Participants were predominantly Cauca-sian (81%) with incomes less than $20,000 a year (54%).While this sample is not representative of Americanconsumers, it is consistent with the demographics of the


population on Twitter. Specifically, 45 percent of onlineyoung adults between the ages of 18–34 use Twitter(Quantcast 2010). According to Quantcast (2012), 67percent of Twitter users are Caucasian and nearly 51percent have at least some college education. In addition,females are the most active users of Twitter (Nielsen2011) and account for 57 percent of Twitter users (Solis2010).

Measures

The questionnaire consisted of six major sections thatassessed: (1) attitude toward celebrity endorsements onTwitter, (2) attitude toward testimonial endorsements onTwitter, (3) perceived expertise, (4) perceived trust, (5)involvement, and (6) endorsement relevance. The mea-sure of attitude toward celebrity endorsements used anestablished six-item, four-point Likert-type scale fromHenthorne, LaTour, and Nataraajan (1993). The mea-sures for perceived trust and expertise were created fromOhanian’s (1990) fifteen-item, seven-point semantic dif-ferential scale to evaluate celebrity endorsers. The mea-sure of involvement used an established three-item, seven-point semantic differential scale (Swinyard 1993). Themeasure of endorsement relevance consisted of an eight-item Likert-type scale from Andrews and Durvasula(1991). The measure for attitude toward celebrity testimo-nial tweets used an established four-item, seven-pointsemantic differential scale from Feick and Higgie (1992).

The questionnaire also featured an open-ended ques-tion that asked respondents to list a celebrity endorser theyfollow on Twitter and what product or service he or shewas endorsing. It also collected information about thenumber of minutes they spend on other social mediawebsites, the number of social media sites to which they

belong, how many followers they have on Twitter, howmany people they follow on Twitter, and which types ofaccounts (e.g., celebrity, athlete, politician, brand, ornews organization) they follow on Twitter. Finally, thesurvey concluded with demographic questions regardingthe respondent’s age, gender, and ethnicity. Respondentstook approximately 25 minutes to complete the entiresurvey. To maintain anonymity, the questionnaires didnot collect the respondents’ names.

RESULTS

Data Analysis

Because the sample was only women students from amajor university, this is reflected in the demographics. Ofthose who participated in the study, 81 percent wereCaucasian, 9 percent were Hispanic American, and 5percent were African American. Of those who responded,66 percent had some college, but no degree. However only1.5 percent responded that they had some graduate schoolexperience, but 16 percent did report that they had at leasta bachelor’s degree.

The reliability assessment of all scales uses Cronbach’salpha, and all exceed the generally accepted guideline of.70 (Hair et al. 1998). Table 1 summarizes the meanscores, variances, and reliability indices.

Because this is an exploratory study, only descriptivestatistics were used. This is intended to provide the re-searchers with a general idea of how women studentsrespond to the celebrity endorsements they encounter onTwitter. Data collected will be used in a future studyexamining attitude toward advertising among Twitterusers.

TABLE 1Mean, Variance, and Cronbach’s Alpha Coefficients

Scale Mean SD A

Perceived Expertise 4.56 1.22 .938

Perceived Trust 4.66 1.17 .873

Endorsement Relevance 3.48 1.23 .955

Involvement 3.72 1.48 .906

Attitude toward Celebrity Endorsement 2.34 .56 .810

Attitude toward Celebrity Testimonial 4.27 1.34 .917


RQ1. In response to RQ1, it appears that women studentTwitter users do not feel more involved whenfollowing celebrities on Twitter. The mean for theinvolvement measure equaled 3.72 (SD = 1.48). A95 percent confidence interval for the mean is 3.42to 4.08.

RQ2. Along the same lines as RQ1, respondents do notfeel that the celebrity endorsements they encounteron Twitter are relevant to their personal needs orinterests. The relevance measure (M = 3.48, SD =1.23) indicates a low-moderate response to therelevance of celebrity endorsements on Twitter.On the relevance measure, responses have a meanyield of 3.20 to 3.78.

RQ3. The attitude toward celebrity endorsements mea-sure (M = 2.34, SD = .56) indicates that womenstudents do have a moderately positive attitudetoward celebrity endorsements they encounter onTwitter. On a four-point scale, the mean hoversbetween “No” and “Yes.” A 95 percent confidenceinterval for the mean is 2.22 to 2.46.

RQ4. Two variables were used to measure credibility incelebrity endorsements on Twitter – perceivedexpertise and trust. While both are key componentsto credibility, perceived trust (M = 4.67, SD = 1.17)scored higher than perceived expertise (M = 4.56,SD = 1.22). Responses on the perceived expertisemeasure have a mean yield of 4.22 to 4.79.Responses on the perceived trust measured havea mean yield of 4.35 to 4.89. This implies thatwomen students trust celebrity endorsements onTwitter, but hold slight reservations about theexpertise of the celebrity endorser.

RQ5. In comparing the means for the attitude towardcelebrity endorsement measure and the attitudetoward celebrity testimonial measure, testimonialsheld a very slight edge. The testimonial measure(M = 4.27, SD = 1.34) was slightly above thescale’s midpoint. This indicates that respondentsare not completely convinced by testimonials bycelebrities, but find them slightly more convincingthan paid celebrity endorsements. A 95 percentconfidence interval for the mean is 3.92 to 4.53.

IMPLICATIONS AND CONCLUSION

The effectiveness of celebrity endorsements has beenstudied in many different ways. However, most of thesestudies have focused on traditional media outlets such asprint and television. This study differs in the sense that itfocuses on social media, which allows the celebrityendorsers to interact with their fans in a more direct

way. The ability to interact with one’s favorite actor orathlete provides a much different experience than lookingat a magazine ad or watching a 30-second ad on television.However, respondents indicated that using Twitter tocommunicate with their favorite celebrity did not makethem feel more involved with that person. On the contrary,respondents did indicate a positive attitude toward celeb-rity endorsements. However, the respondents did not feelthat the endorsements were relevant to their personalneeds or interests. The results indicate that the celebritiesmust be more engaging with their followers if they wishto be effective endorsers. Perhaps if the celebrity showsmore interest in his or her followers, consumers may findtheir tweets and endorsements more relevant.

Source credibility is a key factor in celebrity endorse-ments because it has the ability to influence opinions andconsumer behavior through the internalization process(Kelman 1961). Respondents in this study indicated thatcelebrity endorsers on Twitter are perceived as moretrustworthy than experts at dealing with a certain productor service. Li and Bernoff (2008) believe that socialtechnologies have increased the word-of-mouth dynamic.Therefore, it is likely that the conversational nature ofTwitter causes celebrity endorsers to be perceived as moretrustworthy. This is in turns increases credibility andcould potentially impact purchase intent or product trial.

LIMITATIONS AND FUTURE RESEARCH

Several limitations exist for this research endeavor.The most glaring limitation of this study is the use of aconvenience sample. Unfortunately, the researcher didnot have a comprehensive list of all Twitter users torandomly select from. However, one possible solutionmay be to utilize a survey panel for future research.Additionally, another significant limitation is the focus onwomen students and the use of only descriptive statistics.Again, this is an exploratory study to examine the generalresponse to celebrity endorsements on Twitter.

The data collected from this study will be used toinform a much larger study focusing on many differentaspects of celebrity endorsements found in more tradi-tional outlets and how they translate into an online plat-form like Twitter. One aspect to be studied is the effect ofinformation overload. The Limited Capacity Model (LCM)assumes that individuals have a limited capacity to pro-cess information (Lang 2000). A Twitter user can followa combination of people and be inundated with massiveamounts of information. If a user fails to check theiraccount over a span of several hours, or even days, theycould easily have thousands of tweets. If this were tooccur, it is unlikely the user will scroll through and readthousands of individual messages. The LCM will beutilized to determine if tweet overload can impact hinderthe impact of celebrity endorsements. Either missing a


particular tweet that was posted hours earlier or scrollingright over it without taking notice could do this.

The larger study will also include the impact of thecelebrity endorser’s physical attractiveness. Consideringthat social media is a computer-mediated form of commu-nication, both parties – users creating content and usersconsuming content – are hidden behind a computer screen.Their likeness is represented by a social media accountname and avatar or a profile page. The Internet does,however, give access to other pictures and videos of the

celebrity. However, some celebrities are more popularthan others and appear more often in traditional media.Based on this, fans may be more familiar with somecelebrities than others. A fan’s personal knowledge andexposure to celebrities outside of social media may impacthow they process celebrity endorsements. A-list celebri-ties are more likely to have a greater following thansomeone who is still up and coming in the business.Attractiveness and recognition could affect how someparticipants respond to celebrity endorsements on Twitterand should be examined in more depth.

REFERENCES

“Eva Longoria, Actress – Celebrity Endorsements” (2011),[available at http://www.celebrityendorsementads.com/celebrity-endorsements/celebrities/eva-longoria/].

360i (2010), “Twitter and the Consumer-Marketer Dy-namic,” [available at http://www.360i.com/trk/360i-Twitter-Consumer-Marketer-Dynamic.html].

Abraham, Linda B., Marie P. Mörn, and Andrea Vollman(2010), “Women on the Web: How Women areShaping the Internet” [available at http://www.comscore .co/Press_Events /Presenta t ions_Whitepapers/2010/Women_on_the_Web_How_Women_are_Shaping_the_Internet].

Agrawal, Jagdish and Wagner A. Kamakura (1995), “TheEconomic Worth of Celebrity Endorsers: An EventStudy Analysis,” Journal of Marketing, 59, 56–62.

Akhtar, Iyaz (2011), “Top 10 Celebrity Tech Endorse-ments,” [available at http://www.pcmag.com/slideshow/story/260174/top-10-celebrity-tech-en-dorsements/6#fbid=meTGV1PsmnF].

Alba, Joseph et al. (1997), “Interactive Home Shopping:Consumer, Retailer, and Manufacturer Incentives toParticipate in Electronic Marketplaces,” Journal ofMarketing, 61, 38–53.

Andrews, J.C. and Srinivas Durvasula (1991), “Sugges-tions for Manipulating and Measuring Involvementin Advertising Message Content,” in Advances inConsumer Research, Rebecca H. Holman and MichaelR. Solomon, eds. Provo, UT: Association for Con-sumer Research, 194–201.

Atkin, Charles and Martin Block (1983), “Effectivenessof Celebrity Endorsers,” Journal of Advertising Re-search, 23, 57–61.

Bailey, Ainsworth A. (2007), “Public Information andConsumer Skepticism Effects on Celebrity Endorse-ments: Studies among Young Consumers,” Journalof Marketing Communications, 13, 85–107.

Barnes, Nora G. and Justina Andonian (2011), “The 2011Fortune 500 and Social Media Adoption: Have

America’s Largest Companies Reached a SocialMedia Plateau?” [available at http://www.umassd.edu/media/umassdartmouth/cmr/studiesandresearch/F5002011.pdf].

Conrad, Lauren (2011), “The Future of Lauren Conrad,Inc.” [available at http://www.forbes.com/sites/laurenconrad/2011/09/19/the-future-of-lauren-conrad-inc/].

____________ (2011), “Tuesday Ten: The Working Girl’sStyle Staples,” [available at http://www.laurenconrad.com/post/tuesday-ten-the-work-ing-girls-style-staples].

Feick, Lawrence and Robin A. Higie (1992), “The Effectsof Preference Heterogeneity and Source Characteris-tics on Ad Processing and Judgments about Endors-ers,” Journal of Advertising, 21, 9–24.

Friedman, Hershey H. and Linda Friedman (1979), “En-dorser Effectiveness by Product Type,” Journal ofAdvertising Research, 19, 63–71.

Friestad, Marian and Peter Wright (1994), “The Persua-sion Knowledge Model: How People Cope withPersuasion Attempts,” Journal of Consumer Re-search, 21, 1–31.

Gangadharbatla, Harsha (2009), “Motivations for SocialNetworking Site Adoption,” in Handbook of Re-search on Digital Media and Advertising: User Gen-erated Content Consumption, Neal Burns, TerryDaugherty, and Matthew S. Eastin, eds. Hershey,NY: Information Science Reference.

Hardesty, David M., Jay P. Carlson, and William O.Bearden (2002), “Brand Familiarity and Invoice PriceEffects on Consumer Evaluations: The ModeratingRole of Skepticism Toward Advertising,” Journal ofAdvertising, 29, 43–54.

Haven, Brian and Suresh Vital (2008), Measuring En-gagement. Cambridge, MA: Forrester Research, Inc.

Heil, Bill and Mikolaj J. Piskorski “New Twitter Re-search: Men Follow Men and Nobody Tweets,” [avail-able at http://blogs.hbr.org/cs/2009/06/new_twitter_research_men_follo.html].


Henthorne, Tony L., Michael S. LaTour, and RajanNataraajan (1993), “Fear Appeals in Print Advertis-ing: An Analysis of Arousal and Ad Response,”Journal of Advertising, 22, 59–69.

Katz, Elihu, Jay G. Blumler, and Michael Gurevitch(1973), “Uses and Gratifications Research,” ThePublic Opinion Quarterly, 37, 509–23.

____________ (1974), “Utilization of Mass Communi-cation by the Individual,” in The Uses of Mass Com-munications: Current Perspectives on GratificationResearch, Jay G. Blumler and Elihu Katz, eds. BeverlyHills, CA: Sage, 19–32.

Kelman, Herbert C. (1961), “Process of Opinion Change,”Public Opinion Quarterly, 25, 57–78.

Kim, Yeung-Jo and June-Hee Na (2007), “Effects ofCelebrity Athlete Endorsement on Attitude Towardthe Product: The Role of Credibility, Attractivenessand the Concept of Congruence,” International Jour-nal of Sports Marketing & Sponsorship, 310–13.

Lafferty, Barbara A. and Ronald E. Goldsmith (1999),“Corporate Credibility’s Role in Consumers’ Atti-tudes and Purchase Intentions when a High VersusLow Credibility Endorser Is Used in the Ad,” Journalof Business Research, 44 109–16.

Lang, Annie (2000), “The Limited Capacity Model ofMediated Message Processing,” Journal of Commu-nication, 46–70.

Li, Charlene and Josh Bernoff (2008), Groundswell.Boston: Harvard Business Press.

McGinnies, Elliott and Charles D. Ward (1980), “BetterLiked than Right: Trustworthiness and Expertise asFactors in Credibility,” Personality and Social Psy-chology Bulletin, 6, 467–72.

Miciak, Alan R. and William L. Shanklin (1994), “Choos-ing Celebrity Endorsers,” Marketing Management,3, 51–59.

Ohanian, Roobina (1990), “Construction and Validationof a Scale to Measure Celebrity Endorsers’ PerceivedExpertise, Trustworthiness, and Attractiveness,” Jour-nal of Advertising, 19, 39–52.

Ostrow, Adam (2009), “Social Networking Sites MorePopular than Email,” [available at http://mashable.com/2009/03/09/social-networking-more-popular-than-email/].

Papacharissi, Zizi and Alan M. Rubin (2000), “Predictors

of Internet Use,” Journal of Broadcasting & Elec-tronic Media, 44, 175–96.

Pomerantz, Dorothy (2011), “Tina Fey, Eva Longoria:Hollywood’s Highest-Paid TV Actresses,” [avail-able at http://abcnews.go.com/Business/tina-fey-eva-longoria-hollywoods-highest-paid-tv/story?id=14633523].

Quantcast (2011), “Twitter.Com Traffic and DemographicStatistics,” [available at http://www.quantcast.com/twitter.com#demo].

Rubin, Alan M. (1983), “Television Uses and Gratifica-tions: The Interaction of Viewing Patterns and Moti-vations,” Journal of Broadcasting, 27, 37–52.

Sliburyte, Laimona (2009), “How Celebrities Can BeUsed in Advertising to the Best Advantage,” WorldAcademy of Science, Engineering, and Technology,58, 934–39.

Solis, Brian (2010), “Influence is Bliss: The GenderDivide of Influence on Twitter,” [available at http://www.briansolis.com/2010/08/influence-is-bliss-the-gender-divide-of-influence-on-twitter/].

Swinyard, William R. (1993), “The Effects of Mood,Involvement, and Quality of Store Experience onShopping Intentions,” Journal of Consumer Research,20, 271–80.

The Nielsen Company (2011), “State of the Media: TheSocial Media Report Q3 2011,” [available at http://blog.nielsen.com/nielsenwire/social.].

Tripp, Carolyn, Thomas D. Jensen, and Les Carlson(1994), “The Effect of Multiple Product Endorse-ments by Celebrities on Consumer Attitudes andIntentions,” Journal of Consumer Research, 20, 535–47.

Weigold, Keith and Joe Pulizzi (2010), Engagement:Understanding it, Achieving it, Measuring it. Pa-latine, IL: Junta 42, Nutlug.

Williamson, Debra A. (2011), “Social Network Ad Rev-enues Rising Worldwide,” [available at http://www.emarketer.com/Article.aspx?R=1008213&dsNav=Rpp:25,Ro:3,N:700].

____________ (2011), “Worldwide Social Network AdSpending: 2011 Outlook,” [available at http://www.emarketer.com/Reports/All/Emarketer_2000757.aspx].

For further information contact:

Nicole CunninghamTexas Christian Universit y

TCU Box 298060Fort Worth, TX 76132Phone: 817.257.5251



CONSUMERS’ ONLINE RESPONSES TO THE DEATH OF A CELEBRITY

Scott K. Radford, University of CalgaryPeter H. Bloch, University of Missouri

SUMMARY

With the growth of the Internet and social network-ing, celebrity veneration has become more participatoryadvancing it to even greater prominence. Consumers areincreasingly forming online communities based on ashared devotion or enthusiasm for a product, brand, activ-ity, or cause (Cova and Pace 2006). Such communitiesmay be strongest, however, when based on a sharedfascination with a celebrity or other public figure. Thepresent study examines buyer and seller responses to acelebrity death using data from eBay auctions and differ-ences in response across different death situations.

Celebrities in the 21st century are known not so muchfor their accomplishments but for the lives they lead(McCutcheon et al. 2002) and increasingly every aspect ofcelebrities’ lives are put under the microscope. Althoughcelebrities control their images, many people feel veryclose to these individuals and form vicarious relationshipswith them (Hills 2002; Thomson 2006). In addition, thetangible objects associated with celebrities serve as to-tems for their worship (McCane 1991). While the popularmedia often denigrates celebrity watching as trivial (Jenkins2006) for many fans the connection to a celebrity is animportant component in their definition of self (O’Guinn1991).

The collectible and memorabilia industry accountsfor billions of sales dollars and one setting that hasdramatically facilitated the exchange of celebrity memo-rabilia is eBay (Pitta 2003). The death of a celebrityproduces a grief reaction and increased sacralization ofrelated goods (Belk et al. 1989; Zaidman & Lowengart2001). Demand for celebrity-related products increasesdue to the meaning transfer process and expected profitopportunities stimulate market entry for memorabiliasellers. Until the advent of online auctions, there was aconsiderable time lag before marketers could bring tomarket additional products to slake heightened demand.

The nature of a celebrity’s death and the type ofconnected products also may affect the resulting marketactivity. Grayson and Martinec (2004) distinguished be-tween indexical and iconic authenticity. In the first case,authenticity rests on a link to something else, while in thelatter, it rests solely on design and realism. With death-related sacralization, it is expected that indexical items,such as signed photographs, will receive much moreinterest than iconic items, such as commemorative t-

shirts, and will therefore demonstrate greater death ef-fects. The cause of death may also have an effect on theresponses to auctions. It is likely that a celebrity who diessuddenly and unexpectedly is more likely to garner agreater outpouring of grief than one whose death wasanticipated.

There are a number of possible explanations for thesepost-death effects. First, death is an intense event thatmotivates consumers to engage in meaning transfer as away of compensating for the loss of a loved one (Richins1994). Second, death may produce a need to cling to itemsthat have been contaminated by contact with the deceasedperson, elevating the status of these items (Radford andBloch 2012). Finally, death is something unknowable andfascinating, and often creates a mythology that is greaterthan that experienced during life (Wang 2007).

We used eBay auction data to track exchange activityaround the celebrity’s death. Data were obtained for the14 days before and after the celebrity’s death. During afive-year period between 2001 and 2006, we collecteddata from celebrities who were very popular, their deathwas newsworthy, and had pre-death merchandise sales.This resulted in seven celebrity memorabilia categories:Dale Earnhardt (Die Cast Car) Johnny Cash (AutographedItems), John Ritter (Autographed Items), Ronald Reagan(Autographed Items & DVDs), Johnny Carson(Autographed Items), and Steve Irwin (DVDs).

The data were aggregated daily and we used a piece-wise regression of the following form to examine thedeath effects (Neter et al. 1996):

Yi = β

0 + β

1T

i + β

2(T

iX

i) + β

3X

i + ε

i

Where Yi = Dependent Variable, i.e., new items, bids,

average price.T

i = Time – days before (-) and after (+) event,

intervention day = 0X

i = 1 If date is after intervention (T

i > 0)

= 0 If date is before intervention (Ti < 0)

The present research demonstrated that a celebrity’sdeath exerted a powerful influence on auction activity.Death effects were both substantial in magnitude andextremely quick in their appearance. The nearly instanta-neous surge of activity was generally not sustained and theactivity showed a steady decline. While the decline variedacross different celebrities and memorabilia, it was evi-

(1)


dent that even within two weeks the initial surge indemand was waning. There was some limited evidencethat price was affected by a celebrity’s death. However,pricing results were less conclusive, possibly because ofdiversity within the item category and because the in-creased demand was slaked by an increase in supply. Wefound that indexical items exhibited a greater death effectthan items that were iconic representations of the celeb-rity. The research also showed that the change in market-place activity was greater for celebrities who died unex-pectedly.

Sellers should recognize that online auctions havefacilitated market entry and dramatically reduced time tomarket. Companies who are embedded in traditional retailchannels cannot react quickly enough to take advantage ofthe peak in the sacralization of objects that follows acelebrity death or other significant event. Our data showed

that consumption activity surrounding a celebrity’s deathpeaked no later than nine days after the celebrity’s death,and in most cases the peak occurred on the actual date ofdeath. Traditional retailing takes sellers out of this peakperiod of activity.

We believe that future research related to celebritystature, the consumption of related memorabilia, and theirmanifestation is not just relevant to our understanding ofmodern culture. Celebrity-related research also has thepotential for theoretical contributions to a wide variety ofresearch domains such as product involvement, collect-ing, conformity, the self, consumer fantasy, and sourceeffects, among others. It is hoped that the present researchwill stimulate future work on celebrity effects and creategreater awareness of the opportunities that Internet datamay provide to consumer researchers. References areavailable upon request.

For further information contact:Scott Radford

Haskayne School of BusinessUniversity of Calgary

2500 University Drive NWCalgary, AB T2N 1N4Phone: 403.220.7247



EXTENDING THE ELECTRONIC TECHNOLOGY ACCEPTANCEMODEL: CONSUMER ADOPTION OF AUGMENTED

REALITY-BASED MARKETING TOOL

Mark Yi-Cheon Yim, Canisius College, BuffaloShu-Chuan Chu, DePaul University, Chicago

SUMMARY

Augmented reality (AR) technology has beenrecently attracting high attention from many compa-nies. AR technology allows consumers to virtually inter-act with three-dimensional product visualizations dis-played on users’ screens – consumers can wear diversevirtual products (i.e., “virtual try-on”; Kim and Forsythe2008); they can zoom in, zoom out and investigate virtualproducts on hands; they can even watch fantasies oranimated stories from product packages (e.g., cans forCoke Zero) (Hampp 2009). Due to its uniqueness andentertainment factor, many giant companies, such aseBay, Nike, Adida, and Mini, have been eagerly adoptingthe AR-based product presentations (Groth 2011). Themarket for AR is expected to reach $350 million in 2014,which jumps from $6 million in 2008 (Rachel 2009).

However, considering its high potential and popular-ity as a promotional tool, little is known about how ARtechnology works, what are the components that make itsuccessful, and who is more likely to adopt this newpromotional tool? To answer these questions, we adoptthe electronic Technology Acceptance Model (e-TAM)and extend the model by considering both media anduser characteristics. Specifically, the existing e-TAMhas focused on user evaluations, such as ease of use,usefulness, and enjoyment in determining user technol-ogy acceptance (Heijden 2000). Yet, the proposed modelextends e-TAM by adding perceived interactivity, medianovelty, previous media experience, and users’innovativeness. Given that a growing number ofdiverse technologies are emerging as promotional tools,re-confirming and elaborating the e-TAM would behighly important. A more thorough understanding of themedia and user characteristics of augmented reality isbelieved to nurture online retailers’ knowledge regardinghow to maximize the AR-based promotion strategies.

Referring to the electronic Technology AcceptanceModel (e-TAM), the current study extended the modelincluding users’ previous media experience, media nov-elty, interactivity, and users’ innovativeness. And thefollowing hypotheses were tested:

H1: Consumers’ perceived interactivity is positivelyrelated to usefulness of AR-based promotion.

H2: Consumers’ perceived interactivity is positivelyrelated to enjoyment.

H3: Consumers’ perceived interactivity is positivelyrelated to ease of use in AR-based promotion.

H4: Consumers’ perceived media novelty is positivelyrelated to perceived interactivity.

H5: Consumers’ previous media experience with ARtechnology is negatively related to perceived nov-elty.

H6: Consumers’ previous media experience with ARtechnology is positively related to perceivedinteractivity.

H7: Consumer innovativeness is positively related to theintention to adopt the AR-based promotion.

A structural analysis was conducted using AMOS18.0. The significance of the paths in the model deter-mined whether the proposed hypotheses were supportedor not. All the hypotheses were supported except forHypothesis 1 (interactivity à usefulness: ß = .08, n.s.) andH2 (interactivity à enjoyment: ß = .03, n.s.). Specifically,as prior literature identified (e.g., Kim and Forsythe 2008;Ko, Kim, and Lee 2009), the latent variables constitutingthe e-TAM (e.g., ease of use, usefulness, enjoyment,attitudes) showed all the significant relationships. As forthe extended parts, interactivity unexpectedly did notmake a direct impact on usefulness and enjoyment, whilethey are indirectly linked to each other via ease-of-use(interactivity à ease-of-use à usefulness; interactivity àease-of-use à enjoyment). Thus, it is interpreted that easeof use mediates the impact of interactivity on usefulnessand enjoyment in the process of adopting the AR-basedpromotion. In other words, high interactivity operated inan AR based promotion tool makes consumers perceive itas an easy tool to use. When consumers perceive ease ofuse, they come to feel usefulness and enjoyment, resulting


in high intention to adopt the AR promotion. Addition-ally, it was found that more innovative consumers aremore likely to adopt the AR based promotion. It was inpart because a majority of participants perceived the ARtechnology highly novel (M

novelty = 5.94/7, SD = 1.17,

Skewness = -1.57, Kurtosis = 3.06). However, when themedia novelty of AR technology is diminished, the groupof innovative consumers may not hold their willingness totry it. Rather, it is believed that media novelty andinnovativeness would serve as a moderating impact ontheir adoption of the AR promotion someday after itsnovelty has faded (Tellis 1997); consumers with highinnovativeness are more likely to adopt the AR promotionthan those with low innovativeness when they perceivehigh media novelty, but high and low innovative consum-

ers will not show little difference when they perceive lowmedia novelty.

In summary, our extended model explains that newtechnologies themselves can benefit in persuading con-sumers, and innovative consumers are willing to try themsince they are new, unique, and different. However, thebenefit will not last for a long time (Tellis 1997). Thus, ARtechnology should be able to include machineryinteractivity high enough to make consumers feel com-fortable, which results in ease of use. In turn, it helpsconsumers to explore products from multiple dimensionswith a positive mood state. Finally, consumers come toadopt the new technology-based promotion tool due tothose benefits they perceive. References are availableupon request.

For further information contact:Mark Yi-Cheon Yim

Department of Marketing & Information SystemsThe Richard J. Wehle School of Business

Canisius College2001 Main Street

Buffalo, NY 14208Phone: 716.888.3264



DO PERSONAL RESPONSE SYSTEMS (CLICKERS) ENHANCELEARNING AND RETENTION OF KNOWLEDGE IN HIGHER

EDUCATION: AN EMPIRICAL INVESTIGATION

Erin Cavusgil, University of Michigan – Flint

SUMMARY

Educators face a major challenge in the classroom:engaging and motivating students. As a result, they haveincreased their adoption/use of instructional technology.One such technology includes personal response systems,also referred to as ‘clickers’. These devices, now beingoffered by a myriad of software firms, allow students torespond to questions presented to the class by using theirclicker devices. The aggregate responses of the studentsare then immediately displayed in a graphical form. Themotivation to incorporate such technology is often toengage students, as well as encourage participation throughthe use of clickers when students may be reluctant toverbally contribute. In addition, today’s students areinherently familiar with technology, and require an ac-tive, rather than passive, learning environment (Preis et al.2011; Young 2005). The use of clicker technology canprovide the interactivity instructors seek in the contempo-rary classroom environment.

The few studies that have examined clicker use in theclassroom are exploratory in nature, examining the ben-efits of the use of this technology to both students andinstructors. Challenges that instructors face include: (i)getting students to attend class, (ii) gaining and keepingstudents’ attention, and (iii) encouraging student partici-pation (Lincoln 2008). Several authors provide reviews ofthe literature on the use of personal response systems inthe classroom (Caldwell 2007; Eastman et al. 2011; Fiesand Marshall 2006; Morse et al. 2010). In general, posi-tive outcomes from the use of clickers in classroomsettings have been reported. Such benefits include: imme-diate feedback of students’ learning and comprehension,increased student engagement and alertness, greater recalland understanding of the material, and increased classattendance (Caldwell 2007; Eastman et al. 2009; Hoffman2006; Micheletto 2011; Salemi 2009; Ueltschy 2001).

The present study investigates the advantages ofusing clicker devices in the classroom. A formal experi-ment is implemented with the purpose of seeking defini-tive evidence as to the merits of this instructional technol-ogy. Specifically, the goals of the study were twofold: (1)to examine whether the use of clickers in the classroomwould aid student retention of the class material, and (2)to examine students’ perceptions of the benefits of clickeruse. A key shortcoming of previous studies is that the

evidence has been limited to students’ self-reported per-ceptions of using clickers. The present study goes beyondstudent surveys to additionally test the impact of clickeruse on students’ retention of information.

The study was conducted in an undergraduate Mar-keting Management course at a mid-western university.The total number of students enrolled in the course wasforty. Multiple choice questions were posed to the stu-dents throughout the semester, either at the beginning orthe end of a lecture. Questions were presented to thestudents in two manners: (i) students were asked to usetheir clickers to respond to the question (clicker ques-tions), and (ii) questions were presented without clickers(non-clicker questions), after which a student was askedto volunteer the answer; in other words, no clicker wasused. In the case of (i), individual student responses wererecorded, as students were required to login using theclickers with their student. Therefore, data was recordedas to how each individual student performed on eachclicker question.

A second component of the study involved retestingstudents on all questions posed in earlier lectures, bothclicker and non-clicker questions. This was accomplishedby giving the students a written quiz. Four such quizzeswere conducted throughout the semester. Each quiz con-tained 10 clicker questions and 10 non-clicker questionsthat had been posed to the students during lectures in theprevious 3–4 weeks.

Two measures were employed to assess if the use ofclickers helped students retain the class material. Onemeasure involved comparing students’ performance dur-ing the quiz on the clicker questions vs. non-clickerquestions. The expected result was that using the clickers(posing clicker questions) would help students betterretain the material, compared to not using the clickers(posing non-clicker questions). In other words, studentswere expected to, on average, answer more clicker ques-tions correctly than non-clicker questions during the quiz.

Another measure examined comparing how studentsperformed on the clicker questions during two time-points: (i) when the question was initially posed to thestudents during the lecture using the clickers (t

1), and (ii)

when the clicker question was posed on a later date duringthe re-test written quiz (t

2). The expected result was that


students would perform as well, or better, during t2 com-

pared to t1, if the material was indeed retained.

Lastly, students were asked to participate in a surveyexamining their perceptions of using the clickers. Thesurvey questions related to the students’ perceived benefitand level of enjoyment of clicker use.

The results provide sufficient evidence that the use ofclickers help students to retain class material, compared towhen clickers are not employed. During the first writtenquiz comparing clicker questions to non-clicker ques-tions, the results indicated greater performance for theclicker questions. Overall, we can conclude that use of theclicker tool leads to improved retention of content. Fur-thermore, students consistently achieved a higher score

when quizzed on the clicker questions at a later time point(t

2), compared to their earlier performance (t

1), indicating

retention of the class material over time. Together, theseresults suggest that perhaps the use of clickers increasesstudent attention and engagement, resulting in improvedretention of information.

These encouraging findings are further corroboratedby direct feedback from the students. Students perceivepositive benefits from the use of clicker technology, asindicated by the student survey. They found the use ofclickers to be novel, and appreciated immediate feedback/reinforcement of their knowledge. In sum, this instruc-tional technology does hold much promise for learningreinforcement and retention. References are availableupon request.

For further information contact:Erin Cavusgil

University of Michigan – Flint2126 Riverfront Center West

Flint, MI 48502Phone: 810.237.6544



USING THE RFM MODEL TO RANK DOCTORALMARKETING PROGRAMS

Matt Elbeck, Troy University at DothanBrian A. Vander Schee, Aurora University

ABSTRACT

Students balance perception with scholarly fact toidentify the ‘best’ marketing doctoral program. We rankdoctoral marketing programs using scholarly productiv-ity. The Recency-Frequency-Money (RFM) segmenta-tion model is applied to offer a contemporary measure ofscholarly excellence based on top tier marketing journal90th percentile citation counts from 2006 to 2010.

INTRODUCTION

“Which business school offers the best marketingdoctoral (PhD or DBA) program?” is a question likely tocause pause for thought. A first step toward finding theanswer would be to explore the American MarketingAssociation’s (AMA 2011) list of schools offering doc-toral programs in marketing. Then one could seek guid-ance from the Association to Advance Collegiate Schoolsof Business (AACSB 2011a) on the quality of eachdoctoral program by considering (a) AACSB accredita-tion status, (b) faculty qualifications via posted vitas, (c)faculty scholarship productivity based on quantity andquality of journal article authorship and (d) prior profes-sor recommendations.

It is the third item “faculty scholarship productivitybased on quantity and quality of journal article author-ship” in particular that poses the greatest challenge forstudents to ascertain. The student faces a daunting task ofcataloging masses of information, some of which is diffi-cult to judge, and others may be obsolete. Thus there isvalue in ranking faculty and institutions addressing item(c) that will narrow the field for students to then consideritems (a), (b), and (d) which are much easier to identify.This approach will help students be more efficient anddiscerning in the information search process. Given thesalary offered to marketing doctoral program graduatesranges from $70,000 to $202,000, investigating whereone should embark on doctoral studies is time well spent(AMA 2006 to 2010).

A straightforward analysis focusing on scholars (i.e.,the source of scholarship quality) is needed to identifyinstitutions with outstanding performance in marketingscholarship as a contributing factor in assessing market-ing doctoral program quality. The results of this analysis

can then be presented in a table of institutional ranking tohighlight faculty scholarly productivity as suggested byAACSB in item (c) above.

In contrast with doctoral business programs, rankingsproliferate for MBA programs both in the U.S.(BusinessWeek 2011; U.S. News 2011) and world-wide(Economist 2011; Financial Times 2011). Over the yearsdoctoral rankings have appeared for various businessdisciplines such as finance (Chan, Lung, and Wolfe 2005;Heck 2006), accounting (Chan, Chen, and Cheng 2007;Coyne et al. 2010) and operations management (Vastagand Montabon 2002). In contrast, rankings for doctoralmarketing programs are somewhat outdated (Elbeck 1988;Robinson and Adler 1981), reason enough for a present-day study.

Ranking doctoral marketing programs has a practicaland useful benefit for prospective students. In part moti-vated by the seminal work by Bloom and Coan (1979) thatrecognized that many Ph.D. programs relied on detaileddegree requirement and faculty publication materials;Davis and McCarthy (2005) highlight the critical impor-tance of intangibles such as college rankings and facultyreputations used by prospective marketing doctoral pro-gram students.

To help fellow colleagues and students improve thequality of their marketing doctoral program choice, thepurpose of this study is to focus on one facet of marketingdoctoral programs by offering a contemporary ranking ofdoctoral programs in marketing based on the highestlevels of marketing scholar impact on the marketingliterature. The ranking of scholars and institutions canthen be used in concert with accreditation status, facultyqualifications, and prior professor recommendations tomake a more informed decision.

LITERATURE REVIEW

Rankings of scholars, departments and colleges aretypically dependent on a particular time frame, researchemphasis and set of journals examined (Polonsky 2008),as well as sub-areas in marketing (Baumgartner andPieters 2003) suggesting that regular construction ofrankings will always supersede a one-off study. Neverthe-less, variations in ranking methodology persist. In one


example longitudinal perception-based rankings withquantitative adjustments based on student experience hasshown merit when it comes to MBA programs (Holbrook2007).

Over the years scholars have also discovered variousways to rank PhD programs. Initiatives include incorpo-rating an institution’s accumulated research grants (Koshal,Koshal, and Gupta 1996), graduate placement (Amir andKnauff 2008) and research productivity (Bakir, Vitell,and Rose 2000). However, historical methods to rankprogram quality based on counts of faculty authorships(see University of Texas at Dallas 2011) have beenquestionable due to concerns regarding the scholarlyimpact of such publications (Heesacker and Elliot 2007).A solution to this is a citation count which is often used toinvite candidates to leading journal board membership(Rynes 2006), acting as a proxy for academic excellence.

Citation Analysis

Citation analysis is often used as a quantitative mea-sure to estimate the impact or contribution of a scholarlywork to the literature (Neuhaus and Daniel 2008). Citationanalysis offers investigators a robust metric in its ability toisolate scholarly excellence. Consider “some 90 percentof papers that have been published in academic journalsare never cited. As many as 50 percent of papers are neverread by anyone other than their authors, referees andjournal editors” (Meho 2007, p. 32). Though variouscitation providers exist such as Thompson ISI Web ofScience (WoS) and Elsevier Publishing’s Scopus, empiri-cal evidence shows that citation count correlationsbetween Google Scholar (GS) and WoS, and betweenGS and Scopus are .87 and .97 respectively (Meho andYang 2007) offering confidence in GS citation results.Furthermore, GS will report higher citation countsbecause it is not limited to particular databases such asthe case with WoS and Scopus (Belew 2005) and providesmore global coverage (Touzani and Moussa 2010).

Journal Impact

In a survey of 309 marketing faculty at U.S. institu-tions, the Journal of Marketing (JM), Journal of Market-ing Research (JMR), Journal of Consumer Research(JCR), Marketing Science (MS) and the Journal of theAcademy of Marketing Science (JAMS) ranked as five ofthe top six (Journal of Retailing was the other journal) asthe most important marketing-related journals (Hult, Neese,and Bashaw 1997). Using perceived quality as a factor ina survey of 372 marketing faculty around the world, thejournals listed above ranked as the top four, with JAMS inthe top 10 (Theoharakis and Hirst 2002). These journalsalso accounted for 67 percent of the article citations foundin doctoral marketing program course syllabi (Bauerlyand Johnson 2005).

In a more recent world-wide study involving 629marketing faculty, the same journals listed above wereidentified as the top five by Hult, Reimann, and Schilke(2009) based on marketing faculty perception of impor-tance, prestige, popularity and familiarity of marketingjournals. A similar conclusion was drawn in as citation-based study by Touzani and Moussa (2010) where thesame five journals ranked in the top six (Industrial Mar-keting Management was the other journal) using a citationindex covering 2003 to 2007. Thus there is a consistentpattern over time demonstrating that the top five market-ing journals to utilize in a citation analysis include JM,JMR, JCR, MS, and JAMS. The following section detailsthe methodology used for this study that uses a citationanalysis of the five most influential marketing journals toidentify those scholars with the greatest impact in themarketing literature and according to their institutionalaffiliation, arrive a ranking of the elite colleges anduniversities offering a doctoral program in marketing.

METHOD

Data analysis model selection is based on a popularyet simple quantitative method used by direct marketers topredict future customer behavior based on past purchas-ing patterns (Alencar et al. 2006; Malthouse and Blattberg2005) known as the recency-frequency-money (RFM), orrecency-frequency-monetary value (RFV) model(Cullinan 1977; Hughes 1996). The underlying logic ofthe model is to focus resources on profitable marketsegments. In this study the application is designed toextract outstanding marketing scholars by specifying modelparameters as follows;

• R = recency of publication. We select the five-yearperiod 2006 to 2010 to identify those academicsactively publishing in top tier journals. This timeperiod is consistent with AACSB’s Standard 2 onfaculty intellectual contribution such that “the portfo-lio of intellectual contributions for individual facultymembers, within each discipline, and for the businessschool as a whole . . . used to provide an overall 5-yearsummary of the school’s intellectual contributions”(AACSB 2011b). The bias toward chronologicalrecency may not afford the most recently publishedarticles sufficient time to be cited by others. Howeverit is in line with the respected and often used ImpactFactor which, for a given year considers the averagenumber of citations received per paper published inthat journal during the two preceding years (ThomsonReuters 2011).

• F = frequency of publication. We limit the selectedscholars to those who have published two or morearticles in JM, JMR, JCR, MS, or JAMS in the 2006to 2010-period. These journals consistently rank asthe most competitive as outlined above. At least one


of their articles must also meet the conditions inparameter M below. This allows for a demonstrationof sustained scholarly activity.

• M = total number of citations per author in the fiveselected marketing journals. This metric is furtherrefined to recognize those scholars with the veryhighest levels of scholarly impact such that scholarswere included if one or more of their article(s) met orexceeded the 90th percentile of all citations per jour-nal, signaling the very highest levels of scholarlyquality and impact. If an article included more thanone author, then the total number of citations wasequally divided by the total number of authors, iden-tified as an author’s citation count share.

The first step is to select high quality marketingjournals as the source of excellence in marketing scholar-ship and impact. Citation data for JM, JMR, JCR, MS, andJAMS were captured using Publish or Perish (Harzing2011) employing Google Scholar as the source data. Datawere collected on April 29, 2011. The RFM model appli-cation is such that within the five-year period (2006 to2010), for authors to be selected, they must publish at leasttwo articles in one or more of the top five marketingjournals, with one of the articles in the 90th percentilecitation count from 2006 to 2010. In the case of multipleauthors, the author’s (and their institution’s) impact, re-ferred to as citation count share, is calculated as thearticle’s number of citations divided by the number ofauthors contributing to that article, consistent with theprocedure used by Bakir, Vitell, and Rose (2000). Forexample, assume a three-author article achieves 60 cita-tions; each author is awarded 20 citations based on theassumption of equal author contribution.

For cases where one or more of the authors areconsultants, managers or scholars outside the field ofmarketing, deceased, or retired, they were not included inthe analysis, but each remaining author’s citation countshare was still divided by the total number of contrib-uting authors. Updating a scholar’s institutional affili-ation affirms the truism that “people make institu-tions.” In that spirit author affiliation was updated byexamining online resumes found via Google and institu-tional biography postings during the second week of May2011, a date likely to provide relatively reliable findingsgiven faculty usually change their institutional affiliationeither at the start of the year or around Fall. Finally, eachauthor’s institution was checked against the AmericanMarketing Association’s list of business schools offeringa doctoral program in marketing (AMA 2011).

Institutions with current marketing faculty receivingat least 100 citations were categorized as preeminent.Those with at least 50 but not more than 99 citations werecategorized as superior. A similar approach was used to

rank scholars. Those with at least 100 citations werecategorized as preeminent which is at or above the 98th

percentile while those with between 50 and 99 citationswere categorized as superior which is at or above the 90th

percentile.

RESULTS

There were 1,855 articles published in JM, JMR,JCR, MS, and JAMS from 2006 to 2010 with one or morecitations (range 1 to 301). To meet the RFM modelrequirements, the number of articles was limited to thosewith 45 or more citations (at or above the 90th percentilecitation count for all 1,855 articles), resulting in a total of117 articles (6% of the initial set of 1,855 articles) authoredby 266 scholars and cited 6,661 times.

Descriptive Overview

Eighty-nine universities made up the sample encom-passing 162 scholars with 6,661 citations in the 90th

percentile of all citations for each of the five journals usedin this study. The majority (61%) of the total 6,661citations are attributed to articles published in 2006.Others years showed diminishing contribution with 21percent in 2007, 13 percent in 2008, 5 percent in 2009, andzero percent in 2010. However the Vargo and Lusch(2008) JAMS publication garnered the largest number ofcitations (301) of any article in the period 2006 to 2010.In line with the RFM model application, Vargo and Luschwere each allocated 150.5 citations (301 citations dividedby two authors) referred as each author’s citation countshare. Regarding the distribution of 6,661 citations foreach of the five journals, JM is the dominant outlet (43%),consistent with the findings by Baumgartner and Pieters(2003). MS and JMR clustered into second place with 20percent each, while JAMS (10%) and JCR (9%) make upthe lower positions.

University Ranking

Table 1 displays the preeminent marketing doctoralgranting institutions, each with over 100 citations over thefive-year period. Confirmation regarding whether a schooloffers a PhD in marketing was based on a population of186 institutions provided by the American MarketingAssociation (2011) and examination of college websites.The 21 marketing doctoral program granting universitiesin Table 1 represent 11.3 percent of all 186 marketingdoctoral granting institutions, accounting for 58.3 percent(3,880) of all 6,661 citations. In this group, all but twoprograms (ESMT – Berlin and HEC – Paris) are located inthe U.S. All but one (ESMT – Berlin) are accredited byAACSB (AACSB International 2011c).

Table 2 reports superior institutions offering market-ing doctoral programs earning between 50 and 99 cita-


tions over the five-year period. The 19 marketing doctoralgranting institutions account for 10.2 percent of the 186marketing doctoral granting universities, accounting for18.6 percent (1,238) of all 6,661 citations. In this group,five marketing doctoral granting institutions (26%) arelocated outside of the U.S. (Germany, Singapore, theU.K., and the Netherlands) and all but two (University ofCologne, University of Hamburg) are accredited byAACSB (AACSB International 2011c).

TABLE 1Preeminent Doctoral Marketing Programs (100 or More Citations 2006–2010)

Rank Institution Citations Rank Institution Citations

1 University of Arizona 408 12 Stanford University 156 2 Columbia University 331 13 European School of Mgt &

Technology, Berlin *,** 148= 3 Duke University 278 14 University of Pennsylvania 146= 3 Yale University 278 15 University of Michigan 140 5 University of Texas – Arlington 195 16 University of Wisconsin 136 6 Harvard University 185 17 University of Chicago 131 7 University of Hawaii 184 18 University of Texas – Austin 117 8 UCLA 182 19 USC 116 9 University of Maryland 173 20 CUNY Baruch 10610 University of Oklahoma 169 =21 Emory University 10211 University of Washington 163 =21 HEC – Paris 102

Note: Citations <.4 are rounded down, and >.5 rounded up.* Institution without a doctoral marketing program. ** Institution not AACSB accredited.

Tables 1 and 2 include five universities without adoctoral program in marketing. This demonstrates thathighly productive marketing scholars are not exclusivelymembers of a marketing doctoral program.

Author Ranking

Table 3 displays the nine most accomplished (pre-eminent) scholars by author citation count share from

TABLE 2Superior Doctoral Marketing Programs (50 to 99 citations 2006–2010)

Rank Institution Citations Rank Institution Citations

23 Babson College* 98 34 New York University 6424 University of Minnesota 90 =35 Indiana University 6025 University of North Carolina 78 =35 Old Dominion University 6026 Frankfurt University – Main 76 =35 University of Cologne ** 6027 University of Florida 71 = 35 University of Hamburg ** 6028 MIT 70 39 London Business School 5929 Dartmouth College* 69 40 University of Illinois 5830 University of Mannheim 68 41 University of Utah 5731 Nanyang Technological University 67 42 Tilburg University 5532 Northwestern University 66 43 Georgia State University 5433 University of Iowa 65

Note: Citations <.4 are rounded down, and > .5 rounded up.* Institution without a doctoral marketing program. ** Institution not AACSB accredited.


2006 to 2010. The nine scholars account for 20 percent(1,326) of all 6,661 citations. All but two of the scholarsare located in U.S. institutions.

Table 4 shows a list of 29 highly accomplished(superior) scholars by author citation count share from2006 to 2010. The 29 scholars retain a 27.5 percent(1,830) share of all 6,661 citations in the study’s sample.In this group, all but six scholars (21%) are located atinstitutions within the U.S.

DISCUSSION

There are a number of notable results worth high-lighting. The dominance of the University of Arizona asa center for top flight marketing scholars is unequivocal.The usual suspects populate the highest college ranks andis consistent with the notion that publishing performanceis often higher in large versus small marketing depart-ments (Bakir et al. 2000). The historic dominance of U.S.doctoral programs in marketing may be waning slightlyand may benefit from competition from outside the U.S.such as Germany’s Cologne, Frankfurt, Hamburg, andMannheim; the Netherlands’ Tilburg; the U.K.’s LondonBusiness School; Singapore’s Nanyang TechnologicalUniversity; and France’s HEC – Paris.

When examining author citation count shares inTables 3 and 4, new names listed amidst recognizablemarketing scholars suggests that emerging scholars canexcel alongside than their established peers. Results inTables 3 and 4 also show that 23 percent (38) of 162marketing scholars with citations at or above the 90th

percentile level of citations in the top five journals accountfor 3,381 citations, or 51 percent of all 6,661 citations. The

51:23 ratio approximates the Pareto principle (law of thevital few) suggesting that 38 marketing scholars make aremarkable contribution to the discipline. Furthermore,the contribution of one or two scholars to an institution’sranking (e.g., Luo at UT Arlington, Vargo at Hawaii,Ulaga at HEC – Paris) suggests that top-tier schools seekand hire the very best scholars for their doctoral programs.

The question of whether attending a top tier programduring the study’s five-year time-frame makes a profes-sional difference is based on the AMA’s Who WentWhere reports (AMA, 2010–2006). The survey results arebased on responses from 522 marketing doctoral gradu-ates. Of the 350 graduates from institutions not designatedas preeminent or superior in this study, 329 (94%) werehired by other non-preeminent or superior institutions,whilst of those hired by preeminent or superior institu-tions, 51 (30%) graduated from preeminent or superiorinstitutions. These findings indicate a significant relation-ship (χ2 = 54.25, df = 1, p < .001) indicate that institutionsrecognized for highly productive faculty are more likelyto hire graduates from other recognized institutions.

Application of the RFM segmentation model in thisstudy offers additional value to prior studies using citationdata. By limiting article selection to those articles meetingor exceeding the 90th percentile of all article citations inthe five most influential journals in marketing from 2006to 2010, this study offers a contemporary collection ofoutstanding scholars and their institutional affiliation.This information serves a number of stakeholders.

Consistent with guidance offered by the AMA Doc-toral Special Interest Group (American Marketing Asso-ciation 2008a), doctoral marketing program applicants

TABLE 3Preeminent Scholars (100 or More Citations, 2006–2010)

Rank Scholar Institution Citations

1 Xueming Luo University of Texas – Arlington 195 2 Yong Liu University of Arizona 192= 3 Robert F. Lusch University of Arizona 184= 3 Stephen L. Vargo University of Hawaii 184 5 Dina Mayzlin Yale University 173 6 C.B. Battacharya European School of Mgt & Technology, Berlin *,** 148 7 Donald R. Lehmann Columbia University 112 8 Wolfgang Ulaga HEC, Paris 102 9 Gerard J. Tellis University of Southern California 100

Note: Citations < .4 are rounded down, and > .5 rounded up.* Institution without a doctoral marketing program. ** Institution not AACSB accredited.


TABLE 4Superior Scholars (50 to 99 Citations 2006–2010)

Rank Scholar Institution Citations

= 10 Rajiv P. Dant University of Oklahoma 98= 10 Dhruv Grewal Babson College* 98= 10 Robert W. Palmatier University of Washington 98 13 Sunil Gupta Harvard University 85 14 Ravi Dhar Yale University 84 15 Robert Zeithammer UCLA 83 16 Kathleen D. Vohs University of Minnesota 78 17 Bernd Skiera Frankfurt University 76 18 Kenneth R. Evans University of Oklahoma 71 19 John Hauser MIT 70 20 Christian Homburg University of Mannheim 68= 21 Tuck Siong Chung Nanyang Technological University 67= 21 Roland T. Rust University of Maryland 67 23 Angela Y. Lee Northwestern University 66 24 Uzma Khan Stanford University 63= 25 Jan-Benedict E.M. Steenkamp University of North Carolina 62= 25 Kenneth C. Wilbur Duke University 62= 27 Dominique M. Hanssens UCLA 61= 27 Günter J. Hitsch University of Chicago 61= 27 Ran Kivetz Columbia University 61= 30 Franziska Volckner University of Cologne ** 60= 30 Henrik Sattler University of Hamburg ** 60= 30 Yuping Liu Old Dominion University 60 33 Anja Lambrecht London Business School 59 34 Aric Rindfleisch University of Wisconsin 58 35 Abbie Griffin University of Utah 57 36 Sandy D. Jap Emory University 54 37 Jonah A. Berger University of Pennsylvania 53 38 Olivier Toubia Colombia University 51

Note: Citations < .4 are rounded down, and > .5 rounded up.* Institution without a marketing doctoral program. ** Institution not AACSB accredited.

can use this study’s results to make a more informedchoice about which doctoral program to consider. Theresults allow doctoral applicants to more readily identifythe most productive, contemporary scholars. This under-scores the potentially positive impact an outstandingprogram will have on their career. The AMA’s Who WentWhere surveys validate the results in this study in thatdoctoral marketing programs with highly productive fac-ulty on staff are more likely to hire graduates fromprograms with similar academic standing. Further, this isvery helpful when it comes to selecting a potential dis-sertation committee chair or academic advisor.

The results also highlight marketing educators anumber of present-day high quality scholarly institutions

with an eye to possible job candidates. Attracting produc-tive and influential scholars serves two purposes. Recruit-ing such scholars enhances the reputation of the doctoralmarketing program by having highly cited scholars whopublish in the top tier marketing journals on staff. Sec-ondly, it brings top scholars into contact with the currentfaculty who then have greater connection and promisingcollaboration opportunities. Corporations use the infor-mation to solicit opinions or consultancy work fromscholars in the finest institutions. This enhances the repu-tation of the marketing program, business school, andcollege or university overall. Highly ranked institutionscan also use the information as part of their generalfunding, recruitment, and retention programs for highcaliber students and faculty.


Even with all the aforementioned benefits, there isone main drawback for institutions having doctoral mar-keting programs ranked this way. According to Kotler andKeller (2007), “the smart competitor must design anddeliver offerings for well-defined target markets” (p. 22).This encourages colleges to be proactive, market orientedand to avoid the temptation of resting on their laurelswhich, if left unchecked will tarnish their historic claim tothe heights of marketing excellence. In other words, overtime seminal papers in the marketing discipline will havea shorter lifespan which is consistent with many market-ing innovations given the rapid advancement of researchand technology. Institutions and scholars alike who relyon past accomplishments without an eye for currentimpact may find their prestige waning particularlywith the competitive influence from outside the UnitedStates.

LIMITATIONS AND FUTURE RESEARCH

This study has two limitations meriting discussion.Perhaps the most severe criticism typical to any cross-sectional study is the time period and therefore question-able reliability. Clearly, the longer an article is in circula-tion, the greater the opportunity for it to be cited. Whilstlikely to cause unease, we expect this to mitigate over timeas we plan to offer an updated study of this nature on anannual basis. The RFM model’s simplicity is favored overCHAID or logistic regression (McCarty and Hastak 2007)in spite of concerns about the model’s predictive value(Yang 2004; Rust and Verhoef 2005) that have been

addressed with additional data parsing and weights formore recent transactions (Fader, Hardie, and Lee 2005). Asimilar approach to artificially weigh the more recentcited articles (typically less cited compared to olderarticles) was considered for this study; however, thiswas rejected given this study’s emphasis on qualitativevalue versus predictive value. That is, arbitrarily manipu-lating the number of citations for the most recent articleswould compromise the face validity of the study. Patternsof sustained scholarly productivity will emerge as thisstudy is repeated on an annual basis.

The second issue concerns face validity, in that mostranking methodologies overlook the smaller high qualityprograms due to the limited number of faculty. White et al.(2011) suggest applying a per capita approach to level theplaying field between large and small doctoral programs.Over time we plan to introduce rankings in the variousspecialty fields in doctoral marketing programs. Interestin the topic will serve as a catalyst to develop annualcitation-based doctoral marketing programs ranking withthe likely addition of ranking for sub-areas within themarketing discipline. Over time, repeated studies willoffer a more stable and reliable set of rankings of facultyscholarly productivity. Prospective students can use theresults then, along with the accreditation status providedin this study to develop a more focused search for appro-priate doctoral marketing programs using the other twocriteria cited by AACSB faculty qualifications and priorprofessor recommendations.

REFERENCES

AACSB International (2011a), “Determining a DoctorateDegree Program’s Quality,” (accessed April 2011),[available at http://www.bestbizschools.com/doctor-ate/quality.asp].

____________ (2011b), “Business Accreditation Stan-dards,” (accessed April 2011), [available at http://www.aacsb.edu/accreditation/business/standards/strategic_man/standard2.asp].

____________ (2011c), “Accredited Institutions,” (ac-cessed April 2011), [available at http://www.aacsb.edu/accreditation/AccreditedMembers.asp].

Alencar, Antonio J., Eduardo M. Ribeiro, Armando L.Fereira, Eber A. Schmitz, Pricila M.V. Lima, andFernando S.P. Manso (2006), “Optimized RFV Analy-sis,” Marketing Intelligence & Planning, 24 (2), 106–18.

AMA (2006), “2006 Who Went Where,” (accessed Janu-ary 2012), [available at http://docsig.org/wwwsurvey2006/ WhoWentWhere2006ResultsFINAL.pdf].

____________ (2007), “2007 Who Went Where,” (acces-sed January 2012), [available at www.docsig.org/WWW2007.pdf].

____________ (2008a), “DocSIG Journal Study: Six TopMarketing Journals 2000–2007,” (accessed January2012), [available at http://docsig.org/Study%2008/DocSIG%27s%20Journal%20Study.htm].

____________ (2008b), “2008 Who Went Where,” (acces-sed January 2012), [available at www.docsig.org/WWW2008.pdf].

____________ (2009), “2009 Who Went Where,” (acces-sed January 2012), [available at www.docsig.org/DocSig-WhoWentWhere2009.pdf].

____________ (2010), “2010 Who Went Where,” (acces-sed January 2012), [available at www.docsig.org/FinalWWW2010.pdf].

____________ (2011), “PhD Programs in Marketing,”(accessed May 2011), [available at http://www.marketingpower.com/Community/ARC/Pages/Ca-reer/PhD/Programs.aspx].

Amir, Rahab and Malgorzata Knauff (2008), “Ranking


Economics Departments Worldwide on the Basis ofPhD Placement,” Review of Economics & Statistics,90 (1), 185–90.

Bakir, Aysen, Scott J. Vitell, and Gregory M. Rose (2000),“Publications in Major Marketing Journals: An Analy-sis of Scholars and Marketing Departments,” Journalof Marketing Education, 22 (2), 99–107.

Bauerly, Ronald J. and Don T. Johnson (2005), “AnEvaluation of Journals Used in Doctoral MarketingPrograms,” Journal of the Academy of MarketingScience, 33 (3), 313–29.

Baumgartner, Hans and Rik Pieters (2003), “The Struc-tural Influence of Marketing Journals: An Analysisof Scholars and Marketing Departments,” Journal ofMarketing, 67 (1), 123–39.

Belew, Richard K. (2005), “Scientific Impact Quantityand Quality: Analysis of Two Sources of Biblio-graphic Data,” (accessed May 2011), [available athttp://www.cogsci.ucsd.edu/~rik/papers/belew05-iqq.pdf].

Bloom, Paul N. and George Coan (1979), “ExplainingEducational and Occupational Choices: An Explor-atory Study,” Advances in Consumer Research, 6,617–21.

BusinessWeek (2011), “The Best U.S. MBA Programs bySpecialty 2010,” (accessed February 2011), [avail-able at http://www.businessweek.com/interactive_reports/mba_specialty_2010.html].

Chan, Kam C., Carl R. Chen, and Louis T.W. Cheng(2007), “Global Ranking of Accounting Programmesand the Elite Effect in Accounting Research,” Ac-counting and Finance, 47 (2), 187–220.

____________, Peter P. Lung, and Edward R. Wolfe(2010), “Research Ranking of Finance Departments:A Modified Citation Approach,” (accessed July 2011),[available at http://ssrn.com/abstract=646185].

Coyne, Joshua G., Scott L. Summers, Brady Williams,and David A. Wood (2010), “Accounting ProgramResearch Rankings by Topical Area and Methodol-ogy,” Issues in Accounting Education, 25 (4), 631–54.

Cullinan, G.J. (1977), Picking Them by Their BattingAverages: Recency-Frequency-Monetary Method ofControlling Circulation. New York: Direct Mail/Marketing Association.

Davis, Donna and Teresa T. McCarthy (2005), “TheFuture of Marketing Scholarship: Recruiting for Mar-keting Doctoral Programs,” Journal of MarketingEducation, 27 (1), 14–24.

Economist (2011), “Full-Time MBA Ranking,” (accessedJune 2011), [available at http://www.economist.com/whichmba].

Elbeck, Matthew (1988), “Evaluating Institutional Con-tributions to Scholarly Journals in Marketing, 1977–86,” in Developments in Marketing Science, Vol. XI,1988 Academy of Marketing Science Annual Confer-

ence, K.D. Bahn, ed. Montreal, PQ, (April 27–May1).

Fader, Peter S., Bruce G.S. Hardie, and Ka Lok Lee(2005), “RFM and CLV: Using Iso-Value Curves forCustomer Base Analysis,” Journal of MarketingResearch, 42 (4), 415–30.

Financial Times (2011), “Global MBA Rankings 2011,”(accessed March 2011), [available at http://rankings.ft.com/businessschoolrankings/global-mba-rankings-2011].

Harzing, Anne-Wil (2011), “Publish or Perish,” (acces-sed January 2011), [available at http://www.harzing.com/pop.htm].

Heck, Jean L. (2006), “Establishing a Pecking Order forFinance Academics: Ranking of U.S. Finance Doc-toral Programs,” SSRN, (accessed March 2011),[available at http://ssrn.com/abstract=969413].

Heesacker, Martin and Timothy R. Elliot (2007), “MyDog’s Better than Your Dog: Publication Counts andQuality of Clinical Psychology PhD Training,” Clini-cal Psychology: Science and Practice, 14 (2), 175–78.

Holbrook, Morris B. (2007), “Objective Characteristics,Subjective Evaluations, and Possible DistortingBiases in the Business-School Ranking: The Case ofU.S. News & World Report,” Marketing EducationReview, 19 (2), 1–12.

Hughes, A.M. (1996), “Boosting Response with RFM,”Marketing Tools, 5, 4–10.

Hult, G. Tomas M., William T. Neese, and R. EdwardBashaw (1997), “Faculty Perceptions of MarketingJournals,” Journal of Marketing Education, 19 (1),37–52.

____________, Martin Reimann, and Oliver Schilke(2009), “Worldwide Faculty Perceptions of Market-ing Journals: Rankings, Trends, Comparisons, andSegmentations,” Global Edge Business Review, 3(3), Table 2. (accessed January 2011), [available atht tp : / /g lobaledge.msu.edu/newsandviews/businessreviews/gbr%203-3.pdf].

Koshal, Rajindar K., Manjulika Koshal, and Ashkok K.Gupta (1996), “Research Grants and School Rankingof PhD Granting Institutions: A Simultaneous Equa-tions Model,” Education Economics, 4 (1), 3–10.

Kotler, Philip and Kevin L. Keller (2007), A Frameworkfor Marketing Management (3rd ed.). Upper SaddleRiver, NJ: Pearson Prentice Hall.

Malthouse, Edward C. and Robert C. Blattberg (2005),“Can We Predict Customer Lifetime Value?” Jour-nal of Interactive Marketing, 19 (1), 2–16.

McCarty, John A. and Manoj Hastak (2007), “Segmenta-tion Approaches in Data-Mining: A Comparison ofRFM, CHAID, and Logistic Regression,” Journal ofBusiness Research, 60 (6), 656–62.

Meho, Lokman I. (2007), “The Rise and Rise of CitationAnalysis,” Physics World, 20 (1), 32–36.


____________ and Kiduk Yang (2007), “Impact of DataSources on Citation Counts and Rankings of LISFaculty: Web of Science versus Scopus, and GoogleScholar,” Journal of the American Society of Infor-mation Science Technology, 58 (13), 2105–25.

Neuhaus, Christoph and Hans-Dieter Daniel (2008), “DataSources for Performing Citation Analysis: An Over-view,” Journal of Documentation, 64 (2), 193–210.

Polonsky, Michael J. (2008), “Publishing on Publishing:Streams in the Literature,” European BusinessReview, 20 (5), 401–20.

Robinson, Larry M. and Roy Adler (1981), “Measuringthe Impact of Marketing Scholars and Institutions:An Analysis of Citation Frequency,” Journal of theAcademy of Marketing Science, 9 (2), 147–62.

Rust, Ronald T. and Peter C. Verhoef (2005), “Optimiz-ing the Marketing Interventions Mix in Intermediate-Term CRM,” Marketing Science, 24 (3), 477–89.

Rynes, Sara L. (2006), “Getting on Board” with the AMJ:Balancing Quality and Innovation in the ReviewProcess,” Academy of Management Journal, 49 (6),1097–1102.

Theoharakis, Vasilis and Andrew Hirst (2002), “Percep-tual Differences of Marketing Journals: A World-wide Perspective,” Marketing Letters, 13 (4), 389–402.

Thomson Reuters (2011), “The Thomson Reuters ImpactFactor,” (accessed July 2011), [available at http://thomsonreuters.com/products_services/science/free/essays/impact_factor/].

Touzani, Mourand and Salim Moussa (2010), “RankingMarketing Journals Using the Search Engine GoogleScholar,” Marketing Education Review, 20 (3), 229–47.

University of Texas at Dallas (2011), “The UTD Top 100Business School Research Rankings,” (accessed Janu-ary 2011), [available at http://som.utdallas.edu/top100Ranking/].

U.S. News (2011), “Marketing (MBA) Best BusinessSchools,” (accessed July 2011), [available at http://grad-schools.usnews.rankingsandreviews.com/best-graduate-schools/top-business-schools/marketing-rankings].

Vargo, Stephen L. and Robert F. Lusch (2008), “Service-Dominant Logic: Continuing the Evolution,” Jour-nal of the Academy of Marketing Science, 36 (1), 1–10.

Vastag, Gyula and Frank Montabon (2002), “JournalCharacteristics, Rankings and Social Acculturationin Operations Management,” Omega, 30 (2), 109–27.

White, Roger M., John B. White, and Michael M. Barth(2011), “Removing Size as a Determinant of Quality:A Per Capita Approach to Ranking Doctoral Pro-grams in Finance,” Journal of Education for Busi-ness, 86 (3), 148–54.

Yang, Amy X. (2004), “How to Develop New Approachesto RFM Segmentation,” Journal of Targeting, Mea-surement and Analysis for Marketing, 13 (1), 50–60.

For further information contact:Matt Elbeck

Troy University Dothan500 University Drive

Dothan AL 36303Phone: 334.983.6556, Ext 356

E-Mail: [email protected].


IMPACT DYNAMICS OF MARKETING SCHOLARSHIP: GOINGBEYOND JOURNAL QUALITY

Shibo Li, Indiana UniversityEugene Sivadas, University of Washington, Tacoma

Mark S. Johnson (Deceased)

SUMMARY

Researchers are paying increased attention to theimpact of marketing scholarship (Baumgartner and Pieters2003; Stremersch, Verniers, and Verhoef 2007). A desirefor objective measures of impact of published researchhas sparked interest in citation analysis. While the journallevel measures such as the impact factor of the journal apaper is published in is often used as a proxy for the qualityand potential impact of an individual article, it is a poormeasure of the impact of an individual article (Woodside2009). A second challenge is that evaluations of potentialinfluence of individual articles for tenure and promotiondecisions must be made on the basis of recent publicationsthat have not had sufficient time to gain citation recogni-tion.

In this paper, we suggest that article citation is drivenby its latent dynamic article attractiveness and adopt a newdiscretized Tobit model (Li, Liechty, and Montgomery2005) to estimate future citations potential and potentialimpact of an article. Our model can estimate within anynumber of years of publication, the potential citation-based impact of a published article. Our model captures

key drivers of impact including article, journal, and authoreffects and controls for time. It also accounts for otherarticle-specific variables and unobserved heterogeneitynot captured by previous research (Burrell 2003;Stremersch, Verniers, and Verhoef 2007).

The model is estimated using articles published inseven leading general marketing journals. We present anexpanded conceptualization of the Matthew effect andtest for and find strong presence of the article-levelMatthew effect. We document the Zero Citation Curseeffect which indicates that zero-cited articles become lessand less likely to get cited over time. The Zero CitationEffect is even stronger for “A” journals. There are signifi-cant author, journal, and article effects on article citationdynamics.

Our major contribution is that institutions and schol-ars from all disciplines can use our approach to dynami-cally predict citations impact of recent publications thatmay not have had sufficient time to garner recognition.Contributions to theory and implications for researchersand universities are also identified. References are avail-able upon request.

For further information contact:Eugene Sivadas

University of Washington, TacomaCampus Box 358420

1900 Commerce StreetTacoma, WA 98402Phone: 253.692.4707



THE EFFECTS OF INDIVIDUAL AND TEAM CHARACTERISTICS ONSIMULATION-ENHANCED CRITICAL THINKING:

A MULTILEVEL ANALYSIS

George D. Deitz, The University of MemphisAlexa K. Sullivan, The University of Memphis

Robert Evans, Jr., Texas A&M International University, Laredo

SUMMARY

The marketing education literature has emphasizedthe importance of augmenting traditional lecture-basedteaching approaches with more participatory, group-basedlearning experiences (Frontszcak 1998). Due to theapplied nature of many business disciplines, it has beenproposed that simulations are especially helpful in help-ing students make connections between classroom con-tent and the dynamic contexts in which such knowledge isapplied (Chapman and Serge 1999). It stands to reasonthat in assigning such projects, instructors are often inter-ested in optimizing both individual and team learningoutcomes.

This research presents findings from a multilevelSEM model that examined factors leading to studentperceptions of sim-enhanced critical thinking skills. Atthe within-team level, the study focuses on the role ofstudent motivation and cognition as facilitators of sim-enhanced critical thinking. At the between team level, thestudy examined the effects of team size upon objectiveand subjective learning. In addition, we considered thecross-level moderating effect of team size upon the self-efficacy à simulation-enhanced critical thinking relation-ship.

Hypotheses

According to attribution theory (Kelley 1967), peoplehave an inherent need to understand their environmentand create naïve theories to account for their circum-stances. Based on this view, teams with stronger gains inperformance over the course of the game are likely toattribute their success to themselves. It follows that,relative to students on poor performing teams, such indi-viduals are likely to place greater emphasis on a game’slearning benefits. Further, all things being equal, studentson smaller teams have more direct input into decisionmaking and more opportunity to engage in critical think-ing activities. As smaller team size is typically associatedwith improved objective team performance, it is expected:

H1: Growth in term performance will negatively mediatethe effect of team size upon overall team perceptions

that simulation participation enhanced critical think-ing ability.

Given undergraduates’ relative lack of work or simu-lation experience, simulation participation for class creditmay stir feelings of ambiguity and unease. Researchsuggests self-efficacious students participate more readilyand have fewer adverse emotional reactions when theyencounter difficulties (Bandura 1997). Further, studentself-belief increases commitment to fulfill game-relatedchallenges (Zimmerman 1990). Thus, we predict:

H2: Generalized self-efficacy will be positively associ-ated with individuals’ perceptions that participationin the simulation enhanced their critical thinkingability.

People differ in their tendency to enjoy and take partin challenging cognitive endeavors (Petty and Cacioppo1982). Since simulation-based games offer autonomy forstudents to engage in cognition, this format should beattractive to students who enjoy effortful thought. Fur-ther, high NFC students should be less affected by taskambiguity. This study contends:

H3: NFC will negatively moderate the effect of self-efficacy upon simulation-enhanced critical thinking,such that the relationship will be weaker for studentswith higher NFC.

All else being equal, decision-making in smallerteams is likely to be dominated by a single student. Asteam size grows, however, there is a greater likelihoodthat multiple high aptitude and high self-efficacy studentswill be on the same team. Therefore, we propose:

H4: Team size will negatively moderate the relationshipbetween individual self-efficacy and perceptions thatparticipation in the simulation enhanced critical think-ing ability.

Findings

Students enrolled in five consecutive semesters of aPrinciples of Marketing course participated as teams inthe Marketplace6 simulation (Cadotte 2009). Over the


course of the semester, student respondents completed aseries of surveys that assessed individual traits, learningpreferences, game-related knowledge, and perceptions ofthe simulation activity. A total of 537 usable responseswere received. Growth in team performance, measured bythe slope of a growth model comprised of longitudinalperformance measures over the last four decision rounds,negatively mediated the relationship between team sizeand overall perceptions that the simulation enhancedcritical thinking skills. This supports H1. Students’ gener-alized self-efficacy influenced their perceptions that simu-lation participation enhanced critical thinking ability, insupport of H2. The motivation-driven effects of self-efficacy were weaker for students high in NFC, support-ing H3. However, team size was not a significant predictorof the effects of self-efficacy upon sim-enhanced criticalthinking. Thus, H4 was not supported.

Sim-based learning provides students with the uniqueopportunity to gain applied problem-solving experience

as a part of their coursework. Using a multilevel design,we simultaneously investigated antecedents to studentand team perceptions of sim-enhanced critical thinking.We found that the effect of team size, an instructor-controllable design feature, upon enhanced critical think-ing was negatively mediated by the extent to whichobjective team performance improved over the course ofthe game. As we controlled for the effects of the initialperformance upon performance growth, this indicatedthat student perceptions of learning were channeled throughgains in performance over the course of the game and werenot simply based upon relative performance at the game’sconclusion. This implies that instructors can favorablyinfluence student perceptions of learning by acknowledg-ing performance gains through informal (e.g., encourage-ment) and formal (e.g., grading) mechanisms. Further,given the typically self-directed nature of simulationexercises, the importance of self-efficacy beliefs wasunderscored. The full paper provides a detailed discussionof findings. References are available upon request.

For further information contact:George D. Deitz

302 Fogelman College Admin BuildingThe University of Memphis

Memphis, TN 38152Phone: 901.678.2667



WITHOUT THE VOLUNTEERS THE EVENT CANNOT GO ON:EXAMINING RETENTION WITH A BEHAVIORAL

REASONING THEORY APPROACH

Mya Pronschinske, University of Wyoming, LaramieMark D. Groza, Northern Illinois University, DeKalb

Mark Peterson, University of Wyoming, Laramie

SUMMARY

“I truly appreciate the time, dedication, and commit-ment that you (the volunteers) make to the tournament –it goes without saying that the event could not functionwithout you” (Executive Director of THE PLAYERS).

Without the unpaid work and dedication of volun-teers numerous events worldwide would cease to exist.Retaining and motiving human capital – in this casevolunteers – has become a challenging and increasinglyimportant objective of management (Pfeffer 1994). Thischallenge becomes exacerbated when considering volun-teers are by definition unpaid and therefore not necessar-ily attracted or committed to an organization because ofeconomic reasons. As evidenced in the above quotationfrom a newsletter produced for THE PLAYERS Champi-onship golf tournament, volunteer retention is a keyconcern. Considering this important challenge theresearch addressed here is interested in volunteers’ inten-tion to remain committed to the organization they volun-teer for.

In this paper a conceptual model is developed andempirically tested which examines the retention of volun-teers. Behavioral Reasoning Theory (BRT) (Westaby2005) serves as the guiding theoretical framework for themodel developed in this paper. Specifically, BRT pro-poses that the beliefs and values volunteers hold lead tothe reasons they engage in the behavior of volunteering.It is proposed that these reasons (organizational reputa-tion and corporate community relations), in turn, lead tothe attitudes (pride) the volunteers have toward the orga-nization and these attitudes then, lead to intentions toremain with the organization. With the exception ofBriggs, Peterson, and Gregory (2010), BRT has not beenstudied in a volunteer context. While these authors adoptedBRT, they stopped at the global motives and did notexamine volunteers’ intention to remain with the organi-zation. The current study builds on this research to under-stand the degree to which volunteers’ values, reasons forvolunteering, and the attitude of pride is linked to theirintentions to remain a volunteer with the specific organi-zation.

The conceptual model developed in this paper istested using structural equation model of data collectedfrom volunteers of a large Professional Golf Association(PGA) golf tournament. Following the completion of thetournament, an on-line survey was emailed to the 1,980participating volunteers (n = 718, 36% response rate).Males comprised the majority of the sample (71%) and theaverage age of the respondents was 57 years old. Theaverage tenure or years of volunteering service for the golftournament is six years. Forty-three percent of the respon-dents choose to take personal/vacation time off from workto volunteer for the tournament.

The statistical analysis provided support to seven ofthe ten hypothesized relationships. Results indicate apositive relationship between the values congruence ofthe individual and the organization and organizationalreputation (H

1) (β = .67, p < .001). Similarly, H

1b which

posits a relationship between the values congruence of theindividual and the organization and corporate communityrelations (CCR) is also supported (β = .71, p < .001).Behavioral reason theory suggests that reasons are animportant independent predictor of global motives,beyond expectations and values. The results indicatesupport for H

2a, which links reasons of being involved

with an organization known for having a good reputationto the pride in membership attitude (β = .91, p < .001). H

2b,

which posits a positive link between reasons of CCR andthe global motive of pride in membership however, is notsupported.

BRT also suggests that people rely on their subjectivereasons to form behavioral intentions (Westaby 2005).That is, reasons are expected to predict intention over andabove that which is explained by global motives. It washypothesized that reasons for intending to volunteer wouldbe positively related to intention to return next year andthe number of hours willing to commit in the future.Organizational reputation (H

4a and H

4b) did not have a

significant relationship with the likelihood of returning orhours willing to commit. We did however, find support forH

4c and H

4d, which links CCR to likelihood of returning

and hours willing to commit (β = .10, p < .05 and β = .44,p < .01, respectively). Support was provided for H

3a which


posits a link between the pride attitude and intention toreturn to next year’s event (β = .42, p < .001). We also findsupport for H

3b, a positive relationship was uncovered

linking pride to the number of hours willing to commit tonext year’s event (β = .54, p < .001).

This study offers contributions to theory and ourunderstanding of organizational reputation and corporatecommunity relations. Organizational reputation does notdirectly influence the retention of volunteers; rather it ismediated by pride in the organization. This is similar to thefindings of Boezeman and Ellemers (2008) who foundvolunteer commitment to the organization to be partially

mediated by the pride and respect they derived frommembership. Our research focused on current volunteers(rather than the recruitment of prospective volunteers)which furthers a contribution to theory that organizationalreputation may work to attract volunteers but also for avolunteer to remain involved, they must feel proud of theorganization. On the other hand, corporate communityrelations leads directly to an intention to remain with theorganization. The CCR construct has yet to be tested in thevolunteer context; thus this research begins to illustratethat a volunteer may appreciate when the organizationthey volunteer for acts as a good corporate citizen. Refer-ences are available upon request.

For further information contact:Mya Pronschinske

Management and Marketing, Dept. 3275University of Wyoming1000 E University Ave.

Laramie, WY 82071Phone: 307.766.3124



MOTIVATING FACTORS FOR PARTICIPATION IN NATIONAL ANDOLYMPIC SPORTS IN KAZAKHSTAN

Elmira Bogoviyeva, KIMEP University, Almaty, Kazakhstan

SUMMARY

Academic studies on sport marketing have addres-sed a variety of different issues, including: drivers forinvolvement (Gaston-Gayles 2004; Gillet et al. 2009;Maclean and Hamm 2008); demographic factors (Allenderet al. 2006; Hardin and Greer 2009; Koivula 1999; Green2002); event attendance and fan support (Lee 2002).Recent studies have also focused on the increasinglypopular extreme sports (Park 2004; Ko et al. 2008). Mostof these studies explored traditional or Olympic sports.However, as sport is an important element of nationalculture, governments are also interested in their citizensparticipating in sports that are nation specific. Exploringmotives for participation in National sports can help, notonly in preserving the culture, but developing and dis-seminating a nation’s values and traditions.

Some studies investigate the influence of nationalidentity and national pride among sport fans (Funk et al.2002; Bogdanov 2005) and differences in the degree ofinterest in sport activity, due to ethnicity have beeninvestigated in USA (MacClancy 1996). However,motives that drive participation in National sportsreceived little attention.

The main objective of the research is to investigatethe motives behind participation in the most popularOlympic and National sports in Kazakhstan: tennis, foot-ball, baiga, and kazaksha-kures. Based on this objectivethe following research questions are developed:

RQ1: Are there differences in the factors that motivateparticipation in Olympic and National sports?

RQ2: Are there differences based on gender?

RQ3: What roles do nationalism and patriotism play inparticipation decisions?

A survey approach is selected to complement theexploratory nature of the study. Existing scales on sportparticipant motivation were used (Ko et al. 2008;McDonald et al. 2002). In addition, items on nationalism,patriotism and national identity were added (Kostermanand Feshbach 1989; Bogdanov 2005). A specific scale forparticipation in National sports is developed and vali-dated. Two hundred and sixty-three non-professionals,

who participate in football, tennis, kazakhsha kures(national wrestling) and baiga (national horseback riding)responded to the survey. Analysis of variance (ANOVA)and multiple analyses of variance (MANOVA) wereconducted in the context of Ko et al. (2008) four dimen-sions (sport characteristics, existence, relatedness, andgrowth).

In Kazakhstan the most important motive was foundto be physical fitness (6.02), followed by fun and enjoy-ment (5.75) and risk taking and stress reduction (both5.73). The least important factor for Kazakhstani citizens,who participate in the sports, appears to be aggression(5.11). Nationalism (6.16), and patriotism (6.39) hadscores higher than the physical fitness demonstrating highrelevance to all participants across sports.

For hypothesis one the data partially supports theview that there are differences between the motives forparticipation in Olympic and National sports. There wassignificant difference found between national and Olym-pic sports on eleven out of fourteen motives: self-esteem,achievement, value development, self-actualization,affiliation, social facilitation, skill mastery, aggression,competition, aesthetics and risk taking (Ko et al. 2008).The results reveal significant difference among individu-als, who participate in Olympic sports (football, tennis)and the National sport of baiga. The difference betweenfootball, tennis, kazaksha kures on one hand and baiga onthe other hand was found for almost all motives, exceptthe physical fitness motive between baiga and football.

The second hypothesis related to differences in per-ceived patriotism among participants in Olympic andNational sports; more specifically participants ofNational sports were expected to have a higher degreeof patriotism than participants of Olympic sports. Thishypothesis was partially supported. There was a signifi-cant difference found between individuals playing foot-ball and subjects participating in baiga. Surprisingly,people participating in kazaksha kures had significantlylower level of patriotism than those, who practice baiga.

The third hypothesis stated that there are differ-ences in nationalism among participants of Olympicand National sports; more specifically participants inNational sports have a higher degree of patriotism thanparticipants of Olympic sports. This hypothesis was par-


tially supported. Individuals participating in Nationalsport baiga had significantly higher level of nationalismthan subjects participating in all other sports, includingkazaksha kures.

To summarize, this is a premier study to explore themotives for participation in National sports. Olympic andNational sports’ participants were compared across fourdimensions of sport participation motives (Ko et al. 2008),and on attitudes to nationalism and patriotism. Differ-

ences between National and Olympic sports were foundacross eleven out of fourteen motives. These motivesappear particularly relevant and influential for Baigaparticipants. The study findings revealed males confirm-ing greater relevance of the motives assessed in the study.Comparison between sports revealed motivational differ-ences existing for each specific sport, as well as gender-based differences, thus suggesting careful design of mar-keting programs for each sport and more careful targetingof communications to potential recruits.

For further information contact:Elmira BogoviyevaKIMEP University

2 Abai AvenueAlmaty, Kazakhstan

Phone: +7.727.3763928Fax: +7.727.2704440



SEGMENTING FANS OF A NEW TEAM: A TYPOLOGYOF EARLY ADOPTERS

Heath McDonald, Swinburne University, AustraliaCivilai Leckie, Swinburne University, Australia

Adam Karg, Deakin University, Australia

SUMMARY

New professional sports teams arise through leagueexpansion, relocation or new league creation. In eachcase, the identification of those consumers most likely tobecome fans of the new team, and the development ofmeaningful relationships with those consumers, is para-mount to organizational success. Past research suggests anumber of reasons fans choose to connect with sportsteams. However, fan development in the context of a newteam is rarely examined. This paper looks specifically ata large sample of the first fans (n = 1741) to connect withan elite professional team in its inaugural year of opera-tion, with the specific aim of gaining a clearer understand-ing of who associates with a new team and why.

Sport organizations are faced with increasingly frag-mented customer bases and fierce competition for rev-enue driven by attendance, viewership and other forms ofconsumption. Such changes have heightened calls for agreater focus on understanding and segmenting complexconsumer bases for both participation and spectator levelsof sport. Without proper segmentation and targeting strate-gies sport organizations, including teams, can find itdifficult to effectively leverage limited resources todevelop and maintain relationships with various con-sumer groups. Responding to this need, there have beena number of recently proposed theoretical and practicalapproaches to classifying types of fans. These includeprogressive frameworks such as the Psychological Con-tinuum Model (PCM) (Funk and James 2001), and othercategorical segmentation processes based on motivations(Ogles and Masters 2003), brand associations (Ross 2007),sport consumption (Dwyer and Drayer 2010; Dwyer,Shapiro, and Drayer 2011), sport participation (Taks andScheerder 2012), demographics and skill level (Casper2012). If easily identifiable segments of new fans exist,profiling them should allow the development of moreefficient marketing tactics aimed at expediting loyalty andcommitment. This information could, in turn, add morelight on why new fans come to existing teams, especiallyin cases outside of inherited or socially transmitted fandom.

To examine any heterogeneity amongst early adopt-ers of a new team, data were collected from a survey withfans of the Gold Coast Suns, an expansion team of the

Australian Football League (AFL). The AFL is the high-est profile professional sport in Australia, with the teamjoining in 2011 as the 17th team in the league. Thesampling frame was a database compiled throughout theprocess of launching the new team, where residents of theGold Coast area were encouraged to sign up via a websiteto support the club’s creation. Within this group werepeople with widely varying levels of interest. Some wenton to become season ticket holders in the inaugural yearwhile others had no further contact with the club. All fanson this database with a current email address were invitedto complete an on-line questionnaire, completed fourweeks into the clubs inaugural AFL season. In total, 1741responses were collected, a response rate of 23 percent.Items to develop clusters, as well as output variables toexplore differences between them were developed fromexisting measures in all cases.

Initial consumer identification items were developedfrom existing measures from past sport-related segmenta-tion studies. A six-dimension measure of team identitywas used (Heere and James 2007), with each dimension(private evaluation, public evaluation, interdependence,interconnection, cognitive awareness, and centrality) hav-ing three items. Loyalty and commitment measures wererepresented by resistance to change, level of commitment,stability of preference and degree of resistance a con-sumer has in changing a sport team (Pritchard, Havitz andHoward 1999). Brand loyalty intentions to keep support-ing and purchasing a brand (Chaudhuri and Holbrook2001) were also included. Perceived corporate socialresponsibility utilized Lichtenstein et al.’s (2004) pastwork, attempting to capture those fans engaged with theteam through its community support.

After scale purification and validation, a hierarchicalclustering algorithm was used to determine the appropri-ate number of clusters (Cannon and Perreault 1999; Punjand Stewart 1983). We then employed SPSS K-meanprocedures, using either factor score (Singh 1990) orstandardized mean scores (Wong et al. 2010) to developcentroids and classify the sample. Wong et al.’s (2010)use of cross validation procedures was used to assessinternal validity. The empirical segmentation suggestedthe presence of five clusters, each comprising between 14percent and 28 percent of the sample. The five segmentswere found to differ markedly in terms of brand associa-


tions, satisfaction and behavioral involvement. We namedthem Community Focused, Publicly Detached, HighlyRelational, Lowly Relational, and Passive Observers.

The highly satisfied and involved consumers of thehighly relational segment are the consumers that demon-strated the highest levels of commitment, loyalty andprivate evaluation, and would be a key target for encour-aging positive word of mouth. Community focused fansare moderately involved with the team and identifiedthrough involvement in CSR activities, highlighting theimportance of CSR activities for a new sport team atcommunity level (Walker and Kent 2009). The publiclydetached group represent moderately satisfied customers,who exhibit strong loyalty but distance themselves fromthe team in a public or social sense (Madrigal 1995). Theymay enjoy the drama and excitement the game brings, butdo not have much emotional attachment to the team. Thelowly relational segment captures casual observers withlow levels of commitment and loyalty and poor private

evaluation. Likewise, passive observers have low per-sonal involvement with the team and connectivity, how-ever, despite little personal connection, they follow theactivities of the team closely. These latter two groupspresent a key challenge for the team, given that their lowlevels of identity (and subsequent outcomes of this) meanthey are highly likely to churn as customers or fans,especially if on-field performance is poor.

For the new team in question, the drivers of identityhave formed quickly, with attitudinal and behavioraldifferences between segments suggesting there is signifi-cant value in pursuing segmentation strategies in thiscontext. Instead of segmenting consumers according tosequential links among discrete variables (e.g., the PCM),the approach employed here allowed for the coexistenceof multiple dimensions in individual relationships with asport team. In that way it reflects the complexity of themany ways new fans can connect with a new team.References are available upon request.

For further information contact:Heath McDonald

Swinburne UniversityP.O. Box 218

Hawthorn, Victoria 3122Australia



THE SERVICE PROFIT CHAIN IN A PROFESSIONAL SPORTS SETTING

Adrien Bouchet, The University of TulsaJames J. Zboja, The University of Tulsa

SUMMARY

The service profit chain is a framework used byscholars to describe the antecedents of profit and growth(Heskett et al. 1994). Specifically, the service profit chainexplores the relationship links starting with internal ser-vice quality, employee satisfaction and productivity andonward to customer satisfaction and loyalty, with thedesired outcomes of revenue growth and profitability(Heskett et al. 1997). The underlying premise of theservice profit chain has been studied for years (Bennis1970; Lawler 1973; Blau 1974). However, this stream ofresearch did not begin to come together until Heskett,Sasser, and Hart (1990) identified a “self-reinforcingservice cycle.” Further research into this phenomenon bySchlesinger and Heskett (1991) produced what they calledthe service profit chain (SPC). Prior research has explorednumerous aspects of the relationships among these indi-vidual links. However, there is a paucity of research thatexamines all the contiguous links in the service profitchain. This empirical study accomplishes that by examin-ing data from a single professional sports franchise.

One of the franchises in the National Hockey League(NHL) has adopted a service strategy based on the idea ofthe service profit chain. This franchise has invested con-siderable monetary and human capital in the belief that theservice profit chain will help grow revenue. This studyreports the findings of an empirical study, conducted inconjunction with the management of this NHL franchise,to measure the service profit chain in a sports franchisesetting. While the premise of the model on both theemployee and customer sides is accepted for this specificindustry, it is the contention of the authors that the serviceprofit chain will be broken where the employee variablesmeet the customer variables. Given the combination of theimportance of the overall league brand to driving cus-tomer outcomes and the outsourcing of key customercontact personnel, the importance of the contactemployee to the customer experience is mitigated. Hencewe propose the following:

H1: There will be a break in the service profit chain for a

professional sports franchise at the link betweenemployee variables and customer variables.

The data were collected in conjunction with themanagement of a NHL franchise located in a non-traditional market. The organization sent a link to thecustomer survey to 750 recent buyers of single game

tickets. Meanwhile, the employee survey link was sent to220 frontline workers with game day responsibilities atthe arena. After removing cases for missing items andrandomly matching the two separate samples, the result-ant final sample size was 175 for each.

Study constructs (employee perceived service qual-ity, job satisfaction, employee productivity, customerperceived value, customer perceived relationship satis-faction, and customer loyalty) with the exception of therevenue change item, were all measured with reliable,established scales previously used in the marketing litera-ture. All included scales exhibited sufficient internalconsistency, with Cronbach alphas ranging from .82 to.91. Additionally, tests of convergent and discriminantvalidity (Fornell and Larcker 1981) were administered.

The hypothesized model was tested with an item-level structural equation model (SEM) using variance-based SmartPLS Graph Version 2.0M3 (Ringle, Wende,and Will 2005). The results indicate that all of the paths(except for the path from employee productivity to cus-tomer perceived value) in the structural model were sig-nificant at the p < .05 level, lending support to H

1. The

r-square values indicated that the hypothesized modelaccounted for the following amounts of explainedvariance: 41.9 percent of job satisfaction, 17.7 percent ofemployee productivity, 14.9 percent of relationship satis-faction, 38 percent of customer loyalty, and 3 percent ofrevenue change.

Significant path loadings demonstrated that both theemployee and customer sides of the service profit chainwere indeed intact. Internal service quality is positivelyrelated to work job satisfaction (ß = .648) which was, inturn, positively related to employee productivity (β =.420). At this point, as hypothesized, the service profitchain is broken. For sports at the highest professionallevel, employee productivity does not naturally translateto customer-perceived value, as this path was not found tobe significant. However, the remainder of the serviceprofit chain continues as intended. Customer-perceivedvalue is positively related to relationship satisfaction (β =.386), which is also related to customer loyalty (β = .616).Finally, customer loyalty was found to be positivelyrelated to change in revenue (ß = .160). Taken as a whole,these results demonstrate that, in this context, there arelimitations to the service profit chain concept despite itstheoretical elegance and intuitiveness.


Despite its strength on both the employee and cus-tomer sides of the equation, these results found the serviceprofit chain to be missing a link in the context of aprofessional sports franchise. Although the results mayseem to suggest that customer service levels are notimportant for professional sports franchises, this findingshould be taken with a caveat. That is, it would seemsomewhat reckless to suggest this. It would seem more

appropriate to view customer service levels as a hygienefactor, in the parlance of Herzberg (1987). That is, acertain base level of customer service quality is a neces-sary, albeit not sufficient, antecedent to customer satisfac-tion and, ultimately revenue gains and profitability. While,in itself, its presence may not translate directly to thesedesirable outcomes, it absence would surely negativelyimpact them. References are available upon request.

For further information contact:Adrien Bouchet

The University of Tulsa800 South Tucker DriveTulsa, OK 74104–9700Phone: 918.631.2943



THE EFFECT OF FORMAL AND INFORMAL MARKETING CONTROLSON CUSTOMER CONTACT EMPLOYEE PERFORMANCE

Ryan C. White, University of Wisconsin – La CrosseRoger J. Calantone, Michigan State University, East LansingClay M. Voorhees, Michigan State University, East Lansing

SUMMARY

Managers have made the quality control of customercontact employees a prominent issue as a result of therecognition of the importance and centrality of the cus-tomer contact employee’s performance in forming theservice encounter to the customer (Czepiel, Solomon, andSurprenant 1985). A need still exists, however, to under-stand the effectiveness of specific formal and informalmarketing controls at influencing employee performancewhen simultaneously implemented. This study answersthat need by using hierarchical linear modeling (HLM) toanalyze data collected from 88 managers and 157 cus-tomer contact employees.

Typically, the control of marketing activities, such asthe control of employee behavior, can be thought of asfalling into one of two broad groups, formal and informalcontrols (Jaworski 1988). Formal controls are the written,management-initiated mechanisms that influence the prob-ability that employees will behave in the manner desiredby the firm (Jaworski 1988). Agency theory suggestsoutcome-based contracts and behavior-based contractswith employee monitoring. These solutions, however,cannot by definition address employee extra-role behav-ior as extra-role behavior is the actions of the employeewhich goes beyond role requirements (Bettencourt andBrown 1997).

As a result, the formal systems of the traditionalmarketing control perspective need to be augmented withother forms of control, such as informal controls (Jaworski1988). Briefly, informal controls are the unwritten, usu-ally worker-based mechanisms that influence the indi-vidual employee’s behavior (Jaworski 1988) and differfrom formal controls because they are not formally docu-mented, may not match the firm’s goals, are initiated bythe employees, and are usually only actively controlled bythe employees. Within these informal controls, self con-trols are how individuals establish personal objectives,monitor their attainment of the objectives, and adjust theirbehavior if the objectives are not being met.

With regard to predicting customer contact employeein-role and extra-role performance, the development oftwo employee self controls are important to firms seeking

to achieve a sustained competitive advantage throughtheir customer contact employees, an employee’s cus-tomer orientation (enduring disposition to meet cus-tomer needs) and an employee’s intrapreneurial orien-tation (enduring disposition to behave in an entrepreneur-ial manner, which is to pursue opportunities to be asuccessful customer contact employee without regard tothe resources currently controlled).

Therefore, it is hypothesized that:

H1: The interaction between the agency theory-basedsolutions of behavior-based contracting andemployee monitoring has a positive effect onemployee in-role performance.

H2: The agency theory-based solution of outcome-based contracting has a positive effect onemployee in-role performance.

H3: The agency theory-based solution of behavior-based contracting and employee monitoring has apositive effect on employee extra-role performance.

H4: The agency theory-based solution of outcome-based contracting has a positive effect on emplo-yee extra-role performance.

H5: A customer contact employee’s customer orienta-tion will have a positive effect on employee in-roleperformance.

H6: A customer contact employee’s customer orienta-tion will have a positive effect on employee extra-role performance.

H7: A customer contact employee’s intrapreneurialorientation will have a positive effect on employeein-role performance.

H8: A customer contact employee’s intrapreneurialorientation will have a positive effect on employeeextra-role performance.

H9: The interaction between the agency theory-basedsolutions of behavior-based contracting and


employee monitoring has a relatively weakereffect on employee extra-role performance thanself controls.

H10: The agency theory-based solution of outcome-based contracting has a relatively weaker effect onemployee extra-role performance than self con-trols.

All constructs were assessed using a combination ofextant and new scales, which were developed consistentwith the procedure prescribed by Churchill (1979) andmeasured using a five-point Likert type scale. Each man-ager provided ratings of the formal marketing controlconstructs of behavior-based contracting (MBC), moni-toring (MMON), and outcome-based contracts (MOC).Each employee provided ratings of their own in-roleperformance (EIRP), extra-role performance (EERP),customer orientation (ECO), and intrapreneurial orienta-tion (EIO). Employees were nested within the managerthat they identified they had worked with the most andHLM via HLM 6.08 (Raudenbush and Bryk 2002) was

used to analyze the data. Cronbach’s values ranged from0.78 to 0.95, composite reliability values ranged from0.83 to 0.97, and convergent and discriminant validitywas established per Fornell and Larcker (1981).

Hypotheses 5, 6, 8, 9, and 10 were supported whilehypotheses 1, 2, 3, 4, and 7 were not supported. Thepattern of results offers two key implications regardingthe influence of formal and informal marketing controlson employee performance. First, the relative ineffective-ness of formal marketing controls and the relative effec-tiveness of informal marketing controls at influencingemployee performance are worth noting. Second, theidentification of specific key informal controls whichinfluence on employee performance is also worth noting.This study demonstrates the importance of the establishedinformal control of customer orientation at influencingemployee performance, introduces an additional informalcontrol, intrapreneurial orientation, and is an importantfirst step in identifying an informal control, other thancustomer orientation, that positively influences employeeperformance. References are available upon request.

For further information contact:Ryan C. White

University of Wisconsin – La Crosse1725 State Street

La Crosse, WI 54601Phone: 608.785.8097



BAD CUSTOMERS OR BAD MANAGEMENT? AN EMPIRICALINVESTIGATION OF WHAT DRIVES SERVICE

EMPLOYEES’ DEVIANT BEHAVIOR

Gianfranco Walsh, Friedrich Schiller University of Jena, GermanySimon Brach, Friedrich Schiller University of Jena, Germany

Arne K. Albrecht, Friedrich Schiller University of Jena, GermanyDavid Dose, Friedrich Schiller University of Jena, Germany

Patrick Hille, Friedrich Schiller University of Jena, Germany

SUMMARY

Because of its negative effects on organizationalperformance employee workplace deviance is of greatinterest for services research. In this study, a model for abetter understanding of the relationship between extra-and intra-organizational antecedents and employee’s jobsatisfaction and its link to workplace deviance is devel-oped and tested.

Workplace deviance in service organizations is rel-evant for managers because it harms firm performance inseveral ways. It might create direct costs if employeesdestroy firm property (Schmitt et al. 2003) or lead toindirect costs if it harms the morale of co-workers (Hunget al. 2009) or prevents delivery of appropriate service tocustomers (Patterson and Baron 2010).

Scholars have examined related antecedents ofemployee workplace deviance such as intra-organiza-tional variables within the service firm’s control(Colbert et al. 2004; Thau et al. 2009), as well as intra-organizational variables over which the firm has little orno control (Diefendorff and Mehta 2007; McIntyre andMeloche 1995). In addition, the influence of extra-orga-nizational variables, such as customers’ behavior, hasreceived attention (Walsh 2011; Wegge et al. 2007).However, research so far is missing a structured examina-tion of the links between antecedents and workplacedeviance, together with a careful consideration of poten-tial mediating and moderating effects in the relationsbetween workplace deviance and its drivers. Drawing onprior research, this study closes this gap as it proposes amodel considering multiple pertinent antecedents simul-taneously.

In the proposed model antecedents of employee work-place deviance are structured and conceptualized along acontinuum, from low to medium to high firm control.Three factors represent the continuum: perceived cus-

tomer unfriendliness, service climate and employee roleambiguity. Furthermore, consistent with previous research(e.g., Judge et al. 2006) job satisfaction is included as avariable mediating the antecedents’ influence onemployee workplace deviant behavior. Drawing onErickson and Grove’s (2007) findings that older employ-ees are more likely than younger employees to engage inemotion regulation to manage the negative emotions thatarise from their jobs, the moderating effect of employeeage in the relationship between job satisfaction and work-place deviance is investigated as well.

The proposed model was tested via structural equa-tion modeling. Building on the notion that workplacedeviance results from factors that firms can influence tovarying degrees, the results show that antecedents repre-sentative of low, medium and high organizational controlall affect job satisfaction, which in turn can minimizeworkplace deviance. The more satisfied service employ-ees are, the less likely they are to engage in deviantbehavior (Judge et al. 2006). Moreover, perceived cus-tomer unfriendliness directly affects workplace deviance.Testing for contingency conditions of these job satisfac-tion–workplace deviance relationships shows this effectto be stronger for younger service employees. Youngerservice employees’ job satisfaction exhibited a morenegative association with workplace deviance than didthat of older service employees (Erickson and Grove2007).

The results suggest implications for researchers andpractitioners. To prevent workplace deviance, servicefirms need a clear understanding of its drivers, so they canseek avoidance routes. The proposed model enables man-agers of service organizations to study and measure howdifferent types of internal antecedents, namely, employeerole ambiguity and service climate, as well as the mediat-ing effect of employee satisfaction, engender deviantbehaviors in the workplace. References are availableupon request.


For further information contact:Gianfranco Walsh

Friedrich Schiller University of JenaCarl-Zeiss-Strasse 3

Jena, 07745Germany

Phone: +49(0)3641.943110Fax: +49 (0)3641.9.43112



LINKING IMO WITH DIFFERENT FIT TYPES AND WILLINGNESS TOREPORT SERVICE COMPLAINTS

Achilleas Boukis, Athens University of Economics and Business, GreeceSpiros Gounaris, Strathclyde University, United Kingdom

Kostas Kaminakis, Athens University of Economics and Business, Greece

ABSTRACT

This study integrates the equity and the fit theory inorder to explore the impact of retail store supervisor’sInternal Market Orientation (IMO) on contact employees’IMO, fit with their organization and their supervisor.Moreover, the importance of the aforementioned vari-ables is displayed for contact employees’ willingness toreport service complaints.

INTRODUCTION

Considering contact employees’ pivotal role forretail services (Liao and Chuang 2004; Bitner et al.1994), scholars highlight the benefits stemming frominternal market orientation (IMO) adoption in terms ofleveraging contact employees’ performance (Gounaris2008). IMO constitutes a philosophy which suggests theimportance of having satisfied and motivated contactemployees in order to provide superior customer service(Tortosa et al. 2010). Despite that aligning employeeswith organizational goals and directives and diffusingorganizational values across the firm constitute ultimategoals of IMO adoption (Gummesson et al. 1987), scholarsrarely encapsulate in internal marketing models the im-portance of these issues (Wieseke et al. 2009).

Therefore, this study addresses the role of IMOadoption for enhancing employees’ fit with their environ-ment within a retail context, as scholars produce arestricted understanding of the simultaneous influenceof different fit indices on employee attitudes (Lauver andKristof-Brown 2001) and fail to provide mechanisms ofhow to increase employees’ fit with their organization(Edwards and Cable 2009). Given that employees’ lack offit with the organization can provoke several destructiveoutcomes for the firm, managers are in need of ways toenhance of employees’ fit with their environment(Hofmann et al. 2011; Tak 2011). We focus on more thanone fit types as the work environment is comprised ofdifferent entities with which individuals may fit while fitlevels vary across subcultures that exist within differentorganizational functions (Tak 2011). In particular, wefocus on employee-organization (E-O) fit which yieldssignificant effects on a variety of employee-level out-comes such as commitment and job performance (Kristof-Brown et al. 2005; Vilela et al. 2008) as well as on

employee-supervisor (E-S) fit lies whose value lies uponsupervisor’s indispensable role in retail services as (s)herepresents the firm itself on subordinates’ eyes (Lam et al.2010).

In overall, we develop a conceptual model whichextends the present literature by integrating IMO withinthe fit theory. From an empirical view, we argue for theimportance of IMO for employees’ fit with their organiza-tion and their supervisor by adopting a multilevel researchdesign drawing evidence from both supervisors and con-tact employees in a retail service context. In particular, weinvestigate whether supervisor’s IMO accounts for con-tact employees’ IMO adoption and different fit types (i.e.,E-S and E-O fit) and explore how contact employees’IMO and fit types shape their willingness to report servicecomplaints.

THEORETICAL BACKGROUND

Evidence from several fields is integrated in order toprovide the theoretical background of our study. Fittheory underlies our conceptual model in the sense that itsbasic demarcation assesses how fit with various aspects ofthe work environment influences individuals’ attitudesand behaviors (Kristoff 1996). Although individualsoften engage themselves into jobs that best match theirabilities and interests, however, due to changes in theirworking environment, they may find themselves in amiss-fit situation, which causes stress and low job satis-faction (Edwards 1996). Organizations highly oriented totheir internal market care for employees’ needs and wantsand thus, are more inclined to recover employees’ congru-ence with their environment either by re-establishing jobsatisfaction or by stressing a positive working environ-ment through the formation of a trust climate (Johnstonet al. 1990), higher empowerment levels (Gounaris 2008)and the delivery of higher value for employees (Simberova2007). We also stress the importance of the equity theory(Adams 1963) which underlies the internal market orien-tation (IMO) philosophy (Gounaris et al. 2010). On thebasis of the equity theory, employees evaluate their jobsby balancing their inputs with the associated outputs.Inputs include effort on the job and compliance to organi-zational policies whereas outputs include benefits andrewards (Huseman and Hatfield 1990). Consequently,when service firms deliver higher value for employees


(Simberova 2007), the latter become more willing toreciprocate toward their employer with feelings of obliga-tion, to embrace organizational directives and values andto put the extra effort to serve customers better and deliversuperior service (Gounaris et al. 2010).

LITERATURE REVIEW AND MODELDEVELOPMENT

As IMO implementation requires multilevel manage-ment to continuously align employees with organiza-tional goals (Gummesson 1987), this study adopts amultilevel approach of IMO adoption in order to displaythe importance of cross-level interactions within the su-pervisor-employee dyad and overcome concerns relatedto single-level analyses, considering that by examiningone level at a time prevents one from knowing whetherfactors at one level remain significant predictors afterfactors at the other level are accounted for (Raudenbushand Bryk 2002). Our model proposes supervisor’s IMO asa predictor of contact employees’ E-O fit and E-S fit afteraccounting for individual IMO in order to highlight thatthe aforementioned impact remains significant after con-trolling for employees’ level of IMO (H4, H5) (Figure 1).In addition, we investigate the impact of contact employ-ees’ IMO and fit types on willingness to report servicecomplaints (WRC).

Supervisor IMO and Employee Outcomes

Internal marketing is a philosophy of treating em-ployees as internal customers (Sasser and Arbeit 1976)and enhances the value provided to employees with theaim of encouraging them to enact the firm’s marketing

objectives and thus strengthen its competitive position inthe external market (Ahmed et al. 2003; Lings and Greenley2010). Despite several attempts at the conceptualizationof the internal marketing construct (Gounaris 2008; Lings2004), recent advances in the field of internal marketingidentify specific managerial behaviors associated withenacting this philosophy and their conceptualization as aninternal market orientation (IMO) (Lings 2004). IMOpromotes the need to plan and build effective relation-ships between the company’s employees and manage-ment (Gounaris 2006) so that the company responds toemployees’ needs and wants more effectively, motivatingthem to better satisfy customer needs. However, theinternal marketing literature mainly focuses on individualgains from IMO adoption (Lings and Greenley 2010),despite normative studies emphasizing supervisor’s IMOfor shaping positive employee-level consequences (Ahmedand Rafiq 2003). Moreover, although the main pillars ofthe IMO construct (i.e., management-employee commu-nication) stress the importance of cross-level interactions(Gounaris 2008) between different organizational ech-elons for IMO adoption, no previous research addressesthis issue.

This study argues that IMO adoption within retailfirms can be actualized through two routes. The main oneis through the social learning process (Bandura 1977)which argues that as managers are considered as strongbehavioral models for subordinates (Manz and Sims 1981),subordinates observe the outcomes of their role models’behavior and develop a propensity to engage in a similarbehavior (Waldman and Yammarino 1999). Thus, man-agers with a strong manifestation of IMO can facilitatesubordinates’ adoption of IMO philosophy and drive

Supervisor Level (2) Employee Level (1)

Internal Market

Orientation

Employee-organization fit

Internal Market

Orientation

H1

H2

H3

H4

Employee-supervisor fit

H5

H8

H6

H7

Employee willingness to report complaints

FIGURE 1


them toward developing behaviors consistent with theirpersonal orientation. We complementarily employ thesocial identity theory (Tajfel 1981) which accounts foremployees’ IMO adoption which argues that group iden-tification increases the internalization of group’s aims andgoals and thus the more an employee identifies with agroup the more one will work toward achieving its goals(Ashforth and Mael 1989). Extending the idea of leaders’communicating by deeds in the internal marketing litera-ture (Berry et al. 1976), we argue that manager’s strongmanifestation of IMO-congruent behavior will render thefirm as a more attractive target for subordinates. In thisvein, supervisor’s display of an IMO-consistent behaviorthrough their positional or referent power is likely to exertinfluence on contact employees and render them moresusceptible to the internalization of IMO philosophy. Onthese grounds, we suggest that supervisor’s level of IMOcan enhance contact employees’ IMO (H1).

Fit theory suggests positive employees’ attitudes dueto higher congruence levels (Chatman 1991; O’Reillyet al. 1991). At the heart of this construct is the congru-ence between individual and organizational values (Kristof1996). We argue that the implementation of an internalmarketing program can prove quite beneficial foremployee-organization (E-O) fit as IMO assistsemployees in developing a holistic view of the firm’sphilosophy and instils them the organization’s values andculture (Gronroos 1990) through the formation of effec-tive internal communication between management levels(Lings 1999). Moreover, supervisors through the provi-sion of higher organizational support can create a climateconducive for employees to embrace organizational val-ues as their own and enhance understanding of their rolefor organizational effectiveness (Conduit and Mavondo2001). In this vein, we argue that supervisor’s IMOadoption accounts for employees’ perceptions of a goodfit with the organization (H2), after controlling for indi-vidual level of IMO (H4).

Although traditional fit literature focuses on the E-Ofit (Tak 2011), scholars ignore whether supervisor’sbehavior actually produces positive subordinates’ out-comes such as better value fit or alignment with organiza-tional goals (Ostroff et al. 2005). Supervisors who stronglymanifest IMO-congruent behavior are perceived byemployees as being more committed to the organiza-tional mission (Wieseke et al. 2009) facilitating, there-fore, their adoption of organizational values (Huang et al.2005). Additionally, supervisor’s support also leadsemployees to adopt behaviors that benefit the organi-zation and enhances their consistency with organiza-tional values espoused by the supervisor (O’Hara et al.1991). IMO adoption through effective info exchangeand communication between supervisors and employ-ees can raise employees’ job satisfaction (Gounaris2008) by reducing work-related uncertainty and bal-

ancing problem-solving endeavors (Lings 2004),increas-ing thus, their reciprocal response toward thefirm. As IMO implementation involves ongoing feed-back about employees’ feelings toward their work, thebenefits and their needs within their roles (Lings andGreenley 2010), supervisor’s IMO adoption is anticipatedto enhance employees’ reciprocating with the adoption ofa value orientation congruent to their supervisor’s. Thus,we hypothesize that supervisor’s IMO adoption enhancesemployees’ perceptions of a good fit with their supervisor(H3), after controlling for individual IMO (H5).

Willingness to Report Service Complaints (WRC)

As failures are pervasive in service encounters con-tact employees need to engage in the service recoveryprocess in order to achieve successful service provision(Maxham and Netemeyer 2002; Liao 2007). Despite thevalue of traditional complaint management systems (Hom-burg and F rst 2007; Harris and Ogbonna 2006), severalstudies indicate that in retail services most of the custom-ers who complain submit their complaints only informallyor to contact employees rather than to senior management(Voorhees et al. 2006; Harris and Ogbonna 2010). There-fore, contact employees must be not only suitably trainedand supported (Homburg and Fürst 2005) but also moti-vated and willing to report service complaints to theirimmediate supervisor so as to allow the organization topursue effective service recovery attempts (Luria et al.2009).

As contact employees are often averse to reportcomplaints to their supervisors in a complete and accuratemanner (Gilly et al. 1991), their behavior can lead to thecreation of “organizational black holes in which informa-tion from below gets lost” (Argyris 1990, p. 23). There-fore, the importance of employees’ WRC is considered asa key driver of service recovery efforts (Luria et al. 2009).As the extant literature rarely provides mechanisms whichdeal with customers’ informal complaints (Harris andOgbonna 2010), we propose IMO as a mechanism thatincreases contact employees’ WRC. IMO adoption canprove beneficial for motivating employees to report com-plaints for several reasons. First, IMO enhances contactemployees’ feelings of obligation toward their employerby producing extra value for them (Simberova 2007),leading them to reciprocate with behaviors that benefit theorganization. Second, higher level of organizational sup-port through IMO adoption raises employees’ motivationto perform better, their commitment to customer serviceand their effort in carrying out in-role responsibilitiessuch as reporting complaints (Bell et al. 2004). Highermanagement consideration in terms of enhanced supervi-sory support or increased interest about employees’ needsand wants can lead to extra-role behaviors that benefit theorganization as a whole (Van Yperen et al. 1999). Consid-ering that IMO’s consequences such as employee satis-


faction and motivation are important precursors toincreased customer-oriented behaviors (George 1990),we hypothesize that IMO can positively enhance contactemployees’ willingness to report service complaints (H6).

Despite the value of IMO adoption for enhancingcontact employees’ behaviors during service recovery(Bell et al. 2004), we also highlight the importance ofemployee-supervisor (E-S) fit as an antecedent of employ-ees’ willingness to report service complaints (H7). Thislink can be attributed to the leadership theory (Avolio andBass 1988) which echoes the importance of E-S fit forwork effectiveness (Hofmann et al. 2011). In the processof employees’ value internalization, supervisors act asrole models and influence subordinates who cognitivelyadopt their own value orientation by using informationprovided by them. As supervisors in retail services act asrepresentatives of the firm’s management, their role forraising employees’ fit in this process is critical in the sensethat their manifestation of desired values enhances em-ployees’ value adoption (O’Reilly and Chatman 1986).Additionally, E-S fit can also enhance WRC by promotingcommunication (van Vianen et al. 2004), as open infoexchange among organizational levels establishes a com-mon frame for describing, classifying and interpretingevents (Erdogan et al. 2004). This common frame facili-tates the info exchange and reduces the likelihood ofmisunderstandings (Meglino and Ravlin 1998). Consid-ering that high E-O fit leads to more positive workbehaviors, we propose employees’ WRC as one suchbehavior, as a lack of E-O fit may diminish contactemployees’ motivation to provide feedback regarding tocustomers’ complaints. Based on the above, we hypoth-esize that E-O fit may influence contact employees’ will-ingness to report service complaints (H8).

METHODOLOGY

To test the conceptual framework of our study wechose a retail service setting, considering that the specific

context is appropriate for exploring the employee-supervisor dyad, as supervisor constitutes the onlylinking pin between store employees and top manage-ment. We totally obtained evidence from eighty-ninestores of retail chains which operate in Greece. In particu-lar, we draw evidence from 89 retail store supervisorsthrough personal interviews and 417 contact employees.A hierarchical research design is used with data obtainedfrom supervisors and contact employees. Regarding oursample’s demographic characteristics, the number ofemployees per retail store ranges from 10 to 16 and almostseven out of ten contact employees were women; theaverage age was approximately 27 years old, 68% of themworked full-time while the average tenure is about twoyears. All measures employed in this study were based onextensive review on the services marketing and psychol-ogy literature. All constructs use a 7-point scale, withanchors of strongly disagree (1) and strongly agree (7).Supervisors report on the IMO construct, whereas contactemployees’ sample provides evidence for the IMO, E-Ofit, E-S fit, patronage and WRC constructs. The IMO scaleis adapted from Gounaris’ (2006) scale, whereas the E-Ofit construct was measured based on Netemeyer et al.’sfour-item (1997) scale. For assessing E-S fit, Hofmann’set al. (2011) three-item scale was used. Subjective fitmeasures are used for E-S and E-O fit, as scholars arguethat they have better predictability than objective fitmeasures have (Kristof-Brown et al. 2005). Finally, WRCis adapted from Luria et al. (2009).

DATA ANALYSIS AND RESULTS

Given the nature of our data, a multilevel approach isnecessary to test our hypothesized relationships(Raudenbush and Bryk 2002) and therefore the HLMsoftware is employed. First, we estimated an intra-classcorrelation coefficient (ICC) which represents the amountof variation in an outcome variable due to the store asopposed to another data level. The levels reported inTable 1 tend to be typical of those reported for data

TABLE 1Means, Standard Deviation

Mean SD CFI TLI RMSEA A ICC

S_IMO 4.38 .83 .941 .924 .029 .967 .442

E_IMO 3.91 .90 .928 .909 .014 .952 .481

E-O FIT 4.21 1.08 .998 .989 .002 .709 .576

E-S FIT 4.48 1.04 – – – .778 .528

WRC 4.60 1.13 .991 .982 .036 .896 .595


aggregation. Table 1 above reports on descriptive statis-tics and internal consistency reliabilities. As Table 1suggests, all the measurement scales have reliability indi-ces that exceed the .70 threshold and AVE that is greaterthan .50. Additionally, CFI and TLI indices exceed the .90threshold while the RMSEA index is lower than .08 for allstudy’s measures. Table 2 presents intercorrelations of allstudy’s variables.

Four HLM models were performed in order to verifythe hypothesized effects. To justify the use of higher-levelpredictors, we ran four null models to determine whetherthere was significant between-group variation. These nullmodels are intercept-only models in which no predictorsare selected for higher levels of analysis. For each modelwe first estimated the null model with no predictors ateither level-1 or level-2 in order to partition the varianceinto within and between-stores components for each ofthe dependent variables (Raudenbush and Bryk 2002).The results reveal significant between branch variance for

all dependent variables; employee IMO (x2 = 1103.77, p <.01) E-O fit (x2 = 1754.71, p < .01), E-S fit (x2 = 1438.88p < .01) and WRC (x2 = 1776.82, p < .01). Table 3 providesan overview of the results of the HLM models. Concern-ing cross-level effects, we find strong support for thesupervisor IMO-employee IMO relationship (γ = .63, p <.01) and thus hypothesis H1 is verified. Results alsosuggest that supervisor IMO significantly predictsemployee E-O fit (γ = .54, p < .01), in support of hypoth-esis H2. Furthermore, hypothesis H3 is supported, assupervisor IMO also predicts employee E-S fit (γ = .63,p < .05). Regarding employee-level effects, findings indi-cate that individual IMO significantly accounts for bothE-O fit (γ = .66, p < .01) and E-S fit (γ = .40, p < .05),supporting thus hypotheses H4 and H5. Additionally,individual IMO is positively related to WRC (γ = .37, p <.01), in support of hypotheses H6. E-S fit (γ = .47, p < .05)also E-O fit (γ = .28, p < .01) both come up as drivers ofemployees’ willingness to report service complaints, sup-porting, thus, hypotheses H7 and H8.

TABLE 2Correlation Matrix

E_IMO E-O FIT E-S FIT WRC

E_IMO 1

E-O FIT 642** 1

E-S FIT .607** .608** 1

WRC .458** .527** .796** 1

TABLE 3Model (Dependent)

Model 1 Model 2 Model 3 Model 4E_IMO E-O FIT E-S FIT WRC

Intercept (SE) 3.57** (.03) 4.18** (.06) 4.64** (.06) 4.63** (.08)

γγγγγ (SE)

S_IMO .63** (.04) .54** (.07) .63* (.09) –

E_IMO – .66** (.05) .40* (.08) .37** (.07)

E-O FIT – – – .47* (.05)

E-S FIT – – – .28** (.06)

Model Deviance 453.52 756.18 652.96 764.57

** = p < .01, * = p < .05. E_IMO = contact employee IMO, S_IMO = supervisor IMO, E-O fit = employee-organization fit, E-S fit = employee-supervisor fit, WRC = willingness to report service complaints.


DISCUSSION

This study integrates evidence from a variety ofdisciplines in order to stress the role of IMO adoption foremployees’ better fit with their environment and is in-tended to be a key step in bringing internal marketing andfit research together. As prior research on internal market-ing remains fragmented, this study contributes to theextant literature by suggesting IMO as a mechanismwhich can raise employees’ fit with their environment.Against traditional internal marketing background whichfocuses at the single level of analysis and ignores the roleof cross-level influences for IMO adoption (Gounaris2010), we formally include IMO into multilevel market-ing research and provide an important shift for extantresearch by discussing how cross-level interactions shapeIMO adoption and enhance employees’ congruence withtheir environment. With reference to the view that IMOimplementation requires multilevel management (Wiesekeet al. 2009), this study is the first one that acknowledgesthe importance of supervisor’s complementary role forenhancing employees’ responsiveness to the firm’s inter-nal marketing efforts. The use of a multilevel approachallows us to argue that supervisor’s IMO congruent behav-ior actually shapes positive employees’ behaviors whilemaintaining the appropriate level of analysis. From atheoretical view, we extend fit theory by establishing IMOimplementation as a key strategy for recovering or in-creasing employees’ fit with their environment.

This study also offers some important insights froma practical view. First, retail supervisors are posited at theheart of the IMO diffusion process, as their manifestationof IMO-congruent behavior is a prerequisite for enhanc-ing employees’ responsiveness to internal marketing pro-grams. Second, results suggest that the adoption of IMOphilosophy increases employees’ fit with their organiza-tion and their supervisor. Given the value of informationcontained in informal complaints that customers presentdirectly to contact employees (Harris and Ogbonna 2010),we provide IMO as a mechanism which renders employ-ees more informative and motivates them to report criticalservice incidents to their supervisor. However, it appearsthat service providers’ motivation to report complaintsalso depends on their perceptions of fit with their supervi-sor. Although reporting complaints is considered as an in-role expectation, a lack of E-S fit diminishes their willing-ness to report problems they encounter keeping manage-ment in the dark about emerging service failures. Thisfinding can perhaps be explained by the fact that employ-ees’ willingness to report service complaints is not anautomatic response to organizational directives but is

influenced by the degree that employees’ perceptions arecongruent with supervisor’s ones in terms of a similarvalue orientation.

Considering that supervisor’s level of IMO accountsfor a significant portion of contact employees’ IMOadoption, it is imperative that top management must firstsell the organization itself especially to middle levelmanagers before implementing an internal marketingprogram across the whole firm. Another managerialimplication that stems from our study is related to theimportance of IMO as a mechanism which can actuallyincrease employees’ fit with their environment, departingthus from the view that managers can mainly achieve “fit”into their organization and avoid misfit and high turnoverrates by carefully attracting and selecting individuals(Van Vianen 2000). Internal marketing programs can beemployed as a tool for helping employees to embraceorganizational values and achieve better fit with theirsupervisor. Given the importance of service recovery insome settings, managers should take into account thatIMO increases employees’ WRC to their supervisor. Suchwillingness needs to be fostered but cannot be taken forgranted. Managers should be encouraged to establishinformal communication lines and mechanisms forobtaining customer informal complaints and act toimprove service based on such data. Practically speak-ing, store supervisors by displaying a better under-standing of employees’ wants and needs of employeesand providing more support, they will add value forsubordinates with the aim of reciprocating toward the firmwith different ways.

As with all research, our study has some limitationsthat restrict its interpretation and generalizability. First,the dependent variable (i.e., WRC) is a self-reportedmeasure and there is a concern associated with using aself-reported service performance measure, as employeestend to over report their performance under the influenceof social desirability bias. Second, our conceptual frame-work features only supervisor’s impact on contactemployees, ignoring, thus, the importance of othercontextual drivers of employees’ fit with their environ-ment. In conclusion, the results presented in this studyclearly suggest an important first step in understandingthe importance of IMO adoption for employees’ better fitwith their environment. Moreover, by integrating evi-dence from multiple disciplines, we add significantly tothe extant literature by providing some evidence regard-ing the role of IMO for enhancing employees’ reciprocalresponse toward the firm.


REFERENCES

Adams, J.S. (1963), “Toward an Understanding of Ineq-uity,” Journal of Abnormal and Social Psychology,67, 422–36.

Ahmed, P.K. and M. Rafiq (2003), “Internal MarketingIssues and Challenges,” European Journal of Mar-keting, 37 (9), 1177–86.

Argyris, C. (1990), Overcoming Organizational Defenses–Facilitating Organizational Learning. Boston, MA:Allyn and Bacon.

Ashforth, B.E. and F. Mael (1989), “Social Identity Theoryand the Organization,” Academy of ManagementReview, 14 (1), 20–39.

Bandura, A. (1977), Social Learning Theory. EnglewoodCliffs, NJ: Prentice Hall.

Bell, S.J., B. Mengüç, and S.L. Stefani (2004), “WhenCustomers Disappoint: A Model of Relational Inter-nal Marketing and Customer Complaints,” Journalof the Academy of Marketing Science, 32 (Spring),112–26.

Bitner, M.J., B.H. Booms, and L.A. Mohr (1994), “Criti-cal Service Encounters: The Employee’s Viewpoint,”Journal of Marketing, 58 (4), 95–106.

Chatman, J. (1989), “Improving Interactional Organiza-tional Research: A Model of Person-OrganizationFit,” Academy of Management Review, 14, 333–49.

Chatman, J. (1991), “Matching People and Organiza-tions: Selection and Socialization in Public Account-ing Firms,” Administrative Science Quarterly, 36,459–84.

Conduit, J. and F.T. Mavondo (2001), “How Critical isInternal Customer Orientation to Market Orienta-tion?” Journal of Business Research, 51 (January),11–24.

Edwards, J.R. (1996), “An Examination of CompetingVersions of the Person-Environment Fit Approach toStress,” Academy of Management Journal, 39 (2),292–339.

____________ and D.M. Cable (2009), “The Value ofValue Congruence,” Journal of Applied Psychology,94, 654–77.

Erdogan, B., M.L. Kraimer, and R.C. Liden (2004), “Per-son-Organization Fit and Work Attitudes: The Mod-erating Role of Leader-Member Exchange,” Person-nel Psychology, 57, 305–32.

Gilly, M.C., W.B. Stevenson, and L.J. Yale (1991), “Dyna-mics of Complaint Management in the Service Orga-nization,” Journal of Consumer Affairs, 25 (2), 295–322.

Gounaris, S.P. (2006), “Internal-Market Orientation andIts Measurement,” Journal of Business Research, 59(4), 432–48.

____________ (2008), “Antecedents of Internal Market-ing Practice: Some Preliminary Empirical Evidence,”International Journal of Service Industry Manage-ment, 19 (3), 400–34.

Gummesson, E. (1987), “The New Marketing-Develop-ing Long-Term, Interactive Relationships,” LongRange Planning, 20 (August), 10–20.

Harris, L.C. and E. Ogbonna (2006), “Service Sabotage:A Study of Antecedents and Consequences,” Journalof the Academy of Marketing Science, 34, 543–58

____________ and ____________ (2010), “Hiding Cus-tomer Complaints: Studying the Motivations andForms of Service Employees’ Complaint Conceal-ment Behaviors,” British Journal of Management,21, 262–79.

Hoffman, B.J., B.H. Bynum, R.F. Piccolo, and A.W.Sutton (2011), “Person-Organization Value Congru-ence: How Transformational Leaders Influence WorkGroup Effectiveness,” Academy of Management Jour-nal, 54 (4), 779–96.

Homburg, C. and A. Furst (2005), “How OrganizationalComplaint Handling Drives Customer Loyalty: AnAnalysis of the Mechanistics and the OrganicApproach,” Journal of Marketing, 69 (3), 95–114.

____________ and ____________ (2007), “See No Evil,Hear No Evil, Speak No Evil: A Study of DefensiveOrganizational Behavior Toward Customer Com-plaints,” Journal of the Academy of Marketing Sci-ence, 35, 523–36.

Huang, M.P., B. Cheng, and L. Chou (2005), “Fitting inOrganizational Values: The Mediating Role ofPerson-Organization Fit Between CEO CharismaticLeadership and Employee Outcomes,” InternationalJournal of Manpower, 26 (1), 35-49.

Huseman, R.C. and J.D. Hatfield (1990), “Equity Theoryand the Managerial Matrix,” Training and Develop-ment Journal, 44 (4), 98–102.

Kristof, A.L. (1996), “Person-Organization Fit: An Inte-grative Review of its Conceptualizations, Measure-ment, and Implications,” Personnel Psychology, 49,1–49.

Kristof-Brown, A.L., R.D. Zimmerman, and E.C. Johnson(2005), “Consequences of Individuals’ Fit at Work:A Meta-Analysis of Person-Job, Person-Organiza-tion, Person-Group, and Person-Supervisor Fit,”Personnel Psychology, 58, 281–342.

Lam, S.K., F. Kraus, and A. Ahearne (2010), “The Diffu-sion of Market Orientation Throughout the Organi-zation: A Social LearningTheory Perspective,” Jour-nal of Marketing, 74 (61), (September), 61–79.

Lauver, K.J. and A. Kristof-Brown (2001), “Distinguish-ing Between Employees’ Perceptions of Person-Joband Person-Organization Fit,” Journal of VocationalBehavior, 59, 454–70.

Liao, H. and A. Chuang (2004), “A Multilevel Investiga-tion of Factors Influencing Employee Service Perfor-mance and Customer Outcomes,” Academy of Man-agement Journal, 47 (1), 41–58

____________ (2007), “Do It Right This Time: The Roleof Employee Service Recovery Performance in Cus-tomer-Perceived Justice and Customer Loyalty After


Service Failures,” Journal of Applied Psychology, 92(2), 475–789.

Lings, I.N. (2004), “Internal Marketing Orientation: Con-struct and Consequences,” Journal of BusinessResearch, 57, 405–13.

____________ and G.E. Greenley (2010), “Internal Mar-ket Orientation and Market-Oriented Behaviors,”Journal of Service Management, 21 (3), 321–43.

Luria, G. Gal I. and D. Yagil (2009), “Employees’ Will-ingness to Report Service Complaint,” Journal ofService Research, 12 (2), 156–74.

Manz, C.C. and H.P. Sims, Jr. (1981), “Vicarious Learn-ing: The Influence of Modeling on OrganizationalBehavior,” Academy of Management Review, 6 (1),105–13.

Meglino, B.M. and E.C. Ravlin (1998), “Individual Val-ues in Organizations: Concepts, Controversies, andResearch,” Journal of Management, 24, 351–89.

Netemeyer, R.G., J.S. Boles, D.O. MacKee, andR. MacMurrian (1997), “An Investigation into theAntecedents of Organizational Citizenship Behav-iors in a Personal Selling Context,” Journal of Mar-keting, 61, 85-98.

O’Reilly, C.A., J. Chatman, and D.F. Caldwell (1991),“People and Organizational Culture: A Profile Com-parison Approach to Assessing Person-OrganizationFit,” Academy of Management Journal, 34, 487–516.

Ostroff, C., Y. Shin, and A.J. Kinicki (2005), “MultiplePerspectives of Congruence: Relationships BetweenValue Congruence and Employee Attitudes,” Jour-nal of Organizational Behavior, 26, 591–623.

Raudenbush, S.W. and A.S. Bryk (2002), HierarchicalLinear Models: Applications and Data AnalysisMethods, 2nd ed. Newbury Park, CA: Sage.

Riketta, M. (2005), “Organizational Identification: AMeta-Analysis,” Journal of Vocational Behavior,66, 358–84.

Sasser, W. Earl and P. Arbeit Stephen (1976), “SellingJobs in the Service sector,” Business Horizons, (June),61–65.

Simberova, I. (2007), “Internal Marketing as a Part of

Marketing Culture Supporting Value for ExternalCustomer,” Economics and Management, 15 (1),470–80.

Tajfel, H. (1981), Human Groups and Social Categories.Cambridge, UK: Cambridge University Press.

Tak, J. (2011), “Relationships Between Various Person-Environment Fit Types and Employee WithdrawalBehavior: A Longitudinal Study,” Journal of Voca-tional Behavior, 78, 315–20.

Tortosa-Edo, V., J. Sanchez-Garcia, and M.A. Moliner-Tena (2010), “Internal Market Orientation and itsInfluence on the Satisfaction of Contact Personnel,”The Service Industries Journal, 30, 1279–97.

Van Yperen, N.W., A.E. van den Berg, and M.C. Willering(1999), “Toward a Better Understanding of the LinkBetween Participation in Decision-Making andOrganizational Citizenship Behavior: A Multi-LevelAnalysis,” Journal of Occupational and Organiza-tional Psychology, 72 (September), 377–92.

Van Vianen, A.E.M., I.E. De Parter, A.L. Kristof-Brown,and E.C. Johnson (2004), “Fitting in: Surface- andDeep-Level Cultural Differences and Expatriates’Adjustment,” Academy of Management Journal, 47,697–709.

Vilela, B.B., J.A.V. Gonzalez, and P.F. Ferr n (2008),“Person-Organization Fit, OCB and PerformanceAppraisal: Evidence from Matched Supervisor-Sales-person Data Set in a Spanish Context,” IndustrialMarketing Management, 37, 1005–19.

Voorhees, C.M., M.K. Brady, and D.M. Horowitz (2006),“A Voice from the Silent Masses: An Exploratoryand Comparative Analysis of Noncomplainers,” Jour-nal of the Academy of Marketing Science, 34 (4),514–27.

Waldman, D.A. and F.J. Yammarino (1999), “CEO Char-ismatic Leadership: Levels-of-Management and Lev-els of Analysis Effects,” Academy of ManagementReview, 24 (2), 266–85.

Wieseke, J., M. Ahearne, S.K. Lam, and R. Van Dick(2009), “The Role of Leaders in Internal Marketing,”Journal of Marketing, 73 (March), 123–45.

For further information contact:Achilleas Boukis

Athens University of Economics and BusinessEvelpidwn and Lefkados, 47

Athens, 11263Greece

Phone: 0030.2108203853E.Mail: [email protected] / [email protected]


FIRM SELF-SERVICE TECHNOLOGY READINESS: A SOCIO-TECHNICAL SYSTEMS PERSPECTIVE

B. Ramaseshan, Curtin University, Western AustraliaRussel P.J. Kingshott, Curtin University, Western Australia

SUMMARY

Self-service technologies (SST’s) provide firms anopportunity to offer innovative customer services. Theability to model customer relationships using SST’s is animportant area that needs to be understood because suchtechnology is a key factor in helping to create customervalue (Bitner, Brown, and Meuter 2000; Hunter andPerreault 2007; Meuter et al. 2000; Padgett and Mulvey2007; Zhu et al. 2007). Moreover, SST’s have reshapedservice firm operations (Matthing, Kristensson, andGustafsson 2006; Meuter et al. 2005; Parasuraman 1996)at the same time as offering great potential to help improvefirm performance (De Clercq, Menguc, and Auh 2008).To date the marketing literature has primarily focusedupon customers to help explain successful adoptions (e.g.,Collier and Sherrell 2010; Curran and Meuter 2007;Meuter et al. 2005) but little is known about firm readinessto use SST within their operations. Whilst, many firms useSST there is no marketing literature indicating if the firmis indeed SST ready to adopt and implement SST into theirvalue creation activities. Although existing scholarly workin this area provides some direction (e.g., Davis 1989;Davis-Sramek, Germain, and Iyer 2010; Parasuraman2002; Shum, Bove, and Auh 2006) this fails to fullyexplain the firm’s capabilities in relation to SST usage.This represents a significant gap in the marketing litera-ture and accordingly provides the impetus for our trajec-tory of thinking and the scholarly need to conceptualize anew SST readiness construct. We do so by wideningcurrent thinking in relation to articulating the conceptualdomain of this construct. Accordingly, this manuscriptoutlines the foundations in which firm capability for SSTadoptions can be modeled to indicate the dimensions anySST readiness construct is likely to comprise.

We therefore: (1) present a new firm readiness con-struct, (2) outline its conceptual domain, and, (3) articu-late the operational dimensions it comprises. Overarchingour research is the managerial need for SST’s adoptions tobe regarded as a strategic imperative because bringingtechnological innovation into the firm is deemed a man-agement process not simply a functional activity(Ottenbacher and Harrington 2010). Therefore a firms’capacity to innovate, that is, the ability to successfullyadopt or implement new ideas, processes, or, products(Hurley and Hult 1998) must factor the firms’ environ-ments and wider stakeholders into the firm’s SST adop-tion and implementation processes. Accordingly, we pro-

pose the SST readiness construct to comprise two facetsthat are intrinsically linked to the firm’s strategy planningactivities and these provide a solid basis to indicate thesuccess of technology adoptions. In order to do this wehave grounded our thinking in socio-technical systems(STS) theory (Pasmore 1988; Pasmore and Sherwood1978) because this explains how firm outcomes arederived from the interplay between its technical and socialsystems (Lui, Shah, and Schroeder 2006; Pan andScarbrough 1999). Both of these systems are relevant toSST usage within the firm and we posit that the conceptualdomain of the SST readiness comprises both a strategiccapability and environmental interconnectedness dimen-sions. We label this new construct self-service technologyreadiness and define it simply as the capacity of firms toembrace and successfully use SST in their value creationactivities.

Stemming from an extensive review of the marketingliterature coupled with a number of qualitative interviewswe reveal the need to juxtapose a firms’ SST readinesswith its capacity to enhance the value proposition. Thiscorroborates existing SST literature related to technologyusage patterns within the firm (e.g., Davis 1989; Hom-burg, Wieske, and Kuehnl 2010) – indicating managersneed to understand how to model successfully SST adop-tions from a much wider vantage than customer adoptioncapabilities. It is plausible that the main reason technol-ogy does not always attain expected pay-offs (e.g., Davis-Sramek, Germain, and Iyer 2010) is simply because firmsare not SST ready. Our thinking therefore provides man-agers a suitable platform in which the firm’s environmen-tal factors, intrinsic to the creation of value through SST,that potentially enhance and/or impede successful SSTadoptions can be modeled. From a scholarly viewpoint,we meld existing literatures and provide a new directionof thinking for this critical area of the marketing disciplinein a number of areas. First, a suitable measure can bedeveloped to help model the impact this construct hasupon other variables pertinent to the role of SST’s in thevalue proposition. Second, as adopting SST’s helps re-duce costs, increase customer satisfaction and loyaltylevels, as well as reach new segments (Bitner, Ostrom, andMeuter 2002) we can use the new scale to help model theseoutcomes. Finally, our focus has been upon SST but asthere are many other forms of technology available to thefirm there would be merit in establishing whether firmtechnology readiness can be regarded as a generalizedconstruct. References are available upon request.


For further information contact:B. Ramaseshan

Curtin UniversityKent Street

Perth, Western Australia 6102Phone: +618.9266.4674

Fax:+618.9266.3937E-Mail: [email protected]


DOES TECHNOLOGY ORIENTATION MATTER IN TECHNOLOGYSERVICES ORGANIZATIONS?

Nacef Mouri, George Mason University, FairfaxMaheshkumar P. Joshi, George Mason University, Fairfax

Sidhartha R. Das, George Mason University, Fairfax

SUMMARY

A firm’s strategic orientation is the direction taken bythe firm to gain competitive advantage (Narver and Slater1990), and reflects its philosophy with regard to under-standing customers, competitors, and use of technology(Gatignon and Xuereb 1997; Zhou, Yim, and Tse 2005).Technology orientation is an important type of firm stra-tegic orientation, yet it is sparsely addressed in the litera-ture (Zhou et al. 2005). This study examines the relation-ship between technology orientation and productinnovativeness. Using contingency theory arguments, thestudy proposes that firm aggressiveness and the level ofautonomy that the firm provides individuals within theorganization moderate the relationship between technol-ogy orientation and product innovativeness.

Though recent years have witnessed a strong growthin the service sector, researchers have noted the dearth ofresearch on innovation in the service industry.Gopalakrishnan and Damanpour (2000) observe that“. . . although service organizations and industries are amajor economic force in the contemporary U.S. economy,research and theory exploring innovation in service deliv-ery is rare” (p. 14). Technological advances are often thebasis for radical service innovations (Johnson, Menor,and Roth 2000), and are valuable in firms that provideservices with high levels of technology content. Hence,this research investigates the relationship betweentechnology orientation and product innovativeness inthe context of technology services organizations(TSOs). TSOs are defined as organizations that pro-vide technology-based services in the form of technol-ogy system design, implementation, analyses, studies,plans, management consultations, purchasing advice, andother artifacts and services (Das and Joshi 2007). Thesefirms need to acquire a substantial technological back-ground for deploying technology in their products andservices, and will therefore be characterized by a greaterdegree of technology orientation, providing a suitablecontext to study its relationship with productinnovativeness.

Based on extant literature, we first hypothesize thattechnology orientation and product innovativeness arepositively associated. We then argue that the firm’s levelof aggressiveness positively moderates the relationship

between technology orientation and productinnovativeness. Lumpkin and Dess (2001) define aggres-siveness as an organization being characterized by astrong offensive posture. Aggressiveness is indicated byboth the level of the organization’s deployments of corpo-rate resources (e.g., for product/service development),and the level of risk that these deployments entail (Fombrunand Ginsberg 1990; Khandwalla 1977). Both theseaspects of aggressiveness should enhance the relationshipbetween technology orientation and product innova-tiveness.

We also argue that the relationship between technol-ogy orientation and product innovativeness is negativelymoderated by the level of autonomy provided by the firm.Autonomy is defined as the amount of freedom anddiscretion the organization provides individuals in carry-ing out their tasks (Hackman 1983; Langfred and Moye2004). Due to the diverse and specialized nature of tech-nology related information in a technology-oriented firm,high levels of autonomy can hamper the process ofdeploying technology for the development of new prod-ucts/services. Research on the impact of task characteris-tics on communication patterns shows that more complextasks require more intensive interaction and coordinationto arrive at high quality decisions (Hackman and Vidmar1970; Olson, Walker, and Ruekert 1995). With increasingtask complexity, there is a greater need for interaction andinformation transfer (Tushman 1978). Providing moreautonomy can lead to less information search and knowl-edge utilization, resulting in inferior task outcomes (Corwinand Louis 1982). Therefore, in technology intensive envi-ronments characterized by high task complexity, greaterautonomy can lead to less interaction, communication,and coordination, as well as poor information transfer andknowledge utilization, with deleterious effects on theprocess of using technology for new products and ser-vices. Since higher levels of technology orientation inTSOs entail the performance of technical, knowledge-intensive tasks with high levels of interdependence andcomplexity, increased autonomy will weaken the rela-tionship between technology orientation and productinnovativeness.

Our analyses show that while product innovativenessin TSOs is positively associated with technology orienta-tion, this relationship is enhanced when a firm has an


aggressive posture and as such a propensity to take boldactions in its response to customer and market conditions.We surmise that aggressiveness fosters an organizationalenvironment that boldly encourages R&D and develop-ment of technologies related to exploitation of marketopportunities, and facilitates resource commitments forunproven and risky technology-related projects with prom-ise of high returns, thereby enhancing the relationshipbetween technology orientation and productinnovativeness. We also find that autonomy negatively

moderates the relationship between technology orienta-tion and product innovativeness. While research on orga-nizational behavior suggests that autonomy can have apositive impact on organizational outcomes, higher levelsof technology orientation in TSOs are characterized bygreater degree of task interdependence and task complex-ity, which require high levels of interaction, close coordi-nation, and joint decision making (Wageman 1995) inorder to achieve higher levels of product innovativeness.References are available upon request.

For further information contact:Nacef Mouri

George Mason University4400 University Drive

Fairfax, VA 22030Phone: 703.993.1769



ADAPTING PRINCIPLES OF SERVICE QUALITY TO PERSONALMEDIA MARKETING COMMUNICATION

Todd J. Bacile, Florida State University, TallahasseeCharles F. Hofacker, Florida State University, Tallahassee

SUMMARY

This theoretical conceptualization proposes market-ing communication sent to personal media, such as mobiledevices and social media, will be more effective byadapting principles of service quality. Extant models ofservice quality exist in technology mediated channels,such as e-service quality applicable in an e-commercecontext. However, e-service quality models are largelylimited to quality surrounding economic purchase trans-actions. The present research posits that quality principlesare also applicable to marketing communication that isseparate and apart from a purchase. This quality adapta-tion is made possible through the evolution of media,technology, and an increase in consumers participating indecision making processes with firms. Four specific areasof service quality are extended to personal media market-ing communication. The authors propose a disconfirmationparadigm exists within personal media marketing com-munication, as well as three quality dimensions: process,environment, and outcome.

Firms communicating company updates, productnews, and promotional offers to consumers have a longhistory within marketing. The original 12-element mar-keting mix and the consolidated 4Ps model each includea promotion or advertising component to communicateinformation from firms to consumers (Duncan and Moriarty1998). Here we define marketing communication(MarCom) as planned marketing messages initiated asfirm-to-consumer, but ultimately not limited to one-waycommunication. MarCom makes it possible for a firm-consumer two-way interactive information exchange(Duncan and Moriarty 1998, 2006).

A key differentiating factor for MarCom sent topersonal media compared to advertising sent throughtraditional media is the opportunity for a consumer togrant permission (i.e., opt-in) to become a member of aMarCom audience. This interactive characteristic is notpresent in traditional mass media and represents a dra-matic shift in MarCom. We posit consumers developinitial expectations concerning the future benefits used asan opt-in enticement before receiving or experiencing theactual marketing messages. Expectations are consumerpredictions about what will likely occur in a future trans-action or interaction (Zeithaml, Berry, and Parasuraman1993). Research into the process formation ties expecta-

tions to pretrial beliefs and these beliefs serve as thefoundation for, and the primary antecedent of, attitudeformation used to predict (Oliver 1980; Olson and Dover1979). Consumers enabled to participate with firms inmarketing communication processes suggests adisconfirmation gap is applicable to such communicationsent to personal media.

Extant service and e-service quality models (e.g.,Bauer, Falk, and Hammerschmidt 2006; Brady and Cronin2001; Collier and Bienstock 2006; Dabholkar, Thorpe,and Rentz 1996; Grönroos 1984; Parasuraman, Zeithaml,and Berry 1985; Parasuraman, Zeithaml, and Malhotra2005; Rust and Oliver 1994) suggest quality is a multi-dimensional construct. The present research suggestsprinciples associated with three dimensions of quality areapplicable to marketing communication sent to personalmedia: process quality, environment quality, and out-come quality. Some or all of these dimensions are presentin various forms across several quality models, but thecurrent research’s adaptation of quality for MarCom ismost closely aligned with the Brady and Cronin (2001)model.

Process quality concerns the underlying interactionand exchange between a service provider and a consumerthat is separate from the core service outcome. Thisconceptualization carries over well into personal mediaMarCom. Responsiveness, empathy, frequency, and time-liness of marketing communication dissemination are keysuccess factors within these newer media during theinteractive exchange (Bacile and Goldsmith 2011; Barwiseand Strong 2002; Bowler 2011; Mangold and Faulds2009; Momentus Media 201; Shankar andBalasubramanian 2009).

Environment quality concerns the physical or virtualenvironment or atmosphere consumers are exposed to(Bitner 1992). Virtual environment cues include textdescribing merchandise and promotional details, messagelength, pictures, navigation aids, colors, borders, typestyles,and use of white space (Eroglu, Machleit, and Davis2001). Two personal media MarCom digital cues aremessage design and social interaction with others. Mes-sage design characteristics such as length, verbiage, andvisual aesthetics influence consumer evaluations(eMarketer 2011; Momentus Media 2011; Peters, Amato,and Hollenbeck 2007; Shankar and Balasubramanian


2009). Social interaction with others within personalmedia makes it possible for other consumers to interactwith MarCom. Other consumers can then see social inter-action when viewing the original marketing message.Consumers can signal quality with a recommendation byliking, re-tweeting, commenting on, or electronically for-warding a marketing message.

Outcome quality represents the result of viewingpersonal media MarCom (i.e., what the MarCom pro-duces for a consumer). Outcome quality may be the most

similar to extant traditional media advertising measures.MarCom and advertising traditionally measure consum-ers’ attitude toward the ad or message after exposure (e.g.,Lutz 1985) to capture the tendency to respond in a favor-able or unfavorable manner without accounting for aprocess or environment dimension. This attitudinalassessment of message favorability is what a consumer isleft with after a communication, which produces otherattitudes (i.e., attitude toward a firm), purchase intent,assessment of value, and the usefulness of a marketingcommunication. References are available upon request.

For further information contact:Todd J. Bacile

Florida State University821 Academic WayP.O. Box 3061110

Tallahassee, FL 32306–1110Phone: 850.644.4091



CONSUMER REACTIONS TO AESTHETIC INCONGRUITY:INVESTIGATING AESTHETIC INCONGRUITY IN THE

SERVICE CONSUMPTION CONTEXT

Seonjeong (Ally) Lee, University of Massachusetts, AmherstMiyoung Jeong, University of Massachusetts, Amherst

SUMMARY

Imagine you have booked a nice hotel suite to cel-ebrate your special anniversary. However, after youchecked into the hotel, you find a hotel room is not whatyou expected or what you saw on its website. How wouldyou respond? Visual design has been considered an im-portant factor for a company when creating its website.However, is high quality visual design really better, evenif the actual product or the actual physical environment isnot the same as the visual image provided on the website?If the website presents a product or a service product (i.e.,pictures of the hotel room) nicer than the actual product,consumers might become frustrated or regret their pur-chase, which, in turn, directly influences their futurebehavioral intentions (i.e., return and/or purchase inten-tion).

As illustrated in the example, the Internet has funda-mentally transformed the way tourism-related informa-tion is distributed and the way consumers plan their trip,as well as the way they behave (Buhalis and Law 2008).For instance, travel-related photos and videos havebecome a major content category on many social mediaapplications (i.e., Youtube, Facebook, and Flicker; Xiangand Gretzel 2010). Social media (SM) refers to a group ofInternet-based applications, where user-generated con-tent can be exchanged and shared (Kaplan and Haenlein2010). According to the Travel Industry Association ofAmerica, about two-thirds (64%) of online consumersrely on search engines to obtain destination-related infor-mation, when they plan a trip (TIA 2005).

Despite the importance of aesthetic incongruity inservice consumption, previous research has only investi-gated aesthetic incongruity in product design or in adver-tising context. To address this void, this research concep-tually investigates consumers’ emotional and behavioralresponses to aesthetic incongruity, incorporating con-sumers’ different value orientations as a moderating vari-able, in the context of a service consumption environ-ment. In this study, aesthetic incongruity refers to the

inconsistency that arises from a mismatch between aservice product shown in the picture (i.e., pictures on thewebsite or on the menu) and the actual product (i.e.,physical environment or the service product), consistentwith previous research (Meyers-Levy and Tybout 1989).

The central idea of this paper is that when aestheticincongruity arises in a service consumption environment(i.e., hotel room or restaurant menu), consumers responddifferently to resolve the incongruity, depending on thevalence of the aesthetic incongruity (i.e., positive vs.negative) and consumers’ expectations. Therefore, thisstudy aims to examine the interplay between aestheticincongruity and consumers’ expectations, comparing theservice consumption environment provided in social mediaand the conventional physical environment. In addition,this study investigates an individual difference as a mod-erator that influences the relationship between aestheticincongruity and consumers’ emotional responses, as wellas the relationship between consumers’ expectations andtheir emotions.

To the best of our knowledge, this research is the firstto investigate aesthetic incongruity in the service con-sumption context and its effects on consumers’ responses;thus, this study fills an existing literature gap in theaesthetic literature. This study contributes to a betterunderstanding of how aesthetic incongruity and consum-ers’ expectations interplay to influence their emotionaland behavioral responses. In addition, this study incorpo-rates different consumers’ value orientations that influ-ence the relationship between aesthetic incongruity andtheir emotion. Beyond theoretical contributions of thisstudy, this research helps service providers develop theappropriate visual design strategy in social media. This isimportant, given that consumers tend to build their expec-tations based on viewing pictures of service environ-ments, which, in turn, influence their emotions and behav-iors. It is hoped this study can add the current trends on thesocial media development to provide practical insights forthe service organization’s online marketing. Referencesare available upon request.


For further information contact:Seonjeong (Ally) Lee

Hospitality and Tourism ManagementUniversity of Massachusetts, Amherst

012 Flint LabAmherst, MA 01003Phone: 413.545.5376



ENHANCED REGRET: RUMINATIVE THINKING, MOOD REGULATIONAND SERVICE FAILURE

Kevin Lehnert, Grand Valley State University, Grand RapidsMark Arnold, Saint Louis University

SUMMARY:

This paper investigates the relationship betweenrecurrent thinking and service experiences; where theconsumer’s thoughts persist and focus on the positive ornegative association of the experience (Martin and Tesser1996b). Classified as ruminative thoughts, self-attentivethinking entails recurrent thoughts that revolve around acommon theme, often not progressing toward a goal,traditionally resulting in negative affect and depression(Treynor et al. 2003). When a discrepancy between ourexperiences and the current state occurs, such as in aservice failure, ruminative thoughts are likely to emerge(Martin and Tesser 1996a).

Trapnell and Cambell (1999) identify two types ofrecurrent thought states. The first, rumination, is theneurotic self-attentiveness, motivated by perceived threats,losses, or injuries. This is characterized by recurrent past-oriented, negative thinking. The second state, reflection,is intellectual self-attentiveness, referring to self-focusmotivated by curiosity or interest. Reflection explainsmore intellectual traits relating to self-knowledge andopenness to ideas, experiences and increased elaboration.Reflection reduces the negative affect (Mor and Winquist2002) and moves toward future problem resolution (Mar-tin and Tesser 1996a).

Mood regulation can be used to address and managerecurrent thinking. Consumers are motivated to regulatetheir affect in order to reduce discrepancies between theircurrent state and a perceived “ideal” affective state(Andrade 2005; Gross 1998). They will engage in anumber of cognitive and behavioral strategies to do so,such as reflective thinking. One framework which hasbeen used to describe mood regulation is the concept ofemotional intelligence. As defined by Kidwell et al. (2007,p. 154), consumer emotional intelligence is “a person’sability to skillfully use emotional information to achievea desired consumer outcome.” Emotional intelligenceconsists of three regulatory traits: mood monitoring,mood clarity, and mood repair (Salovey et al. 1995). Mood monitoring is the tendency to take notice andmonitor one’s emotional state (Swinkels and Guiliano1995). Mood clarity is the ability to identify and describespecific moods, beyond knowing one feels simply good orbad (Gohm and Clore 2000). Mood repair concerns a

person’s ability to repair unpleasant moods and maintainpositive ones (Salovey et al. 1995).

Of the three, mood clarity is likely to be a centralcomponent in affect regulation (Gohm 2003). Forexample, if information arising from the affective systemis critical for decision making (Schwarz and Clore 1996),then a person who is unable to understand and effectivelyuse these emotions should exhibit certain biases in cogni-tive processes (e.g., Berenbaum and Irvin 1996; Lischetzkeand Eid 2003). Individuals high in clarity, who are able toidentify and understand their emotions, make self-affirm-ing attributions for good events, but do not let themselvesmake the same attributions for bad events (Gohm andClore 2003).

Investigating the role of recurrent thinking in thisprocess is of particular interest; how we engage in recur-rent thoughts can have a strong influence on our ability tomaintain and control our moods. This ability to refine ourmood can have a strong impact on service responses asthose who can rebound from service failures quicker maybe more likely to return to the store and continue theshopping relationship. This discussion leads us to proposethe following:

H1a: Reflective thinkers have more clarity when it comesto identifying their moods than ruminative think-ers.

H1b: Reflective thinkers are more able to monitor theirmoods than ruminative thinkers.

H1c: Reflective thinkers are more able to repair theirmoods easier than ruminative thinkers.

H1d: Reflective thinkers are less intense regarding theirmoods than ruminative thinkers.

H2: Reflective thinkers are more able to manage theirmoods in face of service failures than ruminativethinkers.

Utilizing shopping scenarios we evaluated how self-attentive thoughts (ruminative or reflective) and moodregulation relate to consumer evaluations of shoppingexperiences and purchase intentions. In general we findthat reflective thinking allows for more efficient mood


monitoring, clarity of mood and ability to address servicefailures (H1a – H1c). Surprisingly the intensity of themood is not influenced by whether one reflects or rumi-nates upon it. The analysis of marginal means for theinteraction between experience and self-reflective thoughtsrevealed that for positive experiences, reflectors had higheroverall mood regulation but held little significant differ-ences over ruminators. In bad experiences reflectors weremore significant indicators of consumer mood regulationthan ruminators supporting H2.

This analysis begins to investigate the relationshipbetween self-attentive thoughts and mood regulation.Results indicate that in the face of positive shoppingexperiences, how we reflect upon those experiences haslittle to do with our emotional state, or even how we work

to maintain those emotions. This has a strong impact formanagers, as it is less about how good the shoppingexperience was, but that it was a positive one in the firstplace. In light of service failures, ruminative thoughtsexacerbate the overall mood reflection, intensity, clarityand ultimately how we repair those moods. This hasdistinct implications for managers as the negative impactof a poor shopping experience tends to stay with rumina-tive customers. Managers should realize that a negativeshopping experience can extend beyond the experienceand can create a long-term deeply emotional response thatcan destroy the brand relationship. Managers can begin toovercome these ruminative thoughts by addressing posi-tive outcomes in the face of service failures, effectivelyhighlighting the silver lining around the dark cloud.References are available upon request.

For further information contactKevin Lehnert

Seidman College of BusinessGrand Valley State University

401 Fulton Street WestGrand Rapids, MI 49504



SERVING IN AN ONLINE WORLD: HOW TO REACT ON NEGATIVEELECTRONIC WORD-OF-MOUTH?

Werner H. Kunz, University of Massachusetts, BostonAndreas Munzel, EM Strasbourg Business School (HuManiS, EA 1347), France

Benedikt Jahn, Ludwig-Maximilians-University Munich, Germany

SUMMARY

Because of the new social media platforms (e.g.,Yelp, Facebook, Twitter), individuals today have manyways to express their opinions (electronic word-of-mouth[eWOM]) about services, brands, companies, andemployees. While positive comments from former cus-tomers can convince potential customers to try a service(Park and Lee 2009), negative online published opinionscan have a significant impact on a consumer’s choice(Laczniak et al. 2001; Senecal and Nantel 2004), revenue(Chevalier and Mayzlin 2006; Dellarocas et al. 2007), afirm’s stock returns (Tirunillai and Tellis 2010), and ismore useful than positive eWOM messages for severalproduct categories (Sen and Lerman 2007). The viralquality of negative eWOM diffusion in new online mediacan ruin a company’s reputation and put a service com-pany out of business. By investigating the effects on thecompany, Chatterjee (2001) found that negative reviewssignificantly deteriorate the service provider’s perceivedcredibility. Our literature review additionally reveals thatonly a few studies have investigated the effects of differ-ent aspects of eWOM on credibility and trust as perceivedby the receiver. Despite the importance of negative eWOM,the question of how to systematically react to unfavorableonline comments has been widely neglected in academia.

Therefore, the goal of this study is to investigate theeffects of company response strategies on consumers’perceptions of negative eWOM. In particular, we areinterested in the effects of two response strategies (orsocial accounts): apologies and excuses (Bies 1987). Tothe best of our knowledge, the trade-off between attrib-uted responsibility and the adequacy of a company’sresponse has not been empirically verified in the newmedia context to date. In addition, we investigate theinfluence of potential customers who observe the interac-tion between a complainant and a service company in twoonline media environments. For this, we execute a seriesof experimental studies that were conceptualized based onattribution and attributional theories (Kelley and Michela1980), conflict theory (Sitkin and Bies 1993), and excusetheory (Snyder and Stukas 1999).

In the first study, we investigate the effects of the tworesponse strategies on the eWOM receiver’s evaluation ofthe service company. To test our hypotheses, we con-

ducted an online experiment including a 2 by 2 by 2between-subjects design: severity of the failure (low,high) x service failure type (process, outcome) x socialaccount type (apology, excuse). We also measure cred-ibility, benevolence, and attitude toward the company –before and after the company’s response – in order tomeasure the dynamics. In the second study, we build onthe findings from the first study and integrate additionallypublic interactions. We used brand pages (e.g., fan pageson Facebook, channels on YouTube, G+ pages onGoogle+) as a forum for integrating repliers who react toinitial negative comments.

Based on our data, we can show that apologizing isalways the most effective eWOM response strategy –regardless of whether a particular company is responsiblefor the failure. Further, an apology has a greater effect oneWOM receivers when a highly severe service problem isreported. Our data explains the influence of repliers’comments within the social media environment, which isknown for its high degree of interactivity between differ-ent actors.

Our research contributes to the existing marketingliterature from several standpoints. First, we highlight theimportance of dealing with negative eWOM communica-tion with social accounts in two contexts: online reviewsites as well as brand pages within social networking sites.Second, we empirically assess the effects of specific typesof complaint management tools (i.e., apologies, excuses)on eWOM receivers, and study the effect of potentialconsumers who are not directly affected by the problem(Bobocel and Zdaniuk 2005). Third, while eWOMresearch on consumer online reviews has received con-siderable attention, eWOM within social networking sites,including the information from other individuals, is widelyneglected in the literature.

Our results have important implications for market-ing management. First, we were able to show the power ofan apology in dealing with negative eWOM. Our resultsshow that the question of who is responsible for theproblem is only of minor importance for the eWOMreceiver; the attribution of responsibility has only limitedeffects on subsequent evaluations of firm. Therefore,even in the event that the cause for a problem lies outsidea company’s responsibility, it is better for the company toapologize and accept responsibility for the problem.


Second, we can show the influence of consensusbetween the initial negative commenter and other socialmedia users on attitude, credibility, and benevolence.This illustrates the relevance of a vibrant fan communityin social media; such a community is capable of defendingthe company when it is attacked. The opinion of otherusers is much more credible than communication from thecompany itself and may help to reduce or even avoidnegative viral effects. But despite the positive effect ofsocial media users’ defensive behaviors, our results revealthat it is better to admit responsibility when a problemoccurs, even if some users may not agree with the initialnegative comment. This is a very important finding,because management often denies responsibility com-pletely (van Laer and de Ruyter 2010).

Future research should also include a company’s lackof response in order to underline the importance of anapology. In addition, the importance of the intensity ofconsensus between the initial negative comment and othersocial media users remains unclear. Therefore, more re-search is needed to understand the effects of a company’sresponse to negative eWOM in social media, as well as itsinteraction effects with other social media users.

New media channels enable customers to articulateexperiences and express themselves in various ways.WOM communication on the Internet provides consum-ers with considerable power. Our study is the first steptoward achieving a better, more comprehensive under-standing of eWOM in the context of new media channels.References are available upon request.

For further information contact:Andreas Munzel

EM Strasbourg Business School (HuManiS EA 1347)61 Avenue de la Foret Noire

67085 Strasbourg CedexFrance

Phone: +33.3.68.85.85.29Fax: +33.3.68.85.85.93



RELINQUISHING THE MOMENT OF TRUTH: A MODEL OF FIRMAND CUSTOMER-LED SERVICE RECOVERY

Sara K. Bahnson, University of Oregon, Eugene

SUMMARY

Given that firms routinely engage their customer baseto provide customer support (Nambisan and Baron 2007;Mathwick et al. 2007) and involve their customers in theservice recovery process (Meuter and Bitner 1998; Donget al. 2008), it is important that this customer-led activityis reflected in existing models of service recovery. How-ever, no existing model considers the customer agent as apotential support provider for the failed customer insearch of a service recovery. In this conceptual research,the focus is to expand on an existing model of servicerecovery by Smith et al. (1999) in order to develop a newframework that incorporates the customer as a viableservice recovery agent. The current research comparesfirm and customer-led service recovery in the context ofonline firm-hosted forums and communities, and appliesthe characteristics of online brand communities to informhow perceptions of justice may generate satisfaction whenthe recovery is customer-led. Attribution theory isemployed to examine the post-recovery attribution set,which is expanded when the customer is the recoveryagent.

Following a review of existing models in the servicerecovery literature (Hess et al. 2003; Kelly and Davis1994; Smith et al. 1999), the model by Smith et al. (1999)was chosen as the most appropriate with which to inte-grate the customer agent, because it examines recoveryengagement as an encounter or interaction rather than asa relationship, and satisfaction rather than perceptions ofservice quality as the key outcome variable. Smith et al.identify four recovery attributes that can be used to restorecustomer satisfaction following a failure: compensation,response speed, apology, and initiation. Informed bysocial exchange theory, these attributes are associatedwith customer perceptions of distributive, procedural, andinteractional justice. Positive perceptions of justice arethen linked to customer satisfaction with the servicerecovery encounter. While this model by Smith et al. issuitable to describe firm-led service recoveries, revisionsmust be made to accommodate customer-led servicerecovery.

The newly enhanced model introduces two recoveryattributes—explanation and empathy – that can be uti-lized during a recovery by both firm and customer recov-ery agents and are essential for understanding the valuablerole a customer agent can play during the recovery pro-cess. Customer choice of the recovery agent is integrated

into the model to make connections between elements ofthe failure context (type: process vs. outcome, and cat-egory: service and product) and recovery attributes. Choiceof recovery agent then mediates the relationship betweenfailure context and recovery attributes, moderated byindividual difference variables (age and prior experience)and failure magnitude. The failure context is extended toinclude failures related to both service and product-related issues, as customer-led service recoveries ofteninvolve problem-solving for technical products, such ascomputers and cell phones.

The addition of explanation and empathy as recoveryattributes available to both firm and customer agents areinformed by characteristics of online brand communities.Online communities and user forums are “vast reservoirsof information” (Madapu and Cooley 2010, p. 364), andmembers of brand communities have a moral responsibil-ity to assist fellow customers with the proper usage of thebrand’s products or services (Muñiz and O’Guinn 2001).Customer agents can therefore provide an accurate expla-nation for a failure due to the extensive information andexperience with the product or service, and can be assufficient as the firm in doing so. Empathy is another vitaladdition to the recovery attribute set of both firm andcustomer agents. Muñiz and O’Guinn (2001) identifyconsciousness of kind as a significant characteristic of abrand community, which influences the members’ abilityto promote empathy, which is distinct from an apology.Bickart and Schindler (2001) found that in comparison tocorporate websites, online forums have a greater ability toevoke empathy, therefore giving the customer agent anedge over the firm agent in this recovery realm. Responsespeed is the final recovery attribute that can be employedby a customer agent, and it is expected that firm andcustomer agents will perform similarly in this regard.

Regarding distributive justice, this dimension is strictlyrelated to a firm-led recovery and is absent from recover-ies provided by a customer agent in the new framework.Therefore, procedural and interactional justice are main-tained on the side of the customer agent, suggesting thatresponse speed influences positive perceptions of proce-dural justice, and empathy and explanation influencepositive perceptions of interactional justice. Holcutt et al.(2006) find that high responsiveness (procedural justice)and courtesy (interactional justice) have a powerful posi-tive effect on customer evaluations, which would lead tosatisfaction with a customer-led recovery. This proposedframework suggests that a customer agent will be moresuccessful at delivering on positive perceptions of inter-


actional justice for the failure customer than a firm agent,due to their ability to provide a necessary explanation anddeliver appropriate empathy during the recovery process.In addition, the locus of the failure attribution during acustomer-led recovery has yet to be studied. Existingresearch on attribution theory suggests that the failedcustomer attributes the failed recovery to the firm, asopposed to the front-line employee. However, the interac-tions between peers and within communities are decid-edly different, and it is not yet understood where theattributions lie and whether expectations of customeragents play into these attributions post-recovery.

Smith and Bolton (1998) describe a service recoveryas the “moment of truth”; it is the opportunity of the firmto restore customer satisfaction or intensify the failure anddrive customers to competitors. Firms that engage theircustomers in providing customer support relinquish thiscritical moment, trusting their customers to provide therequisite due care and timely responsiveness following aservice failure. This research suggests that firms canrecover from service failures without direct notification ofthe failure, and that customer agents can successfully leadthis recovery. References are available upon request.

For further information contact:Sara Bahnson

University of Oregon1208 University of Oregon

Eugene, OR 97403Phone: 541.346.9080

Ee-Mail: [email protected]


THE IMPACT OF PERCEIVED SERVICE RECOVERY JUSTICE ONCUSTOMER AFFECTION, LOYALTY, AND WORD-OF-MOUTH

Beomjoon Choi, California State University, Sacramento

SUMMARY

Service failure is inevitable and when it occurs, it maydeteriorate relationships with customers, possibly caus-ing customer dissatisfaction, negative word-of-mouth,and even customer defection if it is not handled properly.Hence, it is crucial for firms to successfully deal withservice recovery to reestablish and/or maintain positiverelationships with customers (Gustafsson 2009; Rio-Lanza,Vazquez-Casielles, and Diaz-Martin 2009). Customersatisfaction with service recovery has been deemed criti-cal for firms in maintaining positive relationships withcustomers after service failure (Harris et al. 2006; Maxham2001; Maxham and Netemeyer 2002; Stauss 2002). Cus-tomer satisfaction with service recovery, mostly, how-ever, taps the cognitive aspects of consumer behavior(i.e., expectations, disconfirmations, justice perception,etc.) (Harris et al. 2006; McColl-Kennedy, Daus, andSparks 2003; Smith et al. 1999; Wirtz and Mattila 2004).While there is no denying that satisfaction with servicerecovery is a critical factor in ensuring customer loyalty,we suspect service recovery satisfaction may not be suf-ficient enough to keep customers loyal after a servicefailure because service failure and recovery encountersoften evoke strong emotional reactions from customers,which may influence customers’ decision of whether toremain in a relationship with a company directly orindirectly.

While customers’ emotional responses to servicefailure and recovery encounters have received some atten-tion (Weiss et al. 1999; Chebat and Slusarczyk 2005;DeWitt et al. 2008; Namkung and Jang 2010) and someresearchers have begun to investigate customer affectionin the context of service failure and recovery (La and Choi2011), little progress has been made in developing atheoretical understanding of what constitutes customeraffection and its consequences in the context of servicefailure and recovery. Furthermore, no empirical studieshave specifically addressed how customers’ justice per-ception influences customer affection and its impact onbehavioral outcomes such as word-of-mouth.

The primary objective of this study therefore is two-fold. First of all, we examine the effects of customerjustice perception on customer affection. We also test themoderating effect of service failure severity on the rela-tionship between justice perception and customer affec-tion. Second, the current study investigates the effects of

customer affection on customer loyalty and word-of-mouth. While much research has examined the relation-ship between customer perception of service recoveryefforts and aforementioned attitudinal and behavioraloutcomes, little research has been conducted to addressthe role of customer affection as the mediating factorbetween customer justice perceptions and key relation-ship constructs such as loyalty and word-of-mouth. In all,our theoretical model highlights the role of customeraffection as the mediator between customer justice per-ception and word-of-mouth. This model also addressesthe role of customer loyalty as the mediator betweencustomer affection and word-of-mouth.

To test the model, we conducted a survey and col-lected self-administered data for data analysis. The pro-posed relationships were then tested using confirmatoryfactor analysis and structural equation modeling. Resultsshow that procedural and interactional justice perceptionssignificantly influence customer affection, with distribu-tive justice perception being significant only if the sever-ity level is high, and that customer affection greatlyinfluences word-of-mouth both directly and indirectly viacustomer loyalty. Our findings imply that service recov-ery efforts should be viewed as a strategy to build a long-term relationship with customers, rather than just short-term solutions to temporarily ease customer dissatis-faction.

The present research expands our understanding ofthe links between the two important relationship con-structs, namely loyalty and word-of-mouth in the contextof service failure and recovery. Poor service failure recov-ery can often exacerbate the relationship with customers,causing them to rate the firm’s service recovery perfor-mance lower and even switch to other service providersafter service recovery encounters. Then, how can a firmmake customers remain loyal after service failure? Mostresearch on service failure and recovery has suggestedsatisfaction with service recovery as a solution to rebuildloyalty after service failures (Harris et al. 2006; Maxham2001; Maxham and Netemeyer 2002; Stauss 2002),Though there is no denying that satisfaction with recoveryis critical factor in ensuring customer loyalty, recoverysatisfaction may not be sufficient to help customers toremain loyal after a service failure. Recovery satisfactionmostly taps the cognitive components (i.e., expectations,disconfirmations, justice perception) (Harris et al. 2006;McColl-Kennedy, Daus, and Sparks 2003; Smith et al.


1999; Wirtz and Mattila 2004). However, a service failureoften evokes strong emotional responses from customers(e.g., angry, frustrated, irritated, etc.) and the affectiveresponses has a significant influence customers’ evalua-tion of organization’s service recovery effort (Andreassen1999). Therefore, it is highly likely that service recoveryefforts affect customers’ emotional tie with firms.

While important, compared to the constructs such assatisfaction and trust, the role of customer affection hasbeen overlooked in service recovery until very recently(Yim et al. 2008). The results of the present study deep-ened our understanding of the antecedents of loyalty in thecontext of service failure and recovery by finding the linksbetween customer affection and other constructs such asloyalty and word-of-mouth. The results show that strength-ening the emotional tie between customers and companiesis crucial in enhancing customer relationships after ser-vice failure and recovery. Customer affection influ-ences word-of-mouth directly and indirectly via loy-alty. Together, the present research provides insightinto the nature of the relationship between customeraffection and word-of-mouth in the context of servicefailure and recovery.

Future research should focus on several areas toovercome the limitations of the present study. First, themodel used in the current study was tested using a self-administered survey. Provided that recall bias may haveinfluenced the results, the additional test using an experi-ment may help enhance the validity of the current find-ings. However, we believe survey is appropriate in thecurrent context as laboratory experiments, the alternativemethod, also have pitfalls in that they may not reflect thereal world situation appropriately. Second, the presentresearch used cross-sectional sample of undergraduatestudents. Though the use of undergraduate students mightnot be ideal for the study, we argue it was appropriategiven that the majority of participants of the present studyexperienced various service failures (e.g., wireless phoneservice, high speed internet, cable TV, restaurants, etc.).Nonetheless, we admit the use of a homogeneous studentsample may limit the findings’ generalizability. Futureresearch may consider testing the model using a broadrange of customers or customers who experienced similarservice failures and recoveries in specific service catego-ries to deepen our understanding in regards to the issuesaddressed in the present study. References are availableupon request.

For further information contact:Beomjoon Choi

California State University, SacramentoSacramento, CA 95819–6088



CAUGHT BY SURPRISE: THE BEHAVIORAL EFFECTS OF SURPRISEAND DELIGHT ON CONSUMERS IN DIFFERENT INDUSTRIES

Tobias Kraemer, EBS Business School, GermanyAndreas Giese, EBS Business School, Germany

Christopher Bartl, EBS Business School, GermanyNadine Ludwig, EBS Business School, Germany

Matthias Gouthier, EBS Business School, Germany

SUMMARY

The concept of customer delight has increasinglybecome the focus of attention of researches and practitio-ners. This is due to increasing doubt that has been shed onthe impact of customer satisfaction on customer loyalty(e.g., Deming 1988; Leung and Kwong 2009; Schneiderand Bowen 1999). In several different industries it can beobserved that despite high degrees of customer satisfac-tion, a considerable share of customers still switches tocompeting brands for their next purchase. Therefore, it isnot sufficient for companies to strive to satisfy theircustomers, they instead need to achieve delight in order tobenefit from increased and persistent levels of loyalty andrepurchase intentions (e.g., Oliver, Rust, and Varki 1997;Schneider and Bowen 1999).

In order to evoke delight, a company has to providesome kind of unexpected value to the customer. This caneither be based on a surprising level of performance (e.g.,Finn 2005; Rust and Oliver 2000) or a complete surprise(e.g., Arnold et al. 2005; Schneider and Bowen 1999).According to Johnston (2007) the management systemwhich causes customer delight is service excellence.Service excellence occurs when customers perceive aservice to have exceeded their previous expectations ofthe respective service (Horwitz and Neville 1996). Theoverall goal of service excellence is to deliver such a highlevel of service quality that customer expectations can beexceeded which in turn will result in delight (Johnston2004, 2007). This level of service quality was representedby the construct excellent service quality for the purposeof the authors’ study.

Considering the role of delight in premium and massmarkets, one intuitively might assume that delight is ofgreater importance in a premium market setting. Productsand services there are usually associated with high in-volvement. This means that from a customer’s perspec-tive, such goods are of a certain relevance in the customer’slife; possibly, as a status symbol or based on one’s self-conception (e.g., Kim and Sung 2009; Traylor 1981;Zaichkowsky 1985). Following the argument presentedby Oliver et al. (1997), for companies operating in indus-tries which offer low-involvement products and services,

it is hardly possible to evoke the emotion of delight. Suchproducts are most common in mass markets.

The authors conducted a survey focusing on theantecedents and effects of delight in the two differentmarkets in order to gain an empirical insight into therespective role and importance of customer delight as wellas the importance of excellent service quality and surpriseas an antecedent. In order to do so, two services withcompletely opposite degrees of customer involvement,namely premium versus mass-market service products,were identified. On the one hand, representing the pre-mium business segment offering a high-involvement ser-vice, a luxury hotel was chosen. The mass-market servicewith a low-involvement product was a mobile phoneservice offered by a telecommunications provider. In bothsettings, the effect of surprise as well as excellent servicequality on delight was analyzed. To assess the influence ofdelight on costumer loyalty, the effects of delight onrepurchase intentions and recommendation intentions wereanalyzed. Two questionnaires were created whichincluded introductory descriptions of service encoun-ters for the two different market environments.

Before the actual data collection, a convenience pre-test was conducted, in which a total of 17 people workedthrough the respective parts of the questionnaires.Respondents were allocated to a questionnaire on thebasis of a random selection. In total, a sample of 328respondents (average age 26.7; 52.4% female, 47.6%male) completed the survey. One hundred sixty-sevencomplete questionnaires for the high-involvement sce-nario and 161 answers for the low-involvement scenariocould be obtained. For hypotheses testing the authors usedstructural equation modeling.

The effect of excellent service quality and delight onrepurchase intentions could be shown not only in the high-involvement, premium market scenario, but also in thelow-involvement, mass market. Nonetheless, the resultsindicate that there are still major differences between thecases: Considering the recommendation intentions of thecustomers, there is no significant effect of delight in themass market, while there is one in the premium market. Apossible explanation is that customers of customizedpremium market offerings experience a higher degree of


emotional involvement and delight than customers ofstandardized mass market offerings. The degree of emo-tional involvement stimulates the likelihood of consum-ers to establish a relationship with the provider and thustheir willingness to spread word-of-mouth. Secondly,there is no direct effect from surprise on either repurchaseintentions or recommendation intentions. It is mediatedthrough delight as suggested in the literature on emotionswhich describes surprise as a neutral preemotion.

Since managers do want to create customer delight,but are unaware of how to implement such measures inpractice, the identification of two delight triggers, namelysurprise and excellent service quality, is very helpful.However, although surprise is an activator for delight, itsimpact is not equally strong across industries and, com-

pared to their premium-market colleagues, provider ofmass-market services can achieve better results with lesseffort. Expectations in the premium market are high, butif delighted customers become loyal customers the extraeffort is legitimized. Since the effect of surprise differs inthe scenarios presented, managers need to be aware of thesetting they operate in and align their surprise events to it.A strategy of service excellence is not suited for everycompany or every target group. The behavioral intentionsprovoked by the service excellence strategy also need tobe considered to be able to conduct a cost-benefit analysis.A final evaluation of whether the loyalty-induced behav-ioral effects outweigh the costs associated with the sur-prise measures in the long-run has to be assessed by themanagement on case-specific basis. References are avail-able upon request.

For further information contact:Tobias Kraemer

Center for Service ExcellenceServices Marketing

EBS Business SchoolRheingaustr. 1, 65375

Oestrich-WinkelGermany

Phone: +49.611.7102.1479Fax: +49.611.710210.1479



FORGING RELATIONSHIPS TO COPRODUCE: A CONSUMERCOMMITMENT MODEL IN AN EXTENDED

SERVICE ENCOUNTER

Lin Guo, University of New Hampshire, DurhamCuiping Chen, University of Ontario Institute of Technology, Oshawa

Chuanyi Tang, University of Florida, Gainesville

SUMMARY

Although the concept of commitment has been exam-ined in various consumer services contexts, an under-standing of consumer commitment theory is still incom-plete. First, the role of commitment as a stabilizing psy-chological force that can direct one’s behavior has notbeen sufficiently acknowledged. As a result, the investi-gation of behavioral implications of commitment hasbeen exclusively limited to consumers’ switching orreferral intentions (e.g., Bansal et al. 2004; Barksdale,Johnson, and Suh 1997; Verhoef et al. 2002). Second, theunique psychological mechanisms that can foster variousdimensions of consumer commitment have not beenexplored. Consequently, the widely examined ante-cedents of consumer commitment are limited to satisfac-tion, trust, and switching cost (e.g., Bansal et al. 2004;Brown et al. 2005; Jones et al. 2007). Finally, with a fewexceptions (e.g., Barksdale et al. 1997), the mostfrequently studied service encounters in commitmentresearch are characterized by low consumer participation.This is in contrast to the employee-organization contextwhere the commitment construct was originated asemployees usually play an active role in organiza-tions’ production. Our study intends to address theselimitations through reconceptualizing consumer commit-ment, developing a theoretical model of consumer com-mitment, and testing the model in a complex and extendedservice encounter.

We argue that both organizations’ and consumers’investment in the relationship help forge consumers’commitment to the organization, which in turn promotesconsumers’ retention as well as a variety of theircoproduction behaviors. Data were collected via a self-administered survey from 364 clients of a debt manage-ment program in the U.S. A structural equation modeling(SEM) approach was used to establish the constructvalidity and test the hypotheses. Our results showed that,compared with firms’ and consumers’ relationship invest-ment, perceived availability of alternatives has a muchstronger effect on consumers’ calculative commitment. Incontrast, firms’ and consumers’ relationship investmentplay more important roles in forging affective commit-ment. Furthermore, as predicted, calculative commit-

ment, associated with the cost of leaving, only increasesintention to remain, while affective commitment, associ-ated with consumers’ emotional bond to the organizationleads to a broader range of desired behaviors includingindividual initiative and civic virtue. Surprisingly, affec-tive commitment does not influence consumers’ intentionto remain and compliance. Firms’ relationship investment(i.e., perceived organizational support, POS) is the onlypredictor of compliance in the model and its effect is notmediated through affective commitment. Lastly, onlyaffective commitment has a direct impact on civic virtue.The effects of POS and consumers’ relationship invest-ment are fully mediated through affective commitment.

Findings of this study have several implications forconsumer commitment research in services context, espe-cially when the successful service delivery comprises acontinuous stream of transactions and requires custom-ers’ participation. We build on previous consumer com-mitment literature by studying additional behavioral out-comes of commitment other than switching or referralintentions. Our findings also revealed the differentialeffects of different types of commitment on consumerbehavioral consequences. In addition, examining bothorganizations’ and individuals’ relationship investmentas antecedents, our study also suggests additional meansfor developing consumer commitment.

Our findings also provide guidance for managers inthe debt management program and other similar behav-ioral change counseling services such as weight loss andsmoking cessation. Although consumers’ perceived costof exiting the relationship may increase their intention tostay in the program, consumers’ intention to stay is notadequate for their successful completion of the program.In order for consumers to complete the program, they needto follow their service providers’ instructions and complywith their requests. To gain consumers’ compliance, pro-viding social and emotional support to consumers is thekey and cultivating an emotional bond with consumersmay not be necessary. However, in order to engageconsumers to perform beyond the minimally requiredlevel, forging consumers’ emotional attachment becomesas crucial as offering support. References are availableupon request.


For further information contact:Lin Guo

Whittemore School of Business and EconomicsUniversity of New Hampshire

15 Academic WayDurham, NH 03824

Phone: 603.862.2709Fax: 603.862.4468



YOU’RE SUCH AN EMBARRASSMENT! A QUALITATIVE STUDY OFTHE DETERMINANTS AND CONSEQUENCES OF VICARIOUS

EMBARRASSMENT IN CUSTOMER-TO-CUSTOMERINTERACTIONS IN THE SERVICE CONTEXT

Thomas Kilian, University of Koblenz-Landau, GermanyKathrin Greuling, University of Koblenz-Landau, Germany

Eva Hammes, University of Koblenz-Landau, Germany

SUMMARY

Research Topic and Aims

Consumers’ vicarious embarrassment, which occurswhen one “fellow customer” (Wu 2008, p. 1504) feelsembarrassed for another, has not yet been analyzed in theservice context. Krach et al. (2011) recently demonstratedthe existence of vicarious embarrassment subsequent to aviolation of social norms or standards. Nonetheless, theantecedents and consequences of vicarious embarrass-ment in a service context remain unclear and so arebusiness consequences. Thus, our research aims are (1) todemonstrate the existence of vicarious embarrassment,(2) to identify the determinants of vicarious embarrass-ment, and (3) to identify and explain the consequences ofvicarious embarrassment, including the correspondingemotions.

Literature Review

Embarrassment is a self-conscious, socially occur-ring emotion that can be explained by someone’s fear oflosing his or her social identity through others’ evalua-tions (Keltner 1995; Miller 1996). However, the questionis whether the consequences mentioned in embarrassmentand empathic embarrassment literature, such as the inten-tion to flee embarrassing situations, which lead to nega-tive service evaluations (Brumbaugh and Rosa 2009;Dahl, Manchanda, and Argo 2001; Grace 2007; Marcusand Miller 1999; Robbins and Parlavecchio 2006; Shearnet al. 1999; Silver, Sabini, and Parrott 1987), are also truefor cases in which one is simply observing someone elseviolating social norms or being in an embarrassing situa-tion. Vicarious embarrassment could be highly relevant inservice transactions because such encounters often con-sist of interactions among a service provider and custom-ers and other customers present, in which embarrassmentcould be particularly relevant (Richins 1997). Because theinappropriate behavior of employees or other customerscan result in a customer’s dissatisfaction (Bitner, Booms,and Tetreault 1990; Grove and Fisk 1997, Grove et al.1998; Guenzi and Pelloni 2004; Harris and Reynolds2003; Huang, Lin, and Wen 2010; Huang 2008; Martin

1996; Martin and Pranter 1989; Moore et al. 2005; Wu2008), vicarious embarrassment is expected to lead tosimilar effects. Krach et al. (2011) demonstrated that itdoes not matter whether an incident occurs intentionallyor unintentionally, or with the embarrassed customer’sawareness or not.

Method

Critical incidents in which one customer feels embar-rassed for another were collected during 25 interviews, inwhich respondents were asked to recall service experi-ences when they felt vicariously embarrassed for anothercustomer. The respondents were encouraged to recount aspecific story in detail and to relay their feelings andreactions to gain information not only about the incidentsbut also the customers’ emotions, the possible determi-nants and the consequences. Content analysis was used toidentify recurring themes in the interview transcripts.Through an open coding process, a code system wasdeveloped linking the 386 coded interview fragments tothe categories incidents (58), determinants (196), andconsequences (132) of vicarious embarrassment.

Findings

First, we identified verbal and behavioral violationsof social norms to elicit vicarious embarrassment in ser-vice situations. This corresponds to research by Huang(2008), who demonstrated that the customer failures ofpeople talking too loudly, cutting in line, crowding, andverbal and physical abuse were reported most often.Second, the influence of the following determinants waselaborated: work experience in service, embarrassability,the closeness of the relationships, inconvenience, thesituational involvement, the degree of violation of socialnorms, and the service sector. Third, the feeling of vicari-ous embarrassment was concretized through the analysisof emotional, cognitive, and behavioral reactions. Hence,vicarious embarrassment can be described by feelingssuch as discomfort, irritation, agitation, annoyance, com-passion for the service provider, the intention to flee,vicarious shame, or ridiculousness. The reactions tovicarious embarrassment comprised feelings, thoughts,


acting, calming actions, apologies, and ignoring by thefellow customer. Finally, we disclosed relevant businessconsequences of vicarious embarrassment for servicefirms.

The findings confirm that vicarious embarrassment isfelt more frequently by people with work experience inthe service sector. Plausible explanations are that serviceproviders are involved in social transactions more oftenand have higher social competencies, including empathictraits (e.g., Groth, Hennig-Thurau, and Walsh 2009).

Our qualitative analysis affirmed a relationshipbetween personal embarrassability and how frequentlyvicarious embarrassment is perceived. Personalembarrassability refers to the ability to be easily embar-rassed, and consequently, to be more susceptible andmore affected by embarrassing situations.

Closeness of relationship as an antecedent of vicari-ous embarrassment was not supported by our qualitativeinterview data, which contradicts Miller (1987) and Stockset al. (2011). However, our findings are consistent withthe work of Tangney et al. (1996), who found that embar-rassment was more common for strangers, but strongerfeelings such as guilt and shame were more common forloved ones. Hence, we can state that vicarious embarrass-ment is felt in every kind of underlying relationship

between customers and fellow customers; however, theintensity of the emotion and the reactions could differ.

The situational involvement, the degree of violationof social norms and the inconvenience were found asfurther determinants of vicarious embarrassment.

The identified consequences for service providersincluded increased negative word-of-mouth intentionsand negative associations toward the firm. This conse-quence can result in image damages, depending on theseverity of the incidents and the skills of the serviceemployee to handle vicarious embarrassment incidents.Furthermore, the intention to flee the situation was sup-ported.

Contributions

On the basis of our qualitative analysis, we deriveduseful information about how, when, and to what extentvicarious embarrassment as an emotional variable influ-ences service evaluations from the customers’ point ofview and results in consequences for service firms. Wethereby advance the existing knowledge of emotionalprocesses in service transactions from a customer-to-customer perspective and discuss vicarious embar-rassment in the service sector for the first time. Referencesare available upon request.

For further information contact:Katherin Greuling

Institute for ManagementUniversity of Koblenz-LandauUniversitaetsstrasse 1, 56070

Koblenz, GermanyPhone: 00492612871593



TEMPTED BY ANOTHER: HOW CUSTOMER-PERCEIVEDCOMPETITIVE ADVANTAGE INFLUENCES REPURCHASE

INTENTIONS IN SERVICE RELATIONSHIPS?

Martin Mende, University of KentuckyScott Thompson, University of Georgia

Christian Coenen, Zurich University of Applied Sciences, Germany

SUMMARY

In light of hypercompetitive markets, customer loy-alty remains a strategic priority for firms across serviceindustries (Palmatier et al. 2006). Consequently, servicemarketers strive for an ever better understanding of thedrivers of customers’ repurchase intentions. However,surprisingly little research on service relationships hasfocused on the specific impact competitors have on repur-chase intentions. Accordingly, marketing scholars havecalled for a stronger emphasis on understanding the influ-ence of a firm’s competition on loyalty mechanisms(Boulding et al. 2005). Answering this call, this studyexamines the effects of customer-perceived competitiveadvantage (CPCA) – a consumer’s explicit assessment ofthe service firm relative to its competitors—on repurchaseintentions, above and beyond those of other predictorsincluding satisfaction.

Customer satisfaction is widely considered the cen-tral determinant of loyalty intentions (Seiders et al. 2005).However, some authors have argued that the satisfactionconstruct may not paint a complete picture of how compe-tition affects loyalty; thus, they have proposed that cus-tomer assessments of the firm relative to competitors helpprovide a different lens in predicting loyalty (Dick andBasu 1994). This idea of relative firm assessmentsreceived particular attention in research on customervalue (Gale 1994) and attractive alternatives (Ping 1993).However, the extant literature suffers from importantlimitations. To expand current knowledge regarding theinterplay of competition and loyalty intentions, we studyCPCA. This study proposes that CPCA will have a posi-tive effect on repurchase intentions from a focal firm, overand above any effect of satisfaction. Moreover, it hypoth-esizes an interaction between CPCA and satisfaction suchthat with lower levels of satisfaction the positive relation-ship between CPCA and repurchase intentions becomesstronger.

We test the hypotheses using data on actual servicerelationships, employing a survey research design in thecontext of customer-bank relationships. The survey usesa comparative measurement approach that requires cus-tomers to compare a focal service firm (bank) to itscompetitors. This approach deliberately recognizes thatcustomers evaluate and choose companies in a context-

dependent rather than an isolated manner (Tversky andSimonson 1993). The results from an Ordinal LogisticModel demonstrate that accounting for a service firm’sCPCA is fundamental for better understanding and pre-dicting repurchase intentions. Specifically, there is a sys-tematic association between CPCA and repurchase inten-tions, such that higher levels of CPCA are linked toincreased repurchase intentions, even when satisfaction isaccounted for. Indeed, controlling for CPCA renders themain effect of satisfaction on repurchase intentions notsignificant. Additional analyses show that, consistentwith the existing loyalty literature, satisfaction is a strongdriver of repurchase, but only in the absence of CPCA.

In addition, the predicted interaction between CPCAand satisfaction emerges. We show that CPCA interactswith satisfaction to affect repurchase intentions, with theimpact of CPCA increasing as satisfaction decreases.Thus, the impact of CPCA on repurchase intentionsdepends on the level of satisfaction, with CPCA having alarger impact when customers are less satisfied. In otherwords, less satisfied people particularly weigh CPCA asthey decide whether to repurchase. This strongly suggeststhat future researchers should account for CPCA whenstudying repurchase intentions.

Moreover, scholars interested in the concept of cus-tomer value should consider the distinct conceptual roleof CPCA vis-á-vis the consumer’s trade-off betweenbenefits and sacrifices. Given the prominence of customervalue in marketing, our analyses controlled for this vari-able (i.e., its trade-off operationalization). This approachmade our hypothesis tests more conservative and empha-sized that CPCA represents a unique theoretical andempirical contribution to the prediction of repurchaseintentions that the established trade-off conceptualizationof customer value fails to capture.

In summary, given its role in shaping repurchaseintentions, theoretical frameworks in marketing shouldinclude the notion of CPCA as a distinct conceptualantecedent of consumers’ loyalty intentions in serviceindustries. In parallel, service marketing scholars shouldexamine the possibility that satisfaction plays a morecontingency-related role regarding repurchase intentionsthan has been previously assumed.


SELECTED REFERENCES

Boulding, W., R. Staelin, M. Ehret, and W.J. Johnston(2005), “Customer Relationship ManagementRoadmap,” Journal of Marketing, 69 (4), 155–66.

Dick, A.S. and K. Basu (1994), “Customer Loyalty:Toward an Integrated Conceptual Framework,” Jour-nal of the Academy of Marketing Science, 22 (2), 99–113.

Gale, B.T. (1994), Managing Customer Value: CreatingQuality and Service That Customers Can See. NewYork: Free Press.

Palmatier, R.W., R.P. Dant, D. Grewal, and K.R. Evans(2006), “Factors Influencing the Effectiveness ofRelationship Marketing: A Meta-Analysis,” Journalof Marketing, 70 (4), 136–53.

Ping, Jr., R.A. (1993), “The Effects of Satisfaction andStructural Constraints on Retailer Exiting, Voice,Loyalty, Opportunism, and Neglect,” Journal of Re-tailing, 69 (3), 320–52.

Seiders, K., G.B. Voss, D. Grewal, and A.L. Godfrey(2005), “Do Satisfied Customers Buy More? Exam-ining Moderating Influences in a Retailing Context,”Journal of Marketing, 69 (4), 26–43.

Sung, Y. and S.M. Choi (2010), “I Won’t Leave YouAlthough You Disappoint Me: the Interplay BetweenSatisfaction, Investment, and Alternatives in Deter-mining Consumer-brand Relationship Commitment,”Psychology and Marketing, 27 (11), 1050–74.

Tversky, A. and I. Simonson (1993), “Context-Depen-dent Preferences,” Management Science, 39 (10),1179–89.

For further information contact:Scott Thompson

University of GeorgiaAthens, GA 30602–6251



THE SERVICE LEVEL/GRATITUDE/RECIPROCATION RELATIONSHIPAND THE MODERATING IMPACT OF RECIPROCATION WARINESS

Yihui (Elina) Tang, University of Illinois at ChicagoChris Hinsch, Grand Valley State University, Grand Rapids

SUMMARY

Consumer gratitude is by and large an under-researched topic in the field of marketing and consumerbehavior. There are three seminal articles that haverecently dealt with the gratitude construct. First, Morales(2005) suggested that gratitude influenced a consumer’swillingness to pay for products and services. Second,Soscia (2007) found that consumer gratitude was posi-tively linked to word of mouth (WOM) and repurchaseintention. In a more heavily cited article, Palmatier, Jarvis,Bechkoff, and Kardes (2009) found that gratitude medi-ated the relationship between relationship marketinginvestments (RMI) and several desirable outcomes for thefirm (i.e., purchase intention, share of wallet, sales rev-enue and sales growth). In this paper, we extended thefindings of Palmatier et al. (2009) in two ways. First, wereplicated Palmatier’s finding in a services setting usingincreased service level as a relationship marketing invest-ment. We tested the mediating impact of gratitude on therelationship between service level and customer satisfac-tion, reciprocal behaviors and the consumer’s behav-ioral intentions of repurchase and WOM. Second, weaddressed factors that may impact the link betweenrelationship marketing investment (service level) andgratitude. Specifically, does an RMI generate the samelevel of gratitude in all consumers, or are some consumersmore/less receptive to the provision of these benefits? Weproposed that gratitude is not solely dependent on thebenefit received, but a consequence of both the benefit(service level) received and the customer’s psychologicalmakeup. We hypothesized that a subject’s level of recip-rocation wariness, or an individuals’ “general fear ofexploitation in interpersonal relationships” (Cotterrellet al. 1992; Eisenberger et al. 1987) would affect theirreaction to receiving an RMI. People who are high inreciprocation wariness are worried about being exploitedin a relationship; therefore we hypothesized that theywould not respond to high levels of RMI with the samelevels of gratitude as low reciprocation wary individuals.As a result, the same level of RMI will result in betteroutcomes for the firm if the customers are less reciproca-tion wary.

The hypotheses are outlined below:

H1: The higher the service level the more gratitude will befelt by the recipient (consumer).

H2: Gratitude will mediate the relationship between ser-vice level and customer satisfaction level.

H3: Gratitude will mediate the relationship between ser-vice level and reciprocation behavior.

H4: Gratitude will mediate the relationship between ser-vice level and behavioral intention.

H5: Reciprocation wariness will moderate the relation-ship between service level and gratitude, such that therelationship will be more positive for low wary thanhigh wary individuals.

To test the hypotheses, we collected data using anonline survey. There were 3,200 graduate students in aMidwestern university were invited to participate in thesurvey for a chance to win one of five $50 gift cards. Atotal of graduate 234 students (average age > 27 years)completed the survey. Participants were assured of confi-dentiality and were randomly assigned to one of twoservice scenarios based on their month of birth. Onescenario represented adequate (or low) service level. Theother scenario represented high service level. After read-ing through the scenario, the respondents answered ques-tions on gratitude, customer satisfaction level, tippingbehavior, behavioral intention and reciprocation wari-ness.

In analyzing the data, to construct the measurementmodel, we entered all reflective constructs into an EQS6.1 confirmatory factor analysis (Bentler 1995). All fac-tors met the accepted reliability requirements (Nunnallyand Bernstein 1994). We followed the recommendationsof Bagozzi and Yi (1988) in assessing the fit of themeasurement model. The full model had good model fit.Factor loadings were satisfactory. The variances for allitems and factors was statistically significant indicatingconvergent validity (Anderson and Gerbing 1988). Thisfinding was reinforced by the fact that all standardizedfactor loadings were positive, high in magnitude andstatistically significant.

We followed Preacher and Hayes (2004, 2008) to testfor the mediation effects in H2–H4. We found support forall three hypotheses. We used linear regression to test H1and H5 and found support for both. Reciprocation wari-ness negatively moderated the relationship between ser-vice level and gratitude data. At an adequate service level,


both high and low reciprocation wariness subjects reactedwith the same level of gratitude. Gratitude increased withbetter service for subjects in both high and low reciproca-tion wariness. However, the increase for high reciproca-tion wariness individuals was not as steep as it was for lowreciprocation wariness individuals.

This research represents three important contribu-tions to the marketing field. First, this research extendedextant findings about relationship marketing to a servicesscenario. This study replicated the findings of Palmatier etal. (2009) using service level as a relationship marketinginvestment. Second, gratitude was found to mediate rela-

tionships between service level and behavioral intentions,tipping, and customer satisfaction. The third and mostimportant finding from this research comes from uncov-ering the moderating effect of reciprocation wariness thatimpacts the service level/gratitude relationship. Gratitudehas been shown to be an important construct in thedetermination of customers’ post purchase behavior.Revealing factors that impact this relationship will ad-vance our understanding of the black box that is therelationship marketing process. We proposed and demon-strated that customers’ responses to RMI differ accordingto their level of reciprocation wariness. References areavailable upon request.

For further information contact:Yihui (Elina) Tang

Department of Managerial StudiesUniversity of Illinois at Chicago601 S Morgan Street, MC 243

Chicago, IL 60607E-Mail: [email protected]


GIVE-AND-TAKE IN LOYALTY PROGRAMS: THE ASYMMETRICEFFECTS OF MEDIUM MAGNITUDE

Sören Köcher, TU Dortmund University, GermanyMarkus Blut, TU Dortmund University, Germany

SUMMARY

Over the last decade, medium-based loyalty pro-grams have become a popular instrument for service firmsto improve customer retention. These programs offer theirmembers a certain number of medium units (e.g., points,miles, stamps, etc.) for each monetary unit they spend onthe service provider. The collected points in turn can beused to redeem a reward. Whereas most loyalty programsare similar in terms of this reward mechanism, they varyin terms of the amount of credited medium units permonetary unit, and consequently in the number ofmedium units required for redemption. For instance, anairline may either decide to set up a program which offersits customers 1,000 points for every flight and a free flightfor 10,000 points or a program with only 10 points perflight and a free flight for 100 points. Although theexchange rates between trips and loyalty points respec-tively points and rewards differ, from a normative per-spective these differences should be irrelevant as bothprograms require exactly the same effort (10 paid tickets)for the same outcome (a free ticket). Referring to thisexample, the first program is characterized by a highmedium magnitude, whereas the second exhibits a lowmedium magnitude.

Although a broad body of literature investigatesconsumers’ response to different loyalty program schemes(Dorotic, Bijmolt, and Verhoef 2011), research onmedium characteristics is rather scarce. Hence, this studyexamines the effects of medium magnitude on the two keydecisions in a loyalty program membership: the choicebetween (1) participation vs. non-participation in theprogram and (2) redemption vs. non-redemption of theaccumulated points.

Research Hypotheses

Building on prospect theory (Kahneman and Tversky1979; Thaler 1985), research on the effects of irrelevantattributes on consumer choice (e.g., Brown and Carpenter2000; Carpenter, Glazer, and Nakamoto 1994; Hsee 1995;van Osselaer, Alba, and Manchanda 2004), and the ap-proach of reason-based choice (Shafir, Simonson, andTversky 1993) the following hypotheses were theoreti-cally derived and empirically tested:

H1a: Medium magnitude has a positive impact on thelikelihood of joining a loyalty program.

H1b: Medium magnitude has a negative impact on thelikelihood of redeeming accumulated points.

H2a: The economic advantage of participation moder-ates the relationship between medium magnitudeand the likelihood of joining the program. A de-creasing economic advantage strengthens mediummagnitude effects.

H2b: The economic advantage of redemption moderatesthe relationship between medium magnitude andthe likelihood of redeeming accumulated points. Aneutral economic advantage strengthens mediummagnitude effects.

H3: The monetary amount required as partial paymentof rewards moderates the relationship betweenmedium magnitude and the likelihood of redeem-ing accumulated points. An increasing proportionof money in the payment of rewards attenuatesmedium magnitude effects.

Method and Results

To test the hypothesized effects in the participationand redemption context, we conducted three 2x2 full-factorial, between-subject experiments with a fictive rail-way company participants were asked to imagine theyused regularly. In sum, results of analyses of covariance(ANCOVA) reveal that medium magnitude effects areasymmetric: Although medium magnitude has a positiveimpact on the willingness to participate in a loyaltyprogram, it affects redemption behavior negatively. More-over, identified interactions between medium magnitudeand economic advantages improve the understandingof the integration of irrelevant attributes into decision-making. As medium magnitude effects predominantlyappear in the absence of a clearly superior choice option– i.e., participation vs. non-participation and redemptionvs. non-redemption – the irrelevant attribute of mediummagnitude is principally incorporated if the focus onimportant attributes does not provide sufficient reasonsfor choice. Furthermore, our results indicate that a combi-nation of loyalty points and money in the payment ofrewards predominantly reduces perceived costs if thepayment principally comprised loyalty points and thusenhance the tendency to redeem accumulated points.However, in this case medium magnitude is incorporatedinto the decision to redeem a reward very strongly produc-ing high differences in redemption behavior between high


and low magnitude programs. In contrast, if the paymentprincipally comprised money, the negative effect of me-dium magnitude disappears. But the willingness to re-deem a reward in these conditions remains generally low.

Implications

This research’s findings supply important implica-tions for a more efficient usage of medium-based loyaltyprograms in business practices. The acquisition of newmembers for a loyalty program should be realized either(a) via a high economic advantage of participation or (b)

by means of a high medium magnitude design that signifi-cantly improves the participation likelihood in low eco-nomic advantage programs. However, the second imple-mentation strategy holds the risk of a high participationbut low redemption intention. From the company’s per-spective, this is both good and bad news. On the one hand,a high magnitude program is less cost-intensive as mem-bers participate to gain rather than to redeem points. Onthe other hand, such a program might lose its attractive-ness in the long run if customers do not exchange theiraccumulated points for rewards. References are availableupon request.

For further information contact:Sören Köcher

Department of MarketingTU Dortmund University

Otto-Hahn-Straße 644221 Dortmund




THE BRIGHT AND DARK SIDE OF ENDOWED STATUS INHIERARCHICAL LOYALTY PROGRAMS

Andreas Eggert, University of Paderborn, GermanyIna Garnefeld, University of Paderborn, GermanyLena Steinhoff, University of Paderborn, Germany

SUMMARY

Hierarchical loyalty programs are a commonly usedrelationship marketing tool. Within these programs, cus-tomers exceeding a certain spending level are awardedelevated customer status (e.g., “gold membership,” “plati-num customer”). Elevating customer status has beenfound to be beneficial for firms due to increased customerloyalty (Drèze and Nunes 2011; Lacey, Suh, and Morgan2007) and the firm’s possibility to align the cost to servea customer to his value. Examples of hierarchical loyaltyprograms encompass many different industries, includingairlines, hotels, retailers, and banking.

In current business practice, it is observed that cus-tomers are not always awarded elevated customer statusbased on their past spending level. Rather, some compa-nies award some of their customers elevated statusalthough they did not achieve the spending level pre-defined in the loyalty program’s rules. That is, certaincustomers who typically are expected to be more profit-able in the future are elevated to a higher tier withouthaving met the spending level required for status eleva-tion (Kumar and Shah 2004). In this case, elevated statuscompares to the notion of endowed status because itsachievement is beyond customers’ control (Drèze andNunes 2009). Examples of companies endowing elevatedcustomer status include Hilton Hotels & Resorts(HHonors), Starwood Hotels & Resorts (Starwood Pre-ferred Guest), or Hertz Car Rental (Hertz Gold PlusRewards). Firms assume to profit from this practice asthey expect the same loyalty effects as from assigningachieved status.

However, according to literature from social psy-chology, endowment potentially differs from achieve-ment in its impact on customer loyalty. Specifically, weexpect endowed status to entail both a bright and a darkside effect on customer loyalty. On the one hand, based onsocial identity theory (Tajfel and Turner 1979) and theconcept of gratitude (Palmatier et al. 2009), we assumeendowed status to initiate a loyalty-enhancing effect viacustomer gratitude. On the other hand, referring to thetheory of psychological reactance (Brehm 1966), wepropose endowed status to result in a loyalty-reducingeffect via customer skepticism. Comparing endowed withachieved status, we suggest endowed status to entail a

dark side effect that does not occur for achieved status.Further on, we argue that the dark side effect of endowedstatus depends on its specific design. In line with attribu-tion theory (Weiner 1985), we expect that designingendowed status in a way that fosters customers’ attribu-tions of status endowment to their own behavior caninduce skepticism-alleviating mechanisms. Our paperaims at answering three research questions: What is cus-tomers’ loyalty response when they are endowed elevatedstatus, taking into account a bright and a dark side loyaltyeffect? How does the loyalty effect compare to that ofachieved status? How can companies alleviate the darkside loyalty effect of endowed status?

To address these questions, we conducted two labo-ratory experiments in the context of a hierarchical loyaltyprogram of a fictitious hotel chain. Study 1 analyzes thebright and dark side effect of endowed elevated customerstatus on customer loyalty and compares it to the loyaltyeffect of achieved elevated customer status. We employeda posttest control group design and manipulated customerstatus on three levels (endowed elevated status versusachieved elevated status versus no elevated status). A totalof 221 participants took part in the experiment. Respon-dents were randomly assigned to one of the three groups.The mean age of the sample was 33.1 years and 53.3percent were female. Our results show the proposed brightand dark side loyalty effect of endowed status. Via cus-tomer gratitude, endowed status increases customer loy-alty. Via customer skepticism, endowed status decreasescustomer loyalty. As compared to achieved status,endowed status reduces customer loyalty by generatingcustomer skepticism.

In Study 2, we examine whether two design charac-teristics of status endowment, i.e., customers’ freedom ofdecision and their proximity to achieving elevated cus-tomer status, can attenuate the dark side loyalty effect. Weconducted a 2 2 between-subjects factorial design inwhich we manipulated the freedom of decision (activedecision versus no active decision) and proximity toachieving elevated customer status (high proximity ver-sus low proximity). In total, 284 participants took part inthe experiment. The four groups were completely ran-domized. Our sample had a mean age of 32.5 years and50.5 percent of respondents were female. Results revealcustomer skepticism-alleviating effects of both custom-


ers’ freedom of decision and their proximity to achievingelevated customer status, which indirectly increase cus-tomer loyalty.

Our findings allow for the following implications:First, the detected positive impact of endowed status oncustomer loyalty implies that awarding customerselevated status although they did not regularly achieve itcan be an effective loyalty-enhancing instrument. Thus,managers of hierarchical loyalty programs can make useof elevated customer status beyond the established

employment as an award for the heaviest buyers. Thegeneral appeal of status can be applied as a stimulus toelicit gratitude and hence greater loyalty. Second, how-ever, the identified negative impact of endowed status oncustomer loyalty impairs its effectiveness and suggeststhat managers must be cautious when elevating custom-ers’ status through endowment. To alleviate skepticism,companies should give customers the opportunity to takean active decision and select customers that already fea-ture a high proximity to regularly achieving elevatedstatus. References are available upon request.

For further information contact:Lena Steinhoff

Marketing DepartmentUniversity of PaderbornWarburger Strasse 100

33098 PaderbornGermany

Phone: +49.5251.60.30.75Fax: +49.5251.60.34.33



NEW INSIGHTS IN THE MODERATING EFFECT OF SWITCHINGCOSTS ON THE SATISFACTION-LOYALTY LINK

Thomas Rudolph, University of St.Gallen, SwitzerlandLiane Nagengast, University of St.Gallen, Switzerland

Heiner Evanschitzky, Aston Business School, United KingdomMarkus Blut, University of Dortmund, Germany

SUMMARY

Due to empirical evidence questioning the impact ofsatisfaction-related constructs on behavioral loyalty (Mittaland Kamakura 2001; Seiders et al. 2005), research onexplaining customers’ repurchase behaviors has beenbroadened. Most notably, an increasing number of studiesinvestigate the role of switching costs (SC) as moderatorof the relationship between satisfaction and loyalty (e.g.,Jones, Mothersbaugh, and Beatty 2000; Patterson andSmith 2003; Woisetschläger, Lentz, and Evanschitzky2011). As these studies reveal inconclusive results, weconducted an extensive literature review and two empiri-cal studies to investigate the moderating effect of SC onthe link between satisfaction and behavioral loyalty. Themain contribution of this research is to resolve existinginconsistencies by suggesting an inverted u-shaped mod-erating effect of SC.

The literature review reveals empirical evidence forpositive as well as negative moderating effects of SC: 12studies report a negative interaction between satisfactionand SC, five report a positive interaction, and five find noeffects.

Considering theoretical approaches that explain themoderating role of SC, we found evidence for two oppos-ing effects:

First, there is theoretical evidence that SC moderatethe relationship between satisfaction and loyalty in anegative way (“lock-in effect”). Decreasing satisfac-tion levels do not necessarily result in lower loyaltyif SC are high because customers face psychologicaland/or economic effort if they switch (Bell, Auh, andSmalley 2005; Jones, Mothersbaugh, and Beatty2000). The benefits from switching to a more satisfy-ing provider are encumbered by increasing SC. Thus,increasing SC weaken the satisfaction-loyalty link asintentions to switch might be hindered (linear,decreasing curve).

Second, there is theoretical evidence for a positivemoderating effect (“amplifying effect”) which isespecially important for low SC. This effect suggeststhat increasing SC strengthen the relationship

between satisfaction and loyalty (e.g., Blut et al.2007). Research has shown that a high number ofalternatives (i.e., low SC) may lead customers tochoose others than their preferred providers andswitch to less satisfying alternatives (Sela, Berger,and Liu 2009). One explanation for these irrationaldecision behaviors is customers’ need for variety(Ratner, Kahn, and Kahneman 1999; Seetharamanand Che 2009) which induces customers with low SCto switch because switching is so easy. With increas-ing SC, customers will no longer switch to lesssatisfying alternatives. Thus, satisfaction is morelikely to translate into loyalty. In line with existingresults (e.g., van Trijp, Hoyer, and Inman 1996), weassume the satisfaction-loyalty link to strengthenwith increasing SC at a decreasing rate (concave up,increasing curve).

Combining these two effects, the moderating effectof SC can be described as follows: The amplifying and thelock-in effect complement each other depending on theSC-level: For low SC, the amplifying effect overlaps thelock-in effect as the former increases faster. For high SC,the lock-in effect exceeds the amplifying effect as thelatter decreases faster. Thus, the moderating effect of SCis supposed to follow an inverted u-shape. We assume thesatisfaction-loyalty link to be strongest for intermediateSC-levels. Customers with low SC switch retailers due toreasons other than satisfaction (weak link). Customerswith high SC are hardly able to switch regardless of theirsatisfaction (weak link). For intermediate SC, the ampli-fying effect is already strong whereas the lock-in effect isnot yet strong enough to compensate (strong link).

Contrary to existing studies that propose either apositive or a negative linear moderating effect of SC, wehypothesize a non-linear (quadratic) effect:

SC moderate the relationship between satisfactionand loyalty in a curvilinear (inverted u-shape) way.The relationship is strongest for an intermediatelevel of SC.

We empirically tested this hypothesis with data fromtwo retailing companies (food and DIY). Questionnaireswere sent to customers via postal mail. In total, we


received 1,816 (food) and 5,695 (DIY) usable responses.Satisfaction and SC were measured with established scales(Fornell et al. 1996; Ping 1993). To measure loyalty, werefer to actual purchase data (customers’ spending at thefocal retailer). These transaction data from the retailers’loyalty card programs were available for 12 months (food)and 6 months (DIY) after the survey. By including objec-tively measured data, we reduce the common method bias(Siemsen, Roth, and Oliveira 2010).

Using hierarchical regressions, we find positive maineffects of satisfaction (food: b = 359.49, t = 4.94, p < .01;DIY: b = 4.61, t = 2.01, p < .05) and SC (food: b = 97.59,t = 2.80, p < .01; DIY: b = 2.39, t = 2.24, p < .05) onrepurchase behavior (loyalty). We also show that thelinear interaction term between satisfaction and SC isinsignificant, whereas the quadratic interaction term issignificant in both studies (food: b = -88.65, t = -2.19,p < .05; DIY: b = -.14, t = -2.23, p < .05). The negative

sign of the quadratic term indicates that the moderatingeffect is inverted u-shaped. Thus, our results confirm thatthe effect of satisfaction on loyalty is strong for mediumSC and weak for low and high SC-levels.

This has important implications for retail managersregarding customer retention management. Retailersshould consider customers’ SC-levels when definingfuture investments in satisfaction management. Whilefocusing on satisfaction is a viable and rewarding strategyfor customers with medium SC, it is not sufficient forcustomers with low or high SC.

To sum up, our paper offers new theoretical insightsand contributes to a better understanding of the linkbetween satisfaction and behavioral loyalty. It bringstogether the conflicting results on the moderating role ofSC and gives a reasonable explanation for existing incon-sistencies. References are available upon request.

For further information contact:Liane Nagengast

Institute of Retail ManagementUniversity of St. Gallen

Dufourstrasse 40a9000 St.GallenSwitzerlandPhone: +41.71.224.7191



MANAGING SALESFORCE SELLING BEHAVIORS ANDPERFORMANCE: THE INTERACTIVE EFFECTS

OF SALES CONTROL SYSTEMS

C. Fred Miao, Clarkson University, PotsdamKenneth R. Evans, University of Oklahoma, Norman

SUMMARY

For many firms their salesforce is the only organiza-tional unit that generates sales revenues, which is espe-cially true for business-to-business relationships wheresalespeople are key boundary-spanners between the com-pany and its customers. Sales control systems represent animportant managerial tool in directing salesforce sellingbehaviors such as adaptive selling and selling effort.However, majority of prior sales control research hasfocused on their main effects only. While sales forcecontrol systems can be either outcome- or behavior-based, in practice most sales organizations typically use ahybrid control strategy by integrating both behavioral andoutcome control elements in directing their sales forces.Although prior research has suggested potential positivesynergies of behavioral and outcome control styles inaffecting sales performance (Cravens et al. 2004; Jaworskiet al. 1993; Onyemah and Anderson 2009), importantresearch gaps remain. First, these studies take the configu-ration perspective and link hybrid sales control systemsdirectly to sales performance. Therefore, how exactlysuch hybrid sales control systems may bring about posi-tive synergies remains in a black box because thesestudies did not explicitly investigate the mechanismsthrough which such synergies may be created. Second,studies on hybrid sales control systems treat behavioralcontrol as a unidimensional construct without distin-guishing between capability control and activity control.While both embedded in the global domain of behavioralcontrol, capability control and activity control may havedifferential and even opposite psychological and behav-ioral consequences (Evans et al. 2007; Miao et al. 2007).Indeed, prior research on hybrid control systems that didnot distinguish between capability control and activitycontrol produced inconsistent results (Cravens et al. 2004;Jaworski et al. 1993), which begs the question: Do hybridcontrol strategies characterized by high degrees of behav-ioral and outcome control elements always produce posi-tive synergies? By treating capability control and activitycontrol as separate behavioral control constructs, we maybe able to shed light on the seemingly inconsistent find-ings reported in prior research.

Our conceptual framework investigates the differen-tial interactive effects of activity control, capability con-

trol, and outcome control on sales performance via twomediating variables (i.e., adaptive selling behavior andselling effort). This conceptualization is consistent withthe Job Demands – Resources (JD-R) Theory (Bakker andDemerouti 2007; Bakker et al. 2010). Although each jobhas its own associated characteristics, JD-R theory sug-gests that most job-related factors can be grouped into twogeneral categories: job demands and job resources. JD-Rtheory further posits that job demands and job resourcescan interact to affect important job outcomes through twodifferent mediating processes. One process focuses onhealth-impairing effects mediated by job stress/strain, andthe other process investigates performance-enhancingeffects via job engagement (Bakker and Demerouti 2007).Within the current study, we focus on the job engagementmediation process. In the terminology of JD-R, outcomecontrol can be understood as job demands imposed onsalespeople due to its high pressure and risky nature(Anderson and Oliver 1987); in contrast, activity controland capability control can be perceived as job resourcesbecause they are specifically designed to enable sales-people to achieve their work goals and cope with jobdemands (Challagalla and Shervani 1996; Fang et al.2005; Kohli et al. 1998). Moreover, given that adaptiveselling behavior and selling effort reflect the extent towhich salespeople are invested in the selling process, weconsider these selling behaviors as indicators ofsalesperson’s job engagement. Therefore, the causalordering of the constructs in our conceptual model areconsistent with the overarching framework of the JD-Rperspective, which is tested with a cross-sectional sampleof salespeople in the US manufacturing sector.

A Partial Least Squares (PLS) structural equationmodel finds substantive support for most hypothesizedrelationships as predicted by the JD-R theory. Specifi-cally, capability control and outcome control have posi-tive interactive effects on both adaptive selling behaviorand selling effort; activity control and outcome controlhave a positive interactive effect on selling effort but anegative interactive effect on adaptive selling behavior;capability control and activity control have a negativeinteractive effect on selling effort. Noteworthy is that onlythe main effects of outcome control (but not activity orcapability control) on selling behaviors are significantwhen sales control interactions are considered simulta-


neously, which reveals the complementary (moderating)role of activity control and capability control in directingsalesperson’s selling behaviors. These results providecompelling evidence that activity control and capabilitycontrol are two types of qualitatively different job

resources that can motivate salesperson’s job engage-ment in different fashions. Researchers studying hybridsales control systems can benefit by considering theirinteractive effects with outcome control separately. Ref-erences are available upon request.

For further information contact:C. Fred Miao

School of Business362 New Snell HallClarkson University

8 Clarkson AvePotsdam, NY 13699–5765

Phone: 315.268.3879Fax: 315.268.3810



DO SALESPERSON PERCEPTIONS OF MANAGEMENT-DIRECTEDTECHNOLOGY-ENABLED TRANSPARENCY INFLUENCE

THEIR BEHAVIORAL ETHICALITY?

John E. Cicala, Texas A&M University – KingsvilleAlan J. Bush, The University of Memphis

Daniel L. Sherrell, The University of MemphisGeorge D. Deitz, The University of Memphis

SUMMARY

Salespeople are often more at risk to deviate from desiredsocietal and organizational ethical behaviors than are otherfirm employees (Osborn and Hunt 1974; Ferrell and Gresham1985). This lack of direct supervision puts sales agents in aposition to engage in behaviors that may not be consideredvery ethical; which is why it has been suggested that, “decep-tive practices are often found in many areas of sales,” (Carson2001, p. 275). Research shows salespeople experience ethicalconflicts regularly in the performance of their duties (Dubinsky,Berkowitz, and Rudelius 1980). Such struggles can lead tolower job performance, increases in turnover, customer dissat-isfaction, and negative word-of-mouth about the selling orga-nization (Walker, Churchill, and Ford 1977). Schwepker andIngram (1996, pp. 1151–1152) recognize that increasing pres-sure on salespeople to perform could be a catalyst for unethicalbehavior stating, “personal selling is an area of marketing thatis particularly susceptible to ethical dilemmas.”

Salespeople who participated in a recent dyadic qualita-tive study on sales technology and ethics professed the abilityto easily use technology in order to engage in unethicalbehaviors when dealing with clients and supervisors. Respon-dent managers in the same work countered that they have theability to use technology to engage in surveillance of salesper-son behavior, thus increasing salesperson transparency andholding salespeople more accountable for accuracy in theircommunications (Bush et al. 2007). This paper extends exist-ing scholarship by focusing on the effect of salespersonperceptions of management’s technologically enabled behav-ioral monitoring ability on the likelihood of salespeopleengaging in unethical selling behavior.

The construct perceived salesperson transparency isintroduced in this paper as a multi-dimensional concept incor-porating both a salesperson’s perceptions related tomanagement’s ability to access to information and salesper-son perceptions of management’s willingness to use saidaccessed information to influence or manipulate behaviorduring the selling process. Information cannot be used toinfluence or persuade unless it has been accessed (obtained) tobegin with. Similarly, if an individual’s perception is such thatmanagement may be accessing potentially incriminating sales-

person behavioral information but that managementwill not use that information against them, i.e., thereis no threat of punishment or adverse consequences,they may perceive that their behaviors are, for allpractical purposes, the opposite of transparent.

Agency Theory (AT) provides theoretical ground-ing as, “the relationship between a sales manager andsalesperson is an agency relationship,” (Bergen, Dutta,and White 1992, p. 8). It is hypothesized (H

1) that

salesperson perceived use by management of techno-logically accessed behavioral information has a me-diating effect on the relationship between salespersonperceived access by management of behaviorallyrelevant information and the likelihood of unethicalsalesperson behavior. In other words, any ethicalinfluence on behavior due to perceived managerialaccess to information is itself influenced by perceiveduse of that information.

Self-report surveys were completed by 253 busi-ness-to-business salespeople. The sample was foundto be representative of the current population ofsalespeople according to the United States’ Bureau ofLabor Statistics (www.bls.gov). Confirmatory factoranalysis (CFA) using structural equation modeling(SEM) with maximum likelihood estimation wasused to verify both factor loadings (in accordancewith the Kaiser-Gutterman rule) and construct valid-ity. The resulting levels of the standardized itemloading estimates were significant with values rang-ing from .67 to .92. T-values were each greater than2.0, providing further empirical support for conver-gent validity. Discriminant validity was supported, asthe largest value for shared variance between allconstruct pairings is 0.27, which was less than thelowest value for AVE (0.60) (Fornell and Larcker1981).

The hypothesis, H1, stated that salesperson per-

ceived managerial use of information accessed bytechnology-enabled management would have amediating effect on the relationship between sales-person perceived managerial access to information asenabled by technology, and the likelihood of unethi-


cal salesperson behavior. A chi-square difference test wasperformed between the fully mediated and partially medi-ated models to determine if there was a significant differ-ent between the models. Fornell and Larcker (1981)report, “the critical value for a chi-square with one degreeof freedom at the .01 level of statistical significance is6.63,” (p. 44). The results exceed this criteria (χ21 = 16.877p = 0.000), thus providing empirical support for a partiallymediating effect of the salesperson perceived managerialuse of information (Fornell and Larcker 1981; Babin andBoles 1996) and thus indicating empirical support for H

1.

This research helps explain a very important area thathas been greatly affected by technology - salespeople andtheir relationships with management and clients. Theformer relationship is the heart of agency theory, therelationship between principals (i.e., management) andagents (i.e., salespeople). The enabling of management by

technology to monitor more thoroughly its salespeople(access information relevant to their actions and behav-iors) should - in theory - lessen the issue of moral hazard.However, as shown by the results of this study, thatlessening is mediated by the agent’s perception of how theprincipal will use the knowledge obtained by accessinginformation. What is shown by this study is that theagent’s behavior will be predicated not by who benefitsthe most by the behavior in question, but by how theknowledge of their behavior will be used and that its useprovides more benefit to the principal than the agent.Further, this study has important applications for salesmanagers by providing empirical support for the idea thatif salespeople do not perceive any information accessedby management will actually be used by management,they will likely behave as if there is no transparency intheir agency relationship. Reference are available uponrequest.

For further information contact:John E. Cicala

Texas A&M University – KingsvilleCBA #215, 700 University Blvd.

Kingsville, TX 78363Phone: 361.593.4974



A NEW CUSTOMER TYPOLOGY FOR ADAPTIVE SELLING

Jeffrey S. Larson, Brigham Young University, ProvoSterling A. Bone, Brigham Young University, Provo

SUMMARY

Sales researchers have long advocated for tailoringpersonal selling efforts to the needs of the individualcustomer (Weitz, Sujan, and Sujan 1986; Homburg, Müller,and Klarmann 2011). Customers vary in preferences andneeds surrounding the products being sold, and thussalespeople should alter their sales presentations to high-light the products that would best fit each customer’sneeds. Customers also vary in their needs and their dispo-sition surrounding the sales process itself (Merrill andReid 1981; Williams and Spiro 1985). For example, somecustomers prefer to ask more questions in a conversa-tional sales interaction while others prefer a more abruptand decisive sales interaction (Wilson 2003). The effec-tiveness of a sales tactic is contingent upon the customer’sneeds and disposition as pertaining to the sales process(McFarland, Challagalla, and Shervani 2006).

Underlying Dimensions of a Customer’s DispositionToward Salespeople

The customer-salesperson interaction is a complexand ubiquitous aspect of market behavior, set apart by itsinterpersonal nature and the persuasion motives of theseller. We propose that through frequent interaction witha variety of salespeople, most people develop a multifac-eted disposition that guides their behavior in any salesper-son interaction. In specifically examining the personalselling interaction, we take a more focused approach thanthe persuasion knowledge literature, which examinesconsumer reactions to persuasion attempts from a broadbase of marketing agents and media (Friestad and Wright1994). However, our examination also exhibits a greatdeal of breadth, in two ways. First, the disposition weexamine extends across all personal selling interactions.Second, we explore seven unique dimensions of thisdisposition, and not simply customers’ resistance to sales-person persuasion attempts, making it a broader examina-tion in this regard than persuasion knowledge studies.

A major intended contribution of this paper is a broadexamination of the dimensions that make up a customer’sdisposition toward salespeople. As such, we sought to bethorough in determining the makeup of this disposition.The determination of a construct’s makeup should beguided by either an empirical investigation of the con-struct or an investigation of the theory surrounding theconstruct (Mowen and Voss 2008; Whetten 1989). We

employed both methods. The empirical aspect of theconstruct investigation utilized an iterative scale develop-ment and purification process. This empirical procedurewas guided by a thorough investigation of the theorysurrounding the construct, which was taken from twodifferent literature streams: social psychology and mar-keting/personal selling. Our investigation yielded sevendimensions that make up a customer’s disposition towardsalespeople (CDS): convincibility, avoidance, empathy,distrust, relationship seeking, self-presentation, and in-formation seeking.

Overview of Studies

We performed two studies with the goal of demon-strating that enduring differences in CDS have real, mea-surable effects on marketplace behavior. In Study 1, weseek to find whether a person’s enduring dispositiontoward salespeople can predict actual marketplace behav-iors. Specifically, we expect that one’s disposition towardsalespeople will predict consumers’ likelihood to engagethe help of an expert salesperson. We examine this in thecontext of real estate. In Study 2, we examine how CDSaffects observable behavior within a personal sellinginteraction. Prior research has examined how salespeoplecope with various negative emotions induced by the salessituation (Belschak, Verbeke, and Bagozzi 2006; Verbekeand Bagozzi 2002). We complement these prior results byexamining the customer’s behavior in a high-pressurepersonal selling interaction, particularly how CDS deter-mines the customer’s involuntary anxiety and protectivebehaviors.

Research Implications

Our research focus makes a unique contribution tothe sales literature as it is among the first to focus on theimpact of individual differences on the customer side ofthe interaction. By more thoroughly investigating theunderlying dimensions of customers’ orientation towardsalespeople, we generate novel insights into the determi-nants of customer behavior in the customer-salespersoninteraction. Perhaps more important, our research pro-vides a foundation for further inquiry in the area. Bydetermining the structure and dimensionality of theunderlying heterogeneity of customer preferences in thearea of personal selling interactions, future researchershave a framework to generate and test hypotheses in thisarea.


Our findings also contribute to the adaptive sellingliterature by expanding its scope. Adaptive selling usuallyfocuses on adaptive behaviors that a salesperson canundertake within a customer interaction to better meet thecustomer’s needs. But some customer types are betterserved with a more fundamental change to the structure ofthe customer-salesperson interaction. For example, somecustomers are disposed to avoid salesperson interactions

altogether, so a company seeking to adapt to the needs ofthese customers would need to seek a more fundamentalre-structuring of the entire buying process to allow thesecustomers to minimize their personal interaction withsalespeople. Our customer typology enables examinationof customer needs that require more fundamental adapta-tions, thereby providing sellers additional considerationsfor improving their sales efforts.

For further information contact:Jeff Larson

Marriott School of ManagementBrigham Young University

660 TNRBProvo, UT 84602



SALESPERSON’S ACCULTURATION BEHAVIOR AND ITS IMPACTON BUYERS’ COMMITMENT

Halimin Herjanto, Auckland University of Technology, New ZealandSanjaya S. Gaur, Auckland University of Technology, New Zealand

SUMMARY

In today’s plural market, salespersons are not onlyexpected to achieve sales target, but are also required tobuild and maintain a healthy relationship with buyers(Crosby, Evans, and Cowles 1990). To build this fruitfulrelationship, salespersons should have integrity and adapt-ability to buyers’ requirements and have the sense ofcultural sensitivity to match buyers’ culture (Schultz,Evans, and Good 1999). This cultural sensitivity or adapt-ability is known as the process of acculturation (Ebin et al.2001). The acculturation creates better intercultural rela-tionship (Lerman, Maldonado, and Luna 2009) and facili-tates buyers’ tendency to maintain their original con-sumption patterns in relation to their own native culture(Penaloza 1994). Schultz et al. (1999) suggest that afailure to display sufficient cultural sensitivity can jeopar-dize the relationship and ultimately drives buyers awayand therefore, threaten business continuity andsustainability. The concept of acculturation has started togain momentum among marketing scholars. These schol-ars (Quester and Chong 2001; Bojavic and Xu 2006; Pio2007; Lerman, Maldonado, and Luna 2008) have investi-gated the concept of acculturation from buyers’ perspec-tive. Despite their invaluable findings, their study hasbeen very limited to buyers’ acculturation perspective anddo not provide any empirical investigation into the impactof salespersons’ acculturation behaviors on buyers fromthe perspective of salespersons. An attempt is thereforemade in the present study to fill this gap in the research byexamining how these behaviors impact the buyers’ com-mitment. Within this study, salesperson acculturation isbroadly defined as salespersons’ tendency to adjust andadapt their sales approaches to buyers from differentculture, and based on Berry (1997), salespersons accul-turation is classified into four dimensions; assimilation(an acculturation behavior whereby one prefers to havemore involvement with the others’ culture and have littleinterest in maintaining their own culture), separation (anacculturation behavior that occurs when one prefers tomaintain own culture and have little involvement withothers’ culture), integration (an acculturation behaviorthat happens when ones maintaining their own culturalvalues with pride while at the same time respecting theculture of the others) and marginalization (an accultura-tion mode whereby ones lose connection with their ownculture but do not replace it with the dominant culture).We also attempt to enrich the knowledge in this area by

further exploring the mediating role of buyers’ satisfac-tion in the relationship between salesperson acculturationand buyers’ commitment.

Method

Sample for this study was drawn from the populationresiding in Auckland, the most cosmopolitan and the mostpopulated area of New Zealand (NZ). The survey wascarried out under the personal supervision of the one of theauthor. A total of 556 responses were collected, of which178 had to be dropped due to incomplete information fora few questions. Finally, a usable sample of 378 wasobtained. The respondents consisted of 50.6 percentfemales and 49.4 percent males. Seventy-six percent ofrespondents have been living in NZ for more than fiveyears and 46 percent of respondents have been bankingwith NZ banks for more than five years.

The research instrument contained items related to allthe variables and few demographic characteristics. Fourdimensions of acculturation were measured using scalepresented by Barry (2001). Items for measuring buyers’satisfaction with the service firms were drawn from thestudies of Levesque and McDougall (1996) and Macintoshand Lockshin (1997). Buyers’ satisfaction with servicesalespersons was measured using items developed byRamsey and Sohi (1997) and Cohen, Gan, Yong, andChoong (2006).

Major Research Findings

As hypothesized, this study shows that various accul-turation dimensions influence buyers’ satisfaction differ-ently. While assimilation and integration were found to bepositively associated with buyers’ commitment, separa-tion was found to be negatively associated with buyers’commitment. The relationship between marginalizationand buyers’ commitment had no significant effect onbuyers’ commitment. Our study also found that buyers’satisfaction mediates the relationships between assimila-tion, integration and marginalization and buyers’ commit-ment. However, buyers’ satisfaction was found to have nomediating effect on the relationship between separationand buyers ‘commitment.

Overall, our findings have important implications forpractitioners. Our study shows that salespersons shouldexercise assimilation and integration acculturation be-


haviors in order for them to improve intercultural relation-ship and to facilitate buyers’ tendency to maintain theiroriginal consumption patterns in relation to their own

native culture in New Zealand. These acculturation be-haviors then will help salespersons to improve buyers’commitment. References are available upon request.

For further information contact:Sanjaya S. Gaur

AUT School of BusinessAUT Faculty of Business and Law

Auckland University of Technology (AUT)Wellesley Campus, #304, WU Building

46 Wakefield StreetPrivate bag 92006

Auckland 1142New Zealand

Phone: 0064.9.9219999, Ext. 5465Fax: 0064.9.9219629



SOLUTION SELLING TEAMS: A MULTI-PERSPECTIVE REVIEW OFTHE IMPACT OF CROSS-FUNCTIONALITY

Doreen Wienhold, Freiberg University of Technology, GermanyMichael Nippa, Freiberg University of Technology, Germany

SUMMARY

The solution selling literature demonstrates anincreased use of cross-functional solution selling teams(Sharma, Iyer, and Evanschitzky 2008; Johansson,Krishnamurthy, and Schlissberg 2003, p. 121). Forexample, Storbacka et al. (2009) reveal that while sale ischanging from an isolated activity to a cross-functionalone, the case study companies create cross-functionalteams or business units to sell solutions. Cross-function-ality reflects the number of different functions that arerepresented within a team (Gebert, Boerner, and Kearney2006, p. 432). It causes differences in terms of members’information and knowledge bases as well as their experi-ences. It is argued that this kind of diversity leads toimproved problem solving and qualitative better decisionmaking (Williams and O’Reilly 1998), as well as a higherlevel of creativity (Amabile 1997). However, functionallydiverse team composition may also lead to problems(Keller 2001; Horwitz and Horwitz 2007). For example,different goals, mind-sets, and terminology may causeconflict and misunderstanding among team members,which consequently erodes team performance. Surpris-ingly, research in the field of solution selling seeminglyabstains from investigating the effects of cross-functionalselling team composition on selling team performance.This paper addresses this research gap by answering thefollowing research question: How does cross-functionalselling team composition influence solution selling teamperformance and which moderators influence that rela-tionship?

To answer this question, we reviewed firstly theliterature dealing with cross-functionality in small groupdiversity research. To summarize the findings, positiverelationships between functional diversity and outcomedominate negative – or no – relationships (Bell et al. 2011;Joshi and Roh 2009; Horwitz and Horwitz 2007). How-ever, findings are mixed and positive relationships areoften weak, which is why empirical studies increasinglyaccount for moderating effects of the context. Team typeand task characteristics are the most studied moderators.Cross-functional (solution) selling teams areunderrepresented in the empirical studies and so quitenothing is known about that specific team type.

Secondly, we reviewed three theories which are oftencited in small group research and provide the most

insights into team diversity variables and their potentialeffects on team performance. The Attraction Paradigmstates that people are attracted to one another if they aresimilar in their values and attitudes. Furthermore, there isa preference for working with similar people (Byrne1997). Assuming that members of a functional depart-ment are similar to one another, functional diversity islikely to negatively influence affective reactions such assatisfaction, commitment and cohesion (Jackson, Joshi,and Erhardt 2003, p. 809). Another common theoreticalfoundation for research on diversity is the Social IdentityTheory, developed principally by Henri Tajfel and JohnTurner. People tend to classify themselves and others insocial groups, e.g., according to their functional affilia-tion. They strive for a positive social identity, defined as“that part of an individual’s self concept which derivesfrom his knowledge of his membership of a social grouptogether with the value and emotional significanceattached to that membership” (Tajfel 1981, p. 255). Toachieve this positive social identity, people compare theirown group (e.g., marketing department) with relevantout-groups (other functional departments) and seek posi-tive differences among these groups. This ambition oftenresults in rivalry. Assuming that employees identify them-selves with their functional department and view otherfunctional departments as relevant out-groups, conflictsbetween functions might result and cross-functional inte-gration and cooperation within cross-functional teamsmight be hampered (Ashforth and Mael 1989, p. 31). TheCognitive Diversity Hypothesis offers an explanation forthe positive impact of functional diversity. Cognitivediversity is defined in terms of differences in expertise,experiences, perspectives, preferences, and knowledgeand beliefs held by the team members (Horwitz andHorwitz 2007). In particular, task-oriented diversityattributes such as functional background are associatedwith cognitive diversity (Bell et al. 2011; Jackson, Joshi,and Erhardt 2003, p. 804). In this sense, functional diver-sity might be especially beneficial because the functionalbackground offers task-related information and knowl-edge. Different perspectives and experience are assumedto be positively related to problem solving, innovation,and creativity (Horwitz and Horwitz 2007; Amabile 1997).

In summary, according to attraction paradigm andsocial identity theory, cross-functionality may negativelyinfluence team performance, if the team members per-ceive a greater affiliation and similarity with their func-


tional department than with the team. According to thediversity hypothesis, functional diversity has positiveeffects on the outcome if effective team performancerequires creativity, different ideas and knowledge as wellas various experiences. Thus, it depends on the team‘stask and the identification of the team members howfunctional diversity influence team performance. In thepaper, we assume that the solutions characteristics influ-ence the selling task. Furthermore, solution selling’s taskcharacteristics influence the positive effects emanatingfrom cross-functionality. Complex, interdependent, anduncertain tasks mostly profit from functionally diverseinformation, knowledge, and expertise. In this analysiswe came to the conclusion that solutions consisting of ahigh amount of different but interdependent and custom-ized parts, especially, lead to complexity, interdepen-dence and uncertainty in selling. It is proposed that:

P1: The solution’s complexity influences the complexityof the solution selling task. Selling task complexitymoderates the relationship between functional diver-sity and selling team’s performance in such a way thatas the task becomes more complex the influence ofcross-functionality becomes more strongly positive.

P2: The solution’s level of integration influences the taskinterdependence of solution selling. Task interde-pendence moderates the relationship between func-tional diversity and selling team’s performance insuch a way that as the task becomes more interdepen-dent the positive effect of cross-functionalitybecomes more strongly, but the negative effect

becomes more weakly only if goal and reward inter-dependence is also high.

P3: The higher the level of customization, the higher theamount of service, the longer the solution sellingproject, and the newer the customer’s problem, thehigher is the task uncertainty of solution selling.Selling task uncertainty moderates the relationshipbetween functional diversity and selling team’s per-formance in such a way that as the task becomes moreuncertain the influence of cross-functionalitybecomes more strongly positive.

P4: Team identification moderates the relationshipbetween functional diversity and selling team’s per-formance in such a way that as identification with theselling team is high the negative influence of cross-functionality becomes weaker.

For further research in the solution selling businessthe following is needed: Firstly, empirical research isneeded to find evidence for our propositions that solutionselling is a complex, independent, and uncertain task. Thiswould show that cross-functionality is beneficial for solu-tion selling due to the functional specific information andknowledge of the selling team members. Secondly, it isnecessary to find out if solution selling team membersidentify with the team or their functional background andwhich factors influence this identification. Such insightswould lead to important managerial implications abouthow to minimize the negative effects of functional diver-sity in solution selling teams. References are availableupon request.

For further information contact:Doreen Wienhold

Management, Leadership and Human ResourcesBusiness Administration

Technische Universität Bergakademie FreibergLessingstrasse 4509599 Freiberg




THE STRATEGIC ALIGNMENT OF ORGANIZATIONALDEVELOPMENT INTERVENTIONS FOR SALESPERSON

VALUE MANAGEMENT WITH A SALESPERSONLIFECYCLE MANAGEMENT MODEL

Joon-Hee Oh, Georgia State University, Atlanta

SUMMARY

Introduction

As of 2011, more than 11 percent of the U.S.workforce, or 15.08 million individuals, are engagedwithin sales and related occupations (U.S. Bureau ofLabor Statistics 2011). Their average annual turnover ishigh, at 30~40 percent, and their average tenure is onlynine months (Covert 2010). Managing such mobile sales-persons is a challenging task for sales organizations andrequires sound strategies to align their workforce to orga-nizational imperatives. However, as claimed by Attia,Honeycutt, and Leach (2005), a lack of understanding onthe part of organizations of how to measure and evaluatetheir salesperson development interventions has discour-aged their efforts, and often their training objectives andsalesperson development efforts are not aligned with theirorganizational goals. Therefore, a strategic alignment oforganizational development interventions for sales forcedevelopment is a captivating research subject that invitesthe attention of academic as well as practitioners.

Research Background

Earlier studies have discussed on such individualfactors as motivation, retention, and productivity in rela-tion to effective and productive sales force management.This study, in contrast, focuses on organizational factors,such as organizational development interventions, whichare critical in developing and managing salespeople buthave not gained significant research attention (Zoltners,Sinha, and Lorimer 2006). This study, in addition, arguesthat organizations need to continuously enhance theirmethods of managing salespersons because, as salesper-sons evolve, their needs and demands differ in stages(Cron and Slocum 1986), making the alignment of peoplestrategies to organizational imperatives challenging. Inresponse to the research gap and to provide a heuristicalternative to the identified challenges, this study, lever-aging experiential learning theory and the life cycle model,presents a salesperson lifecycle management (SLM) modelfor providing sales organizations with a practical solutionfor a strategic alignment of their sales force developmentinterventions to their organizational imperatives.

SLM Model

The SLM model is conceptually based on the lifecycle model, which depicts a process of change in anentity as a progression through stages or phases (see Vande Ven and Poole 1995). This study makes a conceptualextension in line with the life cycle model to claim thatsalespersons progress through stages while developingvalue perception within an organization, which changesthroughout their professional career. Salesperson value(i.e., salespersons’ net contributions to firms) thus needsto be identified and further measured to determine thestrategic implications for organizational development in-tervention decisions and a validation of the interventionsmade.

The SLM model also leverages experiential learningtheory, which identifies experience as a critical elementfor learning and knowledge development (Kolb 1984). Inthis study, another extension is made from the experientiallearning theory in explaining the knowledge constructionprocess: Salespersons construct knowledge via an experi-ence transfer process. Based on this extension, this studyclaims that salespersons’ experience is not a simple aggre-gation of experiences that they encounter during theirprofessional careers but the result of certain stimuli (e.g.,[advanced] training and developmental programs) trig-gering either the replication or the transformation ofearlier stored experiences for selectively constructingknowledge through a transfer process, which is thenfurther developed into competency or competitive advan-tage (Grant 1996).

The SLM model is a three-stage model, reflecting thechanges in value perception and the knowledge construc-tion process within the stages, and is a function of organi-zational development investment (ODI) and salespersoncontribution (SC, e.g., revenue generated from new salesclosings). This distinction between the stages will lead tothe provision of different organizational developmentinterventions for salespeople at specific stages. In addi-tion, the stage-specific interventions will ensure a betteralignment of organizational strategies for effective andproductive sales force development and management.References are available upon request.


For further information contact:Joon-Hee Oh

Marketing DepartmentJ. Mack Robinson College of Business

Georgia State University35 Broad St. NW

Atlanta, GA 30303Phone: 765.404.9528



DEMYSTIFYING NETWORK STRATEGIES: INCREASING PRODUCTQUALITY, CUSTOMER SATISFACTION, AND PROFITABILITY

THROUGH THE STRATEGIC DEPLOYMENTOF INFLUENTIAL HUBS

Cinthia B. Satornino, Florida State University, TallahasseeMichael K. Brady, Florida State University, TallahasseeMichael Brusco, Florida State University, Tallahassee

Clay Voorhees, Michigan State University, East Lansing

SUMMARY

The deployment of work teams is a popular andeffective way to manage complex and intensive projects.Team effectiveness has been shown to have significantconsequences for both service and physical goods provid-ers that range from, on the negative side, disastrous airlineaccidents, to, on the positive side, sales growth of 864percent and net income growth of 438 percent. It istherefore obvious why firms would want to understandhow to configure teams in a way that maximizes teamperformance.

The decision to use teams and how to configure themto maximize performance is especially critical for servicefirms and sales organizations, because deliverables areoften an intangible solution to a complex problem. Forexample, in a consulting firm, projects are often knowl-edge intensive and complex. Higher complexity andintangibility lead to ambiguity in recognizing and suc-cessfully implementing an optimal solution to a task orproblem.

Despite the numerous studies that have tackled thequestion of team design, companies still struggle to con-figure teams in a way that maximizes performance. Thepresent study suggests that the key to unraveling theproblem may be approaching it from a network perspec-tive. One provocative insight that has come out of thenetwork perspective is the existence of highly connectedindividuals, or influential hubs, that link global networks.These hubs play a powerful role in the flow of informationand resources across a network. The role of hubs has beenof considerable interest to academics in many disciplines,

from communications to medicine to theoretical physics.Because of their powerful role, leveraging these influen-tial hubs is an intuitive and appealing team design strat-egy.

Given that leveraging hubs shows potential toimprove team performance and effective team perfor-mance can be a critical indicator of both product qualityand firm performance, further exploration seems merited.To that end, the present study tackles the two obstacles todeveloping and implementing strategies based on influen-tial hubs: finding influentials and leveraging them appro-priately in teams. By doing so, the study clears the way formanagers to use information about the informal networksin their organization to improve product quality by con-figuring more effective teams.

The authors use diversity theory and the social capitalmetaphor as theoretical foundations for the exploration ofhub-based strategies. Furthermore, we employ multi-level network analysis to identify team characteristics andlocate influential hubs. Finally, the interactions betweenvarious team types and influential hubs in the organiza-tion are examined. Cohesive teams that also include aninfluential hub are hypothesized to be most effective increating a higher quality product; in turn, higher productquality leads to increased customer satisfaction. Cus-tomer satisfaction is then linked to the profitability of theproject. Results provide new insights about the impor-tance of strategically crafting effective teams using inter-nal network structures, and the subsequent effects onproduct quality, customer satisfaction, and ultimately,project profitability. References are available uponrequest.


For further information contact:Cinthia B. Satornino

Florida State UniversityRovetta Business Annex

821 Academic Way, P.O. Box 3061110Tallahassee, FL 32306–1110

Phone: 850.644.3090Fax: 850.644.0915



A NUANCED VIEW OF THE MARKETING-SALES “ACTIVITY”INTERFACE: A CASE OF SMALL B2B FIRMS

Avinash Malshe, University of St. Thomas, MinneapolisWim G. Biemans, University of Groningen, The Netherlands

SUMMARY

The extant literature on market orientation capturesthe central idea that, in order to create superior customervalue, firms need alignment between all parts of theorganization. Marketing and sales should be well-equippedfor effective cooperation: they both serve customers, withmarketing supporting the sales function and building aconsistent brand image, while sales perform tactical taskssuch as contacting customers, executing marketing strat-egies and closing the sale in the field. Unfortunately, it hasbeen shown that the marketing-sales interface is notalways harmonious and constructive. Researchers haveattributed the frictions, animosity and lack of mutualrespect between these two departments to various factors,including goal differences, different perspectives towardthe business environment, cultural and thought worlddifferences, lack of interfunctional integration, physicalseparation and poor communication and poor coordina-tion during planning and goal setting.

Recently, there has been a surge of interest inresearching the marketing-sales interface. Initially, thesestudies focused on the cultural differences between mar-keting and sales, their antecedents and their results interms of interfunctional conflict and organizational effec-tiveness, but later studies elaborated on these differencesbetween marketing and sales and identified taxonomiesdescribing several marketing-sales configurations thatare found in organizations. Unfortunately, the extantliterature is strongly biased toward firms with separatemarketing and sales departments, which are not alwayspresent in B2B firms; especially small B2B firms oftenlack a dedicated marketing department. But that does notmean that these firms fail to “do marketing.” Rather thanhaving a separate marketing department, marketingactivities are typically carried out by employees in severalother departments such as Sales, Business Development,Applications, Customer Service, and Technical SalesSupport. Recent studies into the organization of the mar-keting function emphasize that traditional marketing ac-tivities, such as advertising and pricing, are increasinglyallocated to different functional groups inside the firm andthis dispersion of marketing activities is found to increasefirm performance. One of the primary functions involvedin market-related decisions is sales.

For our study we collected interview data from bothU.S. and Dutch small B2B firms. Firm size ranged

between 20-60 employees and their annual sales were inthe range of $5–35 million. Our firms came from variedindustries such as pharmaceuticals, IT, telecom, industrialequipment, capital equipment, and industrial raw materi-als. Our questions focused on the role of marketing in theinformants’ firms, the key individuals involved in market-ing and sales activities, their backgrounds, interactionpatterns and day-to-day functioning. Based on our find-ings, we argue that marketing-sales interface researchstands to gain from adopting an activity-centric lens andexamining the interplay between firm’s marketing andsales philosophies and activities (in addition to depart-ments) that are espoused/performed interchangeably bythe same group of people who may or may not have formalmarketing training. Our findings are summarized below.

Dispersed Marketing Activities Despite Absent Mar-keting Department. Many informants remarked that whiletheir firms recognized the importance of doing marketing,they lacked a dedicated marketing department. Neverthe-less, these firms were constantly engaged in performingmarketing activities such as customer service, productcustomization, or handling customer complaints. Suchactivities were being performed by organizational mem-bers as part of or in addition to their “regular job” irrespec-tive of their background or job function.

First Forays into Performing Marketing Activities.Two key drivers prompted individuals to engage in extra-mural marketing activities: (a) top management predispo-sition to marketing and (b) presence of marketing ambas-sadors and employee affinity.

Interface Between Marketing and Sales Activities.Employees often struggled to maintain a delicate balancebetween performing marketing activities and insuringthat they were also keeping a track of their everyday salesfigures. Long-term marketing planning was occasionallyhijacked by short-term sales pressures, resulting in aperpetual activity-orientation tug of war. Marketing islargely operationalized as tactical marketing communica-tions and interpersonal relationships function as facilita-tors.

Transition to Marketing-Sales Departmental Inter-face. Several executives were mulling over the idea ofcreating a separate marketing department in the nearfuture; e.g., because of the firm’s growth and increasingcustomer base.


Our study highlights the fact that small B2B firmsexhibit dispersed marketing activities despite the absenceof a formal marketing department. In addition, thesemarketing activities are not a prerogative (or responsibil-ity) of a single department or individual in such firms. Ourfindings also show how the marketing-sales activity inter-face is characterized by a constant tug of war between

short-term sales activities and long-term marketing orien-tation philosophy. In short, we argue that the extantliterature on the marketing-sales interface may benefitfrom using an activity-centric lens in addition to theconventional department-centric lens used in traditionalstudies of the marketing-sales interface. References areavailable upon request.

For further information contact:Avinash Malshe

Opus College of BusinessUniversity of St. Thomas

Mail #TMH4431000 LaSalle Ave

Minneapolis, MN 55403Phone: 651.962.4287



INVESTMENTS IN CUSTOMER RELATIONSHIPS AND RELATIONSHIPSTRENGTH: EVIDENCE FROM INSURANCE INDUSTRY IN CHINA

Guicheng Shi, Macao University of Science and Technology, MacaoYuan Ping, Macao University of Science and Technology, Macao

Yonggui Wang, University of International Business and Economics, ChinaMatthew T Liu, University of Macau, Macao

SUMMARY

Relationship investments have been shown to gener-ate strong customer relationships that enhance customerloyalty (De Wulf, Odekerken, and Robert 2001 Liang,Chen, and Wang 2008; Yoon, Choi, and Sohn 2008). AsYoon, Choi, and Sohn (2008) pointed out that the bestpoint of their models is to use a global construct, relation-ship quality, by a combination of trust, commitment, andsatisfaction, which offers the best assessment of relation-ship strength. However, relationship quality as a constructhas its intrinsic weaknesses. It has no clear definition itselfand different researchers may include different dimen-sions for relationship quality. For example, Crosby, Evans,and Cowles (1990) and Liang, Chen, and Wang (2008)consider relationship satisfaction and trust to be thedimensions of relationship quality, De Wulf, Odekerken,and Robert (2001) and Yoon, Choi, and Sohn (2008) addrelationship commitment besides satisfaction and trust,while Palmatier (2008) includes commitment, trust, reci-procity norms, and exchange efficiency as the indicatorsof relationship quality. Another weakness of the existingstudies of customer relationship investments is the com-mon method bias caused by the fact that all the variablesare measured from the same source in terms of customer’sself report.

With the aforementioned existing gaps in mind, thispaper’s objectives are threefold. First, we want to use amore appropriate construct named relationship strengthas outcome variable instead of relationship quality todetermine whether relationship investments have differ-ential impacts on different dimensions of relationshipstrength. Second, we want to provide empirical evidence

for the impact of relationship investment efforts on per-ceived relationship investment and ultimately on relation-ship strength with much less common method variancethrough dyadic data, one from the seller and the other fromthe buyer. Third, we want to analyze whether the effect ofperceived relationship investment on relationship strengthis contingent on consumer characteristics.

The agent-client relationship in the Chinese lifeinsurance industry was chosen as the research context forthe empirical testing. To avoid common method variance,we have collected data from both agents and clientsalthough it was difficult to do so. A random sample wasgenerated from the database of a major Chinese lifeinsurance company. In the survey, two versions of thequestionnaire were used: an agent questionnaire and aclient version. Both questionnaires contained same items,but presented from different perspectives. In total, 450agent questionnaires and 450 client questionnaires weredistributed. After deleting uncompleted and unmatchedquestionnaires, a total of 354 sets of agent-client question-naires remained and constituted the sample for the mainstudy.

The empirical results indicate that clients value finan-cial effort most followed by social effort and structuraleffort; Perceived relationship investment influencesaffective strength most followed by cognitive strengthand conative strength; Both customer innovativeness andcomplaint propensity acts as moderators of the effective-ness of perceived relationship investment on two of thethree dimensions of relationship strength. References areavailable upon request.

For further information contact:Guicheng Shi

Macao University of Science and TechnologyA413, Ave. Wai Long

Taipa, MacaoPhone: 00853.88972362

Fax: 00853.28880022E-Mail: [email protected]


SALESPERSON’S PERSONALITY AND RELATIONSHIP QUALITY:ARE YOU A FRIEND OR A CUSTOMER?

Kaveh Peighambari, Lulea University of Technology, SwedenSetayesh Sattari, Lulea University of Technology, SwedenMaria Ek Styvén, Lulea University of Technology, SwedenLars Bäckström, Lulea University of Technology, Sweden

ABSTRACT

This study investigates the quality of buyer-sellerrelationships from the seller’s perspective, by addressingthe influence of salespeople’s personality traits on thequality of the relationships with customers as compared tofriends. Results indicate that the personality traits ofsalespeople influence both of these relationships, but indifferent ways.

INTRODUCTION

The past two decades have seen a resurgence ofinterest among academics and practitioners in the de-velopment and maintenance of long-term relationshipsbetween buyers and sellers (Foster and Cadogan 2000;Grönroos 1994; Parsons 2002). Since then, relationshipquality has become a key subject of research. Whileresearchers have begun to investigate the determinants ofthe success or failure in relationships between exchangepartners by looking at both seller characteristics and at thequality of interactions between buyers and sellers (Crosbyet al. 1990; Morgan and Hunt 1994; Parsons 2002), thesefactors have not been considered jointly. Hence, the roleof salespeople in service firms, mainly those of a long-term relational nature, has received limited attention.

Salespeople involved in relationship marketing areoften “relationship managers.” It is, in part, the quality ofthe relationship between the salesperson and the customerthat determines the probability of continued interchange(Crosby et al. 1990). Customer-oriented behavior bysalespersons helps create good-quality buyer-seller rela-tionships (Williams and Attaway 1996). Therefore, dis-patching the right salespersons to manage speci c businessrelationships is critical for businesses.

Furthermore, individual personality traits have beenimportant components of buyer-seller relationships (Dionet al. 1995). As the main in uence on individual behavior,personality as the complex set of unique psychologicalqualities, influences an individual’s characteristic pat-terns of behavior across situations and over time (Morris1996; Zimbardo and Weber 1994). Since buyer-sellerrelationships often require frequent negotiations on price,

product speci cations, quantity, delivery and other terms,this interactive process may require cooperation amongconflicting interests and needs. Thus, the personality ofthe salesperson affects his/her behavior in managing therelationship, thus affecting a customer’s perceptions ofthe quality of service provided by salespersons. Theliterature on personality shows that purchasing behavioris linked to individual personality traits and that similarpersonality traits between buyers and sellers will result ina better and longer-term relationship (Barrick and Mount1991; Dion et al. 1995; Lamont 1977). The first objectiveof this research is therefore to investigate the quality ofbuyer-seller relationships from the seller’s perspectiveand to address the question – what personality traits of asales person influence the quality of buyer-seller relation-ships?

Moreover, from a personal selling perspective, theliterature suggests that good customer relationships, inwhich salespeople and their clients interact frequently andover an extended period, during which they exchangeinformation, develop into close friendships (e.g., Grayson2007; Price and Arnould 1999; Swan et al. 2001). Friend-ships between clients and salespeople are frequently un-questioned phenomena, following the thought that“[o]verall, sheer frequency of interaction and outcomedependency, whatever the setting, promotes friendship”(Price and Arnould 1999, p. 51). Salespeople are oftenencouraged to treat customers like their best friends. Whatif this exhortation rests on a false premise, and customersare not the same as friends?

Accordingly, the second objective of this paper is tocompare the impact of a sample of salespeople’s person-ality traits on the quality of his or her relationships withfriends and customers. The intention is to identify how theinfluence of salespeople’s personality traits on the buyer-seller relationship quality differs between the seller’scustomers and his or her friends. While one would natu-rally expect there to be a difference, it is hoped that thestudy will identify the similarities and the differencesbetween friends and customers. Where there are similari-ties, it follows that salespeople can treat good customersas friends. Where differences exist, salespeople mightwant to be more careful about the simple incorporation ofthe friendship metaphor into their selling strategies.


CONCEPTUAL BACKGROUND ANDHYPOTHESES

Customers, Friendship, and Relationship Quality

Tansuhaj et al. (1988) suggested that firms shouldgive greater consideration to the “employee-customerinteraction” nature of service. Relationship marketingrests on the premise that this interaction between sales-people and their customers has more in common withother human relationships than mere exchanges or trans-actions (cf., Gummesson 1998). Hence, there is someemotional or other connection between the parties in arelationship in the context of service firms, not just asimple commercial exchange.

The categorization of customers as friends then givesinsights into how to manage these relationships to thefirm’s advantage (e.g., Swan et al. 2001). Salespeople andother employees are often encouraged to treat customerslike their best friends (Geller 2006) and to engage infriendships instead of transactions (Tan and Steinberg2007). According to Wang, Liang, and Wu (2006),research on the effectiveness of relationship bondingtactics works is scarce. Wulf et al. (2001) argue thatrelationship-bonding tactics would improve relationshipquality. Thus, a thorough understanding of the character-istics of an interpersonal relationship is essential for theimplementation of relationship marketing programs in asalesperson-customer setting.

Mustillo et al. (2005) examined the factor structureand psychometric properties of the Trusting RelationshipQuestionnaire (TRQ). As cited by Mustillo et al. (2005),Vance and Sanchez (1997) state that the TRQ was devel-oped by clinicians in North Carolina in an attempt toevaluate the relationships between professionals and para-professionals and youth in community-based treatmentself-measurement scale programs. The resulting ques-tionnaire was designed to assess the quality of the adult-child relationship from both the children’s and the adults’perspective (Mustillo et al. 2005).

Hence, the TRQ appears to capture relationship qual-ity (Mustillo et al. 2005). Somewhat surprisingly, theseauthors do not describe how relationship quality isoperationalized; however, in view of their construction ofthe scale, relationship quality can be defined (and there-fore operationalized) as follows: “Relationship qualityhas to do with the perceptions of the two parties to arelationship regarding the value of the information theyshare, the time they spend, the perspectives they share,and their mutual considerations.”

Personality Traits

Personality, a psychological (hypothetical) construct,has been defined as “a stable set of responses that indi-viduals have to their environments” (Dion et al. 1995;Kassarjian 1971; Lamont 1977). Although there are sev-eral approaches to the study of personality, one of the mostcommonly employed is trait theory (Pederson et al. 1988).Traits, according to this theory, are identifiable character-istics that define a person (Dion et al. 1995; Solomon1992). Examples of such characteristics are whether ornot a person is an extrovert or an introvert and whether heor she primarily uses thinking or emotions in problemsolving.

Studies employing trait theory to explore salesper-sons’ personalities have suggested that a salesperson’sperformance may be a function of his or her personalitytraits, meaning that “sales success would be a result of thedegree to which she or he possessed certain specificpersonality traits” (Dion et al. 1995). Accordingly, Barrickand Mount (1991) found that different personalitydimensions may predict variations in job performanceamong different occupations. There is a widely acceptedfive-factor personality classification, the Big Five, in thepersonality literature (Barrick and Mount 1991, 1993;Digman 1990). The Big Five framework suggests thatmost individual differences in human personality can beclassified in terms of five domains (Gosling et al. 2003;John and Srivastava 1999; McCrae et al. 1998). It has beenused in investigations with different theoretical frame-works, with a variety of instruments and samples (includ-ing samples from diverse cultures), and with ratingsobtained from several sources (e.g., Barrick and Mount1993; Barrick et al. 2002; Digman 1990; McCrae andCosta 1985; Norman 1963).

The Big Five factors have distinctive meanings.Extraversion is characterized by being sociable, gre-garious, assertive, talkative, and active (McCrae andCosta 1985; Norman 1963). Agreeableness is defined byflexible, trusting, good-natured, cooperative, forgiving,soft-hearted, and tolerant (Guilford and Zimmerman 1949;McCrae and Costa 1985). Conscientiousness is describedby trustworthiness, as well as being careful, thorough,responsible, organized, resourceful, hardworking, achieve-ment-oriented and persistent (Norman 1963). Opennessto experience is characterized as imaginative, cultured,curious, original, broad-minded, intelligent, and artisti-cally sensitive (Digman 1990; Norman 1963). Finally,Emotional (in)stability (viewed from the negative pole) ischaracterized by being tense, insecure, depressed, angry,embarrassed, worried, anxious and nervous (Barrick and


Mount 1991). These dimensions form a hierarchical modelof personality traits using a self-measured scale whichrepresents personality at the broadest level of abstraction.Each bipolar factor encompasses several other specific,which subsume a large number of even more specifictraits (Gosling et al. 2003).

Salespeople’s personality traits have been consid-ered as important components in buyer-seller relation-ships. This finding has been based on the followingassumptions: (1) certain personality traits have been seenas more suited to a sales process than to others (Barrickand Mount 1991; Dion et al. 1995; Lamont and Lundstrom1977); and, (2) similarities in buyer/seller personalitieslead to positive outcomes in sales performance (Barrickand Mount 1991; Crosby et al. 1990; Dion et al. 1995;Gosling et al. 2003). Our study, therefore, included mea-sures of Big Five personality domains that were used toexamine the impact of personality on the quality ofsalespeople’s relationships with their customers andfriends. To assess this impact and address the objectivesstated in the Introduction, the following hypotheses areformulated:

H1: A salesperson’s personality traits influence the qual-

ity of the relationship she/he has with customers andalso friends.

H2: There is a difference between how a salesperson’s

personality traits impacts his/her relationship withcustomers compared to how these traits affect his/herrelationship with friends.

METHODOLOGY

The Trusting Relationship Questionnaire, or TRQ(Mustillo et al. 2005), was used to measure the perceivedrelationship quality between salespersons and customers.In the original study, both the adult and child versions ofthe questionnaire were found to be reliable and valid.Since this study focused on the seller’s perspective, weused the adult version of the TRQ, using salespeople asrespondents. To more accurately reflect the relationshipthat a salesperson might have with a friend and a customer,the word “child” was replaced with “friend” and “cus-tomer.” For the same reason, we also changed “times ofcrisis” to “times of need,” and added, “when things gowrong” to the two items that concerned apologizing. Asrecommended by Mustillo et al. (2005), four of the itemsin the original scale were excluded due to poor validity.The final scale employed for Relationship Quality thusconsisted of 12 items, scored on a 5-point Likert-typescale anchored by 1 (“Never”), through 5 (“Always”),indicating how well each item characterizes the relation-ship (see Appendix A). The questionnaire included twosets of the Relationship Quality questions; first consider-ing a friend and then considering a customer.

To make the items easy for the respondents to under-stand, friend was conceptualized by asking the respon-dents to “think of a person with whom you have a closeand positive personal friendship. This could be a spouseor partner, a close family member, or a good friend.”

In order to measure salesperson’s personality, the BigFive scale was used, including the personality domains ofExtraversion, Agreeableness, Conscientiousness, Emo-tional stability, and Openness to experience. We used theshort version of the Big Five self-measurement scaledeveloped by Gosling et al. (2003).

Data for hypothesis testing were collected through asurvey. As the unit of analysis was salespeople’s perspec-tives, we administered the questionnaire to the salesforceof a large Swedish company in the financial servicesindustry. We used judgement sampling as this samplefulfilled some important criteria specified by the research-ers (cf., Hair et al. 2010). That is, the firm had employedlarge enough sales force (198) for a sufficient number ofresponses to be gathered; its products involved high-contact personal selling; and the management cooperatedwith the researchers to facilitate access to the sample. Thequestionnaires were distributed to all the salespeople inthe company in the internal mail, accompanied by a letterfrom management explaining the purpose of the study,that participation would be anonymous, and that it wasimportant to answer all the items in the questionnaire.Usable responses were received from 119 salespeople, foran overall response rate of 60.1 percent.

RESULTS AND DISCUSSION

To identify the dimensions of the Relationship Qual-ity construct in this particular context, an exploratoryfactor analysis with Varimax rotation was used on theTRQ items. Four factors were obtained by applying thelatent root criterion of retaining factors with eigenvalues,sum of squared correlations of each factor with its vari-ables, greater than 1, implying than each can be consid-ered as an independent factor (Hair et al. 2010). Factor 1has been labeled “Sharing” and Factor 2 was called“Caring.” Factor 3 concerns whether each party to therelationship considered the other’s point of view; there-fore, it was labeled “Point of View.” Finally, Factor 4 waslabeled “Time Enjoyment” since it concerns how muchsalespeople enjoy spending time with their customers/friends. All items had factor loadings in excess of .60 andeach converged on its separate factors. The factors repre-sented 70.1 percent of the variance of all the analyzedvariables, which is higher than the suggested limit of 60percent (Hair et al. 2010).

To test the internal consistency of the Big Fivepersonality domains, Cronbach’s alpha reliabilities wereestimated. Extraversion had an alpha of .82, Agreeable-


ness .76, Conscientiousness .81, Emotional stability .83,and Openness to experience .80. Hence, values for allconstructs are above the suggested threshold of .70, indi-cating that the Personality scale is reliable.

In order to test the first hypothesis, a regressionanalysis was conducted on the relationship between Rela-tionship Quality and the Big Five Personality Traits. TheR-squares and t-values suggested that all salespeople’spersonality domains, except Openness to experience,influence the quality of their relationships with bothcustomers and friends. H

1 was therefore supported.

Moreover, to test the second hypothesis, regressionanalyses were done separately in order to determine thedifferences and similarities in the ways in whichsalespeople’s personality influenced their relationshipquality with friends and customers. In total, eight regres-sion analyses were performed; i.e., one regression foreach of the four Relationship Quality dimensions, first forfriends and then for customers. Tables including results ofall regressions are provided in Appendix B. Table 1summarizes the results by showing the R2 for the relation-ship among each of the four factors of relationship qualityand the Big Five personality domains.

The observations to be made from the regressionresults can be summarized as follows: A salesperson’spersonality has a partially greater impact on the quality ofrelationship with friends than with customers. In otherwords, the more extraverted, agreeable, conscientious,and emotionally stable a salesperson is, the better therelationship he or she will have with customers andfriends, albeit in a different way. When we consider thefour items that explain how parties contribute to a rela-tionship, it is evident that from the salesperson’s perspec-tive, friends are not the same as customers.

Data provide support for the second hypothesis, asthere is a difference in how the dimensions of Relation-ship Quality are influenced by the personality domains ofsalespeople. Three out of four Relationship Qualitydimensions; i.e., Sharing, Caring, and Time Enjoyment(see Table 1), were more affected by the salesperson’spersonality traits when the relationship concerned friends,than when it concerned customers. Primarily, the relation-ship with friends is influenced by the salesperson’sdegree of extraversion, conscientiousness, and emo-tional stability.

However, for the Point of View dimension, whichpertains to whether one party in the relationship considersthe other’s point of view (see Appendix A), thesalesperson’s personality seems to have stronger influ-ence on the quality of the relationship with customers thanon the relationship with friends. In the field of personalselling, where salespeople are encouraged to treat cus-tomers like their best friends (Geller 2006) and where trustis an important construct of relationship quality (Wanget al. 2006; Wulf et al. 2001), it would be problematic ifsalespeople considered the viewpoints of friends, but notof customers. Therefore, it is very interesting that thepersonality of a salesperson had a significant effect on theextent to which he or she considers the customer’s pointof view, but no significant influence on the Point of Viewdimension when it comes to friends.

Based on the R-squares, one interesting interpreta-tion pertains to Time Enjoyment. As shown in Table 1, thelevel of influence of personality traits in a salesperson onthe time she/he enjoys to spend is much more with friendsthan customers.

Furthermore, as the regression results in Appendix Bshow, the only personality trait that affects the relation-

TABLE 1R2 and Significance Levels

Dimension of Relationship Quality R2 for the Effects of Personality Traits on Relationship Quality

Friend Customer

Sharing .13** .08

Caring .21** .10*

Point of View .04 .18**

Time Enjoyment .43** .07

** p < .01 * p < .05


ship quality with customers is Agreeableness, which,however, does not influence relationships with friends.On other words, the more sympathetic and warm and theless critical and difficult the salesperson is, the better therelationship he or she has with customers. With friends,however, other personality traits than being agreeable aremore important when it comes to the quality of therelationship in terms of sharing time and information,caring about the relationship, and enjoying spending timetogether.

LIMITATIONS AND FURTHER RESEARCH

Like all research, this study has several limitations.The first limitation is the psychometric cost of using shortmeasures as we did for the Big Five measurement scale.Brief measures also have the limitation of being unable tomeasure individual facets of multi-faceted constructs(Gosling et al. 2003).

The original TRQ questionnaire was developed withthe intention of interviewing both parties to a relationshipconcerning their perceptions of its quality; this was notdone in this study. Future academic research would dowell to examine salesperson-customer relationship qual-ity from both sides of the dyad. This study has focusedonly on the issues related to relationships from the seller’sperspective. However, because relationships involve morethan one person, it is also important to look at these issuesfrom the customer’s perspective. It would be worthwhileapplying and studying the TRQ scale by using customersas respondents, in order to compare their perceptions ofthe quality of their relationships with friends to those ofsalespeople.

CONCLUSIONS AND IMPLICATIONS

This study has described the relationships amongsalesperson personality traits and their relationship qual-ity with friends and customers. The results of this studysuggest that in the financial services industry, salespeople’spersonality traits influence the quality of their relation-ships with customers and friends. Managers can benefitfrom taking advantage of this correlation between person-ality and relationship quality. For instance, personalitytests may be useful in the selection of salespeople ifmanagement wishes to make distinctions among indi-viduals. However, less represented personality types canbe also effective in a sales career since other variables canbe significant in understanding sales performance too,especially in our study that concentrated on thesalesperson’s perception of relationship quality.

Moreover, the study found that among five personal-ity traits, a salesperson’s agreeableness has the mostsignificant effect on all four dimensions of the quality of

his or her relationship with customers (cf., Appendix B).This finding suggests that managers can stress the impor-tance of agreeableness while recruiting salespeople, andcan hire agreeable salespersons.

Furthermore, the results reveal that these traits do notaffect the relationship quality with customers in the sameway as they do relationships with friends. This finding iscontrary to much of the literature on the nature of friend-ships in marketing and relationship selling, which hassuggested that good customer relationships develop intoclose friendships (e.g., Grayson 2007; Price and Arnould1999; Swan et al. 2001).

Using the TRQ questionnaire in a business contextfor the first time is the main academic implication of thisstudy. Furthermore, the findings of this study have impli-cations for the practice of personal selling, and also forsales managers. In order for salespeople to be successfulin developing and maintaining good relationships withcustomers, they must know what to expect. This studyindicates that when it comes to the other party’s contribu-tion to the relationship in terms of sharing time andinformation, customers may differ fundamentally fromfriends. In this relationship domain, as well as in thedomains concerning caring about the relationship andenjoying spending time together, attempts by a salesper-son to treat the customer as a friend, and indeed to rely onthis, will probably be less successful. It might even makethe customer suspicious of the salesperson, which in turncould diminish trust and lead to miscommunication. Thefinding that the salesperson’s agreeableness is the onlypersonality trait that affects relationship quality withcustomers underlines this conclusion. Thus, no matterhow good the relationship with the customer is, thesalesperson should always remain professional.

These findings are important for several reasons.First, in line with literature, the results of the current studyfound that individual personality traits must be consideredas important components in trade relationships; thus,managers should pay more attention to the personalities oftheir sales force. Second, the results suggest that instruct-ing salespeople to treat customers as friends may not beworthwhile. Customers may detect insincerity and over-familiarity in the relationship, which at worst might causethem to withdraw from commercial interaction. Finally,for managers, when guiding salespeople in their interac-tions with customers, it may be helpful to alert them that– unlike a friendship, in a business relationship sales-people will have to contribute more and be prepared forthe other party to contribute less. This will happen in anenvironment in which the time spent in the interaction willbe significantly less enjoyable than the time spent withfriends.


REFERENCES

Barrick, Murray R. and Michael K. Mount (1991), “TheBig Five Personality Dimensions and Job Perfor-mance: A Meta-Analysis,” Personnel Psychology,44, 1–26.

____________and ____________ (1993), “Autonomyas a moderator of the relationships between the bigfive personality dimensions and job performance,”Journal of Applied Psychology, 78, 111–18.

____________ Greg L. Stewart, and Mike Piotrowski(2002), “Personality and Job Performance: Test ofthe Mediating Effects of Motivation Among SalesRepresentatives,” Journal of Applied Psychology,87, 43–51.

Crosby, Lawrence A., Kenneth R. Evans, and DeborahCowles (1990), “Relationship Quality in ServicesSelling: An Interpersonal Influence Perspective,”Journal of Marketing, 54 (3), 68–81.

Digman, John M. and Naomi K. Takemoto-Chock (1981),“Factors in the Natural Language of Personality:Re-Analysis Comparison, and Interpretation of SixMajor Studies,” Multivariate Behavioral Research,16, 149–70.

____________ (1990), “Personality Structure: Emergenceof the Five-Factor Model,” Annual Review of Psy-chology, 41, 417–40.

Dion, Paul, Debbie Easterling, and Shirley Jo Miller(1995), “What Is Really Necessary in SuccessfulBuyer/Seller Relationships?” Industrial MarketingManagement, 24, 1–9.

Foster, Brian D. and John W. Cadogan (2000), “Relation-ship Selling and Customer Loyalty: An EmpiricalInvestigation,” Marketing Intelligence and Planning,18 (4), 185–99.

Geller, Lois K. (2006), “Lose the Attitude,” Target Mar-keting, 29 (December), 21–23.

Gosling, Samuel D., Peter J. Rentfrow, and William B.Swann (2003), “A Very Brief Measure of the Big-Five Personality Domains,” Journal of Research inPersonality, 37, 504–28.

Grayson, Kent (2007), “Friendship Versus Business inMarketing Relationships,” Journal of Marketing, 71(4), 121–39.

Grönroos, Christian (1994), “From Marketing Mix toRelationship Marketing: Towards a Paradigm Shiftin Marketing,” Management Decision, 32 (2), 4–20.

Guilford, J.P. and Wayne S. Zimmerman (1949), TheGuilford-Zimmerman Temperament Survey. BeverlyHills: Sheridan Supply.

Gummesson, Evert (1998), “Implementation Requires aRelationship Marketing Paradigm,” Journal of theAcademy of Marketing Science, 26 (3), 242–49.

Hair, Joseph F., William C. Black, Barry J. Babin, andRolph E. Anderson (2010), Multivariate Data Analy-sis: A Global Perspective, 7th Edition. New Jersey:Pearson Education.

John, Oliver P. and Sanjay Srivastava (1999), “The BigFive Trait Taxonomie,” in Handbook of Personality:Theory and Research, L.A. Pervin and O.P. John,eds. New York: Guilford.

Kassarjian, Harold H. (1971), “Personality and ConsumerBehavior: A Review,” Journal of MarketingResearch, 8 (4), 409–18.

Lamont, Lawrence M. and William J. Lundstrom (1977),“Identifying Successful Industrial Salesmen by Per-sonality and Personal Characteristics,” Journal ofMarketing Research, 14, 517– 29.

McCrae, Robert R. and Paul T. Costa (1985), “UpdatingNorman’s Adequate Taxonomy: Intelligence and Per-sonality Dimensions in Natural Language and inQuestionnaires,” Journal of Personality and SocialPsychology, 49, 710–21.

____________, Paul T. Costa, Gregorio H. Del Pilar,Jean-Pierre Rolland, and Wayne D. Parker (1998),“Cross-Cultural Assessment of the Five FactorModel: The Revised NEO Personality Inventory,”Journal of Cross-Cultural Psychology, 29, 171–88.

Morgan, Robert M. and Shelby D. Hunt (1994), “TheCommitment-Trust Theory of Relationship Market-ing,” Journal of Marketing, 58 (3), 20–38.

Morris, Charles G. (1996), Psychology: An Intro-duction, 9th ed. Upper Saddle River: Prentice Hall.

Mustillo, Sarah A., Shannon Dorsey, and Elizabeth M.Z.Farmer (2005), “Quality of Relationships BetweenYouth and Community Service Providers: Reliabil-ity and Validity of the Trusting Relationship Ques-tionnaire,” Journal of Child and Family Studies, 14(4), 577–90.

Norman, Warren T. (1963), “Toward an Adequate Tax-onomy of Personality Attributes: Replicated FactorStructure in Peer Nomination Personality Ratings,”Journal of Abnormal and Social Psychology, 66,574–83.

Parsons, Amy (2002), “What Determines Buyer-SellerRelationship Quality? An Investigation from theBuyer’s Perspective,” Journal of Supply Chain Man-agement, 38 (2), 4–12.

Pederson, Carlton A., Milburn D. Wright, and Barton A.Weitz (1988), Selling: Principles and Methods.Homewood: Irwin.

Price, Linda L. and Eric J. Arnould (1999), “CommercialFriendships: Service Provider-Client Relationshipsin Context,” Journal of Marketing, 63 (4), 38–56.

Solomon, Michael R. (1992), Consumer Behavior: Buy-ing, Having, and Being. Boston: Allyn and Bacon.

Swan, John E., Cathy Goodwin, Michael A. Mayo, andLynne D. Richardsson (2001), “Customer Identities:Customers as Commercial Friends, CustomerCoworkers or Business Acquaintances,” Journal ofPersonal Selling and Sales Management, 21 (1), 29–37.

Tan, Cheryl Lu-Lien and Brian Steinberg (2007),“Penney’s Updated Image, The Sequel,” The WallStreet Journal, (February 13).


Tansuhaj, Patriya, Donna Randall, and Jim McCullough(1988), “A Services Marketing Management Model:Internal and External Marketing Functions,” Journalof Service Marketing, 2 (1), 31–38.

Vance, E. and H. Sanchez (1997), The Trusting Relation-ship Questionnaire. Unpublished assessment mea-sure.

Wang, Wen-Hung, Chiung-Ju Liang, and Yung-De Wu(2006), “Relationship Bonding Tactics, RelationshipQuality and Customer Behavioral Loyalty –Behavioral Sequence in Taiwan’s InformationServices Industry,” Journal of Services Research, 6(1), 31–57.

Williams, Michael R. and Jill S. Attaway (1996),“Exploring Salesperson’s Customer Orientationas a Mediator Organizational Culture’s Influence onBuyer-Seller Relationships,” Journal of PersonalSelling and Sales Management, 16 (4), 33–52.

Wulf, Kristof De, Gaby Odekerken-Schröder, and DawnIacobucci (2001), “Investments in Consumer Rela-tionships: A Cross-Country and Cross-Industry Ex-ploration,” Journal of Marketing, 65 (4), 33–50.

Zimbardo, Philip G. and Ann L. Weber (1994),Psychology. New York: HarperCollins College Pub-lishers.

APPENDIX AScale Items

Relationship Quality Scale

Construct Items

Sharing Does [Other Party]* identify things he/she likes about you?Does [Other Party] talk to you about his/her problems?Does [Other Party] want to spend time with you?Does [Other Party] share information of a personal nature?Does [Other Party] initiate contact with you in times of need?Does [Other Party] tell you they are sorry when things gowrong?

Caring Does [Other Party] seek counseling or advice from you?Do you apologize to [Other Party] when things go wrong?Do you talk to others in a positive way about [Other Party]?

Point of View Does [Other Party] consider your point of view?Do you consider [Other Party’s] point of view?

Time Enjoyment Do you enjoy spending time with [Other Party]?

Personality Traits Scale

Construct Items

I see myself as:

Extraversion Extraverted, enthusiasticReserved, quiet (R)

Agreeableness Critical, quarrelsome (R)Sympathetic, warm

Conscientiousness Dependable, self-disciplinedDisorganized, careless (R)

Emotional stability Anxious, easily upset (R)Calm, emotionally stable

Openness to experiences Open to new experiences, complexConventional, uncreative (R)

*In the Relationship Quality scale, [Other Party] was substituted for friend and customer, respectively.(R) Denotes reverse-coded items.


For further information contact:Kaveh Peighambari

Department of Business Administration, Technology and Social SciencesLuleå University of Technology

971 87 Luleå, SwedenPhone:+46.920.493121

Fax:+46.920.491399E-Mail: [email protected]

APPENDIX BRegression Results

Dimension of Relationship Quality: SharingPersonality Traits βββββ for Friends βββββ for Customers

Extraversion .07* -0.01Agreeableness .04 .17*Conscientiousness .08* -0.01Emotional stability .01 .01Openness to experience -0.02 -0.02

R2 for Model Sharing .13** .08

Dimension of Relationship Quality: CaringPersonality Traits ßβ for Friends β for CustomersExtraversion .06* .00Agreeableness .04 .12*Conscientiousness .02 .03Emotional stability .06* .01Openness to experience -0.02 .04

R2 for Model Caring .21** .10*

Dimension of Relationship Quality: Point of ViewPersonality Traits β for Friends β for CustomersExtraversion .02 .02Agreeableness -0.01 .18**Conscientiousness .05 .00Emotional stability .00 -0.07Openness to experience .02 .10

R2 for Model Point of View .04 .18**

Dimension of Relationship Quality: Time EnjoymentPersonality Traits β for Friends β for CustomersExtraversion .10** .05Agreeableness -.11** .11*Conscientiousness .09** -0.01Emotional stability .14** .05Openness to experience -0.05 .04

R2 for Model Time Enjoyment . 43** .07

** p < .01 * p < .05


KNOWLEDGE TRANSFER ANTECEDENTS ANDCONSEQUENCES: A CONCEPTUAL MODEL

Nicholas Kolenda, Stonehill College, EastonLee McGinnis, Stonehill College, Easton

Brian Glibkowski, Stonehill College, Easton

ABSTRACT

The authors examine antecedents and consequencesof the Knowledge-Transfer Circumplex model. They pro-pose that different forms of intelligence (i.e., emotional,practical, and cognitive) influence the different styles ofcommunication in the KT-Plex (i.e., relational, direct, andanalytical), resulting in different forms of trust (i.e., affec-tive, behavioral, and cognitive).

INTRODUCTION

Since the popularity of emotional intelligence (EI)some 20 years ago, studies of EI have been emerging in themarketing literature. Kernback and Schutte (2005), forexample, found that higher EI in service providers leads togreater customer satisfaction, while Rozel, Pettijohn, andParker (2004) found that EI is significantly related to asalesperson’s level of customer orientation. Recent workfrom Kidwell, Hardesty, Murtha, and Sheng (2011) indi-cates that EI mediates the relationship between customerrelationship and sales performance. Despite the impor-tance of EI in marketing, the reason behind its effective-ness remains unclear. Also unclear is how EI relates withother forms of intelligence and knowledge transfer tech-niques.

This paper sheds light on why EI is effective byarguing that the Knowledge-Transfer Circumplex (KT-Plex), developed by McGinnis, Glibkowski, Gillespie,and Lemmon (2011), mediates the relationship betweenEI and trust. We begin with an explanation of the KT-Plex,followed by an examination of its antecedents. We choseemotional, practical, and cognitive intelligence as ourantecedents because research indicates that these forms ofintelligence are common, distinct, and present in expertcommunicators (McGinnis et al. 2011). We then examineconsequences to the KT-Plex model. Specifically, wechose affective, behavioral, and cognitive trust as ourconsequences because trust has been linked to severalimportant marketing outcomes, including commitment(Morgan and Hunt 1994), loyalty (Sirdeshmukh, Singh,and Sabol 2002), and satisfaction (Singh and Sirdeshmukh2000).

THE KT-PLEX: UNDERLYING DIMENSIONS

In analyzing how top teaching professionals in golftransferred knowledge to their students, McGinnis et al.(2011) revealed six communication modes grouped withinthree styles. The relational style includes metaphoricaland analogical communication, the direct style includesgeneral and experiential communication, and the analyti-cal style includes causal and technical communication.Underlying these styles are two dimensions: locus ofknowledge and comprehension.

A circumplex structure implies that all variables (i.e.,communication modes) are related to each other in a two-dimensional space (Fabrigar, Visser, and Browne 1997).In the KT-Plex, the comprehension dimension refers tothe level of knowledge, which is represented by a con-tinuum defined by use knowledge at the low end andprocedural knowledge at the high end with declarativeknowledge at the intermediate point. The locus of knowl-edge dimension distinguishes between knowledge in theforeground (i.e., the object of understanding) and knowl-edge in the background (i.e., context within which theforeground is understood). While Polanyi (1966) indi-cates in absolute terms that all knowledge is a gestalt (i.e.,foreground and background are indistinguishable), inrelative terms, the foreground and background are distinctand contain unique properties. Polanyi (1966) refers to thebackground as proximal knowledge because it is nearer toa knower and represents internalized experiences that aredifficult to communicate. The foreground representsknowledge that is the focus of attention and epistemologi-cally distal or further from the background (Polanyi1966). There is a functional relationship between thebackground (proximal) and foreground (distal) such thatthe “we know the first term only by relying on ourawareness of it for attending to the second term” (Polanyi1966, p. 10).

The foreground and background are distinguished bythe manner in which knowledge can be integrated to formgestalt meaning. If the starting point of integration is theforeground, Polanyi (1966) describes that “explicit inte-gration” occurs whereby the relation between particulars


is explicitly stated. Conversely, if the starting point ofintegration is the background, Polanyi (1966) describesthat “tacit integration” occurs whereby the particulars(i.e., foreground) are integrated into a coherent whole thatresists explication. In sum, the locus of knowledge dimen-sion refers to knowledge that is achieved through explicitintegration or tacit integration.

As Lange (2008) suggests, a circumplex is useful forunranked typologies. Because the communication stylesin the KT-Plex are not weighted by importance, thecircumplex structure indicates neither a beginning nor anend. Another defining characteristic of a circumplex isthat the proximity between variables indicates the strengthof the relationship between those variables. All communi-cation modes are positively related to each other and thestrength of the association between the variables increasescommensurate to the variable’s proximity to each other(see Fabrigar et al. 1997 for a review). Adjacent variables(e.g., general and experiential communication) are mostrelated to each other, whereas opposite variables (e.g.,experiential and causal communication) are least relatedbecause they are located 180° from each other.

ANTECEDENTS: FORMS OF INTELLIGENCE

Research from McGinnis et al. (2011) indicates thatdifferent forms of intelligence were prevalent among theirsample of expert golf instructors. We disentangle thedifferent forms of intelligence and propose connections tothe communication styles found in the KT-Plex model.The intelligence antecedents are from the sender’s per-spective, and the trust outcomes are from the receiver’sperspective.

Emotional Intelligence

An early definition characterized EI as the ability todeal with emotions in relation to others’ feelings andemotions (Salovey and Mayer 1990). It involves under-standing, expressing, and regulating one’s true emotions,as well as recognizing and interpreting the emotions ofothers (Davies et al. 1998). Adeptness in relationships, akey element of EI (Goleman 1998), involves identifyingelements that motivate another person, which can lead toeffective, persuasive, and mutually satisfying communi-cation. Research indicates that people with high EI pos-sess more social competency, higher quality relation-ships, and greater interpersonal sensitivity (for a completereview, see Mayer, Salovey, and Caruso 2008).

Practical Intelligence

Similar to street smarts and common sense (Sternberg2000), practical intelligence (PI) is the ability to applyknowledge in real-life situations that are different fromthose presented in classrooms or other academic settings

(Sternberg and Wagner 1993). Someone with high cogni-tive intelligence, for example, may struggle to apply thatknowledge in real situations. Sternberg and Wagner (1993)discern the difference between academic intelligence andpractical intelligence through how problems are definedand solved. Academic problems are generally rigidbecause they usually have only one solution and containlittle or no intrinsic interest. Practical problems, however,possess multiple solutions, require recognition and for-mulation, and are embedded in real-life situations. Thefocus of PI is use value as opposed to analytical under-standing.

Cognitive Intelligence

Cognitive Intelligence (CI), also known as Intelli-gence Quotient (IQ), academic intelligence, or simply“g,” is characterized by academic skills, such as reading,mathematics, and problem solving (Armor 2003). Theseskills are divided into two major forms: fluid and crystal-lized intelligence. Fluid intelligence involves reasoningand problem solving abilities, and it is typically measuredby working memory capacity (Conway et al. 2002). Crys-tallized intelligence, on the other hand, represents accu-mulated knowledge over time. It focuses less on reasoningand more on factual information, such as verbal compre-hension (Plemons, Willis, and Baltes 1978). People withhigh CI have a well-rounded intelligence because theypossess reasoning abilities (i.e., fluid intelligence) andfactual knowledge (i.e., crystallized intelligence).

KT-PLEX MODES OF COMMUNICATION

By depicting different communication styles, theKT-Plex provides insights into how different communi-cation modes can be optimized to convey tacit and explicitconcepts associated with the locus of knowledge, as wellas procedural, declarative, and use knowledge associatedwith comprehension. In this section, we define eachcommunication mode and develop propositions betweenthe different forms of intelligence and the various compo-nents of the KT-Plex (see Figure 1).

Relational Style

The relational style of communication, comprised ofthe metaphorical and analogical modes, borrows meaningfrom other domains to relate understanding to a focaldomain. As depicted by the KT-Plex, metaphorical com-munication transfers declarative knowledge, whereasanalogical communication transfers procedural knowl-edge. Because these modes use symbolic language toconvey knowledge, both convey tacit knowledge.

Metaphorical Communication. Metaphorical com-munication is defined as knowledge of one domain expe-rience in terms of a different domain experience. For


example, Kotler and Singh (1980) popularized the meta-phor, “marketing is warfare,” to clarify one domain (mar-keting) by comparing it to a different context (warfare). Inthat metaphor, the domain of marketing is known as thetarget domain because the metaphor clarifies this con-cept. Conversely, the domain of warfare is known as thesource domain because it provides a source of knowledgeto clarify marketing (Gibbs 1993). The inference or con-nection between the two domains is known as the ground(Murphy 1996). From the previous metaphor, one couldinfer that marketing involves two or more competitorsstrategically battling for the consumer’s money(Rindfleisch 1996).

Not only can metaphors effectively communicateproduct information to consumers, but they can alsotransform abstract concepts into tangible representations(Zaltman 2008). By comparing an intangible domain to atangible domain, metaphors can provide a mental repre-sentation of an intangible characteristic, which can be asignificant advantage for service marketers because ofservice intangibility. For example, Zaltman (2008)describes how “life insurance companies use ideas asso-ciated with various symbols such as umbrellas (Travel-ers), rocks (Prudential Insurance Company), and hands(Allstate) to convey qualities of protection, sturdiness,and support” (p. 35).

Analogical Communication. Like metaphors, analo-gies transfer knowledge by relating an unfamiliar domainto a familiar domain. The difference, however, lies in the

relational structuring of each domain. Analogies involvestructural parallelism between the areas being compared,focusing only on the relational characteristics betweendomains (Gentner 1983). Thus, analogies allow for cause-and-effect inferences to be drawn, which can lead togreater understanding of unfamiliar domains. Forexample, a causal inference could be drawn from thecommon “marketing is marriage” analogy. Based on theunderstanding that commitment, trust, and the right choicein partners leads to a successful relationship with a spouse(Hunt and Morgan 1994), one could infer that incorporat-ing those characteristics in marketing exchanges shouldlead to a successful relationship with a consumer.

Relational communication is effective when the com-parison between the target and source domain is relevantto the receiver. Celsi and Olson (1988) found that con-sumers devote more attention to and exhibit greater com-prehension of advertisements when the information ispersonally relevant. Because EI involves understandingdifferent points of view (Goleman 1998), communicatorswith high EI should be able to judge consumers’ knowl-edge, passions, and interests more effectively than com-municators with low EI. Therefore, communicators withhigh EI should be able to structure metaphors and analo-gies that are relevant to consumers, thereby effectivelytransferring knowledge.

P1: Higher levels of emotional intelligence (relative topractical and cognitive intelligence) will lead to greateruse of the relational style of communication.

FIGURE 1Antecedents and Consequences to the Knowledge-Transfer Circumplex

Affect ive Trust

Em otional In tel ligence

Pract ical In tel ligence

Cognitive In tel ligence

Relational Style

Metaphorical Analogical

Direc t Style

General Experient ial

Analytical Causal

Technical

Behavio ral Trust

Cogn it ive Trust

P1

P2

P3

P4

P5

P6


Direct Style

Unlike the relational style, the direct style is mostuseful when a receiver’s pre-existing knowledge of thefocal domain is very basic. Using general and experientialcommunication, the direct style transmits awarenessknowledge so that receivers become more familiar with adomain. General communication conveys explicit knowl-edge, whereas experiential communication conveys tacitknowledge.

General Communication. McGinnis et al. (2011)define general communication as knowledge by symbolsthat represent basic human experiences. The language isviewed as universal because it applies across contexts andconveys the same meaning to people of different back-grounds. Examples of general communication by thesender (i.e., the golf pro) included verbal communication:“right hand over left hand,” “hit it harder,” and “slowdown your motion.”

General communication effectively transfers knowl-edge because it communicates information in its simplestform, thereby creating a comprehensible message that iseasy to understand. Structuring comprehensible messagesis very important for marketers because comprehensionmoderates persuasion (Ratneshwar and Chaiken 1991)and increases ad credibility (Mick 1992).

Experiential Communication. Experiential commu-nication is associated with structured experiences, whereinrecipients gain multi-sensory experience in a domain.Golf professionals, for example, instructed students to hitgolf balls on the driving range to gain direct experience inthe golf domain (McGinnis et al. 2011). By incorporatingexperiential communication into marketing tactics, mar-keters create sensory experiences for consumers, therebyproviding them with pleasure and excitement (Schmitt1999).

One method of appealing to consumers’ senses isthrough experiential marketing, which involves market-ing tactics that provide consumers with an extraordinaryexperience. Schmitt (1999) describes Procter & Gamble’s“Tide Mountain Fresh” campaign used by Tide laundrydetergent, which portrays a particular fragrance as thescent from a mountaintop. Not only does the product’sadvertising use colors that are “cool, vivid, and refresh-ing,” but the images display snow covered mountaintops,suggesting that the scent resembles crisp mountain air(Schmitt 1999, p. 101). Tide even incorporates scratch-and-sniff features so that consumers can receive directpre-purchase experience with Tide’s laundry detergent.

The direct style of communication includes basicknowledge of a domain and real-world experiential exer-cises and simulations. Therefore, those with high practical

intelligence, or real world know-how and aptitude, shouldbe most inclined to use this style when conveying knowl-edge.

P2: Higher levels of practical intelligence (relative toemotional and cognitive intelligence) will lead togreater use of the direct style of communication.

Analytical Style

The analytical style contains both causal and techni-cal communication. Technical communication usesexplicit knowledge to transfer declarative information,whereas causal communication uses explicit knowledgeto transfer procedural information.

Technical Communication. The language used intechnical communication is explicit and more unique tothe focal domain. This precision in language eliminatesunnecessary verbiage, allowing the sender and receiver tocommunicate more efficiently. Because of the domain-specific language, recipients of technical communicationmust possess knowledge of the focal domain to under-stand the sender’s message.

Evidence suggests that marketers can use technicalcommunication in transferring knowledge to expert con-sumers. Graeff (1997) presented expert consumers andnovice consumers with a camera advertisement contain-ing technical information. Although both groups inferredinformation from the advertisement, only expert consum-ers, who had previous experience using cameras, couldinfer benefits from the technical attributes. These infer-ences lead to greater comprehension of the overall adver-tisement.

Causal Communication. Similar to the analogicalmode, causal communication conveys cause-and-effectknowledge within a domain. The difference, however, isthat causal communication conveys explicit knowledgerather than tacit knowledge. Analogies rely on symboliclanguage to transfer knowledge, whereas causal com-munication clearly articulates causal understanding ina domain, an important foundation for developing proce-dural knowledge. One golf instructor in McGinnis et al.(2011) described the importance of cause-and-effectknowledge in golf by saying, “You have to understandwhy the ball goes where it goes.”

Causal communication can convey how productsfulfill their proposed benefits. Although lower knowledgeconsumers lack means-end chain knowledge to under-stand connections between product attributes and respec-tive benefits, higher knowledge consumers can infer causalconnections between product attributes and certain ben-efits, an understanding that leads to greater comprehen-sion (Graeff 1997).


As previously noted, those with high CI have highfluid and crystallized intelligence. Fluid abilities, whichreflect problem solving, reasoning, and other workingmemory tasks, are crucial for understanding intricateinformation, such as causal relationships. Additionally,because crystallized intelligence (i.e., knowledge of fac-tual information) involves verbal comprehension(Plemons, Willis, and Baltes 1978), people with highcrystallized intelligence are more capable of grasping thetechnical language of a specific domain. Therefore, peoplewith high CI should be more inclined to use the analyticalstyle of communication.

P3: Higher levels of cognitive intelligence (relative toemotional and practical intelligence) will lead togreater use of the analytical style of communication.

CONSEQUENCES: FORMS OF TRUST

Although the communication styles in the circumplex(relational, direct, and analytical) can lead to severaldiverse and important marketing outcomes, we focus onwhat we believe to be an important mediator leading totrust. Adler (2001) declared that trust should be consid-ered a central construct given the importance of knowl-edge intensive activities in organizations. In the market-ing literature, trust is defined as the perceived credibilityand benevolence of a target of trust (see Doney andCannon’s 1997 for a review). Johnson and Grayson (2005)describe three forms of trust – affective, cognitive, andbehavioral – which are the forms of trust we examined.

Affective Trust

Affective trust constitutes the feelings and emotionsgenerated from a partner due to felt care and concern, andit is characterized by a sense of security and perceivedrelationship strength (see Johnson and Grayson 2005).Compared with cognitive trust, affective trust is influ-enced by personal relationships, rather than firm-levelrelationships, and it is deepened when service providersfocus on the customer’s best interest.

Because affective trust is influenced by personalrelationships, we anticipate that the relational style ofcommunication (a more personal variable compared toother styles) will have the strongest relationship withaffective trust. Emotional intelligence, which we positedwould have the strongest relationship with the relationalstyle of communication, is also an individual-level con-struct, thus reinforcing our logic. Because expert commu-nicators with high emotional intelligence can build strongrelationships with receivers of knowledge, receivers shouldfeel greater emotional attachment to the sender. Thisemotional attachment should, therefore, produce an emo-tional form of trust.

P4: Greater use of the relational style will lead to higheraffective trust.

Behavioral Trust

Although affective and cognitive trust are character-ized as a “willingness” to trust, behavioral trust exceedsmere willingness because it involves a commitment toaction (Mayer, Davis, and Schoorman 1995). A personwho has behavioral trust in another party commits to anaction, trusting that the other party will fulfill its obliga-tion.

Because behavioral trust requires reliance and de-pendence on another party (Gillespie 2003), risk is signifi-cantly related to behavioral trust (Mayer, et al. 1995). AsMayer et al. (1995) describe, “There is no risk taken in thewillingness to be vulnerable (i.e., to trust), but the risk isinherent in the behavioral manifestation of the willing-ness to be vulnerable” (p. 724). Although a willingness totrust involves little risk because no action is taken, depen-dence on another party (i.e., behavioral trust) is largelyaffected by risk because the action can produce negativeconsequences. Clearly, minimizing risk should increasethe development of behavioral trust.

Because it structures comprehensible messages, gen-eral communication avoids ambiguity and lowers per-ceived risk (Ghosh and Ray 1997). Experiential learningsituations that provide consumers with direct, pre-pur-chase experiences (e.g., product samples or trial periods)can also lower perceived risk (Smith and Swinyard 1983).The direct style of communication, which is encompassedby general and experiential modes, should therefore lowerperceived risk and produce greater behavioral trust.Whereas declarative and procedural knowledge are asso-ciated with “what” and “why” knowledge, general andexperiential communication are the only modes in the KT-Plex that are associated with the use of knowledge toachieve results – i.e., they are associated with “how”outcomes are achieved. Therefore, we maintain the fol-lowing:

P5: Greater use of the direct style will lead to higherbehavioral trust.

Cognitive Trust

Cognitive trust is described as the confidence inanother party’s competence and reliability, which is builtupon a history of transactions with that partner (seeJohnson and Grayson’s 2005 for a review). The golfexperts in McGinnis et al. (2011) demonstrated the great-est competence and reliability when they delved into thetechnical and causal components of the structure. Thesecomponents allowed the experts to relate with their clients


in terms of specific golf mechanics (e.g., grip terms,balance, angle of attack) and provide causal explanationsfor ball flight patterns. As a result, informed clientsdeveloped trust in the professional’s expertise.

In marketing, informed consumers can also developtrust through causal and technical communication.Szulanski, Cappetta, and Jensen (2004) found that causalambiguity moderates trustworthiness. As causal ambigu-ity increases in communicating information, receiversbecome more skeptical regarding the accuracy of themessage, which then decreases the sender’s perceivedtrustworthiness. However, consumers who understand amarketer’s causal communication will likely view thatmarketer as competent and reliable. In addition to produc-ing greater trust in the marketer, this perceived compe-tence should also lead to a justification of that trust (i.e.,cognitive trust).

Not only does technical communication indirectlyaffect trust because it is a building block toward cause-and-effect knowledge, but it also directly affects trustthrough its use of domain-specific language. Becausetechnical communication demonstrates superior knowl-edge of a domain, it can increase perceived expertise(French and Raven 2001). In addition, this perceivedexpertise has been found to influence consumers’ trust inthe marketer, (Busch and Wilson 1976), specifically,cognitive trust (Johnson and Grayson 2005).

P6: Greater use of the analytical style will lead to highercognitive trust.

DISCUSSION/CONCLUSION

Despite the importance of EI in marketing, littleresearch has examined why EI is effective. The presentstudy proposed mediating constructs (relational, direct,and analytical styles of communication in the KT-Plex) toexplain how different forms of intelligence (EI, PI and CI)might produce positive outcomes in marketing (affective,behavioral, and cognitive trust).

The propositions in the present study have severalmanagerial implications. First, our model can help mar-keters pinpoint weaknesses in their ability to build trustwith consumers. For example, marketers who lack cogni-tive trust with consumers (e.g., consumers are unaware of“what” factors are involved and “why” they are impor-tant) should use the analytical style of communicationbecause it leads to greater cognitive trust. The technicalmode can answer the “what” question, and the causalmode can answer the “why” question.

If a trust outcome is still lacking despite using therespective communication style, the salesforce may needtraining in the corresponding intelligence antecedent. If asalesforce lacks affective trust with consumers, thesalesforce may be using the direct style of communicationineffectively, or perhaps not conveying information in ajargon-free, easy-to-understand manner. Because EI, theantecedent to the relational style of communication, canbe improved with time and training (Boyatzis, Stubbs, andTaylor 2002), a company might build greater affectivetrust by training the salesforce in emotional intelligence.After training, the salesforce might use the relational styleof communication more effectively, which may thenproduce greater affective trust with consumers. Con-versely, if cognitive trust is lacking, the sales staff mayneed to focus more on the analytical style of communica-tion, conveying more specifically the inner workings orprocesses of their products relative to competitors’ prod-ucts. This implication may especially be the case withhighly complex technical products.

Though empirical support is needed to validate inclu-sion and determine the extent to which these relationshipsoccur, we feel that the support provided here and theresearch on how experts transfer knowledge will open thedoors to future research opportunities. These opportuni-ties include potential moderating relationships, such asproduct category, sales context (B2B versus B2C), andforms of communication (unmediated versus mediated).We believe that knowing how to communicate is a keyingredient to knowledge transfer, and, ultimately, trust.

REFERENCES

Adler, Paul S. (2001), “Market, Hierarchy, and Trust: TheKnowledge Economy and the Future of Capitalism,”Organization Science, 12 (March/April), 215–34.

Armor, D.J. and S.L. Aud (2003), Maximizing Intelli-gence, New Brunswick, NJ: Transaction Publishers.

Boyatzis, Richard E., Elizabeth C. Stubbs, and Scott N.Taylor (2002), “Learning Cognitive and EmotionalIntelligence Competencies through Graduate Man-

agement Education,” Academy of Management Learn-ing & Education, 1 (December), 150–62.

Busch, Paul and David. T. Wilson (1976), “An Experi-mental Analysis of a Salesman’s Expert and ReferentBases of Social Power in the Buyer-Seller Dyad,”Journal of Marketing Research, 13 (February), 3–11.

Celsi, Richard L. and Jerry C. Olson (1988), “The Role ofInvolvement in Attention and Comprehension Pro-cesses,” Journal of Consumer Research, 15 (Septem-ber), 210–24.


Conway, Andrew R.A., Nelson Cowan, Michael F.Bunting, David J. Therriault, and Scott R.B. Minkoff(2002), “A Latent Variable Analysis of WorkingMemory Capacity, Short-Term Memory Capacity,Processing Speed, and General Fluid Intelligence,”Intelligence, 30, 163–83.

Davies, Michaela, Lazar Stankov, and Richard D. Roberts(1998), “Emotional Intelligence: In Search of anElusive Construct,” Journal of Personality andSocial Psychology, 75 (October), 989–1015.

Doney, Patricia M. and Joseph. P. Cannon (1997), “AnExamination of the Nature of Trust in Buyer-SellerRelationships,” The Journal of Marketing, 61 (April)35–51.

Fabrigar, Leandre R., Penny S. Visser, and Michael W.Browne (1997), “Conceptual and MethodologicalIssues in Testing the Circumplex Structure of Data inPersonality and Social Psychology,” Personality andSocial Psychology Review, 1 (August), 184–203.

French, John R.P. and Bertram Raven (2001), “The Basesof Social Power,” in The Negotiation Sourcebook, IraAsherman and Sandy Asherman, eds. Amherst, MA:HRD Press, 61–74.

Gentner, Dedre (1983), “Structure-Mapping: A Theoreti-cal Framework for Analogy,” Cognitive Science, 7,155–70.

Ghosh, Dipankar and Manash R. Ray (1997), “Risk,Ambiguity, and Decision Choice: Some AdditionalEvidence,” Decision Sciences, 28 (1), 81–104.

Gibbs, Raymond W. (1993), Metaphor and Though. NewYork: Cambridge University Press.

Gillespie, Nicole (2003), “Measuring Trust in WorkingRelationships: The Behavioral Trust Inventory,”Paper presented at the 2003 Annual Meeting of theAcademy of Management, Seattle, Washington.

Goleman, Daniel (1998), Working with Emotional Intel-ligence. New York: Bantam Dell.

Graeff, Timothy R. (1997), “Comprehending ProductAttributes and Benefits: The Role of Product Knowl-edge and Means End Chain Inferences,” Psychologyand Marketing, 14 (March), 163–83.

Hunt, Shelby and Robert Morgan (1994), “RelationshipMarketing in the Era of Network Competition,” Mar-keting Management, 3 (1), 19–28.

Johnson, Devon and Kent Grayson (2005), “Cognitiveand Affective Trust in Service Relationships,” Jour-nal of Business Research, 58, 500–07.

Kernbach, Sally and Nicola S. Schutte (2005), “TheImpact of Service Provider Emotional Intelligenceon Customer Satisfaction,” Journal of Services Mar-keting, 19 (7), 438–44.

Kidwell, Blair, David M. Hardesty, Brian R. Murtha, andShibin Shang (2011), “Emotional Intelligence inMarketing Exchanges,” Journal of Marketing, 75(1), 78–95.

Kotler, Philip and Ravi Singh (2001), “Marketing War-fare in the 1980s,” Journal of Business Strategy, 2(Winter), 30–41.

Lange, Donald (2008), “A MultidimensionalConceptualization of Organizational Corruption Con-trol,” The Academy of Management Review, 33 (3),710–29.

Mayer, John D., Peter Salovey, and David R. Caruso(2008), “Emotional Intelligence: New Ability orEclectic Traits?” American Psychologist, 63 (Sep-tember), 503–17.

Mayer, Roger C., James H. Davis, and F. David Schoorman(1995), “An Integrative Model of OrganizationalTrust,” Academy of Management Review, 20 (July),709–34.

McGinnis, Lee P., Brian C. Glibkowski, James J. Gillespie,and Grace Lemmon (2011), Customer RelationshipLessons from the Top Golf Instructors in the World:Introduction of the KT-Plex®: Semplar, LLC. http://semplar.com/Documents/KT-Plex_Technical_Report.pdf

Mick, David Glen (1992), “Levels of Subjective Compre-hension in Advertising Processing and their Rela-tions to Ad Perceptions, Attitudes, and Memory,”Journal of Consumer Research, 18 (March), 411–24.

Morgan, Robert M. and Shelby D. Hunt (1994), “TheCommitment-Trust Theory of Relationship Market-ing,” The Journal of Marketing, 58 (July), 20–38.

Murphy, Gregory L. (1996), “On Metaphoric Represen-tation,” Cognition, 60, 173–204.

Plemons, Judy K., Sherry L. Willis, and Paul B. Baltes(1978), “Modifiability of Fluid Intelligence in Ag-ing: A Short-Term Longitudinal Training Approach,”Journal of Gerontology, 33 (2) 224–31.

Polanyi, Michael (1966), The Tacit Dimension. London:Routledge & Kegan Paul.

Ratneshwar, Srinivasan and Shelly Chaiken (1991),“Comprehension’s Role in Persuasion: The Case ofits Moderating Effect on the Persuasive Impact ofSource Cues,” Journal of Consumer Research, 18(June), 52–62.

Rindfleisch, Aric (1996), “Marketing as Warfare: Reas-sessing a Dominant Metaphor,” Business Horizons,39 (5), 3–10.

Rozell, Elizabeth J., Charles E. Pettijohn, and R. StephenParker (2004), “Customer Oriented Selling: Explor-ing the Roles of Emotional Intelligence and Organi-zational Commitment,” Psychology and Marketing,21 (June), 405–24.

Salovey, Peter and John D. Mayer (1990), “EmotionalIntelligence,” Imagination, Cognition, and Person-ality, 9 (3), 185–211.

Schmitt, Bernd (1999), Experiential Marketing: How toGet Customers to Sense, Feel, Think, Act, and Relateto Your Company and Brands. New York: The FreePress.

Singh, Jagdip and Deepak Sirdeshmukh (2000), “Agency


and Trust Mechanisms in Consumer Satisfaction andLoyalty Judgments,” Journal of the Academy of Mar-keting Science, 28 (1), 150–67.

Sirdeshmukh, Deepak, Jagdip Singh, and Barry Sabol(2002), “Consumer Trust, Value, and Loyalty inRelational Exchanges,” The Journal of Marketing,66 (January) 15–37.

Smith, R.E. and W.R. Swinyard (1983), “Attitude-Behav-ior Consistency: The Impact of Product Trial VersusAdvertising,” Journal of Marketing Research, 20(August), 257–67.

Sternberg, Robert J. (2000), Practical Intelligence inEveryday Life. New York: Cambridge UniversityPress.

Sternberg, Robert J. and Richard K. Wagner (1993), “Theg-Ocentric View of Intelligence and Job Perfor-mance is Wrong,” Current Directions in Psychologi-cal Science, 2 (February), 1–5.

Szulanski, Gabriel, Rossella Cappetta, and Robert J. Jensen(2004), “When and How Trustworthiness Matters:Knowledge Transfer and the Moderating Effect ofCasual Ambiguity,” Organization Science, 15 (Sep-tember/October), 600–13.

Zaltman, Gerald and Lindsay H. Zaltman (2008), Market-ing Metaphoria: What Seven Deep Metaphors Re-veal about the Minds of Consumers. Harvard Busi-ness Press.

For further information contact:Lee McGinnis

Stonehill College320 Washington Street

North Easton, MA 02357Phone: 508.565.1971



INTRAFIRM INFORMATION ADVANTAGE AND BROKERAGE:EFFECTS ON SALESPERSON PERFORMANCE

Gabriel R. Gonzalez, Thunderbird School of Global Management, GlendaleDanny P. Claro, Insper Education and Research Institute, Brazil

SUMMARY

A continued shift in the role of the sales force towardcustomer relationship management has resulted in therecognition that salespeople are embedded in their ownsales team and within marketing and other intrafirmfunctions (Bradford et al. 2010). Central to this researchstream is the notion that salespeople must form intrafirmties in order to access diverse and unique information thatcan lead to the development of better solutions for cus-tomers and to gain the cooperation needed to implementthose solutions (Steward et al. 2010).

The goal of this research study is to further identifythe strategies salespeople take to embed themselves intheir organization’s intrafirm advice network in order toacquire the knowledge and resources needed to improvetheir performance. Our conceptual framework draws onsocial network literature to identify the ways in whichsalespeople position themselves as a go-between in orderto broker information gained from individuals in otherfunctions such as marketing and product development forexchange with other salespeople in the sales force. Weidentify the impact of the proportion of ties that sales-people maintain with marketing members, product devel-opment members, and higher-ranking members on per-formance. We explore the interaction effects of combin-ing a go-between position in the advice network with ahigh proportion of ties to marketing, product develop-ment, and higher ranking organizational members. Bydrawing on relationship selling and social network theory,we provide a unique perspective from which to under-stand a salesperson’s ability to leverage firm resources inorder to create and deliver value to customers.


Information advantage may impact sales performancein three distinct ways. First, research demonstrates thatstrong communication ties between the sales and market-ing functions are important for managing relationshipswith customers and improving sales performance (Biemansand Rencic 2007; Smith, Gopalakrisna, and Chatterjee2006). In addition, salespeople who actively exchangeinformation with members of their firm’s product devel-opment function improve the performance of their firm’sproducts in the marketplace (Joshi 2010). Last, communi-

cation with hierarchically higher ranking organizationalmembers, who are more aware of and reflect theorganization’s strategies, can enhance performance (Work-man, Homburg, and Jensen 2003). Therefore:

H1: A greater proportion of intrafirm advice ties to those

in (a) marketing, (b) product development, and tothose in (c) higher-ranking positions positivelyaffects salesperson performance.

To influence those in their immediate set of ties to actin a supportive and cooperative fashion salespeople oftenneed influence. Interpersonally, influence has consis-tently been linked to an individual’s ability to brokerinformation in order to gain cooperation (Fernandez andGould 1994). When a salesperson acts as a go-between fora fellow salesperson and another organizational memberin a function other than sales, she is a broker. Because shestands between unconnected actors, the salesperson con-trols the flow of information from the outside function(e.g., marketing) to her fellow salesperson, thus increas-ing the fellow salesperson’s dependence on her. There-fore:

H2

Occupying a go-between position in the advice net-work, between a fellow salesperson and an organiza-tional member outside of sales, positively affectssalesperson performance.

When brokerage positions are established acrossseveral functional boundaries, the composition of theoutside group affiliations that a salesperson maintainsbecomes relevant. Brokering information for one particu-lar function (i.e., marketing and product development)outside of sales may prove to be more advantageous incomparison to others. In order to sustain high valueinformation flow between functional areas, salespeopleshould be connected to higher-ranking management inorder to be able to exchange short term and tacit knowl-edge, resource opportunities, and information about thelong-term direction of the firm and policy issues (Flahertyet al. 2012). Therefore:

H3: The impact of a go-between position on sales perfor-

mance is enhanced by a greater proportion of ties to(a) marketing, (b) product development and (c) higher-ranking people.


Method and Conclusions

We collected data from salespeople of a provider ofIT services, market intelligence, and credit assessment.The survey produced a response rate of 69 percent yield-ing 204 completed questionnaires. The results of OLSregression support H1b, H2, H3a, and H3c. The resultssuggest that marketers should extend relationship market-ing beyond (1) an interfirm perspective by accounting forthe intrafirm relationship effects that occur across a webof relational ties, (2) a focus on relational trust andcommitment as the primary performance-enhancingmechanism to include other informational and coopera-tive mechanisms, and (3) relational assets built between

firms as the key source of competitive advantage toinclude the benefits of relational assets (information &brokerage) built within firms. This study extends thedomain of relationship marketing across all of these areassimultaneously by demonstrating that some knowledgepools within the firm better help salespeople to managerelationships with customers and that being able to brokera wide range of knowledge for fellow salespeople can leadto cooperation which in turn leads to improved perfor-mance. Managers should focus on improving relationalties among coworkers as a strategically viable alternativeto focusing solely on customer-centric relationship pro-grams. References are available upon request.

For further information contact:Gabriel R. Gonzalez

Thunderbird School of Global Management1 Global Place

Glendale, AZ 85306–6000Phone: 602.978.7417



BEFRIENDING SPORT CELEBRITIES THROUGH MEDIATEDRELATIONSHIPS: PARASOCIAL INTERACTIONS AND

RELATIONSHIPS WITH ATHLETES IN SOCIAL MEDIA

Mujde Yuksel, University of Massachusetts, Amherst

SUMMARY

Millions of consumers have a desire to follow thelives of their favorite athletes as a matter of usual practiceand social media sites have nourished this desire byfacilitating the opportunity of keeping up with the lives ofcelebrities. This conceptual paper, employing the sportscontext for theory assessment and enhancement, is anattempt to highlight and integrate the subjects of athletesas a distinct point of attachment, the consumers parasocialinteractions (PSI) and relationships (PSR) with media

personae, and social media. Thus, adopting PSI and PSR,which, to date, have been mostly investigated by commu-nication and psychology literature, the paper focuses onthe athletes as a distinct point of attachment with regard tothe impact of new media, specifically social media, on therelationship between consumers and athletes as sportcelebrities. Particularly, the paper intends to make aneffort to explain a special form of consumer behaviorpertaining to athletes on display through some form ofmass media and how social media enhances this behavior.References are available upon request.

For further information contact:Mujde Yuksel

University of Massachusetts, Amherst20 Hampton Ave, #Apt 206Northampton, MA 01060



THE EFFECT OF SPORT EVENT ADVERTISING ON BRAND ATTITUDE

Wonseok Jang, University of Florida, GainesvilleYong Jae Ko, University of Florida, Gainesville

Songhyun Cho, Pusan National University, South Korea

SUMMARY

To increase advertising effectiveness, event market-ers put substantial effort into the selection of advertisingmessages used for each target group based on their psy-chological differences (Ruiz and Sicilia 2004). Previousliterature has suggested that both types of product (Vaughn1980) and consumers’ involvement level (Petty, Cacioppo,and, Schumann 1983) being significant variables in theselection of advertising appeals to improve attitudetoward the brand (sport event), however insufficientempirical evidence exists to fully understand each condi-tion. Accordingly, the purposes of the current study wasto examine the potential moderating effect of consumer’sinvolvement level and sport event type in the formation ofsport event attitude.

In the context of marketing and advertising, theplanning model suggests that hedonic products are mosteffectively communicates through emotional appeals,while, utilitarian products are most effectively communi-cated through informational appeals (Puto and Wells1984). Although, this planning model is reasonable insupporting the effectiveness of these two different mes-sage strategies, in the context of service, several scholarsfound that when consumers process their information forhedonic consumption such as sport activities, they uti-lized all available messages since their motivation derivedfrom fun and enjoyment compared too utilitarian (i.e.,document and visualization; Hill et al. 2004; Wakefieldand Inman 2003). Furthermore, psychology researchersfound that consumers’ attitude formation process wouldbe different based on their involvement level (Petty et al.1983). Under high-involvement conditions, consumers’utilize more cognitive processing and focus more on theinformational advertising to form their attitude. Mean-while in low-involvement conditions, individuals are likelyto form their attitude with less effort and time (affectiveprocessing), and pay more attention to emotional tone ofmessage (Petty et al. 1983). However, recent studies havefound that cognitive processing can co-occur with affec-tive processing (Ruiz and Sicilia 2004). Although, theauthors found that both product type and consumers’involvement level moderate sport event consumers’ atti-tude formation, no existing empirical studies fully explainthose conditions.

This study employed a 2 (Sporting event: utilitarianvs. hedonic) x 2 (Advertising: emotional vs. informa-tional) x 2 (Involvement: high vs. low) factorial between-group experimental design. The utilitarian sporting eventselected for the experiments was sports job fair and thehedonic event was a golf tournament. The two advertisingappeals employed in the present study were emotional andinformational. Additionally, purchase decision involve-ment (PDI) was used for involvement manipulation (Mittal1989).

The results of current study indicated that for utilitar-ian sporting event, attitude formation was equal acrossadvertising appeals under low involvement, while, highlyinvolved subjects responded favorably to informationalthan emotional appeals. On the other hand, for the hedonicsporting event, under low-involvement conditions, sub-jects rated higher for emotional than informationalappeals on attitude toward the sport event, whereas,highly involved subjects responded favorably to bothinformational and emotional appeals.

The general implication of our study was that thechoice of advertising appeal for sports event advertisingcan make a substantial difference in developing consum-ers’ sports event attitude. The results of our study revealedthat both the dual-process and planning model perspec-tive are appropriate in evaluating hedonic sporting event(Petty et al. 1983; Vaughn 1980). Thus, the authorssuggest that the traditional advertising planning modelmust shift to the co-occurrence approach for use withhighly involved consumers who are forming their attitudetoward the hedonic sporting event. However, in terms ofthe utilitarian sporting event, consistent with the dual-process model, highly involved individuals develop theirsporting event attitude by using the cognitive processing(Petty et al. 1983). However, types of message appeal didnot affect low-involvement consumers’ attitude forma-tion (Samuelsen and Olsen 2010). Overall, current studyoffered numerous meaningful managerial implicationsfor creative advertising agencies. Particularly, advertis-ing agencies can develop favorable consumer attitudetoward their sport event by using different message strat-egies for highly involved consumers. References areavailable upon request.


For further information contact:Wonseok Jang

Department of Tourism, Recreation, and Sport ManagementUniversity of Florida

186A Florida Gym P.O. Box 118208Gainesville, FL 32611–7150

Phone: 352.392.7588Fax: 352.392.7588



BRANDING HIGHER EDUCATION THROUGH PSAS: PRODUCINGGLOBAL CITIZENS FOR THE 21ST CENTURY

Michael J. Clayton, American University, WashingtonKevin V. Cavanagh, College of William & Mary, Williamsburg

SUMMARY

While prices for advertising time during collegeathletics continue to rise, the custom of networks allottinga single ad unit to each of the competing schools hasremained in place. Most college sport fans have grownaccustomed to seeing these thirty second Public ServiceAnnouncements (PSAs) sprinkled in among the clutter ofFortune 500 marketers. The placement of these PSAs insuch prominent media, adjacent to Madison Ave pro-duced commercials has forced American universities toinvest substantial resources in the creation of these videos,which is justified by the free exposure provided throughthis valuable airtime. Recent reports estimate that nearlya quarter of the U.S. population follows college footballregularly (Silver 2011), thus making the audience forthese PSAs extremely diverse, unlike the majority ofinstitutional communications which are clearly targeted atpotential students or alumni.

This research extends work by Tobolowsky andLowery (2006) and Harris (2009) to identify and examinemessage strategies and executional devices used by NCAAFBS (Football Bowl Subdivision) universities in theirPSAs. During the 2010 season college football season theresearches recorded and solicited PSAs for the purpose ofthis study. These collection methods, which are detailed inClayton, Cavanagh, Hettche (in press), yielded activePSAs for 110 of the 120 schools which comprise theNCAA FBS category. A content analysis was performedon the 110 PSAs, with each video being coded by twocoders, with a third coder serving as tie-breaker to resolveany coding discrepancies. Sixteen variables commonlyidentified in institutional messaging were defined in acodebook and were identified as being either present orabsent, from an audio and visual standpoint, in each of thevideos in the dataset. These institutional specific variablesincluded; scenic beauty, students in a classroom, indi-viduals in a lab, NCAA athletics, intramural sports, per-formance arts, graduation imagery, alumni of distinction,faculty of distinction, history/nostalgia, belonging, inter-national reach, study abroad, student scholars, andvolunteerism. Intercoder reliability was satisfactory(Cohen’s kappa > .40) (Fleiss 1981) for all of the variablesexamined in this study.

The authors used the official U.S. News & WorldReport data to perform a median split on the sample to

separate the top schools (“high-rank”) from the bottomschools (“low-rank”). Fifty-six schools were allocated tothe “top half,” while 54 schools resided in the “bottomhalf,” academically. A chi-square analysis was then per-formed to identify what differences existed between thetwo bodies. This analysis identified five variables that hada greater presence in the PSAs of higher-ranked institu-tions: “international reach” and “volunteerism” beingsignificant at p < .01, “faculty of distinction” significant atp < .05, and “study abroad” and “history/nostalgia” mar-ginally significant at p < .10. While the explanation for“faculty of distinction” and “history/nostalgia” may beself-evident, the other three variables are a bit moresurprising given that over 90 percent of all higher educa-tion institutions in the U.S. offer study abroad opportuni-ties (Green, Luu, and Burris 2008).

In the two previous qualitative studies examininginstitutional PSAs, Harris (2009) failed to make anyobservation or statements about international images ofglobal statements, while Tobolowsky and Lowery (2006)mentioned the frequent presence of “changing the world”messaging in their samples during the 2003-2004 and2004-2005 bowl seasons, but collapsed this messaginginto their theme of “embracing the future.” Given that“international reach” was present in over a third of thevideos in this study, and was the fifth most commonvariable present among the 110 PSAs viewed, the omis-sion of this message as a stand-alone theme was a bitsurprising. Given that “international reach” and “studyabroad” are also more prevalent among higher rankedacademic institutions, we believe that this topic deservesfurther discussion, specifically as it relates to extanttheoretical models on institutional messaging.

Extensive research has been conducted on the prolif-eration of study abroad programs in the U.S. and the focuson “global citizenship” among the modern universities(Lewin 2009). We believe the prevalence of “interna-tional reach” and the focus on this and “study abroad”programs by higher-ranked institutions in the U.S. signalsa fundamental shift in the mission/purpose of top univer-sities. Harris (2009) identified mission/purpose as a com-mon theme in institutional PSAs and we see “globalcitizenship” as an evolution of what it means to be a topinstitution of higher education in the U.S.. Future researchshould seek to explore the varying ways in which institu-tions convey this message in their institutional communi-cations. References are available upon request.


For further information contact:Michael J. Clayton

American UniversityKogod School of Business

4400 Massachusetts Avenue NorthwestWashington, D.C. 20016



A CUTTING EDGE APPROACH TO ACHIEVE VIVID DESTINATIONFOR SPORT SPONSORSHIP

Alireza Faed, Curtin University of Technology, AustraliaAfsaneh Ashouri, Curtin University of Technology, AustraliaMorteza Saberi, Curtin University of Technology, Australia

ABSTRACT

Marketing has been always the answer to the compa-nies as; using that, one can rectify and satisfy the needs.The only thing that companies try to stick with is tocompete with their competitors in the marketplace oftoday in a way to win them. Moreover, not only do theyfulfill the requirements of the clients, they must gratify theneeds of their own employees, as well. The purpose of thisstudy is to acquire the knowledge of advantages anddisadvantages of engaging in sports sponsorship activi-ties in electronic marketplace domain. To this aim, theresearcher chose some research questions and three quali-tative case studies have been conducted. The sampleorganizations which questionnaires and interviews havebeen conducted are Iranian. This research clarifies themost significant advantages and disadvantages of sportsponsorship in electronic markets. It is found that imagemaking, public awareness, the brand’s visibility, publicityimage and building good-will are some of the advantages.In addition, bad-will risks, ambush marketing, lack ofstandardization and evaluation are of disadvantagesinvolved in sponsorship procedure which need to beaddressed in sport sponsorship processes. A novelapproach is proposed in present study in order to rankingthe suitability of different industrial companies for adestination of sport sponsorships and the way to targettheir final clients. Twelve pros and six cons as a standardfeature for proper destination in sport sponsorship domainare defined. Positive image association, Eliminate medianoise, Generate awareness, Leverage public preferences,Leverage public preferences, internal motivation, Good-will, Same theme in diverse media, Create exclusivity,Create pride, Difficult to avoid target audience, Createpublicity and Cost effective are mentioned advantages.Also, Ambush marketing, Controversies and negativeattitude, No standardization, Inflexible, Difficulties inevaluation are disadvantages in sport sponsorshipdomain. Three companies are used in order to show theapplicability of proposed approach as a case study. Parselectric as a famous and distinguished Iranian manufac-turer of electronic device fields is considered as the firstcase. The second and third cases are LG and Samsungwhich are well known and pioneer companies in theworld. This is the first study that considers electronicdevice manufacturing companies as a competitive desti-nation for sport sponsorship. Also, regarding to the litera-

ture numerical approach has not been proposed in sportsponsorship domain. Key words: Marketing; Sport Mar-keting; Event Marketing; Sponsorship; Sport sponsor-ship.

SIGNIFICANCE

This paper will explore, describe and possibly start toexplain how Iranian and International companies whichwork in Iran are involve in sports sponsorship activities.The research purpose is partly exploratory, since weexplore our purpose in order to gain a deeper understand-ing of how Iranian and international companies engage insponsoring of sport. There have been quite a fewresearches in this field which companies launch theirsponsorship activities with a single or group of athletes,particular brand or service. However, this is the first studythat considers sport sponsorship as a main marketingcommunication tool in electronic industry.

INTRODUCTION

In cutting edge marketing, organizations are compet-ing strongly to hold existing and to gain new marketshares. To this aim, marketing communication is of highersignificance (Brassington-Pettit 2000). The increased glo-balization has made the marketing mix a necessary ele-ment for organizations to become successful or even tostay in business (Hollensen 2003).

The marketing mix involves four components whichare price, place, product and promotion. The promotionelement is often recognized as a tool which marketers useto communicate with their specific target markets (Jobber2001). According to Vignali (1997), promotion was for-merly comprised of advertising, public relation and salespromotion. However, as sponsorship has leveled-up inimportance, it is engaged as an element of the promotionmix (ibid). It is important to keep the elements in thepromotion mix apart, since they are considerably differ-ent. One difference between sponsorship and advertisingis for example that sponsorship is an indirect promotionaltool and viewed as bringing benefits to society, whileadvertising by contrast is a more direct and forcefulcommunication tool, perceived as beneficial to the com-pany itself (Meenaghan 2001b). Lardinoit and Derbaix(2001) further claim that there is an increased belief that


(2001b) Lardinoit and Derbaix (2001) further claim thatthere is an increased belief that sponsorship is becominga substitute for traditional advertising. In the past, compa-nies sponsorship as an obligation toward the community.Sponsorship was associated with charitable donationsand public relation opportunities (Gwinner 1997). Today,however, companies perceive sponsorship as a marketingcommunication tool, intended to increase the brand andservice image of the company(Madrigal 2001). The defi-nition of sponsorship as perceived by Meenaghan (1983)is “the provision of assistance either financial or in-kindan activity by a commercial organization for the purposeof achieving commercial objectives.” The outstandinggrowth of perceived sponsorship effectiveness hasresulted in increased investments in sponsorship activi-ties among companies all around the world (Meenaghan2001a). The worldwide sponsorship expenditure increasedfrom US$2 billion in 1984 to US$23.16 billion in 1999(Meenaghan 2001b). Sport sponsorship has become aneffective marketing strategy for many corporations and anequally effective revenue producer for sport organiza-tions during to differentiate their products and servicesfrom the growing number of advertisers in the marketplace, and to get more return for their promotional dollars(O’Neal, Finch, Hamilton, and Hammonds 1987).

THE HISTORY OF SPORT SPONSORSHIP

Sponsorship has come to the world in 1600 whichwas the period of patronage followed by the advent ofadvertising in 1632 that was a great step in the enhance-ment of modern day sponsorship. During 1910 to 1970 theearly pioneers came to the world of sponsorship that firstsupports radio programs, motor sports and golf. From1970 to 1984 a real enhancement illustrated and althoughmany managers refused to accept this phenomenon butsport sponsorship come to the sport as an inclusiveapproach. The decade of 90 was the era of value added andsponsors due to the latest technology became moresophisticated. Finally, the technological era which hasinitiated from 2000, and is in the process currently (Wiley2003).

Sponsorship

Sport sponsorship is an evolving area of interest toacademicians and business practitioners. In spite of recentenhancements, scholarly reviews of sponsorship attest toa lack of underlying theories and conceptual foundationson which to base empirical questions to build this proce-dure (Sam, Batty, and Dean 2005). Sponsorship of largesporting and cultural events has become a significantmarketing communication tool, specifically when com-panies acquire exclusive rights and accumulate the hypeassociated with this honor Sponsorship is an increasinglypopular marketing communication tool. Sponsorship is

one of the main elements of the communication mix wherea company prepare some financial support to an entity,which may be an individual, an organization or a group, inorder to allow this entity to pursue its activities and, at thesame time, advantage from this association in terms ofglobal image and consumer awareness of the firm’s mar-ket offerings (Pitt, Parent, Berthon, and Steyn). Meenaghan(1983) reports that the practice of sponsorship is quitedeveloped and has become a standard component of themarketing plan. As sport becomes an ever-more popularentertainment medium for TV and radio broadcasters,sponsorship becomes an ever more popular way for largecorporations to market themselves: it is a very effectiveway of associating their brand or identity with the valuesrepresented by a particular sport. As professional sportbecomes increasingly driven by money (players demand-ing high wages, equipment needing to be the latest tech-nological advance) then sponsors seem to be the only wayto provide the cash necessary to keep the sport running tothe standards that TV viewers demand. Olkkonen et al.(2000) propose that much of current research is empiri-cally driven and shows a serious lack of theory develop-ment. Sponsorship which Tripodi (2001) claims, has thepotential to become the marketing communication tool ofthe twenty-first century is a form of promotion that shouldbe managed on strict commercial lines.

The Role of Sponsorship in the Promotion Mix

Sponsorship normally practices as a facilitator in themiddle of Information and Interaction process and as,both of the processes are vital in organizations, companiesmust concentrate on various ways to absorb data and turnthem into legitimate information. What is more, it is bestto focus on interaction and different ways of optimizinginteraction methods within the company. While the com-pany wants to launch the process of sponsorship theyshould initiate informing different clients via advertisingand different channels of sales promotion. Finally, thecompanies got to prepare an appropriate image for them-selves and best ways to supply their products and services(Eriksson and Hjalmsson 2001).

Problem Definition

Over the past decades, sport sponsorship has beenshown to be an effective tool with which to alter andenhance a company’s image and reputation. The use ofsport sponsorship is becoming more and more common.Most of the companies are sponsoring sport event orteams due to the fact that the teams and events tend to drawa very variable audience and also they can introduce theirlatest products to the market by advertising. However, assport sponsorship is nowadays widespread among com-panies, companies are not aware of advantages and disad-vantages engaged in sport sponsorship process. The ques-


tion is how the companies, in sports sponsorship activi-ties, perceive and describe the advantages and disadvan-tages?

Demarcations of this Study

Due to the time limitation, it is beyond the scope ofmy study to cover all the aspects of the research purpose.So, I restricted my research to answer my research ques-tion from the sponsoring companies’ perspectives. Mysample companies include the domestic and internationalfirms, which are working in Iran.

Objectives of Sponsorship

A common objective for sponsors is to create a closeassociation between the brand or the company and theevent sponsored (Quester and Farrelly 1998). Sponsor-ship objectives can be divided into two groups: namelydirect or indirect objectives. Direct sponsorship objec-tives pay attention to sales increase and have a short terminfluence on customer attitude. Indirect sponsorshipobjectives are, on the other hand, objectives that ulti-mately result in the desired goal of increasing sales, suchas for example to meet competition, build relations, createawareness, improve the image, as well as to get in touchwith new target markets (Shank 1999). Other researchershave also described objectives for corporations involvingin sponsorship, and these will therefore be presentedbelow.

Public awareness: This objective is mainly used toincrease awareness of a company or its products, in otherwords, its aim is to generate exposure (Mullin 2000). Thisis further strengthened by Meenaghan (1983), who statesthat sport sponsorship has been widely employed toincrease a company’s public awareness.

Corporate objectives: Maintaining and image build-ing is one the most outstanding objectives of sport spon-sorship. The opportunity for sponsors is to associate thecompany or its brands with the positive images producedby the exclusive characteristics of the sporting events.Before deciding upon an event, it is important to conductan analysis of the unique features of the event, as well asof the product or brand (Shank 1999).

Public perception: By connecting itself to a specificsport or event, a company can benefit from image associa-tion or image transfer. The choice of the sponsored activ-ity can help to improve a company’s image, as well asreinforce or change the consumer’s perception of thecompany and its product (Mullin, Hardy, and Sutton2000). Client entertainment: The opportunity to hostclients at an event, especially those for which it is almostimpossible to get tickets, sometimes pays for the entire

sponsorship (Skinner and Rukavina 2003). Employeerelations: According to Meenaghan (1983), sponsorshipcan enhance relations with employees, as the engagementcan result in that employees feel proud of belonging to thecompany.

Reach target market: The main aim of marketingobjective with sport sponsorship is to reach target mar-kets. Sport sponsorship has the capacity to reach peoplethat share a mutual interest sports. The opportunities forsponsors are to target disable people and women thatparticipate in sports (Shank 1999).

Increase sales: Sponsorship itself with other promo-tional activities is usually seen as an element to influencethe buyers to purchase a product or service. At a result, thefinal marketing goal for firms is to raise their sales orprofitability. To increase the sales is an objective that allsponsoring companies have. Companies would not spendtheir money on a specific sport s event if they did not feelthat they would get some kind of return on their invest-ment (Shank 1999).

Compete with other companies: In a long term com-mitment or when the sponsorship fee is high, companiescan create product or category exclusively. This impliesthat a particular product is the only of its kind to beassociated with the event (Shank 1999).

RESEARCH APPROACH

For this research, we have employed a deductiveresearch approach. We have based the empirical studywork on already existing theories and models, which Ilater will compare with the reality. This way of approach-ing a study corresponds to what the theory calls a deduc-tive research approach. For this thesis, I have employed adeductive research approach. I have based the empiricalstudy work on already existing theories and models,which I later will compare with the reality. This way ofapproaching a study corresponds to what the theory callsa deductive research approach.

Step 1: Data Collection

For this paper, archival records, observation, andartifacts are not utilized since they do not fit into the copeof the study (i.e., Archival records are precise and quan-titative which does not fit a qualitative study, observationsare too expensive and time consuming, as well as they maybe biased, and the study does not need evidence oncultural features, thus artifacts are excluded). Conse-quently, the methods of data collection in present paperare interviews, which is considered primary data, anddocumentation, which is secondary data. Interviews aresuitable when in-depth information is desired, and this


kind of information is required in order to fulfill thepurpose of this research. The detail of interviews processand raw questionnaires are explained in Appendix I.

Step 2: Data Analysis

The data collected for each research question isreduced and simplified through use of within caseanalysis.

The within case analysis is performed by comparingthe empirical findings with the already existing theoriesand concepts brought up in the conceptual framework.

A cross-case analysis is also performed through dis-playing and reducing the data from each research questionand comparing it between the two selected cases in orderto find similarities and differences in between them.

Once the data reduction and the data display are donethrough within- and cross-case analysis respectively, con-clusions on each research question are drawn based on thefindings of the study.

Numerical Data Analysis. Regarding to the obtainedresult from above step, the numerical analysis will bedone. The yes answers in advantage status are consideredas an advantage raw score for a given company, then forscaling the score, raw score is divided by total advantagefeatures. Similarly, the negative answers in disadvantagestatus are considered as a disadvantage raw score for agiven company. Scaling operation will be done similar toabove. Following formulas show the calculation of rawand scaled scores in both advantage and disadvantagestatus.

SA = NPA/NASDA = NNDA/NDATSS = SA+SDAThat:SA: Score of sport sponsorship on selected case basedon advantage features (SA).NPA: Number of positive advantage in selected case.NA: Number of advantage features.SDA: Score of selected case based on disadvantagefeatures.NNDA: Number of negative disadvantage in selectedcase.NDA: Number of disadvantage featuresTSS: Total scaled score.

Above formulas are utilized to compare chosen com-panies and finally select the best destination for a givensport sponsorship. Some of the sport sponsorship compa-nies have not equal preference for both advantage and

disadvantage features; therefore, it is better to improveabove formulas to consider their preference. Followingformulas shows the improved version:

TWSS = WA*SA+WDA*SDAThat:WA: preference of given sport sponsorship foradvantage features.WDA: preference of given sport sponsorship fordisadvantage features.TWSS: total weighted scale scored.

Determining critical advantages and disadvantagesfeatures: After calculating the rank of destination compa-nies, determining the features that cause mentioned ranksis important and useful from management view. Deter-mining is done in two ways for advantage and disadvan-tage section. For each feature in advantage section, thenumber of ‘no’ in destination companies is the grade offeature. Finally, the features are ranked regarding to theirgrades and critical features is determined as well. Aboveprocedure is done in disadvantage section with this differ-ence that use ‘yes’ for calculating grade of given feature.

CASE STUDIES

Case 1: Pars Electric

Pars electric Mfg Co. was established in 1963 andlocated in Tehran. Pars Electric’s activities were begunwith the objective of establishing an industrial factory forthe assembly of TV and radio sets. Today it cooperateswith some well known companies in the world and ben-efitting from the latest technology. www.parselectric.com.

Sport sponsorship increase the visibility of the brandincreasing the sole and creating the development of imagein every art that is why firm wants players and coaches towear its brand one of the most important advantages of thefirm is that it generate the perception of the brand imageas well also to create an association to the sport. Also,sport sponsorship can eliminate media noise better thanother promotional tools. Other very important advantagesof sport sponsorship are creating publicity and it buildsgoodwill which shows positive attitudes and feeling to-ward a company. Controversies are regarded as disadvan-tages. He claims that if they have not standardization theywill have a major disadvantage. Also he mentioned thatinflexibility as a disadvantage, so we must consider thesponsorship as a flexible promotional method as long asthere are so many sports and leagues that need to besponsored. Additionally he says that, there is no difficultyfor evaluation because there are many ways for measure-ments like surveys and estimating the sales. Also, some-times the bad will risks make the sponsors to drawback.


Case 2: Samsung’s Perceived Advantages and Disad-vantages of Sport Sponsorship

Samsung is a Korean company which was founded in1938 it has go about 1200 employees all around the worldof who about 450 people work in Iran. About 10 percentof the entire marketing budget allocates to the sportsponsorship. www.samsung.com.

Media visibility, building an image and reaching tothe target market as well as competing with vigorouscompetitors. The company has chosen the repetitiveexposure to increase its brand’s image and achieve apositive image. Elimination of media noise is anotheradvantage of sponsorship. Brand awareness is anotherimportant advantage which will meet by sponsoring dif-ferent kinds of sports. The marketing manager claims that,these days a big worry for sport sponsorship is the mana-gerial problem in Big and prominent Teams (A-Teams).Ambush marketing and controversies are two disadvan-tages that the marketing managers are encountering thesesdays. Lack of standardization is another major disadvan-tage that provides some bans in sponsorship process.

Case 3: LG’s Perceived Advantages and Disadvan-tages of Sports Sponsorship

LG Electronics, Inc. was established in 1958 as thepioneer in the Korean consumer electronics market. Thecompany is a global force in electronics and informa-tion and communication products with more than66,000 employees working in 16 subsidiaries in 39 coun-tries around the world. In Iran headquarter is located inTehran and the marketing section comprise of 30 employ-ees. The company’s turnover in 2004 was US$38 billion.www.LGE.com.

Sponsorship procedure generates media visibility,image making and reaching the target market. One of themajor advantages of the company is building a positiveimage association. Other very important advantages arebuilding goodwill same theme in diverse media and createexclusivity, as the company tries to create unique methodsto publicize their products and while the brand onlyinvolves as a main sponsor of variety of sports. Addition-ally, the public preferences as an advantage as well asreceiving internal motivation, generating publicityalways is an advantage for the firm and has a positiveeffect on the brand, also they should be careful that, toomuch advertising may, sometimes, decreases the valuesof the brand, and its image. He says that cost – effectivepromotional method is an advantage for LG. Negativepublicity and too much advertisement in inappropriateplace create negative image for the firm. According to themarketing manager the ambush marketing has not consid-ered as a disadvantage for LG’s sponsorship activities.The brand sometimes faces with ambush marketing, aswell.

Case Analysis Regarding Advantages and Disadvan-tages of Sport Sponsorship:

Examination of Table 1 shows that in the majority ofthe factors respondents have the same opinion, exceptfrom being cost-effective which is regarded as a criticalcomponent in sponsorship bases.

Ambush marketing defined as attempts by competi-tors to exploit the event has also increased in prominence.Moreover, The term ambush marketing was first coinedby Bayless (1988) to describe the purposeful and falseassociation by a company not sponsoring an event towardthe end of deriving benefits similar to those affordedofficial sponsors (Pitt et al.).

TABLE 1Case Analysis of Pars Electric Pros

Number Advantage Situation

1 Positive image association Yes2 Eliminate media noise Yes3 Generate awareness Yes4 Leverage public preferences Yes5 Internal motivation Yes6 Goodwill Yes7 Same theme in diverse media Yes8 Create exclusivity Yes9 Create pride Yes10 Difficult to avoid target audience Yes11 Create publicity Yes12 Cost effective No


TABLE 2Case Analysis of Pars Electric Cons

Disadvantages Situation

1 Ambush marketing Yes2 Controversies and negative attitude Yes3 No standardization Yes4 Inflexible No5 Difficulties in evaluation Yes6 Time consuming Yes

TABLE 3Case analysis of Samsung Pros

Advantage Situation

Positive image association YesEliminate media noise YesGenerate awareness Yes

Leverage public preferences YesInternal motivation Yes

Goodwill YesSame theme in diverse media Yes

Create exclusivity YesCreate pride Yes

Difficult to avoid target audience YesCreate publicity YesCost effective No

TABLE 4Case Analysis of Samsung Cons


Ambush marketing NoControversies and negative attitude No

No standardization YesInflexible Yes

Difficulties in evaluation YesTime consuming Yes


TABLE 5Case Analysis for LG Pros

Advantage Situation

Positive image association YesEliminate media noise YesGenerate awareness Yes

Leverage public preferences YesInternal motivation Yes

Goodwill YesSame theme in diverse media Yes

Create exclusivity YesCreate pride Yes

Difficult to avoid target audience NoCreate publicity NoCost effective Yes

TABLE 6Case Analysis for LG Cons


Ambush marketing NoControversies and negative attitude Yes

No standardization YesInflexible Yes

Difficulties in evaluation YesTime consuming No

TABLE 7Score of Destination Companies Based on Advantage and Disadvantage Features

Pars Electric Samsung LG

SA 0.91 0.91 0.83SDA 0.16 0.33 0.33TSS 1.08 1.25 1.16

TWSS

TABLE 8Expert Priority of Advantage and Disadvantage Impact on Destination Company

WA WDA

0.6 0.4


Critical advantages and disadvantages features:

Table 9 and 10 shows the critical features in both advan-tage and disadvantage section.

Exploring of Table 9 shows that the grade of costeffective feature is the highest grade among other advan-tage features. Therefore, destination companies need toimprove . . . in theirs organization. More assessing theTable shows that Create publicity and Difficult to avoidtarget audience with grade 1 is other important advantagefeatures. Thus, it is an acceptable strategy to improvethemselves in mentioned features. Finally, other advan-tage features regarding to present case studies has beenequal.

Exploring of Table 10 shows that the grade of Nostandardization, Inflexible and Difficulties in evaluationfeatures are the highest grade among other advantagefeatures. Therefore, destination companies need to im-prove . . . in theirs organization. More assessing the Tableshows that Controversies and negative attitude with

grade 2 is other important advantage features. Thus, it isa acceptable strategy to improve themselves in mentionedfeature.

RESULTS AND CONCLUSIONS

The results and outcomes of this research have ob-tained from above tables. The main common advantagesare that it increases the visibility also the popularity ofsports that influence on the companies and attract them tosponsorship involvement. The three sample cases regardfollowing Items as advantages involve in sports sponsor-ship: – Builds positive image association – Eliminatesmedia internal motivation-Build goodwill – same themein diverse media – creates exclusivity and publicity andthese advantages are the consequence of preferences forthe companies products against other competitors. More-over, Pars Electric and Samsung consider that creatingpride and difficult to avoid target audience as advantages.In addition Samsung and LG consider the cost effectivepromotional method but Pars Electrics reject this idea,

TABLE 9Critically Grade of Advantage Features

Cost effective 2Create publicity 1

Difficult to avoid target audience 1Positive image association 0

Eliminate media noise 0Generate awareness 0

Leverage public preferences 0Internal motivation 0

Goodwill 0Same theme in diverse media 0

Create exclusivity 0Create pride 0

TABLE 10Critically Grade of Disadvantage Features

Disadvantages Grade

1 No standardization 32 Inflexible 33 Difficulties in evaluation 34 Controversies and negative attitude 25 Time consuming 16 Ambush marketing 0


CROSS-CASE ANALYSIS

In this part the cases that has been investigated, will be compared through a cross case analysis:

Advantages Pars Electric Samsung LG

Main Advantages` Increase the` -Media visibility Generatevisibility -Builds an image media visibility

-Developing the association -Image makingvisibility -Generates -Reach to the

-Popularity of sport awareness target market-Popularity of -Building

sport positive imageassociation

-Popularity ofthe Sports

Builds positive image association Yes Yes YesEliminate media noise Yes Yes YesGenerate awareness Yes Yes Yes

Influence public preferences Yes Yes YesAchieves internal motivation Yes Yes Yes

Build goodwill Yes Yes YesSame theme in diverse media Yes Yes Yes

Creates exclusivity Yes Yes YesCreates pride Yes Yes Yes

Difficult to avoid target audience No No NoGenerates publicity No No No

Disadvantages Pars Electric Samsung LG

Main Disadvantages -Restrictions from -Inflexible -Negativeclubs or teams -No standardization publicity because

-Ambush marketing -Negative image of too much-Negative image association advertisement

association -Negative image-unwanted and badbehavior of the fans

-Lack of cliententertainments

Ambush marketing Yes No NoControversies and negative attitude Yes No Yes

No standardization Yes Yes YesInflexible No Yes Yes

Difficulties in evaluation Yes Yes YesTime consuming Yes Yes No

Ambush marketing Yes No NoControversies and negative attitude Yes No Yes


because of some difficulties in effectiveness of sportsponsorship measurement. Regarding disadvantagesinvolved through the process, negative image associationis a main disadvantage in the process of sponsorship. Thisis some how because of the Bad-will risks and inappropri-ate advertising. Ambush marketing is one the disadvan-tages that is exclusive for pars electric as a domesticcompany. Regarding the controversies and negative atti-

tude involved in sports sponsorship in Electronic Indus-tries, LG considers it as a Major drawback and ParsElectric Regards it as a disadvantages and put less empha-sis on it. Also LG, Samsung, and Pars Electric found thelack of standardization as a common Major disadvantageamong three companies, but put less emphasis on ParsElectric. Ultimately, Difficulties in evaluation were alsoperceived as a disadvantage in three companies.

REFERENCES

Bennett, R. (1999), “Sports Sponsorship, Spectator Re-call and False Consensus,” European Journal ofMarketing, 33 (3/4), 291.

Cornwell, B. (1995), “Sponsorship-Linked MarketingDevelopment,” Sport Marketing Quarterly, IV (4),13–24.

____________ (1997), “The Use of Sponsorship-LinkedMarketing by Tobacco Firms: International PublicPolicy Issues,” Journal of Consumer Affairs, 31 (2),238–54.

Faed, A. (2007), “Strategic Assessment of Sport Sponsor-ship as a Marketing Communication Tool in Elec-tronic Industries of Iran,” [epubl.ltu.se].

Gwinner, K. (1997), “A Model of Image Creation andImage Transfer in Event Sponsorship,” InternationalMarketing Review, 14 (3), 145.

Magnus Hultman, Olof Lindgren-Sport sponsorship inSweden (2001).

Meenaghan, J.A. (1983), “Commercial Sponsorship,”European Journal of Marketing, 17 (7), 5–73.

Meenaghan, T. (1991), “Sponsorship – Legitimising theMedium,” European Journal of Marketing, 25 (11),5.

____________ (1991a), “The Role of Sponsorship in theMarketing Communication Mix,” International Jour-nal of Advertising, 10 (1), 35–47.

____________ (2001), “Sponsorship and Advertising: AComparison of Consumer Perceptions,” Psychology

& Marketing, 18 (2), 191.____________ (2001), “Understanding Sponsorship

Effects,” Psychology and Marketing, 18 (2), 95–122.Mullin, B.J., S. Hardy, and W.A. Sutton (1993), Sport

Marketing. Champaign, IL: Human Kinetics Pub-lishers.

Pitt, L., M. Parent, P. Berthon, and P.G. Steyn, “EventSponsorship and Ambush Marketing: Lessons fromthe Beijing Olympics,” Business Horizons, In Press,Corrected Proof.

Pope, N. (1998), “Overview of Current SponsorshipThought,” The Cyber Journal of Sport Marketing, 2(1). Available at: [http://Pandora.nla.gov.au/parchive/h1998-sep-2].

Quester, P.G. and B. Thompson (2001), “Evidence of theImpact of Advertising and Promotion Leverage onArts Sponsorship Effectiveness,” Journal of Adver-tising Research, 41 (1), 33–47.

Sam, M.P., R. Batty, and R.G.K. Dean (2005), “A Trans-action Cost Approach to Sport Sponsorship,” SportManagement Review, 8 (1), 1–17.

Shank, M.D. (2004), Sports Marketing: A Strategic Per-spective. New Jersey: Prentice Hall.

Skinner, Bruce E. and Vladimir Rukavina (2003), EventSponsorship. John Wiley & Sons, Inc.

Speed, R. (2000), “Determinants of Sports SponsorshipResponse,” Journal of the Academy of MarketingScience, 28 (2), 226.

Taranto, R. (1998), “Special Events: A Unique Tool ofMarketing and Promotion,” Business Date, 6 (1), 5.

APPENDIX I

Websites

www.Parselectric.comwww.samsung.comwww.LGE.com

Interviews

Mr. Kouhi, Marketing Manager of Pars ElectricMr. Shahram Bozorgi, Marketing Manager of LGMrs. Nazila Arbabiyan, Marketing Manager ofSamsung


For further information contact:Alireza Faed

Curtin University of TechnologyPerth, Australia

Phone: 0061425879722E-Mail: [email protected] / [email protected]


AUTHOR INDEX

Acevedo, Claudia Rosa 266Adjei, Mavis T. 401Agarwal, James 42Aiken, K. Damon 326, 353Akdeniz, M. Billur 289Albrecht, Arne K. 452Albrecht, Carmen-Maria 235Allen, Alexis M. 233Allmann, Jan 19Andrews, Demetra 208, 233Andrews, Michelle 345Anokhin, Sergey 347Argouslidis, Paraskevas 301Arif, Farrah 224Arnold, Mark 242, 470Arunachalam, S. 332Asare, Anthony K. 78Ashouri, Afsaneh 534Babakus, Emin 14Bacile, Todd J. 466Bäckström, Lars 511Bahnson, Sara K. 474Baker, Andrew M. 133Baltas, George 301Barber, Nelson A. 119Barretta, Paul G. 385, 403Bartl, Christopher 478Basuroy, Suman 296, 349Bauer, Hans H. 12, 235Bearden, William O. 236Becerra, Enrique P. 117Bechara, Antoine 205Bergner, Kai 235Beuk, Frederik 343Biemans, Wim G. 281, 508Bishop, Melissa M. 119Blair, Edward 233Bloch, Peter H. 424Blut, Markus 488, 492Bogoviyeva, Elmira 444Bone, Sterling A. 498Bornemann, Torsten 257Bose, Mousumi 240Bouchet, Adrien 448Boukis, Achilleas 454Brach, Simon 452Brady, Michael K. 506Brashear, Thomas G. 78Brettel, Malte 79, 96, 283, 391Bright, Laura F. 182, 416Brock, Meagan 121Brown, Brian P. 382

Bruns, Andrea 303Brusco, Michael 506Bulmer, Sandy 249Burke, Paul F. 194Burke, Sandra J. 194Burmann, Christoph 156Burmester, Alexa 299Bush, Alan J. 496Byun, Kyung-Ah 98Calantone, Roger J. 450Castafio, Raquel 205Castaldo, Sandro 328Cavanagh, Kevin V. 532Cavusgil, Erin 428Chaiy, Clarissa 271Chaiy, Seoil 271Chakravarti, Amitav 170Chapa, Sindy 117Chen, Cuiping 480Chen, Ivy S.N. 40Chen, Qimei 150Chen, Rong 100Cherian, Joseph 45Chiou, Jyh-Shen 370Cho, Songhyun 530Choi, Beomjoon 476Choi, Hyeong-Gyu 343Choi, Laee 405Chou, Szu-Yu 370Chu, Shu-Chuan 426Cicala, John E. 496Cicchirillo, Vincent J. 92Clark, Terry 364Claro, Danny P. 176, 527Claro, Priscila B.O. 176Claus, Bart 206Clayton, Michael J. 532Clement, Michel 299Coenen, Christian 484Cook, Laurel Aynne 262Cooley, Delonia O. 117Cornwell, T. Bettina 103Cui, Annie 260Cunningham, Nicole 416Czerwinski, David 355Dadzie, Charlene A. 47Dadzie, Kofi Q. 47Das, Sidhartha R. 464David, Meredith E. 236Dawes, John 17DeCarlo, Thomas E. 14, 330Deitz, George D. 14, 440, 496


Dennis, Noel 395Dewitte, Siegfried 206Dimitrova, Boryana V. 351Donthu, Naveen 133Dose, David 452Driesener, Carl 17Drumheller, Kristina 121Drumwright, Minette E. 92Eastlick, Mary Ann 178Eggers, Felix 10, 299Eggert, Andreas 490Ehrich, Kristine 109Eisingerich, Andreas B. 135Ekebas, Ceren 113Elbeck, Matt 430Elliot, Esi Abbam 45Evans, Robert, Jr. 440Evans, Kenneth R. 494Evanschitzky, Heiner 492Fabrize, Robert O. 238Faed, Alireza 534Falk, Tomas 297Fan, Pengdong 16Fan, Xiaoping 137Fang, Zheng 90Fitzgerald, M. Paula 260Flint, Daniel J. 311Ford, John 165Forkmann, Sebastian 74Foster, Mary 180, 270, 389Fox, Gavin L. 98France, Karen Russo 260Friesen, Daniel 365Froman, Adam 180Furchheim, Pia 186Garnefeld, Ina 303, 490Gaur, Sanjaya S. 500Gehrt, Kenneth C. 355Gerlich, R. Nicholas 121Geyskens, Kelly 206Giese, Andreas 478Gill-Simmen, Lucy 135Gironda, John T. 387Glibkowski, Brian 519Gogia, Tamara 235Gonzalez, Gabriel R. 527Gonzalez, Richard 407Gounaris, Spiros 454Gouthier, Matthias 478Gregory-Smith, Diana 169Gretz, Richard T. 349Greuling, Kathrin 482Griffith, David A. 25Grosso, Monica 328Groza, Mark D. 442Gu, Haodong 131

Guo, Lin 480Haas, Alexander 27Habel, Johannes 334Hammerschmidt, Maik 12, 21, 268, 397Hammes, Eva 482Hampson, Daniel P. 325Handrich, Matthias 294, 399Hansen, John D. 14, 330Hansen-Addy, Joseph 45Harmeling, Colleen 242Harris, Milton 296Hattula, Stefan 12Hawkins, Matthew A. 143He, Yi 150Heidenreich, Sven 294, 399Heinberg, Martin 46Heinrich, Daniel 235Henneberg, Stephan C. 23, 74Hennigs, Nadine 129, 152, 245Herjanto, Halimin 500Herm, Steffen 259Hiler, Jacob L. 212Hille, Patrick 452Hinsch, Chris 486Hirschman, Elizabeth C. 195Hofacker, Charles F. 466Hofmeister, Agnes 264Homburg, Christian 19, 257, 334Hoppner, Jessica J. 25Hoyer, Wayne D. 224Hsiao, Daniel F.J. 317Hsu, Chun-Kai Tommy 105Hulland, John 141Ivens, Björn S. 139Iyanna, Shilpa 244Izquierdo-Yusta, Alicia 315Jahn, Benedikt 384, 472Jahn, Steffen 186, 230Jang, Wonseok 530Jensen, Susan 290Jeong, Miyoung 468Jevons, Colin 154Jin, Yao 262Johnson, Jean L. 285Johnson, Jeff S. 338Johnson, Mark S. 439Joshi, Maheshkumar P. 464Jun, Chi-Hyuck 165Jung, Jae Min 222Kale, Sudhir H. 172Kaminakis, Kostas 454Kang, Jun 78Kanitz, Christopher 156Karande, Kira 368Karande, Kiran 113Karg, Adam 446


Kauri, Vamsi K. 345Kay, Mark J. 309Keith, Megan E. 90Kendall, David L. 195Kennett-Hensel, Pamela 210Keysan, Gulberk 170Kiessling, Tina 230Kilian, Thomas 245, 482Killian, Ginger 141Kim, Hye Young 379Kim, Junghwan 98Kim, Kyungdoc 165Kim, Sang Yong 36Kim, Seongjin 271Kingshott, Russel P.J. 462Kinsky, Emily 121Kishiya, Kazuhiro 371Klarmann, Martin 334Kleiser, Susan Bardi 182Ko, Yong Jae 530Köcher, Sören 488Kohlbacher, Florian 264Kolenda, Nicholas 519Korgaonkar, Pradeep K. 94, 387Kraemer, Tobias 478Krasnikov, Alexander 39Krishen, Anjala 102Krishnan, Vijaykumar 360Krupka, Zoran 43Ku, Edward 228Kulviwat, Songpol 364Kumar, V. 133Kunz, Werner H. 381, 384, 472Kusch, Sergey 39Lacey, Russell 210Laczniak, Russell 109Lan, Ping 34Landreth-Grau, Stacy 182Langer, Josef 43Langner, Sascha 152Larson, Jeffrey S. 498Lawry, Charles Aaron 405Leckie, Civilai 446Lee, Ruby P. 150Lee, Seonjeong (Ally) 468Lee, Sung-Won 379Lehnert, Kevin 470Leischnig, Alexander 323Lenz, Isabell 268Leung, Ada 290, 357Levin, Avner 270, 389Li, Shibo 439Liang, Beichen 111Lim, Chae-Un 81Liu, Feng 137Liu, Matthew T. 510

Llonch, Joan 393Look, Christoph 409Loveland, James M. 29Ludwig, Nadine 478Luk, Sherriff T.K. 40Lund, Donald J. 330Luo, Xueming 16, 90, 345Luthans, Kyle 290Lyes, Amy 249Magnusson, Peter 360Maity, Devdeep 313Malshe, Avinash 281, 508Mangleburg, Tamara 94Mangus, Stephanie M. 212Mantrala, Murali 336Manzon, Elliott 407Martínez-Ruiz, María Pilar 315Massiah, Carolyn 307Matthes, Joseph 340Mau, Gunnar 139McCarthy, Killian J. 51McDonald, Heath 446McGinnis, Lee 519McGoldrick, Peter J. 325McLarty, Benjamin D. 212Megicks, Phil 315Mende, Martin 484Meng, Juan (Gloria) 364Merlo, Omar 224Miao, C. Fred 494Michon, Richard 49Millet, Kobe 206Milne, George R. 382Min, Kyeong Sam 222Minor, Michael S. 385Minten, Bart 37Mishra, Anubha 178Mohanty, Praggyan (Pam) 226Möller, Jana 259Morgan, Todd 347Mouri, Nacef 464Mrozek, Alexander J. 297Mudambi, Susan M. 37Muehling, Darrel 109Muenkhoff, Eva 303Muniz, Marcello 266Munzel, Andreas 472Nagengast, Liane 492Naudé, Peter 23, 74Naveh-Benjamin, Moshe 226Neto, Silvio A. Laban 176Nguyen, Adam 72Nichols, Bridget Satinover 311Nippa, Michael 502Nohara, Jouliana Jordan 266O’Brien, Matthew 355


Oakes, Helen 395Oakes, Steve 395Oh, Joon-Hee 504Oneto, Stephanie 208Orr, Linda M. 343Painesis, Grigorios 301Palakshappa, Nitha 249Palmeira, Mauricio 154Pappu, Ravi 103Park, Jang-Ho 165Park, Jieun 362Pascal, Vincent J. 326, 353Peighambari, Kaveh 511Peltier, James 102Pentina, Iryna 107Peterson, Mark 442Petrescu, Maria 94Pick, Doreén 322Ping, Yuan 510Pontes, Nicolas 154Preacher, Kristopher J. 14Pronschinske, Mya 442Prostka, Tim 299Qiu, Tianjiao 359Qu, Riliang 354Radford, Scott K. 424Rajagopal, Priyali 174Rajagopalan, Kirti 332Rajan, Mahesh N. 355Raju, Sekar 174Ramaseshan, B. 462Ramaswami, Sridhar N. 332Raska, David 311Ratneshwar, S. 226Rauschnabel, Philipp A. 139Ravaja, Niklas 305Ravid, S. Abraham 296Reardon, Thomas 37Rebiazina, Vera 39Reich, Jens-Christian 96, 391Reimann, Martin 205Rialp, Josep 393Robben, Henry 393Root, Ann 94Rosenbaum, Mark S. 307Rosenbloom, Bert 351Roswinanto, Widyarso 167Rudolph, Thomas 492Rungie, Cam M. 1, 17Russell, La Toya M. 188Ruvio, Ayalla A. 184Ryu, Kisang 222Saberi, Morteza 534Saini, Amit 338, 340Satornino, Cinthia B. 506Sattari, Setayesh 511

Schade, Michael 156Scheer, Lisa K. 330Schielke, Thomas 318Schmidt, Steffen 129, 152Schmitz, Christian 31Schweitzer, Fiona 292Schwertfeger, Marko 322, 323Seifert, Colleen M. 407Seshadria, Sukanya 381Shainesh, G. 355Sharma, Piyush 40Shen, Chung-Chi 228, 370Sherrell, Daniel L. 496Shi, Guicheng 510Shin, Sohyoun 36, 326, 353Singh, Jatinder Jit 143Singh, Nitish 367Sinha, Ashish 131Sisodiya, Sanjay R. 285Sivadas, Eugene 439Smirnova, Maria 39Smith, Andrew 169Smith, Ronn J. 262Sollosy, Marc 121Somer, Eli 184Somervuori, Outi 305Song, Ji Hee 379Song, Yiping 16Spears, Nancy 231, 238Sprott, David E. 235Steinbacher, Eva K. 31Steinhoff, Lena 490Stierl, Marcel 257Stuebiger, Nina 27Styvén, Maria Ek 511Sudbury-Riley, Lynn 264Suh, Taewon 165Sullivan, Alexa K. 440Sun, Wenbin 247Sutcliffe, Alistair 74Swani, Kunal 115, 382Swilley, Esther 342Talay, M. Berk 289Talwar, Vishal 170Tam, Leona 105, 150Tang, Chuanyi 480Tang, Yihui (Elina) 336, 486Taube, Markus 46Tavani, Zhaleh Najafi 23Taylor, David G. 107Thompson, Scott 484Thorson, Esther 336Totzek, Dirk 19Trinh, Giang 17Ulaga, Wolfgang 29Uribe, Felipe 393


Van Steenburg, Eric 238Vander Schee, Brian A. 430Vicdan, Handan 76Vijayalakshmi, Akshaya 109Vinhas, Alberto Sa 33Vlasic, Goran 43Völckner, Franziska 10Voorhees, Clay M. 450, 506Wachter, Michael 362Wagner, Stephen M. 285Walsh, Gianfranco 452Wang, Di 74Wang, Liz C. 317Wang, Yonggui 510Warner, Anne 180Waschto, Andreas 79, 283Weiger, Welf 397Weinberger, Marc 115West, Bettina 270, 389Westjohn, Stanford A. 360Wetzel, Hauke A. 12, 21, 268, 397White, Ryan C. 450Wiedmann, Klaus-Peter 129, 152Wienhold, Doreen 502Winklhofer, Heidi 169Winston, Evelyn M. 47Wise, Chelsea 194Wozniak, Richard 307Wright, Malcolm 17Wu, Terry 42Wuestefeld, Thomas 129Xie, Jeffrey Hu 320Xu, Huimin 290, 357Yamamoto, Nao 371

Yan, Terry 357Yan, Tony 158Yang, Chun-Ming 228, 370Yang, Jing 78Yang, Song 213Yao, Qing 100Yaprak, Attila 365Yazdanparast, Atefeh 231Ye, Jun 287Ye, Lei 240Yi, Ho-Taek 81Yim, Mark Yi-Cheon 92, 426Yu, Jun 367Yu, Kevin 213Yuan, Xina 36Yuksel, Mujde 529Zablah, Alex R. 21Zaefarian, Ghasem 23Zahay-Blatz, Debra 102Zaichkowsky, Judith L. 205Zanger, Cornelia 186, 230Zboja, James J. 448Zdravkovic, Srdan 360Zehetner, Andreas 63Zhang, Jia 137Zhang, Nan 401Zhao, Xiaomin 34Zhou, Joyce X. 367Zhu, Tao 51Zielke, Stephan 318Zlatevska, Natalina 172Zou, Yueming 368Zupancic, Dirk 31

The Power of a Tweet: An Exploratory Study Measuring the Female Perception of Celebrity Endorsements...

Documents

Transcript of The Power of a Tweet: An Exploratory Study Measuring the Female Perception of Celebrity Endorsements...