LMX quality, supervisorycommunication and

team-oriented commitmentA multilevel analysis approach

Hassan Abu Bakar, Che Su Mustaffa and Bahtiar MohamadCommunication Department, College of Arts and Science,

Universiti Utara Malaysia, Sintok, Malaysia

Abstract

Purpose – Researches have documented the impact of dyadic communication and relationships onindividual behavior in workgroups. However, communication remains as the background element inleadership and management literature as opposed to being the primary process in the leader-memberrelationships development. The purpose of this paper is to establish and interpret the appropriate levelof analysis based on the correlation between leader-member exchange (LMX) quality, supervisorycommunication and team-oriented commitment in a Malaysian organization setting.

Design/methodology/approach – A survey from a Malaysian organization (n ¼ 201) is analyzedon the relationships between LMX quality, supervisory communication and commitment using withinand between analysis (WABA).

Findings – The individual dyad relationships and communication correlates with team-orientedcommitment at the group level. Therefore, LMX quality and supervisory communication influenceoverall team-oriented commitment in a work group.

Research limitations/implications – Given that the analyses are based on self-report in oneorganization, these results have to be handled carefully.

Practical implications – The results implied that the relationship between dyadic relationshipsquality, supervisory communication and team-orientated commitment is best connected if individualworkers in work group are encouraged to communicate their needs to supervisors. As such, theworker’s ability to communicate mutually about relationships (LMX quality and positive relationshipscommunication) and work (upward openness and job relevant communication) with their immediatesupervisor implicates both personal fit and work group functioning.

Originality/value – The results extend the authors’ understanding of LMX quality, supervisorycommunication and team-oriented commitment by identifying the specific form levels of analysis in aMalaysian organization setting.

Keywords Malaysia, Team working, Employee behaviour, Team leaders, Communication

Paper type Research paper

The current issue and full text archive of this journal is available at

www.emeraldinsight.com/1356-3289.htm

A previous version of this paper was presented at the 57th International CommunicationAssociation Annual Conference, Organizational Communication Division, in San Francisco,California.

The authors are indebted to Professor Gail Fairhurst from University of Cincinnati andProfessor James Honeycutt from Louisiana State University for their feedback on an earlier draft.The authors would also like to thank reviewers of this paper for their many helpful comments.Finally, the authors would like to acknowledge the support of the Universiti Utara Malaysia,Research and Innovation Management Center.

LMX quality,communication

and commitment

11

Received June 2008Revised July 2008,

October 2008Accepted October 2008

Corporate Communications: AnInternational Journal

Vol. 14 No. 1, 2009pp. 11-33

q Emerald Group Publishing Limited1356-3289

DOI 10.1108/13563280910931054

Management and communication scholars have long support the importance ofcommunication between superior and their subordinates (Dansereau and Markham,1987; Jablin, 1979). Within an organization for example, supervisor plays monumentalrole as information provider to his or her subordinates at various levels (Andrews andKacmar, 2001; Connaughton and Daly, 2004). Several studies also have examined thesupervisory communication within the framework of leader-member exchange (LMX)theory. These studies mainly focus on antecedences and longitudinal explorations ofLMX-communication links (Fairhurst, 2001; Fairhurst and Chandler, 1989; Lee, 1997;Lee and Jablin, 1995). Evidently, however, little research has been pursued towardsunderstanding the supervisory communication and dyadic relationship that influenceindividual workers behavior in work group. This is important despite the fact thatcommunication is the fundamental component in social relationships (i.e. development,maintenance, and demise (Duck, 1994)). However, surprisingly communication remainsas the background element in LMX and management literature, as opposed to beingthe primary process in the LMX development (Lee, 2005).

In organization theory research, attempts to understand the behavior of individualworkers in organizations are focused on organizational commitment as thecritical psychological factor. Mathieu and Zajac (1990) meta-analysis of organizationalcommitment uncover two main issues. First, the affective involvement in organizationalcommitment proposed by Allen and Meyer (1990) is the most relevant as a behavioralpredictor of individual in an organization (Mathieu and Zajac, 1990). The instrumentdeveloped by Allen and Meyer (1996) (Allen and Meyer, 1990; Meyer and Allen, 1991)has been frequently used in organizational commitment research. Of the threecomponents they distinguish, affective organizational commitment, that is, the extent towhich people experience a sense of identification and involvement with an organization,appears to be the most related to various work aspects.

A second point that emerged from Mathieu and Zajac’s (1990) meta-analysis is thatfocus of the commitment measures might be better suited to predict behavior thanbroad measures. The results of various individual studies seem to point to theconclusion that particular forms of commitment may be related to specific behavior atwork (Randall et al., 1990). Accordingly, in a theoretical analysis, Reichers (1985) pointsout that although the concept of commitment refers to acceptance of the goals andvalues of an organization, it is important to bear in mind that organizations usuallyencompass many different constituencies that may have conflicting goals. To thatextent, the degree of commitment is defined as willingness to dedicate oneself toparticular values and goals. Therefore, it seems essential to specify the nature of thesevalues and goals in order to predict members of organization’s behavior in theirrespective work group (Ellemers et al., 2004, 1998, 1999, 1997; Reichers, 1985).

The aim of the present study is to establish and interpret the appropriate level ofanalysis based on the correlation between LMX quality, supervisory communicationand team-oriented commitment in a Malaysian organization setting. We intend toexamine whether specific forms of correlation between LMX quality-supervisorycommunications; LMX quality-team-oriented commitment and supervisorycommunication-team oriented commitment could be distinguished from each otherat the individual or higher level of analysis. We present results from an individual,dyad and group sample of a Malaysian organization. We relate the measures of LMXquality, supervisory communication and team-oriented commitment from individual

CCIJ14,1

12

embedded in a specific workgroup and to reflect individual relationships quality andcommunication with their immediate supervisor as possible antecedents toteam-oriented commitment.

Although previous investigations have underlined the usefulness of investigating theeffects of the specific rather than broad commitment measures, in line with Reicher’stheoretical analysis, such research efforts have only mainly tried to assess the extent towhich people feel committed to different organizational constituencies (Scott, 1997; Scottet al., 1999; Stinglhamber and Vandenberghe, 2003). In addition, the communicationconcepts are generally lacking in studies associated with team-oriented commitment(Scott et al., 1999). Jablin (1987) claims that the reason for the scarcity of communicationvariables in commitment literature is the complexity of the communication-commitmentdynamic; for that reasons, he offers a preliminary model that specifies communicationvariables as possible antecedents of commitment. In Jablin’s (1987) model the mostsalient variable is the supervisory communication and relationship. Additionally,researchers have found that employee relationships and communication withsupervisors are important as antecedent of team-oriented commitment (Eisenbergeret al., 2002; Kacmar et al., 2003; Stinglhamber and Vandenberghe, 2003).

Our approach in the present contribution, however, was slightly different fromprevious studies, in that we aimed to explore the nature and possible consequences ofteam-oriented commitment based on dyadic activities (LMX quality and supervisorycommunication) where we tested both at individual and group level analysis. Dansereauand colleagues suggest the importance and relevance of the group context in which thedyad resides. They argued that all behaviors take place between dyad do effect thewithin context of one or more larger systems. Additionally, according to them, within thegroup behavior, it seems either the superior or member alone, nor the group alone, butthe individual within the group or a parts effects known as frog-pond effect (Dansereau,1995; Dansereau et al., 1995a, b; Yammarino et al., 2001). Based on this argument, thedyadic activities that involve LMX quality and supervisory communication do notdevelop in a vacuum, but involve characteristics of the leader and the subordinate, theirinteraction, and the situation in which their interaction develops. Therefore, it seemsapparent to specify the appropriate level for theorizing LMX quality, supervisorycommunication and team-oriented link. The current study also draws on concepts fromLMX and Jablin’s (1987) model of communication-turnover, by including LMX qualityby Liden and Maslyn (1998) and supervisory communication behavior by Miles et al.(1996). In addition to this, whereas most studies on antecedent to team-orientedcommitment have been conducted in North America and Europe, this study considersLMX and supervisory communication in a Malaysian organization (Lee, 2005; Walteret al., 2005). Additionally, much is still not known on how level of analysis in specificcultures may affect the LMX-communication-team-oriented commitment link.Intercultural understanding between leader and member in certain cultural context issalient as manifested in the business globalization.

Review of literatureLMX theoryThe LMX model proposed by Graen and his colleagues explain that the relationshipbetween superior and subordinate develops because of their workplace interactions(Graen and Scandura, 1987; Graen and Uhl-Bien, 1995; Graen and Wakabayashi, 1994;

LMX quality,communication

and commitment

13

Graen et al., 1990; Liden and Graen, 1980). This model speculates, that because of timepressures, the leader can develop close relationships with only a few of his/her keysubordinate(s) (high-quality LMX), while, sustaining a formal relationship with the restof his/her group (the low-quality LMX) on the continuum basis. However, one implicitassumption of research exploring the LMX theory is that, once developed, the qualityof LMX remains relatively stable. Empirical findings indicated that the quality of LMXcould stabilize in as early as two weeks (Liden et al., 1993) or two months (Dansereauet al., 1975).

High quality working relationship is sometimes referred to as cadre; or partnershipdemonstrated by a high degree of mutual positive affect, loyalty, respect andproficiency in their work. In contrast, the low quality working relationships are mainlygoverned by their work contract (Liden and Maslyn, 1998). Research suggests that,superior-subordinate in high quality working relationships received superior’s supportand guidance. Furthermore, subordinates who have high quality working relationshipswith their superior demonstrate higher levels of subordinate satisfaction andperformance, lower levels of turnover and most importantly, better quality ofassignments. On the other hand, superior-subordinate in low quality workingrelationships may result in simple contractual relations, higher levels of supervisorycontrol and directives, lower levels of subordinate satisfaction, higher levelsof subordinate turnover and less desired assignments (Liden and Graen, 1980;Liden et al., 1982).

Within the communication literature, findings from several empirical studiessuggest that LMX quality has a strong impact on the superior-subordinate interaction.The differential qualities of LMX have been found to affect variety of communicationbehaviors between superior and subordinate (Fairhurst, 1993; Krone, 1992; Mueller andLee, 2002; Waldron, 1991; Yrle et al., 2002). Earlier research explicated how the qualityof LMX affects subordinates’ and superiors’ communication areas such as discoursepatterns, upward influence, communication expectations, cooperative communication,perceived organizational justice, and decision-making practices (Fairhurst, 1993;Fairhurst and Chandler, 1989; Krone, 1992; Lee, 1997, 2001; Mueller and Lee, 2002;Waldron, 1991; Yukl and Fu, 1999). In addition to this, a recent study indicates thatLMX and the magnitude of change interact to predict employee rationality (Olufowoteet al., 2005). Thus, as LMX quality grows, communication between superior andsubordinate improves too.

Supervisory communicationAccording to Jablin (1987) communication-turnover model, eight communicationvariables may be the antecedents to the intent to leave or commitment variables. Theseeight communication variables are and according to Jablin (1987) supervisorycommunication is critical antecedence for communication-turnover model. Thecommunication variables are:

(1) organization wide communication;

(2) organizational characteristics;

(3) integration in emergent communication networks;

(4) supervisor communication relationship;

(5) co-worker communication relationship;

CCIJ14,1

14

(6) communication-related work expectations;

(7) perceived role ambiguity and conflict; and

(8) communication-related traits and competencies.

The initial concept of supervisory communication is based on role theory; whereKatz and Kahn (1978) conceptualize and expand the basic components ofcommunication (source, receiver, channel and message), while specifying thedirection of information flow in terms of superior-subordinate relationships. Theysuggest that communication between supervisor to subordinate contains five typesof information:

(1) job instruction;

(2) job rationale;

(3) procedures and practices;

(4) feedback; and

(5) indoctrination of goals.

Mean while, communication from subordinate to their superior mainly containsinformation about:

. themselves, their performance and their problems;

. their co-workers’ problems;

. organizational practices and policies; and

. what needs to be done and how it can be done (Katz and Kahn, 1978).

Based on Katz et al.’s (1978) notion of supervisory communication, Huseman et al.(1980) through series of qualitative and quantitative studies developed seven types ofcommunication that occurs in superior-subordinate relationships, namely:

(1) direction;

(2) information;

(3) rationale;

(4) feedback;

(5) positive expression;

(6) negative expression; and

(7) participation.

Hatfield and Huseman (1982) later tested these types of superior-subordinatecommunication and they found that these seven types of superior-subordinatecommunication have significant impact on subordinates’ job satisfaction.Miles et al. (1996) employed and retested Huseman’s et al. (1980) seven types ofsuperior-subordinate communication and found four separate dimensions ofsupervisory communication behaviors that can reflect working and socialcommunication in superior-subordinate relationships, namely:

(1) positive relationship communication;

(2) upward openness communication;

LMX quality,communication

and commitment

15

(3) negative relationship communication; and

(4) job-relevant communication.

Numerous studies have explored supervisory communication as a variable thatinfluences various organizational outcomes (Yrle et al., 2002, 2003). For example,several studies have demonstrated that superior-subordinate communication has apositive impact on job satisfaction and organizational commitment (Goldhaber et al.,1978; Hatfield and Huseman, 1982; Huseman et al., 1980; Miles et al., 1996; Schwigerand Denisi, 1991; Wilhelm et al., 1993), performance appraisals (Nathan et al., 1991),and influence on relationships between managers and workers (Page and Wiseman,1993). Additionally, van Vuuren et al. (2007) also found the direct and indirect effects ofsupervisor communication on commitment. Consistent with these studies, this studyadopts and intends to explore superior-subordinate communication as independentvariable that will influent members of the work group’s commitment. That is, theincreasing quality of supervisory communication will improve the subordinate’steam-oriented commitment.

Team-oriented commitmentThe concept of commitment is one of the major factors in determining the relationshipbetween individuals and an organization (Mowday et al., 1982). A review of theliterature suggests that there are various distinct approaches to defining commitment.Commitment has been defined as a strong desire to maintain membership in anorganization (Mowday et al., 1979). It has also been defined as identification with goalsand values between an individual and the organization (Buchanan, 1974) or anexchange of behavior to get benefits that will be appreciated by others (Meyer andAllen, 1984). These definitions focus on the psychological relationships thatindividuals have with an organization.

As mentioned earlier, Mathieu and Zajac (1990) meta-analysis reveals thataffective commitment is the most relevant as a behavioral predictor. This conceptrefers to an attitudinal construct rather than a calculative investment in theorganization response to the extent into which the organization invests in itsemployees. Thus, the affective commitment can be referred as the extent in whichpeople experience a sense of identification and involvement with an organization.In trying to determine what makes people exert them at work, or how they chooseto devote their energies, we argue that goal specification should be made withinteamwork. This is because, according to Allen and Meyer (1996), employees withstrong affective commitment remain with the group because they want to. Studyalso shows that the extent in which workers are available, take initiative or areprepared to help their coworkers in order to foster the achievement of commonteam goals (Ammeter et al., 2004). Therefore, in the present investigation, weexamined the extent into which people felt committed to common team goals asdependent variable.

Cross-cultural supervisory relationships and hypothesesA great deal of cross-cultural analysis has been based on the seminal work of Hofstede,in which he examined over 50 different countries searching for cultural differences and

CCIJ14,1

16

similarities. Based on his research, Hofstede has proposed five major dimensions wherecultures differ:

(1) power distance;

(2) uncertainty avoidance;

(3) individualism vs collectivism;

(4) masculinity vs femininity; and

(5) long-term vs short-term orientation.

Many of these cultural traits are clearly relevant to the study of superior-subordinaterelationships (Hofstede, 1984, 2003). Hofstede’s concepts of power distance andmasculinity vs femininity dimensions, for example, are used to identify culturalexpectations of superior-subordinate dynamic. Hofstede (2003) suggests thatMalaysian organizations’ culture indicates high scores for power distance andmasculinity-femininity dimensions when compared to Australia, UK and USA.Additionally, Hofstede’s also illustrates Malaysian as more collectivist nature society,meaning that there is close ties among individuals and a greater tolerance for a varietyof opinions. This result implies superior and subordinate in Malaysia exhibit greateracceptance of autocratic and paternalistic leadership behaviors. Connected toHofstede’s cultural dimensions, Asma and Lim (2001) and Lim (2001) examine thesecultural dimensions in various private and public organizations in Malaysia and theyfound similar patterns with Hofstede’s work (Asma and Lim, 2001; Lim, 2001) withrelative high levels of power distance and high levels of the collectivist nature withinthe Malaysian organizations.

Another significant, cross-cultural study, explicitly examining cultural differencesand their relationship with leadership effectiveness, has also been recently released.The Global Leadership and Organizational Behavior Effectiveness (GLOBE) study(Ashkanasy, 2002; Kennedy, 2002) elaborates and expands upon Hofstede’s findings.This study is even more exhaustive, collecting data from 62 different societies over aseven-year period, and examining differences over similar cultural dimensions,including power distance. However, Kennedy (2002) argues that acceptance of powerdistance in Malaysia is less extreme than Hofstede’s (1984) original work and Asma’sand Lim’s (2001) and Lim’s (2001) when compared to other countries involved in theGLOBE study. Kennedy (2002) further argues that even though Malaysia can beconsidered as a culture with high-power distance, it is balanced with strong humanorientation in superior-subordinate relationship. Furthermore, effective leaders inMalaysian organizations are expected to show compassion while using more of anautocratic, rather than a participative style (Kennedy, 2002). However, consistent withHofstede’s work, the GLOBE study also shows collectivist nature in Malaysianorganization and this implies preference of Malaysian employees to work as a group.For example, Malaysian employees are more likely to use coordination to integratetheir work tasks, and use team workflows to deal with task uncertainty (Pearson andChong, 1997). There is also a high preference for teamwork goals rather than individualgoals (Chan and Pearson, 2002) and they tend to be more idealistic in-groupperformance (Karande et al., 2002).

Within the framework of LMX on Malaysian studies, Lo et al. (2006) and Ansari et al.(2007) have demonstrated the link between LMX quality and work outcomes. Both

LMX quality,communication

and commitment

17

studies demonstrated a significant impact of LMX on commitment and satisfaction.LMX quality has a positive direct impact on organizational citizenship behavior andindirect positif effect of LMX on satisfaction and commitment through delegation(Ansari et al., 2007; Lo et al., 2006). However, both studies did not investigate andinterpret the level of analysis on the correlation of LMX and work outcomes. Perhaps,one study that can mirror Malaysian organization sample is a study conducted byPerizade and Sulaiman (2005), which conducted WABA analysis based on Indonesiansample. This study shows that aggregation data of LMX, satisfaction and commitmentvariables do not indicate a group or dyad level of analysis. However, in Perizade andSulaiman (2005) study, the interpretation of correlation between LMX and workoutcomes based on level of analysis was not conducted. However, in this study weexamine the relationship between LMX, supervisory communication and groupcommitment. Furthermore, from the above discussion, the following are evident forMalaysian employees:

. collectivism in nature and they emphasizes the importance of the group;

. high power distance emphasizes the importance of the leader and his or herstatus and power difference in respect of the group; and

. group-based rewards emphasize the importance of group work and performance(Yammarino and Jung, 1998).

Therefore, we advance the following hypotheses:

H1. Positive relationship between LMX quality and supervisory communicationis group based and involves the entire group.

H2. Positive relationship between LMX quality and team-oriented commitmentsis group based and involves the entire group.

H3. Positive relationship between supervisory communication and team-orientedcommitments is group based and involves the entire group.

Based on these hypotheses, we believe that in Malaysian organization setting, there aredifferences between leaders as persons, followers are viewed as a group, and there is aperson-group link. Therefore, the LMX quality, supervisory communication and teamoriented commitment link display between groups’ differences.

MethodParticipantsParticipants in this study are executives reporting to a specific manager in theirrespective work group in an organization involved in an airport management servicesthroughout Malaysia. From the total sample of 874 executives, representing 175 dyadsin seven teams, only 201 executives, representing 41 dyads, embedded in the teamsreturned the survey (23 percent). These seven teams represent units within twodepartments. Three units were from human resources department (training, hiring,salary and promotion), four units from finance department (accounting, purchasing,internal audit and procurement). Approximately, 72 percent (n ¼ 144) are male and28 percent (n ¼ 57) are female. This sample distribution reflects the industry norm forservice sector in Malaysia. Approximately, 15 percent (n ¼ 30) participants haveworked for the organization between three and six years, 55.2 percent (n ¼ 110)

CCIJ14,1

18

worked between six and ten years and 34.8 percent (n ¼ 70) worked more than tenyears in this organization. Approximately, 16 percent (n ¼ 32) participants haveworked for their current mangers for three to five years, 54 percent (n ¼ 108) workedfor six to eight years and 30 percent (n ¼ 61) worked for nine to 11 years under theircurrent manager.

ProcedureIn addressing the concerns over common source variance or common ratter effects inmeasuring LMX, superior-subordinate communication and group commitmentconstructs, we follow procedures proposed by Podsakoff et al. (2003). Firstly, it isachieved through obtaining an understanding of LMX and supervisorycommunication practices constructs from subordinate perspectives via aquestionnaire. Meanwhile the team-oriented commitment constructs were obtainedfrom manager. Secondly, we employed a time lag in obtaining data for LMX,supervisory communication and team-oriented commitment. In doing this, threesessions of questions and answer of the constructs are conducted and the lag betweenthe sessions is a week. Theses approaches are commonly applied to minimize thecommon source variance in cross-level studies (Podsakoff et al., 2003). Theseapproaches were also adopted by Ansari et al. (2007) to minimize common methodbiases in Malaysia organization setting.

InstrumentationThe English language version of LMX by Liden and Maslyn (1998), supervisorycommunication by Miles et al. (1996) and team-oriented commitment by Meyer andAllen (1991) were used to obtain data. This follows the preference of other researcherswho have also used English language questionnaires instead of other local languageson Malaysian subjects (Bochner, 1994; Furnham and Muhiudeen, 1984; Schumaker andBarraclough, 1989). The reason is that Malaysians, especially those involved in thebusiness sector, are fluent in the English language (Lim, 2001). Details of instrumentsused in this study are as follows:

. LMX. We measured perceptions of relationship quality with 12-items known asLMX-MDM, developed by Liden and Maslyn (1998). These items consisting fourdimensions namely the perceived contribution (a ¼ 0.89), affect (a ¼ 0.87),loyalty (a ¼ 0.89), and professional respect (a ¼ 0.85).

. Supervisory communication. To measure superior-subordinate communicationwe used Miles et al. (1996) 24-items. These items represent eight types ofmessages developed by Husemen et al. (1980) consisting of four dimensionsnamely the positive relationships communication (a ¼ 0.84), upward opennesscommunication (a ¼ 0.82), negative relationships communication (a ¼ 0.81) andjob relevant communication (a ¼ 0.86).

. Team-oriented commitment. The managers completed the six-items (a ¼ 0.82)assessing their subordinates’ affective commitment to the group. In doing this weinclude the subordinates’ name in the questionnaire accordingly to the managersto assess each of his or her subordinates’ commitment to the team. Theteam-oriented commitment items were selected from Meyer and Allen (1991)affective commitment scale and modified by Ellemers et al. (1998) to assessemployees’ commitment to their work group. All these items are measured and

LMX quality,communication

and commitment

19

operationalized using a five-point Likert-type ranging from strongly agree tostrongly disagree.

Statistical analysisBecause the major focus of the current study was to test multiple or cross level ofanalysis in addition to assessing both direct effects variables, a variety of analyticaltechniques were employed. Variables means, standard deviations, coefficient ainternal consistency reliability estimates, and Pearson product-moment variableinter-correlations were first computed.

An analytic technique that can test for the presence of either, both, or neither withinor between group effects was warranted in the current research. Therefore, the withinand between analysis (WABA) were employed in the current research. WABA,developed by Dansereau et al. (1984), assesses both variation and co-variation invariables within and between levels of analysis. Different from the traditionalmethodologies of correlation, which used raw score data alone, WABA tests aphenomenon’s level of analysis, and can lead to three different interfaces (Dansereauet al., 1984; Yammarino and Markham, 1992) using a software called DETECT.Furthermore, WABA assesses the variables and the relationships between variablesbased on both statistical and practical significance of. In addition to this, the practicalsignificance in WABA is geometrically based and is not influenced by sample sizecompared to other multilevel techniques such as hierarchical linear modeling (HLM).However, the practical significance in WABA is influenced by group size (Castro,2002).

The WABA approaches can be summarized as follows: first, the E and F testsperformed under WABA I indicate whether the variance is either between or withingroups or whether both or neither within or between group levels are the appropriateinference. Second, in WABA II, A, Z, R, and t-tests are used to decide at which level thecovariance occurs. Lastly, the inferences from WABA I and WABA II are combined,and examination of within and between group correlation components is made using Atest. Based on these analyses, if the result indicates that within group rather thanbetween group effects are present, then the effect is within group and the importance ofindividuals within the context group is stressed. On the other hand, if there is variationand co-variation both within and between groups, this indicates the phenomenon isoperating at both levels. In addition to this, if the tests indicate a lack of significant onthe variation and co-variation, in both within and between groups, this is considered asneither within nor between groups (Yammarino et al., 2001; Yammarino and Markham,1992). Based on this analysis, the DETECT program will indicate the analysis aswholes or parts. Wholes means that individuals are viewed as belonged or intact to thegroup and each person is assigned one score on each variable. While parts means thatindividuals are viewed as process within them (interdependent), therefore each personis assigned multiple scores on each variable.

Means, standard deviations, reliabilities and correlations for all variables appear inTable I.

ResultsLatent composite structural equation modeling (SEM) was used to test themeasurement model for confirmatory factor analysis. This approach is preferred

CCIJ14,1

20

Var

iab

les

M(S

D)

12

34

56

78

9

1.P

erce

ived

con

trib

uti

on3.

80(0

.85)

(0.8

9)2.

Aff

ect

3.84

(0.6

7)0.

43*

**

(0.8

7)3.

Loy

alty

3.97

(0.6

4)0.

38*

**

0.73

**

*(0

.89)

4.P

rofe

ssio

nal

resp

ect

3.89

(0.5

8)0.

50*

**

0.46

**

*0.

60*

**

(0.8

5)5.

Pos

itiv

ere

lati

onsh

ips

com

mu

nic

atio

n4.

01(0

.48)

0.44

**

*0.

40*

**

0.61

**

*0.

44*

**

(0.8

4)6.

Up

war

dop

enn

ess

com

mu

nic

atio

n4.

06(0

.62)

0.37

**

*0.

36*

**

0.59

**

*0.

36*

**

0.68

**

*(0

.82)

7.N

egat

ive

rela

tion

ship

sco

mm

un

icat

ion

4.40

(0.7

9)2

0.35

**

*2

0.35

**

*2

0.30

**

*2

0.25

**

*2

0.26

**

*2

0.32

**

*(0

.81)

8.Jo

bre

lev

ant

com

mu

nic

atio

n4.

34(0

.51)

0.37

**

*0.

49*

**

0.55

**

*0.

49*

0.70

**

*0.

72*

**

234

**

*(0

.80)

9.T

eam

-ori

ente

dco

mm

itm

ent

3.82

(0.7

6)0.

48*

**

0.36

**

*0.

52*

**

0.36

**

0.66

**

*0.

56*

**

20.

27*

**

0.38

**

*(0

.85)

Notes:

* p,

0.01

;*

** p

,0.

05;

Rel

iab

ilit

ies

(Cro

nb

ach

’sa

)of

the

scal

esap

pea

ron

the

dia

gon

al

Table I.Descriptive, correlations

and reliabilities

LMX quality,communication

and commitment

21

over a principal component analysis because SEM approach allows for the estimationof measurement error (Hu and Bentler, 1999). The measurement model was tested withAMOS 5.0. Model fit was assessed with fit indices recommended by Hu and Bentler(1999). The measurement model was estimated, to which the scale indicators wereloaded onto their respective variables; LMX quality, supervisory communication andteam-oriented commitment. The measurement model generated excellent fit,X 2 ¼ 120.74, p ¼ 0.088, CFI ¼ 0.99, NFI ¼ 0.99, RMSEA ¼ 0.03 (CI: 0.00; 0.04).Based on our confirmatory factor analysis, we include only items which load arestatistically significant ( p , 0.001, Table II) to test our hypotheses.

WABA resultsThe WABA results are presented in Tables III (WABA I), IV and V (WABA II) and VI(summary and overall inferences). The decision rules for WABA I, WABA II, andoverall inferences are adapted from Dansereau et al. (1984), who provided a set ofguidelines for interpreting WABA findings based on the results obtained in theWABA I and WABA II.

WABA I resultsWABA I is performed to test whether the total deviation scores for each variable arebetter presented by within cell or between cell scores. As shown in Table III, all of therelationships are tested by the practically significant E test values, while statisticallysignificant F test values tested each variable. Our analyses suggest that the variationbetween groups is significantly larger than the within group variation. Therefore, eachperson is assigned one score on LMX, supervisory communication and team-orientedcommitment.

WABA II resultsWABA II permits a decision about the level of analysis that may underlie relationshipsamong all variables taken two at a time. WABA II involves difference and magnitudetest. The WABA II difference test results are presented in Table IV. The practicalsignificance of the difference given by the value for A is in column three. The A-testresult indicates that, between cells correlation is significantly greater than within cellcorrelation. Similarly, the statistical significant for the Z-test were also conducted andit shows significant difference between the within and between cell correlations. Basedon A-test and Z-test, H1, H2 and H3 were supported. Thus, based on WABA IIdifference test, we infer that groups-based differences lay beneath theLMX-supervisory communication link, and LMX and supervisorycommunication-team-oriented commitment link.

WABA II also includes a magnitude test that provide values for the tests of thepractical (R test) and statistical (t test) significance of the magnitude of the between andwithin cell correlations separately. Results of WABA II magnitude test are shown inTable V. The R-test shows that R-value is by 308 tests. This result indicates thatcorrelation values between LMX, supervisory communication and team-orientedcommitment are greater than 0.50 resulting in an induction of significant greater than30 degrees at the between cells. The t-test also shows that it is statistically significantat the between cells. Therefore, based on WABA II magnitude test, H1-H3 weresupported. This information supports our previous assumption that groups-based

CCIJ14,1

22

Indicator Factor loading

AffectI like my superior very much as a person 0.82 *

I think my superior is the kind of person I would like to have as a friend 0.88 *

I think my superior is a lot of fun to work with 0.83 *

LoyaltyI think my superior defends my work actions to a superior, even without completeknowledge of the issue in question

0.80 *

I think my superior would defend me if I were “attacked” by others 0.80 *

I think my superior would defend me to others in the organization if I made anhonest mistake

0.83 *

Perceived contributionI carry out work tasks for my superior that go beyond what is specified in my jobdescription

0.87 *

I am willing to apply extra effort, beyond that normally required, to further theinterest of the work group

0.87 *

I do not mind working my hardest for my superior 0.86 *

Professional respectI am impressed with my superior’s knowledge of his/her job 0.84 *

I respect my superior’s knowledge of and competency on the job 0.88 *

I admire my superior’s professional skills 0.90 *

Positive relationship communicationI think that my superior jokes good-naturedly with me 0.21I think that my superior asks for my suggestions about how each work task couldbe done 0.72 *

I think my superior asks me about my interests outside of work 0.89 *

I think my superior seeks my input on important decisions 0.13I think my superior strikes up casual conversations with me 0.87 *

I think my superior asks me for suggestions for improvements in my group 0.88 *

Upward openness communicationI question my superior’s instructions when I do not understand them 0.90 *

I tell my superior when I think things are being done wrong 0.95 *

I question my superior’s instructions when I think he/she is wrong 0.88 *

I make suggestion to my superior about how work could be done 0.83 *

I think my superior asks for my suggestion about how work tasks could bedone 0.42I tell my superior about my work problems 0.35Negative relationship communicationI think my superior ridicules or make fun of me 0.65 *

I think my superior criticizes my work in front of others 0.72 *

I think my superior is critical of me as a person 0.27I think my superior’s asks me to do thing rather than tells me 0.13I think my superior tells me how he/she disciplines workers 0.33I think my superior admits to his/her mistakes 0.89 *

Job relevant communicationI think my superior gives me recognition for good work 0.89 *

I think my superior lets me know why changes are made in work assignments 0.85 *

I think my superior keeps me informed about rules and policies 0.82 *

I think my superior gives clear instructions to me 0.28I think my superior informs me about future plan for me in the group 0.38I think my superior tells me the reasons for work schedules 0.31

(continued )

Table II.Standardized factor

loadings for the latentconstructs

LMX quality,communication

and commitment

23

Indicator Factor loading

Team-oriented commitmentMy subordinate prepared to do additional tasks, when this benefits my team 0.90 *

My subordinate feels at home among my team member at work 0.88 *

My subordinate tries to invest effort into a good atmosphere in my team 0.92 *

I let my subordinate be guided by the goals of my team 0.75 *

When there is social activity with my team, my subordinate usually helps toorganize it 0.35My subordinate thinks that he/she could easily become as attached to my team 0.60 *

Notes: *All factor loadings are significant at p , 0.001. N ¼ 193Table II.

h Correlation F ratioVariable Between Within E ratio Between Within Sig F Induction

LMX 1.00 0.00 99.00 * 99.00 * * – 0.00 WholesSupervisory communication 0.72 0.48 1.49 * 1.45 * * – 0.00 WholesTeam-oriented commitment 0.80 0.62 1.27 * 1.70 * * – 0.00 Wholes

Notes: *Significant by 308 tests; * *p # 0.01, degrees of freedom are 98, 102. Wholes means that theindividuals in a group are viewed as homogenized, therefore, each group is assigned one score on eachvariable

Table III.WABA I result

CorrelationRelationship and variables Between Within A-value Z-value Induction

LMX and supervisory communication 0.64 0.00 0.70 * 5.39 * * WholesLMX and team-oriented commitment 0.69 0.00 0.76 * 6.00 * * WholesSupervisory communication and team-orientedcommitment 0.92 0.01 1.15 * 11.05 * * Wholes

Notes: *Significant by 308 tests; * *p # 0.01 Wholes means that the individuals in a group are viewedas homogenized, therefore, each group is assigned one score on each variable

Table IV.WABA II difference test

Between WithinVariable and relationship R-test T-test Sig-t R-test T-test Sig-t Induction

LMX and supervisory communication 0.85 * 0.84 * * 0.00 0.00 0.00 1.00 WholesLMX and team-oriented commitment 0.97 * 9.52 * * 0.00 0.00 0.00 1.00 WholesSupervisory communication andteam-oriented commitment 2.12 * 20.89 * * 0.00 0.00 0.00 1.00 Wholes

Notes: * Significant by 308 tests; * *p # 0.01. Degree of freedom for t-test (between) 1, 97, degree offreedom for t-test (within) 1, 101. Wholes means that the individuals in a group are viewed ashomogenized, therefore, each group is assigned one score on each variable

Table V.WABA II magnitudetests

CCIJ14,1

24

differences lie beneath the LMX-supervisory communication relationships, and LMX,supervisory communication-team-oriented commitment relationships.

WABA I and II resultsFinally, the total correlations that are based on the total deviation scores followed Rand t values testing for practical and statistical significance. Results are shown inTable VI. Based on this test, H1-H3 were supported. Therefore, based on WABA I andWABA II total correlation and component analysis, the group conditions are based onthe between h correlation and between cell correlations. Thus, we conclude thatgroups-based differences caused the relationships between LMX and supervisorycommunication, and the relationships between LMX, supervisory communication andteam-oriented commitment.

DiscussionIn summary, the results of all correlations tested were consistent with the theoreticalmodel of work unit context and LMX. As predicted, the correlation between LMXquality supervisory communication and team-oriented commitment exists at the grouplevel. All the hypotheses are accepted and WABA I and WABA II tests indicated thatcorrelation between LMX quality, supervisory communication and team-orientedcommitment are operating at the group level in Malaysian organization setting.Simultaneously, correlation between supervisory communication and team-orientedcommitment too, commission at the group level. The WABA analyses performed inthis study demonstrated the importance of level of analysis for theorizing andhypothesis testing in Malaysian organization setting. The WABA results clearlyshowed that all variables examined in this study had an interpretation of group effects.

Based on this study, it clearly shows that all behaviors takes place between dyad doaffect one or more larger systems in organizations. Our findings support theproposition by communication scholars that the dyadic activities within a group doesaffect the overall group behavior (Kacmar et al., 2003; Kramer, 2004; Lee, 2005). Ourfinding shows that it seems relevant for managers to encourage his/her subordinates’commitment to the team, as commitment to team is related to higher level ofsuperior-subordinate relationships quality and communication. As the WABA resultsshow this kind of relationship occurring between groups, the emphasis should beplaced on group process activities among work group members. Thus, within thegroup behavior (dyad) seems to have the whole group effects (Dansereau et al., 1995a,b; Yammarino et al., 2001; Yammarino and Jung, 1998).

A study by Liden et al. (2000) based on US organization setting showed that, theLMX quality and work outcomes relationships are interpreted at the individual level ofanalysis. In another study in US organization by Cogliser and Schriesheim (2000), itshowed that the LMX quality, group cohesiveness and organizational climate exist atboth individual and group level of analysis. Similarly, study by Herold et al. (2008)showed that the link between transformational leadership and organizationalcommitment displayed at both individual and group level. Within the Europeanorganizations, comparable evident were found with US organizations, for example,study by Vandenberghe et al. (2007) in Belgium fast-food firm found the link betweenperceived supervisor support and commitment exists both at individual and grouplevel of analysis, while study by van Yperen et al. (1999) on superior-subordinate

LMX quality,communication

and commitment

25

Com

pon

ents

Var

iab

les

Tot

alco

rrel

atio

nR

-tes

tt-

test

Sig

-tB

etw

een

Wit

hin

Ind

uct

ion

LM

Xan

dsu

per

vis

ory

com

mu

nic

atio

n0.

420.

27*

3.88

**

0.00

0.26

0.00

Wh

oles

LM

Xan

dte

am-o

rien

ted

com

mit

men

t0.

830.

65*

9.23

**

0.00

0.55

0.00

Wh

oles

Su

per

vis

ory

com

mu

nic

atio

nan

dte

am-o

rien

ted

com

mit

men

t0.

721.

12*

15.8

7*

*0.

000.

550.

01W

hol

es

Notes:

* Sig

nifi

can

tb

y308

test

s;*

* p#

0.01

.Deg

rees

offr

eed

omar

e1,

199.

Wh

oles

mea

ns

that

the

ind

ivid

ual

sin

ag

rou

par

ev

iew

edas

hom

ogen

ized

,th

eref

ore,

each

gro

up

isas

sig

ned

one

scor

eon

each

var

iab

le

Table VI.WABA I and WABA IItotal correlation andcomponent analysis

CCIJ14,1

26

relationships in The Netherlands medium-sized organization found that perceivedsupervisor support and organizational citizenship behavior related both at individualand department level.

Within the Asian context, a study by Hui et al. (2007) based on Chineseorganizations shows that supervisor’s behavior and employee services quality onlyoperates at the individual level. Similarly, another study by Liao and Chung (2007) inTaiwan organization found that transformational leadership was positively related toemployee service performance at individual analysis. The current investigationhowever, suggests that the LMX quality, supervisory communication andteam-oriented commitment in a Malaysia organization exist at the group level.Unfortunately, the limitation of the current study is that it is one organization andtaking into account the sensitivity of WABA to team size, making inferences about thedirectionality of that relationship must be made with caution: of whether high LMXand supervisory communication quality do encourage commitment among individualin the team as a whole. However, based on the findings in this study on a Malaysianorganization setting, we may conclude that the correlations between LMX quality(perceived contribution, affect, loyalty and professional respect), supervisorycommunication (positive relationship communication, upward openness and jobrelevant communication) and team-oriented commitment were at the group level(interpreted as each group-differences effect).

Our findings have implications for practices. Results of this study clearlydemonstrated that supervisor-subordinate relationships quality and supervisorycommunication are important processes in influencing individual commitment in workgroup. Given the interdependent nature of relationships and communication, thisprocess is best connected if individual workers in work group are encouraged tocommunicate their needs to supervisors. As such, the worker’s ability to communicateabout relationships (positive relationships communication) and work (upwardopenness and job relevant communication) implicate both personal fit and workgroup functioning. In communicating such desires, the supervisor needs to managerelationships and job-related communication with their immediate subordinate in workgroup especially for the benefit of the whole work group. The responsibility lies onsupervisor to facilitate openness in communication, emphasize the importance ofindividual dyad relationships quality and work group goals, and discourage the typesof communication (negative relationships communication) that leads to lowercommitment to work group especially in a collective organization such as in Malaysia.

Limitation and future directionsPerhaps, the main weakness of the study is the focus of commitment. Currentinvestigations limit to team-oriented commitment. Thus, we do not know the directionof level of analysis for LMX quality and supervisory communication withorganizational commitment or commitment to their respective superior in Malaysiaorganization setting. It would be desirable for future studies to combine commitmentto organization and superior. Secondly, the current investigation limited to onlyMalaysian participants. As mentioned earlier, current description on Malaysianparticipants justify the hypotheses advanced in this study where the LMX-quality,supervisory communication and team-oriented commitment exist at the group level of

LMX quality,communication

and commitment

27

analysis. Therefore, a comparison study between high- and low-context culture shouldbe considered.

Finally, the results extend our understanding of LMX quality, supervisorycommunication and team-oriented commitment by identifying the specific form levelsof analysis. Researchers have already documented the direct effects of LMX qualityand supervisory communication on work outcome (Mueller and Lee, 2002), yet in orderto continue providing knowledge useful for managers, researchers must continue theirefforts to identify specific form levels of analysis on LMX quality and communicationbehaviors within superior-subordinate relationship on work outcome. Of course, theresults of this study should be interpreted with caution given the inherent limitationsof the research design. Recent research views relationship development andcommunication activities within a dyadic relationship as interdependent complexprocess that is grounded within a group (Liden et al., 2000), and thus, a key limitationon this report is the cross-sectional nature of the data. Clearly, longitudinal researchthat tracks relationships development and communication activities within andbetween dyad is needed. Likewise, the use of self-report methods and the homogenoussample (e.g. Malaysian and government link corporation) warrant caution. The dyadrepresented in this report may under-represent the actual dyad population at large.In addition, statements of correlations based on the results of statistical techniques areuseful for making inferences, but must be treated with caution given the sensitivity ofWABA to group size.

References

Allen, N.J. and Meyer, J.P. (1990), “The measurement and antecedents of affective, continuanceand normative commitment to the organization”, Journal of Occupational Psychology,Vol. 63 No. 1, pp. 1-18.

Allen, N.J. and Meyer, J.P. (1996), “Affective, continuance, and normative commitment to theorganization: an examination of construct validity”, Journal of Vocational Behavior, Vol. 49No. 3, pp. 252-76.

Ammeter, A.P., Douglas, C., Ferris, G.R. and Goka, H. (2004), “A social relationshipconceptualization of trust and accountability in organizations”, Human ResourceManagement Review, Vol. 14 No. 1, pp. 47-65.

Andrews, M.C. and Kacmar, K.M. (2001), “Confirmation and extension of the sources of feedbackscale in service-based organizations”, Journal of Business Communication, Vol. 38 No. 2,pp. 206-26.

Ansari, M.A., Lee, B.B. and Aafaqi, R. (2007), “LMX and work outcomes: the mediating roledelegation in the Malaysian business context”, Academy of Management Proceedings,pp. 1-6.

Ashkanasy, N.M. (2002), “Leadership in the Asian century: lessons from GLOBE”, InternationalJournal of Organizational Behavior, Vol. 5 No. 3, pp. 150-63.

Asma, A. and Lim, L. (2001), “Cultural dimensions of Anglos, Australians and Malaysians”,Malaysian Management Review, Vol. 36 No. 2, pp. 1-17.

Bochner, S. (1994), “Cross-cultural differences in the self concept: a test of Hofstede’sindividualism/collectivism distinction”, Journal of Cross-Cultural Psychology, Vol. 25 No. 2,pp. 273-83.

Buchanan, B. (1974), “Building organizational commitment: the socialization of managers inwork organizations”, Administrative Science Quarterly, Vol. 19 No. 4, pp. 533-46.

CCIJ14,1

28

Castro, S.L. (2002), “Data analytic methods for the analysis of multilevel questions: a comparison ofinterclass correlation coefficients, rwg(j), hierarchical linear modeling, within and betweenanalysis and random group resampling”, Leadership Quarterly, Vol. 13 No. 1, pp. 69-93.

Chan, C.C.A. and Pearson, C.A.L. (2002), “Comparison of managerial work goals amongBruneian, Malaysian and Singaporean managers”, Journal of Management Development,Vol. 21 No. 7, pp. 545-56.

Cogliser, C.C. and Schriesheim, C.A. (2000), “Exploring work unit context and leader-memberexchange: a multi-level perspective”, Journal of Organizational Behavior, Vol. 21 No. 5,pp. 487-511.

Connaughton, S.L. and Daly, J.A. (2004), “Identification with leader: a comparison of perceptionsof identification among geographically dispersed and co-located teams”, CorporateCommunications: An International Journal, Vol. 9 No. 2, pp. 89-103.

Dansereau, F. (1995), “A dyadic approach to leadership: creating and nurturing this approachunder fire”, Leadership Quarterly, Vol. 6 No. 4, pp. 479-90.

Dansereau, F. and Markham, S.E. (1987), “Superior-subordinate communication”, in Jablin, F.,Putnam, L., Roberts, K. and Porter, L. (Eds), Handbook of Organizational Communication:An Interdisciplinary Perspective, Sage, Newbury Park, CA, pp. 343-88.

Dansereau, F., Alutto, J.A. and Yammarino, F.J. (1984), Theory Testing in OrganizationalBehavior: The Varient Approach, Prentice-Hall, Englewood Cliffs, NJ.

Dansereau, F., Graen, G. and Haga, W.J. (1975), “A vertical dyad linkage approach to leadershipwithin formal organizations: a longitudinal investigation of the role making process”,Organizational Behavior and Human Performance, Vol. 13 No. 1, pp. 46-78.

Dansereau, F., Yammarino, F.J. and Markham, S.E. (1995), “Leadership: the multiple-levelapproaches”, Leadership Quarterly, Vol. 6 No. 3, pp. 251-63.

Dansereau, F., Yammarino, F.J., Markham, S.E., Alutto, J.A., Newman, J. and Dumas, M. (1995),“Individualized leadership: a new multiple-level approach”, Leadership Quarterly, Vol. 6No. 3, pp. 413-50.

Duck, S. (1994), Meaningful Relationships: Talking, Sense, and Relating, Sage, ThousandOaks, CA.

Eisenberger, R., Stinglhamber, F., Vandenberghe, C., Sucharski, I.L. and Rhoades, L. (2002),“Perceived supervisor support: contributions to perceived organizational support andemployee retention”, Journal of Applied Psychology, Vol. 87 No. 3, pp. 565-73.

Ellemers, N., Gilder, D.D. and Haslam, A.S. (2004), “Motivating individuals and groups as work:a social identity perspective on leadership and group performance”, Academy ofManagement Review, Vol. 29 No. 3, pp. 459-78.

Ellemers, N., Gilder, D.D. and Heuvel, D.V.D. (1998), “Career-oriented versus team-orientedcommitment and behavior at work”, Journal of Applied Psychology, Vol. 83 No. 5, pp. 717-30.

Ellemers, N., Kortekaas, P. and Ouwerkerk, J.W. (1999), “Self-categorization, commitment to thegroup and group self-esteem as related but distinct aspects of social identity”, EuropeanJournal of Social Psychology, Vol. 29 No. 2, pp. 371-89.

Ellemers, N., Spears, R. and Doosje, B. (1997), “Sticking together of falling apart: in-groupidentification as a psychological determinant of group commitment versus individualmobility”, Journal of Personality and Social Psychology, Vol. 72 No. 3, pp. 617-26.

Ellemers, N., Rijswijk, V.W., Bruins, J. and Gilder, D.D. (1998), “Group commitment as amoderator of attributional and behavioral response to power use”, European Journal ofSocial Psychology, Vol. 28 No. 4, pp. 555-73.

LMX quality,communication

and commitment

29

Fairhurst, G.T. (1993), “The leader-member exchange patterns of woman leaders in industry:a discourse analysis”, Communication Monographs, Vol. 60 No. 4, pp. 321-51.

Fairhurst, G.T. (2001), “Dualism in leadership research”, in Jablin, F.M. and Putnam, L.L. (Eds),The New Handbook of Organizational Communication: Advances in Theory, Research andMethods, Sage, Thousand Oaks, CA, pp. 379-439.

Fairhurst, G.T. and Chandler, T.A. (1989), “Social structure in leader-member interaction”,Communication Monographs, Vol. 56 No. 3, pp. 215-39.

Furnham, A. and Muhiudeen, C. (1984), “The Protestant work ethic in Britain and Malaysia”,Journal of Social Psychology, Vol. 122 No. 2, pp. 157-61.

Goldhaber, G.M., Yates, M.P., Porter, D.T. and Lesniak, R. (1978), “Organizationalcommunication: state of the art”, Human Communication Research, Vol. 5 No. 1, pp. 76-96.

Graen, G.B. and Scandura, T.A. (1987), “Toward psychology of dyadic organizing”, inCummings, S.B.M. and Cummings, L.L. (Eds), Research in Organizational Behavior, Vol. 9,JAI Press, Greenwich, CT, pp. 175-208.

Graen, G.B. and Uhl-Bien, M. (1995), “Relationship-based approach to leadership: development ofleader-member exchange (LMX) theory of leadership over 25 years: applying a multi-levelmulti-domain perspective”, Leadership Quarterly, Vol. 6 No. 2, pp. 219-47.

Graen, G.B. and Wakabayashi, M. (1994), “Cross-cultural leadership making: bridging Americanand Japanese diversity for team advantage”, in Triandis, H.C., Dunnette, M.D. andHough, L.M. (Eds), Handbook of Industrial and Organizational Psychology, Vol. 4,Consulting Psychology Press, New York, NY, pp. 415-46.

Graen, G.B., Wakabayashi, M., Graen, M.R. and Graen, M.G. (1990), “Internationalgeneralizability of American hypotheses about Japanese management progress: a stronginference investigation”, Leadership Quarterly, Vol. 1 No. 1, pp. 1-23.

Hatfield, J.D. and Huseman, R.C. (1982), “Perceptual congruence about communication as relatedto satisfaction: moderating effects of individual characteristics”, Academy of ManagementJournal, Vol. 25 No. 2, pp. 349-58.

Herold, D.M., Fedor, D.B., Caldwell, S. and Liu, Y. (2008), “The effects of transformational andchange leadership on employees’ commitment to change: a multilevel analysis”, Journal ofApplied Psychology, Vol. 93 No. 2, pp. 346-57.

Hofstede, G. (1984), Culture’s Consequences: International Differences in Work Related Values,Sage, Beverly Hills, CA.

Hofstede, G. (2003), Culture’s Consequences, Comparing Values, Behaviors, Institutions, andOrganizations across Nation, 2nd ed., Sage, Newbury Park, CA.

Hu, L. and Bentler, P.M. (1999), “Cutoff criteria for fit indexes in covariance structure analysis:conventional criteria versus new alternatives”, Structural Equation Modeling, Vol. 6 No. 1,pp. 1-55.

Hui, C.H., Chiu, C.K.W., Yu, L.H.P., Cheng, K. and Tse, H.M.H. (2007), “The effects of serviceclimate and the effective leadership behavior of supervisors on frontline employee servicequality: a multi-level analysis”, Journal of Occupational & Organizational Psychology,Vol. 80 No. 1, pp. 151-72.

Huseman, R.C., Hatfield, J.D., Boulton, W.R. and Gatewood, R.D. (1980), “Development of aconceptual framework for analyzing the communication-performance relationship”,Proceedings of the Academy of Management, pp. 178-82.

Jablin, F.M. (1979), “Superior-subordinate communication: the state of art”, Psychological Bulletin,Vol. 86 No. 9, pp. 1201-22.

CCIJ14,1

30

Jablin, F.M. (1987), “Organizational entry, assimilation, and exit”, in Jablin, F.M., Putnam, L.L.,Roberts, K.H. and Porter, L.W. (Eds), Handbook of Organizational Communication: AnInterdisciplinary Perspective, Sage, Newbury Park, CA, pp. 679-740.

Kacmar, K.M., Witt, L.A., Zivnuska, S. and Gully, S.M. (2003), “The interactive effect ofleader-member exchange and communication frequency on performance ratings”,Journal of Applied Psychology, Vol. 88 No. 4, pp. 764-72.

Karande, K., Rao, C.P. and Singhapakdi, A. (2002), “Moral philosophies of marketing managers:a comparison of American, Australian and Malaysian cultures”, European Journal ofMarketing, Vol. 36 No. 8, pp. 768-954.

Katz, D. and Kahn, R.L. (1978),TheSocial PsychologyofOrganizations, 2nd ed., Wiley, New York, NY.

Kennedy, J.C. (2002), “Leadership in Malaysia: traditional values, international outlook”,Academy of Management Executive, Vol. 16 No. 3, pp. 15-26.

Kramer, M.W. (2004), “The complexity of communication in leader-member exchanges”,in Graen, G. (Ed.), New Frontiers of Leadership. A Volume in LMX Leadership: The Series,Information Age Publishing, Greenwich, CT, pp. 167-91.

Krone, K.J. (1992), “A comparison of organizational, structural, and relationship effects onsubordinates’ upward influence choices”, Communication Quarterly, Vol. 40 No. 1, pp. 1-15.

Lee, J. (1997), “Leader-member exchange, the ‘Pelz effects’ and cooperative communication betweengroup members”, Management Communication Quarterly, Vol. 11 No. 2, pp. 266-87.

Lee, J. (2001), “Leader-member exchange, perceived organizational justice, and cooperativecommunication”, Management Communication Quarterly, Vol. 14 No. 4, pp. 574-89.

Lee, J. (2005), “Communication as antecedents and consequences of LMX development globally:a new strong inference approach”, in Graen, G. and Graen, J.A. (Eds), Global OrganizingDesigns. A Volume in LMX Leadership: The series, Information Age Publishing,Greenwich, CT, pp. 1-41.

Lee, J. and Jablin, F.M. (1995), “Maintenance communication in superior-subordinate workrelationship”, Human Communication Research, Vol. 22 No. 2, pp. 220-58.

Liao, H. and Chuang, A. (2007), “Transforming service employee and climate: a multilevel,multisource examination of transformational leadership in building long-term servicerelationship”, Journal of Applied Psychology, Vol. 92 No. 4, pp. 1006-19.

Liden, R.C. and Graen, G.B. (1980), “Generalizability of the vertical dyad linkage model ofleadership”, Academy of Management Journal, Vol. 23 No. 3, pp. 451-65.

Liden, R.C. and Maslyn, J.M. (1998), “Multidimensionality of leader-member exchange:an empirical assessment through scale development”, Journal of Management, Vol. 24No. 1, pp. 43-73.

Liden, R.C., Graen, G.B. and Hoel, W. (1982), “Role of leadership in the employee withdrawalprocess”, Journal of Applied Psychology, Vol. 67 No. 6, pp. 662-74.

Liden, R.C., Wayne, S.J. and Sparrowe, R.T. (2000), “An examination of the mediating role ofpsychological empowerment on the relations between the job, interpersonal relationships,and work outcomes”, Journal of Applied Psychology, Vol. 85 No. 3, pp. 407-16.

Liden, R.C., Wayne, S.J. and Stilwell, D. (1993), “A longitudinal study on the early development ofleader-member exchanges”, Journal of Applied Psychology, Vol. 78 No. 4, pp. 662-74.

Lim, L. (2001), “Work-related values of Malays and Chinese Malaysians”, International Journal ofCross Cultural Management, Vol. 1 No. 2, pp. 209-26.

Lo, M.C., Ramayah, T. and Hui, K.S.J. (2006), “An investigation of leader-member exchangeeffects on organizational citizenship behavior in Malaysia”, Journal of Business andManagement, Vol. 12 No. 1, pp. 5-23.

LMX quality,communication

and commitment

31

Mathieu, J.E. and Zajac, D.M. (1990), “A review and meta-analysis of the antecedents, correlated, andconsequences of organizational commitment”,PsychologicalBulletin, Vol. 108 No. 2, pp. 171-94.

Meyer, J.P. and Allen, N.J. (1991), “A three-component conceptualization of organizationalcommitment”, Human Resource Management Review, Vol. 1 No. 1, pp. 61-89.

Meyer, J.P. and Allen, N.J. (1984), “The measurement and antecedents of affective, continuanceand normative commitment to the organization”, Journal of Occupational Psychology,Vol. 63 No. 3, pp. 1-18.

Miles, E.W., Patrick, S.L. and King, W.C. (1996), “Job level as systematic variable in predictingthe relationship between supervisory communication and job satisfaction”, Journal ofOccupational & Organizational Psychology, Vol. 69 No. 3, pp. 277-89.

Mowday, R.T., Porter, L.W. and Steers, R.M. (1979), “The measurement of organizationalcommitment”, Journal of Vocational Behavior, Vol. 14 No. 4, pp. 224-47.

Mowday, R.T., Porter, L.W. and Steers, R.M. (1982), Employee-organizational Linkages: ThePsychology of Commitment, Absenteeism and Turnover, Academic Press, New York, NY.

Mueller, B.H. and Lee, J. (2002), “Leader-member exchange and organizational communicationsatisfaction in multiple contexts”, Journal ofBusinessCommunication, Vol. 39 No. 1, pp. 220-44.

Nathan, B.R., Mohrman, A.M.J. and Milliman, J. (1991), “Interpersonal relations as a context forthe effects of appraisal interviews on performance and satisfaction: a longitudinal study”,Academy of Management Journal, Vol. 34 No. 2, pp. 352-69.

Olufowote, J.O., Miller, V.D. and Wilson, S.R. (2005), “The interactive effects of role change goalsand relational exchanges on employee upward influence tactics”, ManagementCommunication Quarterly, Vol. 18 No. 3, pp. 385-403.

Page, N.R. and Wiseman, R.L. (1993), “Supervisory behavior and worker satisfaction in theUnited State, Mexico and Spain”, Journal of Business Communication, Vol. 30 No. 2,pp. 161-74.

Pearson, C.A.L. and Chong, J. (1997), “Contributions of job content and social information onorganizational commitment and job satisfaction: an exploration in a Malaysian nursingcontext”, Journal of Occupational & Organizational Psychology, Vol. 70 No. 4, pp. 357-74.

Perizade, B. and Sulaiman, M. (2005), “Leader-member exchange and leadership effectiveness ofchief executive officers in South Sumatra Indonesia”, The Business Review Cambridge,Vol. 4 No. 1, pp. 226-33.

Podsakoff, P.M., MacKenzie, S.B., Lee, J.Y. and Podsakoff, N.P. (2003), “Common method biasesin behavioral research: a critical review of the literature and recommended remedies”,Journal of Applied Psychology, Vol. 88 No. 5, pp. 879-903.

Randall, D.M., Fedor, D.B. and Longenecker, C.O. (1990), “The behavioral expression oforganizational commitment”, Journal of Vocational Behavior, Vol. 36 No. 2, pp. 210-24.

Reichers, A.E. (1985), “A review and reconceptualization of organizational commitment”,Academy of Management Review, Vol. 10 No. 3, pp. 465-76.

Schumaker, J.F. and Barraclough, R.A. (1989), “Protective self-presentation in Malaysian andAustralian individuals”, Journal of Cross-cultural Psychology, Vol. 20 No. 1, pp. 54-63.

Schwiger, D.M. and Denisi, A.S. (1991), “Communication with employee following a merger:a longitudinal field experiment”, Academy of Management Journal, Vol. 34 No. 1, pp. 110-35.

Scott, C.R. (1997), “Identification with multiple targets in a geographically dispersedorganization”, Management Communication Quarterly, Vol. 10 No. 4, pp. 491-522.

Scott, C.R., Connaughton, S.L., Diaz-Saenz, H.R., Maguire, K., Ramirez, R. and Richardson, B. et al.,(1999), “The impacts of communication and multiple identifications on intent to leave”,Management Communication Quarterly, Vol. 12 No. 3, pp. 400-35.

CCIJ14,1

32

Stinglhamber, F. and Vandenberghe, C. (2003), “Organizations and supervisors as sources ofsupport and targets of commitment: a longitudinal study”, Journal of OrganizationalBehavior, Vol. 24 No. 3, pp. 251-70.

van Vuuren, M., de Jong, M.D.T. and Seydel, E.R. (2007), “Direct and indirect effects ofsupervisor communication on organizational commitment”, Corporate Communications:An International Journal, Vol. 12 No. 2, pp. 116-28.

van Yperen, N.W., van Den Berg, A.E. and Willering, M.C. (1999), “Towards a betterunderstanding of link between participation in decision-making and organizationalcitizenship behavior: a multilevel analysis”, Journal of Occupational & OrganizationalPsychology, Vol. 72 No. 3, pp. 377-92.

Vandenberghe, C., Bentein, K., Michon, R., Tremblay, M. and Chebat, J.C. (2007),“An examination of the role of perceived support and employee commitment inemployee-customer encounters”, Journal of Applied Psychology, Vol. 92 No. 4, pp. 1177-87.

Waldron, V.R. (1991), “Achieving communication goals in superior-subordinate relationships:the multi-functionality of upward maintenance tactics”, Communication Monographs,Vol. 58 No. 3, pp. 289-306.

Walter, H.L., Anderson, C.M. and Martin, M.M. (2005), “How subordinates’ Machiavellianism andmotives relate to satisfaction with superiors”, Communication Quarterly, Vol. 53 No. 1,pp. 57-70.

Wilhelm, C.C., Herd, A.M. and Steiner, D.D. (1993), “Attribution conflict between managers andsubordinates: an investigation of leader-member exchange effects”, Journal ofOrganizational Behavior, Vol. 14 No. 6, pp. 531-44.

Yammarino, F.J. and Jung, D.I. (1998), “Asian Americans and leadership: a level of analysisperspective”, Journal of Applied Behavioral Science, Vol. 34 No. 1, pp. 47-67.

Yammarino, F.J. and Markham, S.E. (1992), “On the application of within and between analysis:are absence and affect really group based phenomena?”, Journal of Applied Psychology,Vol. 77 No. 2, pp. 168-76.

Yammarino, F.J., Dansereau, F. and Kennedy, C.J. (2001), “A multiple-level multidimensionalapproach to leadership: viewing leadership through an elephant’s eye”, OrganizationalDynamics, Vol. 29 No. 3, pp. 149-63.

Yrle, A.C., Hartman, S.J. and Galle, W.P. (2002), “An investigation of relationship betweencommunication style and leader-member exchange”, Journal of CommunicationManagement, Vol. 6 No. 3, pp. 257-69.

Yrle, A.C., Hartman, S.J. and Galle, W.P. (2003), “Examining communication style andleader-member exchange: considerations and concerns for managers”, InternationalJournal of Management, Vol. 20 No. 1, pp. 92-7.

Yukl, G. and Fu, P.P. (1999), “Determinants of delegation and consultation by managers”,Journal of Organizational Behavior, Vol. 20 No. 2, pp. 219-32.

Further reading

Honeycutt, J.M. and McCann, R.M. (2008), “Predicting intrapersonal communication satisfactionon the basis of imagined interactions in the Pacific Rim”, Journal of InterculturalCommunication Research, Vol. 37 No. 1, pp. 25-42.

Corresponding authorHassan Abu Bakar can be contacted at: abhassan@uum.edu.my

LMX quality,communication

and commitment

33

To purchase reprints of this article please e-mail: reprints@emeraldinsight.comOr visit our web site for further details: www.emeraldinsight.com/reprints