This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

ANALYSIS

Consumers' motivational associations favoring free-rangemeat or less meat

Joop de Boer⁎, Jan J. Boersema, Harry AikingInstitute for Environmental Studies, VU University Amsterdam, De Boelelaan 1087, 1081 HV Amsterdam, The Netherlands

A R T I C L E I N F O A B S T R A C T

Article history:Received 25 May 2007Received in revised form 1 July 2008Accepted 1 July 2008Available online 31 July 2008

The present paper analyzed the motivational orientations of consumers who choose to eat(1) small portions of meat or (2) ethically distinctive meat, such as free-range meat, inrelation to the motivational orientations of their opposites. Going beyond the conventionalapproach to consumer behavior, our work builds on recent insights in motivationalpsychology about the ways in which people may approachmatches or avoid mismatches tothe desired end-state of “getting enough nourishment by eating the right food”. Consumerswho tend to approach matches are often focused on choosing the best alternatives fromtheir choice set (chronic promotion focus). Consumers who tend to avoid mismatches areoften focused on rejecting unacceptable alternatives from their choice set (chronicprevention focus). Distinguishing these two motivational orientations provides a scientificbasis for the aim to foster more sustainable food consumption and production patterns. Ourapproach involves a systematic analysis of consumers' goal orientations regarding meatchoices. We examined how a sample of Dutch consumers (n=939) described their chronicmotivational orientations regarding food, their own meat choices and, about two weekslater, their promotion- and prevention-oriented associations favoring either small portionsof meat and free-range meat or their opposites. Largely in line with our hypotheses wefound that consumers with a chronic prevention orientation avoided themismatch of “largeportions”. Also, those of them who paid the premium price of free-range meat consideredeating “meat produced by intensive farming“ a mismatch. Conversely, if consumers with achronic promotion orientation paid the premium price of free-range meat, they consideredthis the best alternative from their choice set. Accordingly, choosing a small portion of meatwas often approached with a prevention orientation and choosing free-range meat witheither a prevention or a promotion orientation. These differences in motivationalorientation underline that the pursuit of sustainability requires careful consideration ofnot just undesirable but also of desirable alternatives.

© 2008 Elsevier B.V. All rights reserved.

Keywords:Food choicesSustainabilityFree-range meatPromotion and preventionorientations

1. Introduction

Modern patterns of food consumption are overusing ournatural resources (Steinfeld et al., 2006). Particularly relevant

here are people's meat choices. Food production will causemuch less pressure on crucial resources (i.e. energy, water,biodiversity), human health and animal welfare, if people inWestern countries choose to eat smaller quantities of meat as

E C O L O G I C A L E C O N O M I C S 6 8 ( 2 0 0 9 ) 8 5 0 – 8 6 0

⁎ Corresponding author. Tel.: +31 20 5989555; fax: +31 20 5989553.E-mail address: joop.de.boer@ivm.falw.vu.nl (J. de Boer).

0921-8009/$ – see front matter © 2008 Elsevier B.V. All rights reserved.doi:10.1016/j.ecolecon.2008.07.001

ava i l ab l e a t www.sc i enced i rec t . com

www.e l sev i e r. com/ l oca te /eco l econ

Author's personal copy

well as types of meat that are produced in a more responsibleway, such as organic or free-range meat (Aiking et al., 2006;Smil, 2000). However, many consumers, both in the UnitedStates (Heller and Keoleian, 2003) and in Europe (Bernués et al.,2003), seem to give little thought to the links between theirconsumption behaviors and the process of food production.Moreover, insights and instruments based on conventionaleconomics fail to improve consumers' ability to live better byconsuming less and reduce their impact on the environmentin the process (Jackson, 2005). One of the prerequisites for sucha change is a better understanding of their underlying mo-tivations. A new theory (Higgins, 1997, 1998, 2000) about theregulation of goal directed behavior provides an interestingscientific basis for this purpose. It can explain how consumersmay get the experience of “feeling right” about what they aredoing if they consume less or opt for a product at a premiumprice (Higgins et al., 2003). This involves the two basicmotivational orientations, termed promotion and prevention,which underlie people's concerns with obtaining nurturance(e.g. nourishing food) and avoiding harm (e.g. seeing to moraland health aspects of eating) respectively. Because it is vital toobtain more insight into how the notion of food sustainabilitycan be made more appealing, the present paper takes a closerlook at promotion- and prevention-orientations of Dutchconsumers favoring either free-range meat or small portionsofmeat versus their opposites (i.e. meat produced by intensivefarming and large portions of meat).

Promotion and prevention can be seen as the dual mo-tivational underpinnings of the sustainability concept. Be-cause a food system is sustainable only if it can bemaintainedat a desirable quality level for a very long time, the twoorientations are crucial for creatively maximizing a system'squalities (e.g. promotion goals) in a responsible and harmo-nious way (e.g. prevention of deterioration). At the individuallevel, a promotion focus makes the person sensitive topositive outcomes that may be gained, for example, throughaccomplishments, aspirations, and ideals (Higgins, 1997, 1998,2000). Also, a person with a prevention focus becomessensitive to negative outcomes that have to be avoided, forexample, by fulfilling one's moral obligations and responsi-bilities. Both orientations can be relevant for a person'sstrategies to reach a desired end-state, either by ensuringthe presence of positive outcomes that match the desired end-state (promotion focus) or by ensuring the absence of negativeoutcomes that mismatch the desired end-state (preventionfocus). If consumers' desired end-state is “getting enoughnourishment by eating the right food”, their meat choicesmaybe motivated by matches they would like to attain, such aslinks with free-ranging animals, or mismatches they wouldlike to avoid, such as links with environmental deterioration.

Higgins' theory specifies how people's experience of“feeling right” about what they are doing results from thepsychological “fit” between their goal orientations (promotionor prevention), their strategies to reach the goal (eagerapproach or vigilant avoidance), and goal-relevant attributesof the choice options (promotion-related or prevention-related associations). Various experiments (Spiegel et al.,2004) show that a promotion focus is sustained by eagernessand doing extra things, whereas a prevention focus issustained by vigilance and being careful. The experience of

“fit” increases the value of what people are doing. With regardto consumption, a “fitting” combination of goal orientation,strategies to reach the goal and goal-relevant attributes of thechoice options may become linked with a product. That is,consumers can learn to associate products with eitherpromotion or prevention (Zhou and Pham, 2004). For example,consumers with a promotion orientationmay prefer a piece offruit that they see as an energizer, whereas consumers with aprevention orientation may prefer fruit that they associatewith health precautions (Florack and Scarabis, 2006; Spiegel etal., 2004). The difference between promotion and prevention isalso related to the decision rule that consumers apply to achoice set; selection of the best alternative fits a promotionfocus and rejection of unacceptable alternatives a preventionfocus (Chernev, 2004). As summarized in Fig. 1, the processends with different emotional appraisals of success andfailure; people feel cheerful if their approach strategy issuccessful, but disappointment if it fails; they feel relaxed iftheir avoidance strategy is successful, but concerned if it fails.

A person's momentary focus on promotion or preventionwill depend on circumstances induced by the situation athand in combination with his or her personal history andcultural background. In the case of food choices, a promotionorientation may include all the social and culinary motivesthat emphasize the importance of food as a positive force inlife—a pattern that is quite common in countries such asFrance and Belgium, but less so in the United States and Japan(Rozin et al., 1999). In contrast, a prevention orientation mayput much decision weight on those food choice criteria thatensure protection from personally felt threats, such as criteriaon the moral and health aspects of eating. The importance

Fig. 1 – Main elements of Higgins' motivation theory appliedto consumer behavior.

851E C O L O G I C A L E C O N O M I C S 6 8 ( 2 0 0 9 ) 8 5 0 – 8 6 0

Author's personal copy

attributed to health is often associated with prevention moti-vation (Zhou and Pham, 2004). Researchers are only beginningto examine the relationship between Higgins' motivationalorientations and preferences for organic products (Grankvistet al., 2004). Several findings suggest that organic buyers oftenshow characteristics of a chronic prevention orientation, suchas wanting control over all aspects of their lives (Homer andKahle, 1988), avoiding risks (Schifferstein andOude Kamphuis,1998), being inclined to reflection (Torjusen et al., 2001) andvaluing a good conscience (Magnusson et al., 2003). From asustainability perspective, it is also relevant that moderateconsumption of redmeat corresponds with attributing greaterimportance to health, naturalness of the food, weight controland ethical considerations (Pollard et al., 1998). These preven-tion-related associations are very interesting for research intosustainability andmeat consumption, but farmore evidence isrequired about the motivational aspects of the options.

Taking this work one step further our aim in the presentpaper is to gain an understanding of the motivational asso-ciations of two sets of relevant behaviors, namely (1a) “eatingsmall portions of meat,” and (1b) “eating large portions ofmeat,” (2a) “eating ethically distinctive meat, such as organicor free-range meat” and (2b) “eating meat produced byintensive farming.” According to Higgins' theory, consumersmay have learned to associate each of these opposing targetbehaviors with promotion or prevention, depending on the fitbetween their chronicmotivational orientation regarding foodand their usual behavior. As to the first pair of opposites,careful preventionmay fit small portions and eager promotionlarge ones. Therefore, consumers with a prevention focusmayevaluate “eating large portions of meat” as a mismatch thathas to be avoided because it would make them feel nervousand uneasy; from their perspective “eating small portions ofmeat” is something that makes one feel relaxed and safe (i.e.positive prevention associations with avoiding mismatches).Consumers with a promotion orientation, however, mayassociate “eating large portions ofmeat”with cheerful feelingsas an accomplishment to be proud of (i.e. positive promotionassociation with approaching a match) and “eating smallportions of meat” with disappointment (i.e. negative promo-tion association with failures to approach a match).

The second pair of behaviors refers to “free-range meat”and “meat produced by intensive farming.” In principle, bothbehaviors may be included in a consumer's choice set. How-ever, “free-range meat” is connected with a premium pricethat not many consumers may want to pay and “meat pro-duced by intensive farming” may be criticized for the impactsof intensive farming methods. Given the ways in whichprevention and promotion-oriented consumers form theirchoice sets, they will deal differently with perceived matchesand mismatches to the desired end-state. Consumers with aprevention orientation will pay the premium price for “free-rangemeat” especially if they see “meat produced by intensivefarming” as a mismatch that they should reject from theirchoice set (i.e. the idea of eating this meat would imply afailure to avoid a mismatch). In contrast, consumers with apromotion orientation will pay the premium price for “free-range meat” especially if they see it as the best alternative inthe choice set (i.e. the one with the strongest positive pro-motion associations).

Accordingly, our hypotheses involve the relationship be-tween chronic goal orientations regarding food, own beha-viors, target behaviors and the evaluative associations that thelatter generate. The first hypothesis (1a) is that consumerswith a chronic prevention orientation will respond withnegative prevention associations to “eating large portions ofmeat” and with positive prevention associations to “eatingsmall portions of meat.” Similarly (1b), consumers with achronic promotion orientation will respond with positivepromotion associations to “eating large portions of meat”and with negative promotion associations to “eating smallportions of meat”. Hypotheses 1a and 1b mean that chronicgoal orientation regarding food is a predictor of different pro-files of the associations to the target behaviors. These asso-ciations will be stronger if there is a fit between their chronicgoal orientation regarding food and their usual behavior.Secondly (2a), if consumers with a chronic prevention orienta-tion eat “free-rangemeat”, they have rejected “meat producedby intensive farming” from the choice set, as reflected by itsstrong negative prevention associations. Also (2b), if consu-mers with a chronic promotion orientation eat “free-rangemeat”, they have identified it as the best alternative in thechoice set, as reflected by its strong positive promotionassociations. Methodologically, hypotheses 2a and 2b implythat chronic goal orientation regarding food is expected tofunction as a moderator of the relationship between ownbehavior, target behavior and evaluative associations.

In testing these predictions we collected our data via aquestionnaire filled in by a sample that was taken from thegeneral population. By using a sample of Dutch consumers,our research can stand as a good example of the NorthWestern European situation (de Boer et al., 2006). A typicalcharacteristic is the country's central position in an extendednetwork of international trade, which has enabled the Nether-lands to develop an extremely intensive (high input) agricul-ture in a very vulnerable delta. Generally, national productionmay also stimulate national consumption, but Dutch con-sumers spend a relatively small part of their household budgeton meat (de Boer et al., 2006). Although annual data about theconsumption of meat are difficult to interpret, due to therecent health crises and scandals in the meat sector, thefigures over the past five years indicate that the per capitagross household consumption is slightly decreasing at a rateof about 1% per year (PVE, 2007). With regard to organic salesper capita, the Netherlands has an intermediary positionamong the Western European countries after forerunnerssuch as Denmark and Austria, and before France and theUnited Kingdom (Wier and Calverley, 2002). About 80% of thehouseholds bought at least one organic product in the year2004, but the total number of organic purchases is low;although the sector is growing, its market share is still about2% (Biologica, 2006). In fact, most Dutch consumers are used tobuy pork and poultry that originate from intensive farming.

In sum, the paper looks at the way in which consumersrespond to promotion- and prevention-oriented associationsregarding the four target behaviors, designed as two pairs ofopposites. A point of attention is the way in which the goalorientations can be measured. The multi-items scales pre-sented in the literature relate to situations experienced inchildhood (Higgins et al., 2001) or during college years

852 E C O L O G I C A L E C O N O M I C S 6 8 ( 2 0 0 9 ) 8 5 0 – 8 6 0

Author's personal copy

(Lockwood et al., 2002) and make no references to food. Analternative approach that is more appropriate for a generalpopulation uses the set of value measures developed bySchwartz et al. (2001). This approach is based on theassumption that that promotion orientation is linked tovalue priorities that emphasize stimulating experiences andprevention orientation to value priorities that emphasizesafety (Fellner et al., 2007; Kluger et al., 2004; van Dijk andKluger, 2004). Because these value measures make noreference to food, we used them to validate a new instrumentthat makes a distinction between chronic prevention andpromotion orientations regarding food.

2. Method

2.1. Design

The relevant measures were incorporated into two differentquestionnaires that were used in subsequent waves of aconsumer survey among the general population. Wave 1included questions about reported behavior (eating small orlarge portions, buying free-range meat or meat from intensivefarming) and a set of items tomeasure chronic promotion andprevention orientations regarding food choices; two weekslater, wave 2 included the evaluations of the four targetbehaviors in terms of promotion-related or prevention-relatedassociations. One of the other modules of wave 1 was the40 item Portrait Value Questionnaire (PVQ) developed bySchwartz and his co-workers to measure the values peoplefind important in their life (Schwartz et al., 2001).

The survey was organized among consumers with Internetaccess. In 2005 this category included 78% of the Dutchhouseholds in the population under 75 years of age (CBS,2005). In June2005, a call to fill inaquestionnairewasmailed toastratified sample drawn from a large panel of persons who arewilling to participate in web-based research for a small fee. Theprocedure resulted in 1530 completed questionnaires of wave 1(response rate 71%of 2150) and 939 ofwave 2 (response rate 78%of 1200). As a result of the stratified sampling procedure, thedata showed an adequate distribution of themain demographiccharacteristics, i.e. gender (51% female), age (between18and89),and level of education (25% higher education).

2.2. Own behavior

The two pairs of opposing behaviors, eating small or largeportions of meat and buying free-range meat or meat fromintensive farming, were measured through two single itemscales. The first scale was based on a self-categorization interms of eating mainly large portions of meat (includingpoultry, excluding fish), sometimes large/sometimes smallportions, mostly average, or mainly small portions, with theadditional option of eating no meat at all. The participantswho said not to eat meat (1.6%) were left out of the analysis.The second scale was based on the question whether oneusually buys meat (including poultry) produced by intensivefarming or free-range (including organic) meat, with theadditional option of not knowing where one's meat comesfrom. To check the consistency of the answers, some ad-

ditional questions asked for the point of sale (specialty shop,up market or down-market supermarket chains), the numberof days per week that meat is part of one's hot meal, and thenumber of days per week that one eats a meat-substitute.

Each of the two opposing ends of eating mainly “large”versus “small” portions of meat was endorsed by about 20% ofthe participants, “sometimes large/sometimes small” by 20%and “mostly average” by 40%. Eating mainly “small” portionscorrelated negatively with the number of days per week thatmeat is part of one's hot meal (r=− .32) and positively with thenumber of days per week that one eats a meat-substitute(r=.23). In testing hypotheses 1a and 1b, the two key answercategories (i.e. eating mainly “large” or “small” portions) wereused to find out whether the fit between own behavior andchronic goal orientation made a difference in predicting theassociations to the target behaviors. Regarding the secondscale, only a small percentage (7%) said they usually ate free-range (or organic) meat, a somewhat larger percentage (25%)said they sometimes ate free-rangemeat and sometimesmeatfrom intensive farming, and 11% said they usually ate meatproduced by intensive farming. Remarkably, 50% said not toknow where their meat comes from; 7% gave other answers.About a third of the consumers who said they usually ate free-range meat reported to have bought this product in greenspecialty shops and about half referred to up market super-markets; those who said sometimes to eat free-range meatmore often referred to supermarkets. These ratios are in linewithmarketing reports (Biologica, 2006) on consumer behavior(to be distinguished from sheer sales volumes). Hence, the 4-point scale comprises the categories: usually eats free-rangemeat, sometimes eats free-range meat, don't know (or other)and usually eats meat produced by intensive farming.

2.3. Chronic promotion or prevention orientationsregarding food choices

Chronic promotion and prevention orientations regardingfood choices should be measured in a way that allows fordistinguishing these orientations from the degree to whichfood is relevant to the person (Avnet and Higgins, 2006). Thelatter refers to differences between consumers who are highlyinterested in this product category and those who are not (Belland Marshall, 2003; Verbeke and Vackier, 2004). In severalrounds of pilot work we developed a number of positivelyworded portraits of persons who show different degrees ofinvolvement in food, both in promotion-oriented and preven-tion-oriented ways (de Boer et al., 2007). High involvement, forexample, is expressed by the preference to vary one's meal;low involvement means that meals are not consideredimportant. Promotion-oriented motives include enjoying eat-ing well (in case of high involvement) or eating plenty of foods(in case of lower involvement). Prevention-oriented motivesinclude a preference for natural products (in case of highinvolvement) or ordinary meals (in case of lower involve-ment). The items were written in a format derived fromSchwartz's value questionnaire (Schwartz et al., 2001). Eachitem comprises two sentences in which short portraits aredescribed to characterize a person's main food choice motive.The participants were asked to compare each portrait tothemselves and to rate on a 6-point scale “howmuch like you”

853E C O L O G I C A L E C O N O M I C S 6 8 ( 2 0 0 9 ) 8 5 0 – 8 6 0

Author's personal copy

the person is. The motivational structure of Schwartz's valuetheory was used to check the directional differences betweenthe food choice motives.

The 11 items that we developed are described in Table 1;their typical relation to involvement and focus is shown inFig. 2. This figure shows the positions of the items in the two-dimensional space of a multidimensional scaling analysis(PROXSCALE, model=interval, normalized raw Stress=.007).The great advantage of this analysis is that the results areinvariant with respect to the average level of each person'ssimilarity ratings. The horizontal dimension can be inter-preted in terms of high versus low involvement in food. Thevertical dimension separates the items into, on the one hand,promotion-oriented motives, and, on the other hand, preven-tion-oriented motives. The two-dimensional configuration ofFig. 2 was checked by a principal component analysis. Asrecommended for this type of research (Schwartz et al., 2001),the itemswere centered to correct for individual differences inaverage similarity ratings. Table 1 presents the two-dimen-sional result of a principal component analysis before andafter Varimax rotation. The two unrotated components (α=.79and .55) can be interpreted as the degree of food involvementand the direction of motivational focus, respectively. The tworotated components show particular combinations of involve-ment in food and motivational focus. For analytical purposes,the unrotated components are more important as theydistinguish motivational focus from involvement. That is,both consumers who were reflective about food and suppor-ters of an ordinary meal were in fact prevention-oriented.Similarly, both consumers who were taste-oriented and bigeaters were promotion-oriented. This interpretation is sup-ported by the motivational structure of Schwartz's valuetheory; the second unrotated component correlated positivelywith the prevention-oriented values Security (r=.31), Con-formity (r=.24) and Tradition (r=.27) and negatively with thepromotion-oriented values Self-direction (r=− .09), Stimula-tion (r=− .31), Hedonism (r=− .39), Achievement (r=− .30), andPower (r=− .31) (in all cases, pb .001). Higgins et al. (2001) reportsimilar relationships in a student sample: promotion scorescorrelated positively and prevention scores negatively withitems related to eagerness in pursuing things and willingness

to take risks. To differentiate between relatively more promo-tion and prevention-oriented participants, Higgins et al. (2001)made amedian split on the difference between promotion andprevention scores. Using the same logic, we conclude that theunrotated component score is a valid measure of goalorientation that can divide consumers into those with achronic promotion orientation and those with a chronicprevention orientation regarding food. In testing its impactsas predictor (hypothesis 1) ormoderator (hypothesis 2), we usethis measure after a transformation into four categories (1highest level of promotion (more than one unit of standarddeviation below the mean), 2 tends to promotion (below themean), 3 tends to prevention (above the mean), 4 highest levelof prevention (more than one unit of standard deviation abovethe mean)).

2.4. Ratings of promotion- or prevention-relatedassociations

The participants were asked to consider each of the targetbehaviors hypothetically and to grade their evaluations in

Fig. 2 – The food choice motives in the two-dimensionalspace of high versus low involvement in food (horizontal) andpromotion- versus prevention-oriented motives (vertical).

Table 1 – Loadings of the food choice motives before and after Varimax rotation

(Items⁎) Unrotated Rotated

1 2 1 2

She enjoys eating well. In her view every meal should be festive. .60 − .44 .74 − .10She eats because she has to. Meals are not important to her. − .68 .24 − .72 − .12She likes to vary her meal. She is curious about new tastes. .68 − .24 .71 .11She feels proud of her taste. She believes that her food choices are very attractive. .62 − .26 .67 .07She prefers an ordinary meal. She is happy with meat and two vegetables. − .59 .32 − .67 − .00Food does not bother her. She has no special demands on it. − .80 − .08 − .67 − .46She prefers natural products. She would really like her food fresh from the garden. .45 .59 .11 .74She is very mindful of food. She wants to eat sensibly. .58 .43 .30 .66She is grateful for her meal. In her view everything that is edible deserves respect. .25 .58 − .06 .63She is easy about cooking. She uses a lot of ready-made products in her meals. − .50 − .41 − .24 − .60She is a big eater. She loves to have plenty of palatable foods. − .03 − .68 .30 − .61Variance explained (%) 32 18 29 21Alpha .79 .55 .75 .63

⁎All items have been centered (rating scale: 1=not like me at all, 6=very much like me).

854 E C O L O G I C A L E C O N O M I C S 6 8 ( 2 0 0 9 ) 8 5 0 – 8 6 0

Author's personal copy

terms of eight promotion- and prevention-oriented associa-tions. The associations were derived from key descriptions(e.g. Higgins, 1997, 1998) of promotion- and prevention-relatedconcerns as well as the corresponding emotions that comewith success and failure. The four promotion-related itemsrefer to “fits perfectly intomy ideal diet”, “givesme good valuefor money”, “is something to be proud of” (positive form) and“makes me feel disappointed” (negative form). The fourprevention-related items refer to “fits a sensible life style”,“makes me feel safe and familiar” (positive form), “makes mefeel concerned about what I take in” and “is bad for theenvironment” (negative form). As these evaluative associa-tions refer to general psychological processes, each of theassociations is routine for all participants. However, therepresentation of the associations in a verbal form mayartificially create responses that overlap, for example, due tothe common positive or negative wording of the statements.This is a well-known methodological problem that can bedealt with by statistical techniques to remove the varianceshared between particular associations (Boldero et al., 2005;Higgins et al., 1997). The target behaviors referred to “eating”or “buying” in a way that could be evaluated by all consumers(i.e. “If I ate small portions of meat, I would match my idealdiet”). The items were rated on a 7-point scale, ranging fromfully agree to fully disagree. For the analysis, the ratings wererecoded into scores from +3 to −3, with reverse scores for thenegatively formulated items. As a result, a positive score onthe scale means a positive promotion- or prevention-relatedevaluation; a negative score means a negative promotion- orprevention-related evaluation. The design of two pairs ofopposing target behaviors, each rated on four promotion-related and four prevention-related items, produced 8 scalesof four items (Cronbach's alpha between .59 and .73, seeTable 2).

2.5. Analysis

The data were subjected to analyses of variance, correlationand regression. This implies that the variables were consid-ered on interval scales, which has to be donewith caution. Theanalysis of variance focused on the profiles of prevention andpromotion-related evaluations of the target behaviors, whichwere treated as within-subjects factors. This is in line with an

extension of the repeated measures design; this approachcannot only be applied to a situation where one variable ismeasured at different times but also to a situation wheredifferent variables are all measured at one time (Tabachnickand Fidell, 2007; p. 311). Profile analysis is a special applicationof multivariate analysis of variance that can be used tocompare profiles of two or more groups that may have thesame pattern of means on a set of measures. When using theprofile approach to repeated-measures ANOVA, the test ofparallelism is the test of interaction. The groups whoseprofiles were compared differed in chronic prevention orpromotion orientations regarding food (a between-subjectsfactor) and own habitual behavior (also a between-subjectsfactor). In testing hypotheses 1a and 1b, the between-subjectsfactor was high or low fit between own habitual behavior andchronic goal orientation. Notably, fit is not a measuredvariable but derived from other variables; high fit wasassumed in case the participants ate large portions and hada promotion orientation or small portions and a preventionorientation (versus low fit otherwise). The tests of hypotheses2a and 2b focused on the moderating impacts of chronicpromotion or prevention orientation regarding food, whichwere elaborated in regression analyses.

3. Results

3.1. Small and large portions of meat

Our first hypothesis stated (1a) that consumers with a chronicprevention orientation will respond with negative preventionassociations to “eating large portions of meat” and withpositive prevention associations to “eating small portions ofmeat,” and (1b) that consumers with a chronic promotionorientation will respond with positive promotion associationsto “eating large portions of meat” and with negative promo-tion associations to “eating small portions ofmeat”. The upperhalf of Table 2 reports how the target behaviors were rated. Ifboth parts of the hypothesiswere true, the analysis of variancewould show a significant interaction effect of the factorsPrevention or promotion association, Target behavior, Ownbehavior and Chronic goal orientation. However, this four-wayinteraction was not significant (F(3,911)=1.41, pN .10). As Fig. 3demonstrates, there were significant differences between theprofiles of the groups, but these agreed with hypothesis 1aonly. Consumers with a chronic prevention orientation madelarger differences between their associations to “eating smallportions” and “eating large portions” (interaction of Target×Chronic orientation, F(3,911)=70.20, partial η2= .19, pb .001). Ingeneral, more positive ratings were made with a preventionassociation than with a promotion association (effect of Asso-ciation, F(1,911)=328.85, partial η2= .27, pb .001) and “eatingsmall portions” was rated more positively than “eating largeportions” (effect of Target, F(1,911)=127.43, partial η2= .12,pb .001), especially through prevention-related associations(interaction of Association×Target, F(1,911)=90.30, partialη2= .09, pb .001).

Theoretically, the predictive impact of chronic goal orien-tation may be stronger among participants who experience a“fit” between their chronic orientation and their usual

Table 2 – Ratings of the target behaviors on the eightscales

Scales Alpha Mean SD

Prevention associationsEating small portions of meat .59 .54 1.20Eating large portions of meat .61 − .42 1.27

Promotion associationsEating small portions of meat .70 − .17 1.35Eating large portions of meat .68 − .63 1.28

Prevention associationsMeat produced by intensive farming .73 − .31 1.28Free-range meat .60 1.10 1.05

Promotion associationsMeat produced by intensive farming .67 − .43 1.13Free-range meat .67 .59 1.14

855E C O L O G I C A L E C O N O M I C S 6 8 ( 2 0 0 9 ) 8 5 0 – 8 6 0

Author's personal copy

behavior (i.e. high fit was assumed in the case of participantswho eat large portions and have a promotion orientation oreat small portions and have a prevention orientation; versuslow fit otherwise). High or low fit had a large impact onthe profiles (interaction of Target×Chronic orientation×Fit,F(3,911)=55.87, partial η2=.16, pb .001). Fig. 4 shows that thedifferences between the profiles were only found among theparticipants who had a high fit. Consumers who had a chronicprevention orientation and used to eat small portions hadstrong negative prevention associations to “large portions” andstrong positive prevention associations to “small” ones. Incontrast, consumers who had a chronic promotion orientationand used to eat large portions had relatively neutral promotionassociations to “large portions”. Their most salient character-istic was that they had strong negative promotion associations

to “small portions”. Although these consumers considered“small portions” a mismatch, the data did not support thehypothesized link between promotion orientations and largeportions of meat.

3.2. Meat produced by intensive farming or free-rangemeat

Our second hypothesis implies that chronic goal orientationregarding food functions as a moderator of the relationshipbetween own behavior, target behavior and evaluative asso-ciations. That is, (2a) if consumers with a chronic preventionorientation eat “free-range meat”, they have rejected “meatproduced by intensive farming” from the choice set asreflected by its strong negative prevention associations. Also

Fig. 4 – Mean ratings of eating “small” and “large” portions of meat in relation to high or low fit between own behavior andchronic goal orientation regarding food.

Fig. 3 – Mean ratings of eating “small” and “large” portions of meat in relation to chronic goal orientation regarding food.

856 E C O L O G I C A L E C O N O M I C S 6 8 ( 2 0 0 9 ) 8 5 0 – 8 6 0

Author's personal copy

(2b), if consumers with a chronic promotion orientation eat“free-range meat”, they have identified it as the best alter-native in the choice set as reflected by its strong positivepromotion associations. The bottom half of Table 2 displayshow the target behaviors “eating meat produced by intensivefarming” and “eating free-range meat” were rated on thepromotion- and prevention-related associations. If both partsof the hypothesis were true, the analysis of variance wouldshow a significant interaction effect of the factors Preventionor promotion association, Target behavior, Own behaviorand Chronic goal orientation. This interaction was sig-nificant, although it was not very strong (four-way interaction,F(9,903)=2.38, partial η2= .02, pb .05). As Table 2 demonstrates,there were many differences between the mean ratings:prevention associations were more positive than promotionassociations (effect of Association, F(1,903)=73.19, partialη2= .08, pb .001), and “eating free-range meat” generatedmuch higher positive associations than “eating meat pro-duced by intensive farming” (effect of Target, F(1,903)=308.56,partial η2= .26, pb .001), especially in terms of preventionassociations (interaction of Association×Target, F(1,903)=43.14, partial η2= .05, pb .001). The profileswere also dependentof the participants' own behavior and this relationship waselaborated in four regression analyses. The hypothesizedmoderating impacts of chronic promotion or preventionorientation were elaborated by regressing the participants'own behavior on the promotion and prevention associationsfor each of the four categories of chronic goal orientationseparately. The results are presented in Table 3. The regres-sion analysis revealed that those who were most prevention-oriented and used to eat free-range meat were not differentfrom the others because of their positive prevention associa-tions to “free-range meat” but because of their negativeprevention associations to “meat produced by intensivefarming” (i.e. negative sign of prevention associations, ß=− .41). This is in line with the hypothesis (2a) that theseconsumers were motivated by rejecting unacceptable alter-natives from their choice set. As noted above, positiveprevention associations to “free-range meat” were quitecommon, also among those who did not buy this type ofmeat, but positive promotion orientations made a differencein all the four categories of chronic goal orientation. Inaddition, the analysis indicated that consumers who weremost promotion-orientated could have positive reasons to buy

meat produced by intensive farming (positive promotionassociations, ß=− .26) or to buy free-range meat (positivepromotion associations, ß=.31). This agrees with the hypoth-esis (2b): if consumers with a chronic promotion orientationeat “free-range meat”, they have identified it as the bestalternative in the choice set as reflected by its strong positivepromotion associations.

In addition to the supportive evidence for the hypotheses,some other results should be mentioned. An important pointis the large percentage (50%) who did not know where theirmeat comes from. Like most Dutch consumers, this groupmight in fact buy meats that originate from intensive farmingwithout being explicitly informed about this, for examplethrough on-package information. These circumstances makeit easy for them to remain unaware of the methods of meatproduction. The mean ratings of promotion- or prevention-related associations made by this group were not significantlydifferent from those of the participants who said to eat meatproduced by intensive farming (Mann–Whitney U test, fourtimes pN .10). Nevertheless, if the “don't know” or “otheranswers” category was left out of the analysis, the correlationand regression coefficients increased somewhat. Similardifferences in effect-size are presented in Table 3; unlike themost prevention- and the most promotion-oriented category,the participants in the two intermediate categories did notproduce very outspoken results. For many participants, foodchoices were probably not a salient part of their daily life andthat was a limiting factor for the strength of the effects.

A final point is the relationship between the two pairs ofopposing behaviors, type of meat and quantity of meat.Especially because of the premium price that consumershave to pay for free-range meat, some of them might havedeveloped the strategy to combine a higher quality with asmaller quantity. There was a significant but small correlationbetween (1) buying free-range meat (versus meat fromintensive farming) and (2) eating small (versus large) portionsof meat (r=.10, pb .01). This correlation became somewhatstronger by taking into account not just the portion size butalso other measures of being low or high on meat, such as thenumber of days per week that meat is part of one's hot mealand the number of days per week that one eats a meat-substitute. However, even then the coefficient is still low(r=.18, pb .001). After leaving out the participants who did notknow where their meat comes from, the correlation became

Table 3 – Regression of the participants' own behavior a on the associations to the target behaviors in the four categories ofchronic goal orientation (standardized ßs)

Chronic goal orientation category

Most prevention Most promotion

1 2 3 4

Prevention associations to “eating meat produced by intensive farming” − .41⁎⁎ − .02 − .09 .06Prevention associations to “eating free-range meat” .03 − .05 .01 − .13Promotion associations to “eating meat produced by intensive farming” .13 − .18⁎ − .04 − .26⁎

Promotion associations to “eating free-range meat” .31⁎⁎ .28⁎⁎⁎ .20⁎ .31⁎⁎

R square .25 .12 .07 .12N 133 289 354 143

⁎pb .05; ⁎⁎pb .01; ⁎⁎⁎pb .001.a 4-point scale: 1: usually eats meat produced by intensive farming, 2: don't know, 3: sometimes eats free-rangemeat, 4: usually eats free-rangemeat.

857E C O L O G I C A L E C O N O M I C S 6 8 ( 2 0 0 9 ) 8 5 0 – 8 6 0

Author's personal copy

again somewhat stronger (r=.21, pb .001). Apparently, thistype of trade-off was weakly supported by our data.

4. Discussion

In line with our hypotheses we found that consumers with achronicpreventionorientationoftenmadetheirmeat choices ina way that ensured the absence of negative outcomes to theirdesired end-state. Consistent with hypothesis 1a, these con-sumers used to eat small portions ofmeat, which they saw as amatch to their desired end-state, helping them to avoid themismatchof “large portions”. Consistentwithhypothesis 2a,wefound that consumers with a chronic prevention orientationused to eat free-rangemeat (and paid its premiumprice) if theyconsidered “meat produced by intensive farming” a mismatch.Conversely, the meat choices of consumers with a chronicpromotion orientation were often motivated by the matchesthey would like to attain. The link between chronic promotionorientation and “large portions” was less strong than expectedinhypothesis 1b, however.Although eating “large portions”wasnot seen as a stimulating accomplishment, “small portions”were rated as a disappointment. Consistent with hypothesis 2b,we found that consumerswith a chronic prevention orientationused to eat free-rangemeat (and paid its premiumprice) if theyconsidered it the best from their choice set.

Formany participants, meat choices were not a salient partof their daily life and it underlines the power of Higgins' self-regulation theory that we found supporting evidence for thetwo hypotheses in this context still. The impacts of a chronicprevention orientation agree with the literature about organicfood consumers, who are known to avoid risks and to value agood conscience (Homer and Kahle, 1988; Magnusson et al.,2003; Schifferstein and Oude Kamphuis, 1998; Torjusen et al.,2001). In this context, Higgins' self-regulation theory isextremely important as it provides a coherent framework tounderstand the relationships between the motives of theseconsumers. The theory explains, for example, why consumersmay combine the avoidance of risks to their health with thefulfillment of moral responsibilities. The point is that bothmotives fit well into a prevention focus, as is illustrated by thepositive loadings of being “very mindful of food” and being“grateful for her meal” on the second unrotated component inTable 1. In addition, it is interesting to note that free-rangemeat also had an appeal to consumers with a chronicpromotion orientation regarding food. These consumers useto be driven more by taste than by moral and health issues(Florack and Scarabis, 2006; Spiegel et al., 2004) and they mayassociate free-range meat with sensory quality, for example.

Overall, “eating free-range meat” generated more positiveassociations than “eating meat produced by intensive farm-ing”, especially in terms of prevention associations. To put itsimply, many consumers were not very enthusiastic aboutintensive farming methods and they considered free-rangemeat more desirable. The latter consideration, however, wasnot a decisive factor to buy free-rangemeat. This type of resultresembles the well-known discrepancy that has often beendescribed in the literature as the weak link between a pro-environmental attitude and pro-environmental behavior (Bieland Dahlstrand, 2005; Tanner et al., 2004; Thøgersen and

Ölander, 2006). However, based on Higgins' theory about theregulation of goal directed behavior in terms of matches andmismatches we can take this work one step further. The pointis that those who were prevention-oriented and used to eatfree-range meat were not different from the others because oftheir positive prevention associations to “free-rangemeat” butbecause of their negative prevention associations to “meatproduced by intensive farming”. For them, eating the lattertype of meat would be the equivalent of a failure to avoid amismatch. In other words, knowing that consumers have apro-environmental attitude is not enough to gain an under-standing of what they see asmatches andmismatches to theirdesired end-state. Next, it is crucial to assess whether theytend to use approach or avoidance strategies to reach theirgoals. As our results demonstrated, both types of consumersmay have quite different reasons for choosing the sameproduct and paying its premium price.

In our analysis we assumed that “getting enough nourish-ment by eating the right food” is a desired end-state for manyconsumers and that their meat choices can be understood interms of matches and mismatches to this broadly denotedstate. It is feasible that some consumers had more specificgoals in which any link with conventional agriculture isdeemed undesirable. Such an anti-goal as end-state givesrise to discrepancy-amplifying (instead of discrepancy-redu-cing) inclinations. The consequences of discrepancy-amplify-ing inclinations depend on the person's chronic goalorientation (Higgins, 1997, 1998, 2000). If these consumershave a chronic prevention orientation, their preferred strategyis to avoid any potential matches to the anti-goal, which mayresult in avoiding all kinds of meat. Indeed, a few participantswith a chronic prevention orientation had to be left out of theanalysis because they did not eat meat. If a link withconventional agriculture is an undesirable end-state to con-sumers with a chronic promotion orientation, they may focustheir preferred approach strategy on mismatches to the anti-goal. Theoretically, choosing free-range meat might be theresult of negative reasons (i.e. as a mismatch to the undesiredend-state of a link with conventional agriculture). Althoughwe cannot rule out this interpretation, the positive impacts ofthe promotion associations presented in Table 3 seem to bemore inspired by a strategy to reduce discrepancies with adesired end-state. Generally, undesirable end-states provideunstable feedback loops if they just call for enlarged dis-crepancies between the present state and the aversivereference values, without goals to head towards (Carver,2006; Carver and Scheier, 2002).

The practical implications of our work are closely related tothe different ways in which consumers may try to achievedesirable end-states. Policy-makers in government and indus-try can make the notion of food sustainability more appealingby taking into account how consumers with a prevention- anda promotion-orientation consider their choice sets. For pre-vention-oriented consumers it is important to have foodoptions that agree with a lot of constraints, including quality,health and moral aspects. For promotion-oriented consumersit is important that themore sustainable option in their choiceset is also the best option in terms of other promotion-relatedaspects, such as taste. Interestingly, our results indicated that“large portions of meat” might not be especially attractive to

858 E C O L O G I C A L E C O N O M I C S 6 8 ( 2 0 0 9 ) 8 5 0 – 8 6 0

Author's personal copy

these consumers. This might support the strategy to combinea higher quality with a smaller quantity. Policy-makers canuse these considerations to identify and breakdown barriersfor the pursuit of food sustainability, to reinforce “greenmarketing” and to improve consumers' satisfaction with theirconsumption. Enabling people to get the experience of “feelingright” about their choices may improve their ability to livebetter by consuming less and reduce their impact on theenvironment in the process (Jackson, 2005).

One of the limitations of this paper is that themotivationalorientations were not under experimental control. The resultswere dependent on the degree to which the participants wereable and willing to understand the subtle differences betweenthe questions. Due to the common positive wording of thestatements, prevention-related and promotion-related asso-ciations were highly correlated, which reduces the power ofthe analytical methods (Boldero et al., 2005; Higgins et al.,1997). A closely related issue is the validation of thedifferences between promotion and prevention. Our approachwas based on the assumption that that promotion orientationis linked to value priorities that emphasize stimulatingexperiences and prevention orientation to value prioritiesthat emphasize safety (Fellner et al., 2007; Kluger et al., 2004;van Dijk and Kluger, 2004). Although our results were quitesatisfactory and Higgins' theory has proven its value in manysettings (Higgins, 1997, 1998, 2000), further work is necessaryto improve thesemeasures for research in the field of food andsustainability. Because the alpha values of several scales werenot very high (i.e. .55 being the lowest), it is crucial to increasethe number of items that measure the direction of motiva-tional orientations regarding food. Further research shouldalso examine the relationship between promotion, preventionand quality of life (Grant and Higgins, 2003) and the relevanceof this motivational approach to other areas of consumption.

Another limitation is that this study is based on single-country cross-sectional data, i.e. on consumers in the Nether-lands. Generalization of the findings to the broader populationmay be limited by characteristics of the sample, the samplingmethod and the geographical scope of the study. In particular,the cross-cultural validity of themotivational associations thatfavor free-range meat or less meat has yet to be determined.

5. Conclusion

The insights of Higgins on goal orientations are crucial for theaim to foster more sustainable food consumption andproduction patterns. Our results demonstrated that choosinga small portion of meat was often approached with a pre-vention orientation and choosing free-range meat with eithera prevention or a promotion orientation. These differences inmotivational orientation underline that the pursuit of sustain-ability requires careful consideration of not just undesirablebut also desirable end-states. Promotion and preventionorientations may function as complementary motivationalunderpinnings of food sustainability, because both can in-volve quite different segments of consumers who may havequite different reasons for choosing the same product andpaying its premiumprice. Involvingmore consumers is crucialfor food sustainability, becausemany of themappeared to give

little thought to the links between their consumption beha-viors and the way in which their meat had been produced.Increasing their involvement can be a matter of highlightingthe matches they might like to attain, such as links with free-ranging animals, and themismatches theymight like to avoid,such as links with environmental deterioration. However,making food production more transparent will only be a firststep to maximize consumer motivation. In addition, theyshould be enabled to experience a psychological “fit” betweentheir chronic promotion or prevention orientation regardingfood and their means of goal pursuit by being eager or beingcareful, respectively. It is this “fit” that can give them theexperience of “feeling right” about what they are doing.

Acknowledgements

This paper is based on work funded by the ZonMw (Nether-lands Organization for Health Research and Development)under its program “Nutrition: Health, Safety and Sustainabil-ity” (grant number 014-12-002). Eefje Cuppen and the anon-ymous reviewers provided valuable comments.

R E F E R E N C E S

Aiking, H., de Boer, J., Vereijken, J.M., 2006. Sustainable ProteinProduction and Consumption: Pigs or Peas? vol. 45. Springer,Environment & Policy, Dordrecht.

Avnet, T., Higgins, E.T., 2006. Response to comments on “Howregulatory fit affects value in consumer choices and opinions”.J. Market. Res. 43, 24–27.

Bell, R., Marshall, D.W., 2003. The construct of food involvement inbehavioral research: scale development and validation.Appetite 40, 235–244.

Bernués, A., Olaizola, A., Corcoran, K., 2003. Labelling informationdemanded by European consumers and relationships withpurchasing motives, quality and safety of meat. Meat Sci. 65,1095–1106.

Biel, A., Dahlstrand, U., 2005. Values and habits: a dual-processmodel. In: Krarup, S., Russell, C.S. (Eds.), Environment,Information and Consumer Behaviour. Edward Elgar,Cheltenham, UK, pp. 35–52.

Biologica, 2006. EKO monitor, cijfers en trends, jaarrapport 2005(Annual monitor of the organic sector in the Netherlands).Biologica, available at www.biologica.nl, Utrecht.

Boldero, J.M., Moretti, M.M., Bell, R.C., Francis, J.J., 2005.Self-discrepancies and negative affect: a primer on when tolook for specificity, and how to find it. Aust. J. Psychol. 57,139–147.

Carver, C.S., 2006. Approach, avoidance, and the self-regulation ofaffect and action. Motiv. Emot. 30, 105–110.

Carver, C.S., Scheier, M.F., 2002. Control processes andself-organization as complementary principles underlyingbehavior. Pers. Soc. Psychol. Rev. 6, 304–315.

CBS, 2005. De Digitale Economie 2005. Centraal Bureau voor deStatistiek (Statistics Netherlands), Voorburg.

Chernev, A., 2004. Goal–attribute compatibility in consumerchoice. J. Consum. Psychol. 14, 141–150.

de Boer, J., Helms, M., Aiking, H., 2006. Protein consumptionand sustainability: diet diversity in EU-15. Ecol. Econ. 59,267–274.

de Boer, J., Hoogland, C.T., Boersema, J.J., 2007. Towards moresustainable food choices: value priorities and motivationalorientations. Food Qual. Prefer. 18, 985–996.

859E C O L O G I C A L E C O N O M I C S 6 8 ( 2 0 0 9 ) 8 5 0 – 8 6 0

Author's personal copy

Fellner, B., Holler, M., Kirchler, E., Schabmann, A., 2007. RegulatoryFocus-Scale (RFS): development of a scale to recorddispositional regulatory focus. Swiss J. Psychol. 66, 109–116.

Florack, A., Scarabis, M., 2006. How advertising claims affect brandpreferences and category-brand associations: the role ofregulatory fit. Psychol. Market. 23, 741–755.

Grankvist, G., Dahlstrand, U., Biel, A., 2004. The impact ofenvironmental labelling on consumer preference: negative vs.positive labels. J. Consum. Policy 27, 213–230.

Grant, H., Higgins, E.T., 2003. Optimism, promotion pride, andprevention pride as predictors of quality of life. Pers. Soc.Psychol. Bull. 29, 1521–1532.

Heller, M.C., Keoleian, G.A., 2003. Assessing the sustainability ofthe US food system: a life cycle perspective. Agric. Syst. 76,1007–1041.

Higgins, E.T., 1997. Beyond pleasure and pain. Am. Psychol. 52,1280–1300.

Higgins, E.T., 1998. Promotion and prevention: regulatory focus asa motivational principle. In: Zanna, M.P. (Ed.), Advances inExperimental Social Psychology. Academic Press, San Diego,CA, pp. 1–46.

Higgins, E.T., 2000. Making a good decision: value from fit. Am.Psychol. 55, 1217–1230.

Higgins, E.T., Shah, J.Y., Friedman, R., 1997. Emotional responses togoal attainment: strength of regulatory focus as moderator.J. Pers. Soc. Psychol. 72, 515–525.

Higgins, E.T., Friedman, R.S., Harlow, R.E., Idson, L.C., Ayduk, O.N.,Taylor, A., 2001. Achievement orientations from subjectivehistories of success: promotion pride versus prevention pride.Eur. J. Soc. Psychol. 31, 3–23.

Higgins, E.T., Idson, L.C., Freitas, A.L., Spiegel, S., Molden, D.C.,2003. Transfer of value from fit. J. Pers. Soc. Psychol. 84,1140–1153.

Homer, P.M., Kahle, L.R., 1988. A structural equation test of thevalue-attitude-behavior hierarchy. J. Pers. Soc. Psychol. 54,638–646.

Jackson, T., 2005. Live better by consuming less? Is there a “doubledividend” in sustainable consumption? J. Ind. Ecol. 9, 19–36.

Kluger, A.N., Stephan, E., Ganzach, Y., Hershkovitz, M., 2004. Theeffect of regulatory focus on the shape of probability-weightingfunction: evidence from a cross-modality matching method.Org. Behav. Human Dec. Process. 95, 20–39.

Lockwood, P., Jordan, C.H., Kunda, Z., 2002. Motivation by positiveor negative role models: regulatory focus determines who willbest inspire us. J. Pers. Soc. Psychol. 83, 854–864.

Magnusson, M.K., Arvola, A., Koivisto Hursti, U.-K., Åberg, L.,Sjödén, P.O., 2003. Choice of organic foods is related toperceived consequences for human health and toenvironmentally friendly behaviour. Appetite 40, 109–117.

Pollard, T.M., Steptoe, A., Wardle, J., 1998. Motives underlyinghealthy eating: using the Food Choice Questionnaire to explainvariation in dietary intake. J. Biosoc. Sci. 30, 165–179.

PVE, 2007. De Nederlandse vee-, vlees- en eiersector in cijfers. Hetjaar 2006 voorlopig (Livestock, meat and eggs in theNetherlands). Productschappen Vee, Vlees en Eieren (ProductBoards for Livestock, Meat and Eggs), Zoetermeer.

Rozin, P., Fischler, C., Imada, S., Sarubin, A., Wrzesniewski, A.,1999. Attitudes to food and the role of food in life in the U.S.A.,Japan, Flemish Belgium and France: possible implications forthe diet-health debate. Appetite 33, 163–180.

Schifferstein, H.N.J., Oude Kamphuis, P.A.M., 1998. Health-relateddeterminants of organic food consumption in the Netherlands.Food Qual. Prefer. 9, 119–133.

Schwartz, S.H., Melech, G., Lehmann, A., Burgess, S., Harris, M.,Owens, V., 2001. Extending the cross-cultural validity of thetheory of basic human values with a different method ofmeasurement. J. Cross-Cult. Psychol. 32, 519–542.

Smil, V., 2000. Feeding theWorld: A Challenge for the Twenty-firstCentury. The MIT Press, Cambridge, MA.

Spiegel, S., Grant-Pillow, H., Higgins, E.T., 2004. How regulatory fitenhancesmotivational strength during goal pursuit. Eur. J. Soc.Psychol. 34, 39–54.

Steinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M., deHaan, C., 2006. Livestock's Long Shadow; Environmental Issuesand Options. Food and Agriculture Organization of the UnitedNations (FAO), Rome.

Tabachnick, B.G., Fidell, L.S., 2007. Using Multivariate Statistics.Pearson/Allyn and Bacon, Boston, MA.

Tanner, C., Kaiser, F.G., Wolfing Kast, S., 2004. Contextualconditions of ecological consumerism—a food-purchasingsurvey. Environ. Behav. 36, 94–111.

Thøgersen, J., Ölander, F., 2006. To what degree areenvironmentally beneficial choices reflective of a generalconservation stance? Environ. Behav. 38, 550–569.

Torjusen, H., Lieblein, G., Wandel, M., Francis, C.A., 2001. Foodsystem orientation and quality perception among consumersand producers of organic food in Hedmark County, Norway.Food Qual. Prefer. 12, 207–216.

van Dijk, D., Kluger, A.N., 2004. Feedback sign effect onmotivation:is it moderated by regulatory focus? Appl. Psychol.: Int. Rev. 53,113–135.

Verbeke, W., Vackier, I., 2004. Profile and effects of consumerinvolvement in fresh meat. Meat Sci. 67, 159–168.

Wier, M., Calverley, C., 2002. Market potential for organic foods inEurope. Br. Food J. 104, 45–62.

Zhou, R.R., Pham, M.T., 2004. Promotion and prevention acrossmental accounts: when financial products dictate consumers'investment goals. J. Consum. Res. 31, 125–135.

860 E C O L O G I C A L E C O N O M I C S 6 8 ( 2 0 0 9 ) 8 5 0 – 8 6 0