Are All Out-Groups Created Equal? ConsumerIdentity and Dissociative Influence

Journal of Consumer Research, Inc. Are All Out‐Groups Created Equal? Consumer Identity and Dissociative Influence Author(s): Katherine White and Darren W. Dahl Reviewed work(s): Source: Journal of Consumer Research, Vol. 34, No. 4 (December 2007), pp. 525-536 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/10.1086/520077 . Accessed: 28/08/2012 23:08 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org. . The University of Chicago Press and Journal of Consumer Research, Inc. are collaborating with JSTOR to digitize, preserve and extend access to Journal of Consumer Research. http://www.jstor.org 525
2007 by JOURNAL OF CONSUMER RESEARCH, Inc. ● Vol. 34 ● December 2007 All rights reserved. 0093-5301/2007/3404-0008$10.00 Are All Out-Groups Created Equal? Consumer Identity and Dissociative Influence KATHERINE WHITE DARREN W. DAHL* Past research finds that consumers exhibit weak self-brand connections to brands associated with out-groups. We extend this work by demonstrating that products associated with dissociative reference groups have a greater impact on consumers’ negative self-brand connections, product evaluations, and choices than do products associated with out-groups more generally. In addition, both situational priming and chronic identification with one’s in-group moderate the avoidance of products associated with dissociative reference groups. Further, we demonstrate the conditions under which dissociative influence does not occur and discuss the implications of the research. Marketers often wish to position their brand in a way that accentuates the positive aspects of the brand yet highlights the negative aspects of a competitor’s brand. One strategy for achieving this is to link the competitor’s brand with a group that the consumer does not wish to be associated with. For example, in their award-winning “Shady Acres” advertisement, Pepsi juxtaposed a fraternity of Pepsi drinkers against a group of nursing-home residents drinking Coke. It was only when the soft-drink delivery was switched that the stereotyped behaviors changed. This tactic utilizes the dissociative reference group—a group that the individual is particularly motivated to avoid being associated with. Presumably, nursing-home patients represent a dissociative group for Pepsi’s target market, and Pepsi hopes that consumers will instead choose the alternative, nondissociative option—Pepsi. Past research has largely focused on positive reference groups (i.e., those groups that individuals wish to be associated with), identifying the central role they can play in determining attitudes and behaviors (see White and Dahl *Katherine White is assistant professor of marketing at Haskayne School of Business, University of Calgary, 2500 University Dr. N.W., Calgary, AB, Canada, T2N 1N4 (kate.white@haskayne.ucalgary.ca). Darren W. Dahl is Fred H. Siller Professor of Marketing at Sauder School of Business, University of British Columbia, Vancouver, BC, Canada, V6T 1Z2 (darren.dahl@sauder.ubc.ca). The authors would like to thank Peter Darke, Dale Griffin, JoAndrea Hoegg, Andrea Morales, Jaideep Sengupta, and Leaf Van Boven for helpful comments on an earlier version of this article. The financial support from the Haskayne School of Business and the Social Sciences and the Humanities Research Council of Canada is gratefully acknowledged. John Deighton served as editor and Jennifer Aaker served as associate editor for this article. Electronically published June 28, 2007 2006). Indeed, consumers often are influenced by members of their own group (e.g., Bearden and Etzel 1982; Moschis 1976; Whittler and Spira 2002) and those they aspire to be like (e.g., Escalas and Bettman 2003; Sen, Gu¨rhan-Canli, and Morwitz 2001). The current research focuses on the dissociative reference group, which is a group with which the individual wishes to avoid being associated and feels a sense of disidentification (e.g., Englis and Solomon 1995). Although the role of dissociative reference groups in consumer behavior has received relatively little attention in the literature (cf. White and Dahl 2006), anecdotal evidence suggests that consumers often are influenced by the desire to avoid particular group memberships—baby boomers will not use products associated with being “elderly,” men do not want to dress in clothing that makes them look “feminine,” and teenagers do not wish to be seen wearing accessories that are associated with the “uncool” group. We focus on the implications such dissociative reference groups have for consumers’ self-brand connections, evaluations, and choices. Recently, Escalas and Bettman (2005) examined the differentiation between the in-group and the out-group as reference groups related to consumer self-brand connections (i.e., the extent to which individuals have incorporated brands into their self-concepts; Escalas 2004). These researchers demonstrated that consumers have stronger selfbrand connections to brands consistent with an in-group than brands inconsistent with an in-group and weaker self-brand connections to brands consistent with an out-group than brands inconsistent with an out-group. Further, these effects were more pronounced for brands that were relatively more symbolic (i.e., brands that communicated something to others about the user’s self-identity). Escalas and Bettman (2005) demonstrated an important point: groups to which 526 JOURNAL OF CONSUMER RESEARCH consumers do not belong can have implications for consumer outcomes such as negative self-brand connections. We extend this research by proposing that consumers are not always motivated to avoid out-group memberships and that it is dissociative reference groups that will most strongly influence negative self-brand connections as well as consumer evaluations and choices. We believe it is necessary to differentiate dissociative reference groups from out-groups more generally because, while there are surely many out-groups that people are not concerned about (e.g., I am not a soccer player, but that group does not have motivational implications for me) and aspire to be members of (e.g., I am not a model, but I wish I were), dissociative groups are out-groups people are motivated to avoid being associated with (e.g., I am not a Goth and I wish to avoid being associated with that group). Thus, we believe that, rather than examining out-groups more generally, it is more telling to examine the effects of specific types of out-groups. SOCIAL IDENTITY THEORY AND REFERENCE GROUP INFLUENCE Social identity theory (Tajfel and Turner 1979) and social categorization theory (Turner 1985) propose that identity comprises both personal identity (i.e., derived from an individual sense of self) and social identity (i.e., related to groups to which one belongs or is affiliated). Different contexts can cause temporary shifts in identity, such that the individual categorizes the self in terms of one group membership (Tajfel and Turner 1986). That is, depending on the context, the individual may see the self in terms of one of several possible social memberships (e.g., Australian, sister, student, soccer player); come to more strongly identify with the activated identity; and think, feel, and act as a group member rather than as an individual (e.g., Markus and Kunda 1986; Tajfel and Turner 1986). Importantly, one’s group memberships can become a vital part of the selfconcept, and people are motivated to see the self and, as a result, their group memberships positively. One way consumers might maintain positive views of themselves is by exhibiting negative self-brand connections to brands associated with out-groups, as Escalas and Bettman (2005) found. People often differentiate themselves from others (Berger and Heath 2007) and distinguish between in-groups and out-groups (Marques, Abrams, and Paez 1998). We believe, however, that consumers are not only motivated to differentiate themselves from out-groups but, rather, that they are motivated to positively differentiate the self from out-groups. That people strive for positive distinctiveness from out-groups is a key postulate of social identity theory. People often are motivated to achieve a positive social identity (and avoid a negative social identity) and use a variety of strategies to do so, such as decreasing affiliations with groups that do not confer positive associations (e.g., Jackson et al. 1996), evaluating the in-group more positively (e.g., Brewer 1979; Jackson et al. 1996), and avoiding products associated with negatively viewed social identities (Tepper 1994; White and Argo 2007; White and Dahl 2006). Thus, we believe that consumers will be particularly motivated to avoid dissociative associations. THE CURRENT RESEARCH Past research does not elucidate whether certain outgroups exert a greater influence on consumers than others. While Escalas and Bettman (2005) looked at out-groups more generally, White and Dahl (2006) have provided preliminary evidence that a specific type of out-group—the dissociative reference group—can influence consumer evaluations and choices in a context where self-presentation concerns are relevant. In particular, men avoided the dissociative associations of a product named the “ladies’ cut steak,” particularly when it was to be consumed in public and when the consumer was high in public self-consciousness. In addition, although not framed in terms of dissociative reference groups, Tepper’s (1994) research found that consumers with younger age identities viewed being labeled as a senior citizen as being stigmatizing and avoided offerings with the presumably dissociative label “senior citizen discount.” Taken together, the findings of Tepper (1994) and White and Dahl (2006) suggest that dissociative reference groups may be particularly influential on consumer evaluations and choices. The goal of the current research is to clarify and extend the results of Escalas and Bettman (2005) and to integrate their findings with those of White and Dahl (2006). In doing so, we make several contributions to the literature. First, we show that not all out-groups are created equal and that dissociative reference groups have a stronger influence on consumers than do out-groups more generally. Second, we not only demonstrate this with regard to consumers’ self-brand connections but also with respect to product evaluations and choices. We investigate this by using self-identified brands (study 1) as well as by manipulating reference groups in a more involving setting, where consumers evaluate real products (studies 2, 3, and 4). Third, we demonstrate that although the tendency to avoid dissociative reference groups is heightened when the product is more symbolic in nature, this effect continues to persist when the product is relatively less symbolic. Furthermore, we extend past research suggesting that the mechanism underlying dissociative influence is self-presentation concerns (White and Dahl 2006) and highlight the finding that private self-identity is also an important determinant of the motivation to avoid products associated with dissociative groups. That is, we go beyond past work showing that impression management motivates dissociative influence to demonstrate that the importance of private self-identity and situational shifts in private selfidentity also are key determinants of dissociative influence. Finally, we demonstrate the dynamic nature of dissociative influence by showing the conditions under which it does not occur. CONSUMER IDENTITY AND DISSOCIATIVE INFLUENCE 527 STUDY 1 In study 1, we examine whether dissociative reference groups are more influential than out-groups more generally in determining consumer self-brand connections and evaluations. Specifically, we predicted that H1: Participants will report weaker self-brand connections and more negative evaluations regarding a brand associated with a dissociative reference group than a brand associated with an out-group more generally. In order to make a complete comparison, participants also evaluated brands associated with in-groups and neutral groups. We anticipate that neutral brands will be viewed similarly to the out-group brands in terms of self-brand connections and evaluations and that only the dissociative brands will differ from all others in terms of negative selfbrand connections and product evaluations. It also seems likely that participants will have the most positive self-brand connections and evaluations regarding the in-group brands. In addition, we predict that people will be more motivated to avoid products associated with dissociative reference groups when the brand is highly symbolic in nature (i.e., when the brand communicates information about the self to others). More formally: H2: Participants will report weaker self-brand connections and more negative evaluations of the dissociative brand when the brand is viewed as being relatively more symbolic. Method Participants. Fifty-five undergraduates from the University of Calgary took part. Data from 10 original participants were removed because of difficulty thinking of one or more of the brand categories, leaving a total of 45 participants in the analyses (see Escalas and Bettman [2005] for a similar procedure). Procedure. Participants took part using computers with MediaLab software, which allowed the stimuli presented to be customized based on their responses. Participants were first asked to identify three different types of groups (i.e., an in-group, an out-group, and a dissociative group) and a brand that corresponded to each group. The procedure used was adapted from Escalas and Bettman (2005; see the appendix for the instructions). The order of the in-group, outgroup, and dissociative prompts was counterbalanced. Participants were then given a filler task that included some demographic and personality measures. After this task, participants completed ratings of each of the three identified brands in counterbalanced order along with two brands that pretested as being neutral with regard to students’ self-concepts (Avis and United Airlines). The purpose of the neutral brands was to decrease demand and to serve as a neutral comparison point. In order to analyze the results for the neutral brands, averages were calculated across the two neutral brands for all measures. Manipulation checks confirmed that our manipulation of the reference group was successful. The first dependent variable was a seven-item measure of self-brand connections that was completed on nine-point scales (Escalas and Bettman 2005; e.g., “This brand reflects who I am”). The second dependent variable was a measure of consumer evaluations (White and Dahl 2006), which consisted of three items completed on nine-point scales (unfavorable/favorable, dislike/like, and bad/good). Finally, participants completed a two-item measure of how symbolic the brand was perceived as being on nine-point scales (Escalas and Bettman 2005). Participants completed the dependent measures for each brand identified. Finally, participants were given a suspicion probe, which indicated that participants were unaware of the study’s hypotheses. Results and Discussion Brands Identified. Groups and the corresponding brands identified by participants were very idiosyncratic; for example: “jocks/Nike,” “rich girls/Lululemon,” and “smokers/ Players” for dissociative; “my friends/Tim Hortons,” “international students/Puma,” and “dance community/Lululemon” for in-group; and “first-year girls/Roxy,” “graduate students/ Tim Hortons,” “business students/Banana Republic” for outgroups. Thus, for different people the same brand carried different meanings. While one individual found Tim Hortons to relate to an out-group, another found it to relate to an ingroup; while one individual found Lululemon to relate to a dissociative group, another found this brand to relate to an in-group. Self-Brand Connections. Because each participant completed evaluations of each brand category, a repeated measures ANOVA was conducted on the measure of selfbrand connections (a’s ranging from .92 to .95). A significant main effect for the reference group emerged (F(3, 44) p 53.70, , Cohen’s ), revealing stronger self- p ! .001 d p 2.21 brand connections to in-group brands ( ) than to M p 5.07 neutral brands ( ), out-group brands ( ), M p 2.58 M p 2.57 and dissociative brands ( ). Consumers reported M p 1.92 more negative self-brand connections to the dissociative brands than to both out-group brands ( , t(44) p 2.03 p ! .05, ) and neutral brands ( , , d p .61 t(44) p 2.06 p ! .05 d p .62). Interestingly, participants’ self-brand connections did not differ between the out-group brands and the neutral brands ( , NS; refer to table 1). t(44) p .03 Brand Evaluations. A repeated measures ANOVA on evaluations (a’s from .88 to .92) revealed a significant main effect ( , , ). Participants F(1, 44) p 60.21 p ! .001 d p 2.34 reported more positive evaluations of in-group brands ( ) than of neutral ( ), out-group ( M p 7.50 M p 5.46 M p 6.02), and dissociative ( ) brands. Also, partici- M p 4.24 pants evaluated the dissociative brand more negatively than both the neutral ( , , ) and out- t(44) p 3.39 p ! .01 d p 1.02 group brands ( , , ). Out- t(44) p 5.80 p ! .001 d p 1.75 528 JOURNAL OF CONSUMER RESEARCH TABLE 1 SELF-BRAND CONNECTIONS AND BRAND EVALUATIONS AS A FUNCTION OF REFERENCE GROUP-BRAND ASSOCIATIONS (STUDY 1) In-group Group association Neutral Out-group Dissociative Self-brand connections M p 5.07a M p 2.58b M p 2.57b M p 1.92c Evaluations M p 7.50a M p 5.46b M p 6.02c M p 4.24d NOTE.—Within rows, means with differing subscripts differ at the level. p ! .05 group brands were evaluated somewhat more positively than the neutral brands ( , , ), con- t(44) p 1.82 p ! .08 d p .55 firming that it was only the dissociative brands that elicited more negative evaluations than all other conditions. Degree of Brand Symbolism. It was predicted that dissociative brands would be related to weaker self-brand connections and more negative evaluations when they were perceived as being more symbolic. We computed a symbolism index for each type of brand (Pearson r’s ranging from .78 to .92, all ). As predicted, self-brand con- p ! .001 nections were weaker for dissociative brands when the product was seen as being more symbolic ( , r p
.29 p p .05), and dissociative brands were rated more negatively when the product was viewed as being more symbolic ( , ). When evaluating in-group r p
.36 p ! .02 brands, there was a positive correlation between self-brand connections and symbolism ( , ) and no sig- r p .44 p ! .01 nificant correlation between evaluations and symbolism ( , ). For out-group brands, the correlations r p .14 p ! .40 between symbolism and self-brand connections ( , r p .26 p ! .09) and evaluations ( , ) did not reach r p .28 p ! .08 significance. However, participants showed a trend toward evaluating the out-group brands more positively when the brands were perceived as being relatively more symbolic. The results suggest that dissociative reference groups are important in determining negative self-brand connections and negative evaluations. Further, out-group brands were rated no differently than neutral brands on the measure of self-brand connections. It was only dissociative brands that were negatively differentiated from the other types of brands across both measures. These findings support our proposition that it is dissociative reference groups that are particularly relevant for consumers’ negative self-brand connections and evaluations and that consumers are not significantly influenced by groups of which they are simply not a part. As anticipated, consumers demonstrated the weakest selfbrand connections and most negative evaluations of a brand associated with a dissociative reference group when the brand was more symbolic in nature. Out-group brands were not evaluated more negatively when they were relatively more symbolic. This suggests that dissociative reference groups operate differently than out-groups more generally and that dissociative influence is more pronounced when the brand conveys information to others about the consumer’s identity. Because White and Dahl (2006) and study 1 provide evidence that dissociative influence is related to presenting a particular self-image to others, in study 2 we examine the role of private self-identity in determining dissociative influence. STUDY 2 In study 2 we seek to broaden our investigation by using nationality to manipulate reference group. In addition, we hold product type constant and utilize a relatively nonsymbolic product to examine the conditions under which the private self relates to dissociative influence. We differentiate the private self from the public self because research shows that consumers are motivated to feel positively about both the public and the private self and that these two concerns are distinct (e.g., Argo, White, and Dahl 2006). Specifically, we test whether salience of in-group identity moderates dissociative influence. A key prediction of social identity theory is that it is only when a particular aspect of social identity becomes salient that people begin to think, feel, and behave on the basis of the activated identity. Social identity research has found that people are more likely to be persuaded by in-group members (McGarty et al. 1994) and to self-stereotype (Hogg and Turner 1987) when ingroup identity is salient. In the consumer domain, identity salience has been shown to relate to the persuasiveness of spokespersons (Deshpande´ and Stayman 1994), advertising response (Forehand and Deshpande´ 2001; Forehand, Deshpande´, and Reed 2002), and consumer preferences (LeBoeuf, Shafir, and Belyavsky 2007). We predict that dissociative influence will be most pronounced when the consumer’s in-group identity is primed. We use Canadian identity as the in-group and operationalize a dissociative identity as being American. It has been suggested that the distinctiveness of Canadian identity is often defined in contrast to Americans (Lalonde 2002; Smith 1994). Thus, we anticipate that Canadians will attempt to differentiate themselves from Americans and will do so by reporting more negative attitudes toward a product that highlights a dissociative (i.e., American) nationality than one that highlights an out-group identity or a neutral identity. However, we expect this effect to be more pronounced when national identity is primed. In particular: H3: Canadians will rate a product that is associated with a dissociative reference group (i.e., American) more negatively than a product associated CONSUMER IDENTITY AND DISSOCIATIVE INFLUENCE 529 FIGURE 1 PEN EVALUATIONS AS A FUNCTION OF REFERENCE GROUP LABEL AND IDENTITY PRIMING (STUDY 2) with an out-group or a neutral product, particularly when their own national identity is primed. Method Procedure. Two hundred eighteen undergraduates from the University of British Columbia took part. Participants took part in small groups and were asked to make evaluations of a variety of stationery products. This study used a 3 (reference group label: dissociative vs. out-group vs. neutral) # 2 (priming: identity prime vs. no identity prime) between-subjects experimental design. Participants took part in a session in which they were told that they would be completing two separate studies. At the beginning of the session, participants completed a questionnaire that was ostensibly the first study and served as our priming manipulation. Participants in the identity prime condition completed a task to prime their Canadian identity, whereas those in the no identity prime condition completed a neutral task (see the appendix). Participants then completed the second portion of the experiment, which involved evaluating a series of stationery products (e.g., mechanical pencils, highlighters), including a pen. The pens used were thick mechanical pens that were fairly inexpensive, and two versions of the similar styles of pen in the same color (blue) were used (and the styles were counterbalanced). Based on a pretest, we identified product labels that were associated with different types of groups: “American” pen was viewed as being associated with an out-group that was dissociative, “Belgian” pen represented an out-group that was nondissociative, and “vintage” pen represented a pen that was not associated with an out-group but was neutral with regard to dissociative concerns. Thus, participants evaluated the vintage pen in the neutral reference condition, the Belgian pen in the out-group condition, and the American pen in the dissociative condition. The stimuli were parallel across treatments except for the manipulated label. Participants evaluated the pen (and the other stationery items used in the task) using the same evaluation measures as in study 1 ( ). In addition, participants a p .92 completed a suspicion probe, which indicated that participants were not aware of the purpose of the priming manipulation or our hypotheses. Results and Discussion A 3 (reference group label: dissociative vs. out-group vs. neutral) # 2 (priming: identity prime vs. no identity prime) ANOVA on the pen evaluations revealed the expected interaction ( , , ; refer to fig. F(2, 212) p 3.85 p ! .03 d p .27 1). Those in the neutral condition rated the pen similarly in both priming conditions ( and Mprime no prime p 6.33 M p 6.03; , , ). Moreover, those in t(212) p .86 p ! .40 d p .12 the out-group condition also rated both pens similarly across the two priming conditions ( and Mprime no prime p 6.25 M p 5.73; , , ). However, in the dis- t(212) p .93 p ! .36 d p .13 sociative condition, those who received the identity prime rated the pen more negatively ( ) than those who M p 5.27 did not receive the identity prime ( ; M p 6.18 t(212) p 2.60, , ). As predicted, among those who re- p ! .02 d p .36 ceived the identity prime, the dissociative pen was rated more negatively than the neutral ( , , t(212) p 3.03 p ! .01 d p .42) and out-group ( , , ) pens. t(212) p 2.18 p ! .04 d p .30 The main effects for reference group label (F(2, 212) p 1.68, , ) and priming ( , p ! .19 d p .23 F(1, 212) p .01 p ! .91, ) were nonsignificant. d p .12 Participants were once again motivated to avoid a product associated with a dissociative reference group. A pen was evaluated more negatively when it was associated with a dissociative reference group than when it was neutral or associated with an out-group. This finding was only true when participants’ own identity was primed. When participants’ identity was not salient, they did not differentially evaluate the neutral, out-group, and dissociative label products. It appears then, that the activation of one’s own identity may be an important precursor for dissociative influence to occur when the product is relatively nonsymbolic in nature. STUDY 3 Study 2 suggests that activation of the private self-concept is an important boundary condition in determining when dissociative influence will occur. In study 3, we examine the role of the private self in reactions to dissociative reference groups in a different way—by examining the moderating role of chronic in-group identification. Past research finds that compared to low identifiers, those who are high in in-group identification are more likely to perceive withingroup similarity (Doosje, Ellemers, and Spears 1995), to exhibit greater group commitment (Ellemers, Spears, and Doosje 1997), to provide more positive evaluations of ingroup members (Crocker and Luhtanen 1990), and to der- 530 JOURNAL OF CONSUMER RESEARCH ogate the out-group (Branscombe and Wann 1994). That is, individuals high in in-group identification often engage in behaviors that allow them to view the in-group in a positive light. We anticipate that consumers will only be motivated to avoid a product associated with a dissociative reference group when they highly identify with the in-group: H4a: As in-group identification increases, consumers will be more likely to rate a product associated with a dissociative reference group more negatively than a neutral product but will rate a neutral product as being similarly positive to an out-group product. H4b: As in-group identification increases, consumers will be more likely to choose a neutral product than a product associated with a dissociative reference group but will not show a preference for a neutral product when it is compared with another neutral product or a product associated with an out-group. Finally, we predict that private self-disidentification (i.e., the tendency to disidentify the self with the product associated with a dissociative group) will mediate the influence of the interaction between in-group identification and reference group label on consumer evaluations. In addition, because we use a product that is relatively nonsymbolic in nature (i.e., a product that is not used to communicate information about self-image to others), we predict that concerns about public self-image will not mediate our effects. More formally: H5: Private self-disidentification will statistically mediate the relationship between the interaction between reference group and Canadian identity on product evaluations. Method Procedure. One hundred twenty-six undergraduates from the University of British Columbia took part. Participants took part in small groups and were asked to make evaluations of a variety of stationery products, including two pens. In the dissociative group condition, the focal (i.e., manipulated) pen was labeled “American,” in the out-group condition the focal pen was labeled “Belgian,” and in the neutral condition the focal pen was labeled “vintage.” In addition, in a pretest we also identified the name “classic” as a name that was neutral and was viewed equally positively as the label “vintage.” In each condition, the participant viewed one of the manipulated pen options and the control pen that was labeled “classic.” This study used a 3 (reference group label: dissociative vs. out-group vs. neutral) # 2 (ingroup identification) # 2 (pen evaluations: manipulated pen vs. classic pen) mixed-model design. Because participants were able to make evaluations of both the manipulated pen and the classic pen, this variable was used as the repeated measure (e.g., White and Dahl 2006). Participants evaluated several pairs of products (e.g., two mechanical pencils, two highlighter sets, two pens) and chose which alternative in each product set they preferred. They were told that at the end of the study they would get one of the chosen products and were given the pen that they chose. The same styles of pens were used across all conditions, and the orders of presentation of stationery items, pen style, and pen name were all counterbalanced. Participants then completed a measure of Canadian identification (“Being Canadian has a great deal to do with how I feel about myself”; “Being Canadian is an important part of my self-image”; “Being Canadian is important to my sense of the kind of person I am”; “I have a strong sense of belonging to Canada”; and “I strongly identify with being Canadian”; ) on seven-point scales (strongly dis- a p .94 agree/strongly agree). A measure of private self-disidentification was completed to determine if participants’ choices related to a desire to avoid associating the self with a dissociative group: “I dislike the name associations of this product”; “I want to avoid being associated with this product”; “This product reflects who I do not want to be”; “I would avoid identifying with this product”; “This product reflects who I am” (reverse scored); “I can identify with this product” (reverse scored); “I feel a personal connection to this product” (reverse scored); “This product suits me well” (reverse scored; ). Participants completed a measure a p .80 of public self-image concerns: “To avoid looking foolish”; “To avoid looking badly in front of others”; “To look like I made an intelligent choice”; “To make myself look good to others” ( ). These items were completed on seven- a p .84 point scales for each product option. Results and Discussion Consumer Evaluations. A difference score was computed between evaluations of the manipulated and control pens, such that a higher score reflected a preference for the manipulated pen. To test our predictions, two dummy-coded variables were created; one variable where the out-group label was given the value of one and a second where the dissociative group label was given the value of one (refer to Aiken and West 1991). We used multiple regression to examine the interaction between each dummy variable and the centered in-group identification index, while simultaneously entering the main effects as predictors in the analysis, using the difference score as the dependent variable (West, Aiken, and Krull 1996). As predicted, the interaction between the variable coded to represent the dissociative label and ingroup identification was statistically significant (t(116) p 3.25, , ). The interaction between the vari- p ! .002 b p .41 able dummy coded to represent the out-group label and ingroup identification was not significant ( , t(116) p .76 p ! .45, ). As can be seen in figure 2, those who were b p .13 high in in-group identification had the least favorable evaluations of the American pen. Importantly, the difference CONSUMER IDENTITY AND DISSOCIATIVE INFLUENCE 531 FIGURE 2 PEN EVALUATIONS AS A FUNCTION OF REFERENCE GROUP LABEL AND IN-GROUP IDENTIFICATION (STUDY 3) NOTE.—Lines represent the effect of a reference group label for persons one standard deviation unit below the mean, at the mean, and one standard deviation unit above the mean on the measure of in-group identification. Higher values indicate more positive impressions of the manipulated pen (vs. the classic pen). between responses to the dissociative and neutral label was significant for those high in group identification (for those one standard deviation above the mean; , t(116) p 3.25 p ! .01, ) but not significant for those moderate or low b p .47 in in-group identification (both , ; West et t’s ! 1.43 p’s 1 .17 al. 1996). Consumer Choice. Binary logistic regression was conducted with the two dummy-coded variables described above, the continuous mean-centered measure of in-group identification, and the two interaction terms as the independent variables, and pen choice (manipulated pen or control pen) as the dependent variable. As predicted, the interaction term between the dummy variable coded for the dissociative label and in-group identification significantly predicted pen choice ( , , ). A main effect for b p .75 Wald p 5.83 p ! .02 in-group identification also emerged ( , b p .65 Wald p 6.60, ). To enhance the ease of presentation, a median p ! .02 split was conducted on the in-group identification measure. Participants high in in-group identification were less likely to choose the manipulated pen over the classic pen in the American pen (36%) condition, as opposed to those in the Belgian pen (68%) or the vintage pen (72%) conditions ( , ). Among those low in in-group iden- 2 x (2) p 6.55 p ! .04 tification, the differences between those choosing the American pen over the classic pen (52%), the Belgian pen over the classic pen (57%), and the vintage pen over the classic pen (75%) were nonsignificant ( , ). 2 x (2) p 2.50 p ! .29 Mediational Analyses. Regression analysis revealed that the interaction between the centered in-group identification index and the dummy-coded variable representing the dissociative label (including the main effects in the analysis) predicting private self-disidentification was significant ( , , ). When private t(117) p 2.93 p ! .01 b p .37 self-disidentification was included in the regression analysis using the interaction between in-group identification and reference group to predict the difference score, private self-disidentification significantly predicted evaluations (t(115) p 4.01, , ). A Sobel’s test indicated that al- p ! .001 b p .35 though the interaction between in-group identification and reference group continued to predict evaluations when private self-disidentification was included in the analysis ( , , ), there was a significant t(115) p 2.28 p ! .03 b p .28 reduction in the influence of the interaction on evaluations ( , ). Thus, private self-disidentification par- Z p 2.36 p ! .02 tially mediated the effects in this study. The interaction between the centered in-group identification index and the dummy-coded variable reflecting the dissociative reference group label (including the main effects in the analysis) predicting the public self-image concerns was not statistically significant ( , , ). t(117) p 1.07 p ! .29 b p .14 Further, when public self-image concerns were included in the regression analysis using the interaction between ingroup identification and reference group to predict the evaluations difference score, public self-image concerns did not significantly predict evaluations ( , , t(115) p .53 p ! .60 b p .05), and the interaction continued to predict evaluations ( , , ). Thus, public self- t(115) p 3.18 p ! .01 b p .40 image concerns did not play a mediational role in this study. The results revealed that it was only when consumers were high in in-group identification that they had more negative evaluations of and were less likely to choose the American pen. Consumers did not show a tendency to avoid products associated with out-groups more generally. These effects were partially mediated by a desire to differentiate the self from dissociative reference groups but not by public self-image concerns. These results support our contention that the private self is important in determining when dissociative influence will occur. At first glance, our results may seem to contradict the findings of White and Dahl (2006), who found that dissociative influence was related to a desire to present a positive public self-image to others. In that research, participants only demonstrated a desire to avoid dissociative associations when consumption was to occur in public, and this tendency was most pronounced among those with heightened public self-image concerns. In those studies, the public nature of consumption was made very salient to participants, and it is publicly consumed goods that carry the most symbolic meaning (i.e., they convey information about the self to others). In the current studies 2 and 3 we wished to isolate the role that private self-concerns play in determining dissociative influence, and so we utilized products that were relatively nonsymbolic in nature. Such products do not communicate much information about the user to others and, as such, are not readily used to present a positive self-image 532 JOURNAL OF CONSUMER RESEARCH to others.1 Thus, the results show that the desire to avoid dissociative associations persists when the product is nonsymbolic in nature but only under certain boundary conditions—when consumer identity is primed or when the consumer highly identifies with the in-group. We suspect that when the product is more symbolic in nature, public selfimage concerns will increase in importance in determining dissociative influence. STUDY 4 Taken together, the results of studies 1–3 highlight the conditions under which dissociative influence is most pronounced. While studies 1 and 3 demonstrate a chronic tendency to avoid dissociative associations, study 2 demonstrates that a situational manipulation of in-group membership salience can influence people’s evaluations of an alternative associated with a dissociative reference group. In study 4, we further examine the dynamic nature of dissociative influence by exploring the conditions under which dissociative influence does not occur. In particular, we identify situational constraints as a moderator of dissociative influence. Situational constraints refer to temporary circumstances (such as cues that induce selfpresentation concerns) that constrain the individual to act in a certain way, even if the behavior is not necessarily consistent with underlying attitudes (see Wallace et al. [2005] for a review). Consumer behavior research shows that situational constraints that activate self-presentation concerns can influence coupon usage (Ashworth, Darke, and Schaller 2005), product choice (White and Dahl 2006), and lying behavior (Sengupta, Dahl, and Gorn 2002). We examine the role of situational constraints by presenting participants with a situation where there is either strong or weak social pressure to behave in a particular way. We predict that H6: When situational constraints are weak, the dissociative option will be evaluated more negatively than the out-group option. When situational constraints are strong, differences in evaluations of the dissociative and out-group options will be mitigated. Method Procedure. Eighty-one undergraduates from the University of British Columia took part. As in study 2, partic1 To confirm that pens were viewed as relatively nonsymbolic, participants in study 3 were asked to rate a variety of products at the end of the study. For each product they were asked to evaluate on a nine-point scale: “To what extent does —— communicate something about the person who uses it?” (does not communicate a lot/communicates a lot). Participants rated the pen as less symbolic than the scale midpoint ( , M p 3.09 t(117) p 4.29, ), as well as less symbolic than other products such p ! .001 as a T-shirt ( , , ), a purse ( , M p 5.59 t(116) p 10.89 p ! .001 M p 5.57 t(116) p 11.55, ), a car ( , , ), or p ! .001 M p 5.67 t(116) p 12.13 p ! .001 jewelry ( , , ). M p 5.97 t(116) p 12.67 p ! .001 ipants took part in small groups and were asked to make evaluations of a variety of stationery products. This study used a 2 (reference group label: dissociative vs. out-group) # 2 (situational constraints: strong vs. weak) design. First, all participants received the stimuli to prime in-group identity as in study 2. Participants were then told that a new requirement by the research ethics board was that they be provided not only with information about the study itself but also with background information about the experimenter in the form of a brief written description (e.g., Sengupta et al. 2002). In the high situational constraints condition, participants read that their experimenter was born and raised in Madison, Wisconsin, United States of America; completed her undergraduate degree in marketing at the University of Wisconsin; and that she played soccer for the U.S. national under-19 team. In the low-situational constraints condition, participants learned that their experimenter was born and raised in Peterborough, Ontario, Canada; completed her undergraduate degree in marketing at the Wilfred Laurier University; and that she played soccer for Canada’s national under-19 team. The rest of the information about the experimenter remained constant across the two conditions, and participants read that the researcher was pretesting products for a future study and so would be examining participants’ evaluations of the products before the participants left the study. Reference group condition was manipulated as in study 2, with the exception that only the dissociative and out-group conditions were used. That is, the pen was called “American” in the dissociative condition and “Belgian” in the out-group condition. We predicted that when the researcher was believed to be American, this situational constraint would eliminate dissociative effects. The pens were evaluated on the same scales as in the previous studies. Results and Discussion A 2 (reference group label: dissociative vs. out-group) # 2 (situational constraints: strong vs. weak) ANOVA on the pen evaluations revealed the predicted interaction (F(1, 77) p 7.48, , ; refer to fig. 3). As anticipated, partic- p ! .01 d p .62 ipants who experienced weak situational constraints demonstrated more negative evaluations in the dissociative condition ( ) than in the out-group condition ( ; M p 5.43 M p 6.88 t(77) p 3.67, , ). Conversely, participants p ! .001 d p .84 who experienced strong situational constraints demonstrated similar evaluations in the dissociative ( ) and out- M p 6.12 group ( ) conditions ( , , M p 6.04 t(77) p .22 p ! .83 d p .19). Moreover, those under weak situational constraints rated the dissociative option marginally more negatively than those under strong situational constraints ( , , t(77) p 1.80 p ! .08 d p .41). Interestingly, those under strong situational constraints evaluated the out-group pen more negatively than did those under weak situational constraints ( , t(77) p 2.10 p ! .05, ). d p .48 These results show the conditions under which dissociative influence does not arise. When situational constraints were strong, consumers did not differentially evaluate the CONSUMER IDENTITY AND DISSOCIATIVE INFLUENCE 533 FIGURE 3 PEN EVALUATIONS AS A FUNCTION OF REFERENCE GROUP LABEL AND SITUATIONAL CONSTRAINTS (STUDY 4) out-group and dissociative options. Further, participants demonstrated somewhat more positive evaluations of the dissociative option when situational constraints were strong rather than weak. Although dissociative influence was significantly reduced under strong constraints, people also lowered evaluations of the out-group pen under these conditions. It may be that participants under strong constraints evaluated the Belgian pen more negatively because they believed that Americans might look unfavorably upon the Belgian label. This may be due to media reports regarding political and ideological differences between Americans and Europeans (e.g., “Europeans: From Venus,” New York Post, July 16, 2002; “Don’t Buy American,” Slate, January 15, 2005; and “Mad at America,” Time, January 20, 2003). In effect, high situational constraints led Canadians to behave as though Belgian had become a dissociative group, which although unexpected, further highlights the dynamic nature of dissociative influence. GENERAL DISCUSSION Across four studies we demonstrate that dissociative reference groups have important implications for consumer self-brand connections, evaluations, and choices. Consumers showed a greater tendency to avoid products associated with dissociative reference groups than with out-groups, more generally. In study 1, participants self-identified brands associated with different group memberships and exhibited the most negative self-brand connections and evaluations toward the dissociative brand. In studies 2 and 3, participants evaluated real products and showed a tendency to avoid the product associated with a dissociative brand (and not an outgroup brand) in terms of evaluations (studies 2 and 3) and choices (study 3). This was only evident when the individual’s private self-identity was primed or when the consumer was high in in-group identification. Finally, study 4 showed that strong situational constraints can lead consumers to prefer a previously dissociative option even when their own identity is salient. The current studies demonstrate that not all out-groups are created equal and that it is dissociative reference groups that have important implications for consumers. We go beyond past research by showing the importance of dissociative influence in determining not only self-brand connections but also consumer evaluations and choices. We show that although these dissociative effects are pronounced when the brand is more symbolic in nature, these effects continue to persist when the product is less symbolic. Further, we extend past research that highlights the role of public selfimage concerns in determining dissociative influence to show that the private self also relates to dissociative influence. Finally, we show conditions under which consumers will prefer a dissociative option even when their own group identity is salient. We believe the effects in the current studies are driven by a desire to avoid the negative associations of the dissociative referent. Indeed, in study 3 the results were partially mediated by private self-disidentification. However, it would be interesting to further elucidate the mechanism underlying the desire to avoid products related to dissociative reference groups. A likely candidate is that, by virtue of being linked to a dissociative out-group, such products become imbued with negative affective associations, lead to negative affect, and are thus evaluated more negatively. Future research might profitably examine this possibility. One interesting implication of this research is that dissociative reference groups might be utilized in marketing communications. An advertiser might highlight the dissociative associations of not using their brand or the dissociative associations of using a competitor’s brand. Apple has used this strategy in promoting Mac computers. By depicting the PC as an older, outdated, and dowdy referent, Apple hopes that consumers will be motivated to avoid these dissociative associations and instead choose a Mac. However, the current research suggests that such a strategy may be most effective if the advertiser can also situationally activate the in-group membership; appeal to an aspect of identity that is highly identified with; or highlight identities that are chronically accessible, frequently activated, or that represent highly schematic traits (i.e., well-articulated and highly salient aspects of the self; e.g., Aaker [1999]). An additional implication is that when marketers utilize a differentiated marketing strategy that targets multiple and distinct markets, they should be cognizant that one market segment might have dissociative associations regarding another segment. For example, Sketchers was well known for targeting teenaged girls with their line of trendy footwear. More recently, however, Sketchers has expanded to also target younger boys, among other segments. It seems likely that, if the segment of teenaged girls were to associate the brand with the much younger and more immature boys, then 534 JOURNAL OF CONSUMER RESEARCH the girls might dissociate themselves from the brand. Thus, marketers using such an approach should be careful to promote the brand in ways that do not allow for negative crossover effects (e.g., promote the boy’s products via outlets where the girls are unlikely to see them). Such considerations may be warranted when marketers differentially target based on gender, age, ethnicity, as well as other social group memberships. A limitation of the current research is that we developed a study design that would allow us to deal with concepts that are fairly abstract in nature. Thus, in studies 2–4 we examined one particular operationalization of reference group and manipulated this using one product type. Although study 1 adds to the generalizability of the results by showing that people see a variety of different groups and brands as being dissociative, future research could examine dissociative influence based on other types of groups and using different types of products. A promising direction for future research is to examine the conditions under which consumers shift their true private beliefs about dissociative reference groups. In study 4 we found that strong situational constraints mitigated dissociative influence. However, it seems likely that although participants altered their evaluations of the pens in the presence of strong situational constraints, their actual identification with the dissociative group did not change. It would be interesting to examine the conditions under which consumers would truly alter their private beliefs about a product previously viewed as carrying dissociative associations. One possibility is that if consumers are made aware of their self-identity at a broader level of inclusiveness (e.g., Levine et al. 2005), they might be more likely to identify with a previously dissociative product. For example, if Canadians’ identities as North Americans were made salient, they might truly prefer the previously dissociative product because it becomes associated with a broader level of identity. It may be that marketers who can highlight the relevant level of identity to their target market will be better able to encourage positive connections to products that might otherwise be viewed as dissociative. Furthermore, it may be that consumers are more likely to switch their affiliations with certain identities than others. While research often examines the role of stable consumer identities (e.g., gender, nationality) in determining consumer outcomes, other identities that dictate consumer behavior may be more malleable in nature (e.g., country music fan, skateboarder, yoga student). An additional direction for future research could be to examine whether the desire to avoid dissociative reference groups represents a conscious or an unconscious motivation (e.g., Bargh 2002). The results of study 3 suggest that consumers are aware of dissociative associations because they can report them, but further research could investigate whether the avoidance of dissociative options is a conscious process or whether dissociative influence can occur in the absence of conscious awareness. Finally, the consequences of wishing to avoid (or even attain) past and possible future group memberships would be an interesting avenue for future research. Indeed, research shows that social identity maintenance is often influenced by thoughts about past and possible future group memberships (Cinnirella 1998) and that possible selves (i.e., potential selves that the individual might become, would like to become, or is afraid of becoming) have implications for self-concept, motivation, and behavior (Markus and Nurius 1986). Although past research has confirmed that consumers often choose products and brands that represent who they are (Sirgy 1982), the current research suggests that consumers also choose products in ways that demonstrate who they are not. Future research could profitably explore the avoidance of products in ways that confirm who consumers do not wish to become. APPENDIX PROMPTS FOR GROUPS AND RELATED BRANDS (STUDY 1) The in-group prompt read: In the box below, we would like you to type the name of a group on campus that you belong to and feel a part of. You should feel you are this type of person and that you fit in with these people. This group should be a tightly knit group, consisting of individuals who are very similar to one another. In the box below, we would like you to type in a brand that is consistent with the group that you belong to. This can be a brand that members of the group actually use, or it can be a brand that shares the same image as the group. A brand is considered a name or symbol that distinguishes one seller’s goods from another’s. The dissociative prompt read: In the box below, we would like you to type the name of a group on campus that you wish to avoid being associated with. You should feel you are not this type of person and that you wish to avoid being associated with these people. This group should be a tightly knit group, consisting of individuals who are very similar to one another. In the box below, we would like you to type in a brand that is consistent with the group that you wish to avoid being associated with. This can be a brand that members of the group actually use or it can be a brand that shares the same image as the group. A brand is considered a name or symbol that distinguishes one seller’s goods from another’s. The out-group prompt read: In the box below, we would like you to type the name of a group on campus that you do not belong to and do not feel a part of. You should feel you are not this type of person CONSUMER IDENTITY AND DISSOCIATIVE INFLUENCE 535 and that you do not fit in with these people. This group should be a tightly knit group, consisting of individuals who are very similar to one another. In the box below, we would like you to type in a brand that is consistent with the group that you do not belong to. This can be a brand that members of the group actually use or it can be a brand that shares the same image as the group. A brand is considered a name or symbol that distinguishes one seller’s goods from another’s. PRIMING MANIPULATION (STUDY 2) Priming condition (adapted from LeBoeuf et al. 2007): “Please state your country of birth”; “What city do you live in at the moment?”; “Name a Canadian celebrity that you admire”; “Name a Canadian attraction/city that you would like to visit”; “Name a Canadian musical artist/band whose music you listen to”; “Please list five words/brief phrases that you associate with Canada. Don’t think too hard about these; just tell us whatever comes to mind first.” Neutral condition (adapted from Shih, Pittinsky, and Nalini [1999]): “What is your gender?”; “Have you declared your major yet?”; “If yes, what is your major?”; “Do you live on or off campus?”; “Do you subscribe to cable television?”; “If no, would you consider subscribing to cable television?”; “How much would you be willing to pay per month for cable television?”; “Please list five reasons for and/or against subscribing to cable television. Don’t think too hard about these; just tell us whatever comes to mind first.” REFERENCES Aaker, Jennifer L. (1999), “The Malleable Self: The Role of SelfExpression in Persuasion,” Journal of Marketing Research, 36 (1), 45–57. Aiken, Leona S. and Stephen G. West (1991), Multiple Regression: Testing and Interpreting Interactions, Thousand Oaks, CA: Sage. Argo, Jennifer J., Katherine White, and Darren W. Dahl (2006), “Social Comparison Theory and Deception in the Interpersonal Exchange of Consumption Information,” Journal of Consumer Research, 33 (2), 99–108. Ashworth, Laurence, Peter R. Darke, and Mark Schaller (2005), “No One Wants to Look Cheap: Trade-Offs between Social Disincentives and the Economic and Psychological Incentives to Redeem Coupons,” Journal of Consumer Psychology, 15 (4), 295–306. Bargh, John A. (2002), “Losing Consciousness: Automatic Influences on Consumer Judgment, Behavior, and Motivation,” Journal of Consumer Research, 29 (2), 280–85. Bearden, William O. and Michael J. Etzel (1982), “Reference Group Influence on Product and Brand Purchase Decisions,” Journal of Consumer Research, 9 (2), 183–94. Berger, Jonah and Chip Heath (2007), “Where Consumers Diverge from Others: Identity-Signaling and Product Domains,” Journal of Consumer Research, electronically published June 13. Branscombe, Nyla R. and Daniel L. Wann (1994), “Collective SelfEsteem Consequences of Outgroup Derogation When a Valued Social Identity Is on Trial,” European Journal of Social Psychology, 24 (6), 641–57. Brewer, Marilynn B. (1979), “Ingroup Bias in the Minimal Group Situation: A Cognitive-Motivational Analysis,” Psychological Bulletin, 86 (2), 307–24. Cinnirella, Marco (1998), “Exploring Temporal Aspects of Social Identity: The Concept of Possible Social Identities,” European Journal of Social Psychology, 28 (2), 227–48. Crocker, Jennifer and Rene Luhtanen (1990), “Collective SelfEsteem and Ingroup Bias,” Journal of Personality and Social Psychology, 58 (1), 60–67. Deshpande´, Rohit and Douglas M. Stayman (1994), “A Tale of Two Cities: Distinctiveness Theory and Advertising Effectiveness,” Journal of Marketing Research, 3 (1), 57–64. Doosje, Bertjan, Naomi Ellemers, and Russell Spears (1995), “Perceived Intragroup Variability as a Function of Group Status and Identification,” Journal of Experimental Social Psychology, 31 (5), 410–36. Ellemers, Naomi, Russell Spears, and Bertjan Doosje (1997), “Sticking Together or Falling Apart: In-Group Identification as a Psychological Determinant of Group Commitment versus Individual Mobility,” Journal of Personality and Social Psychology, 72 (3), 617–26. Englis, Basil G., and Michael R. Solomon (1995), “To Be and Not to Be: Lifestyle Imagery, Reference Groups, and the Clustering of America,” Journal of Advertising, 24 (1), 13–28. Escalas, Jennifer Edson (2004), “Narrative Processing: Building Consumer Connections to Brands,” Journal of Consumer Psychology, 14 (1–2), 168–80. Escalas, Jennifer Edson and James R. Bettman (2003), “You Are What They Eat: The Influence of Reference Groups on Consumers’ Connections to Brands,” Journal of Consumer Psychology, 13 (3), 339–48. ——— (2005), “Self-Construal, Reference Groups, and Brand Meaning,” Journal of Consumer Research, 32 (3), 378–89. Forehand, Mark R. and Rohit Deshpande´ (2001), “What We See Makes Us Who We Are: Priming Ethnic Self-Awareness and Advertising Response,” Journal of Marketing Research, 38 (3), 336–48. Forehand, Mark R., Rohit Deshpande´, and Americus Reed II (2002), “Identity Salience and the Influence of Differential Activation of the Social Self-Schema on Advertising Responses,” Journal of Applied Psychology, 87 (6), 1086–99. Hogg, Michael A. and John C. Turner (1987), “Intergroup Behaviour, Self-Stereotyping and the Salience of Social Categories,” British Journal of Social Psychology, 26 (4), 325–40. Jackson, Linda A., Linda A. Sullivan, Richard Harnish, and Carole N. Hodge (1996), “Achieving Positive Social Identity: Social Mobility, Social Creativity, and Permeability of Group Boundaries,” Journal of Personality and Social Psychology, 70 (2), 241–54. Lalonde, Richard N. (2002), “Testing the Social Identity Differentiation Hypothesis: ‘We’re Not American Eh!’,” British Journal of Social Psychology, 41 (4), 611–30. LeBoeuf, Robyn A., Eldar Shafir, and Julia Belyavsky (2007), “The Conflicting Choices of Alternating Selves,” working paper, Warrington College of Business, University of Florida. Levine, Mark, Amy Prosser, David Evans, and Stephen Reicher (2005), “Identity and Emergency Intervention: How Social 536 JOURNAL OF CONSUMER RESEARCH Group Membership and Inclusiveness of Group Boundaries Shape Helping Behavior,” Personality and Social Psychology Bulletin, 31 (4), 443–53. Markus, Hazel and Ziva Kunda (1986), “Stability and Malleability of the Self-Concept,” Journal of Personality and Social Psychology, 51 (4), 858–66. Markus, Hazel and Paula Nurius (1986), “Possible Selves,” American Psychologist, 41 (9), 954–69. Marques, Jose, Dominic Abrams, and Dario Paez (1998), “The Role of Categorization and In-Group Norms in Judgments of Groups and Their Members,” Journal of Personality and Social Psychology, 75 (4), 976–88. McGarty, Craig, S. Alexander Haslam, Karen J. Hutchinson, and John C. Turner (1994), “The Effects of Salient Group Memberships on Persuasion,” Small Group Research, 25 (2), 267–93. Moschis, George P. (1976), “Social Comparison and Informal Group Influence,” Journal of Marketing Research, 13 (August), 237–44. Sen, Sankar, Zeynep Gu¨rhan-Canli, and Vicki Morwitz (2001), “Withholding Consumption: A Social Dilemma Perspective on Consumer Boycotts,” Journal of Consumer Research, 28 (3), 399–417. Sengupta, Jaideep, Darren W. Dahl, and Gerald G. Gorn (2002), “Misrepresentation in the Consumer Context,” Journal of Consumer Psychology, 12 (2), 69–79. Shih, Margaret, Todd L. Pittinsky, and Nalini Ambaday (1999), “Stereotype Susceptibility: Identity Salience and Shifts in Quantitative Performance,” Psyhological Science, 10 (1), 80–83. Sirgy, M. Joseph (1982), “Self-Concept in Consumer Behavior: A Critical Review,” Journal of Consumer Research, 9 (3), 287–300. Smith, Allan (1994), Canada—an American Nation? Essays on Continentalism, Identity, and the Canadian Frame of Mind, Montreal: McGill-Queen’s University Press. Tajfel, Henri and John C. Turner (1979), An Integrative Theory of Intergroup Conflict, Monterey, CA: Brooks/Cole. ——— (1986), The Social Identity Theory of Intergroup Behavior, Chicago: Nelson. Tepper, Kelly (1994), “The Role of Labeling Processes in Elderly Consumers’ Responses to Age Segmentation Cues,” Journal of Consumer Research, 20 (4), 503–19. Turner, John C. (1985), Social Categorization and the Self-Concept: A Social Cognitive Theory of Group Behavior, Greenwich, CT: JAI. Wallace, David S., Rene M. Paulson, Charles G. Lord, and Charles F. Bond Jr. (2005), “Which Behaviors Do Attitudes Predict? Meta-Analyzing the Effects of Social Pressure and Perceived Difficulty,” Review of General Psychology, 9 (3), 214–27. West, Stephen G., Leona S. Aiken, and Jennifer L. Krull (1996), “Experimental Personality Designs: Analyzing Categorical by Continuous Variable Interactions,” Journal of Personality, 64 (1), 1–48. White, Katherine and Jennifer J. Argo (2007), “Social Identity Threat and Consumer Preferences,” working paper, Haskayne School of Business, University of Calgary. White, Katherine and Darren W. Dahl (2006), “To Be or Not Be: The Influence of Dissociative Reference Groups on Consumer Preferences,” Journal of Consumer Psychology, 16 (4), 404–13. Whittler, Tommy E. and Joan Scattone Spira (2002), “Model’s Race: A Peripheral Cue in Advertising Messages,” Journal of Consumer Psychology, 12 (4), 291–301.

week4-8.(2007) white & dahl.pdf

0.29MB

https://smartstore.naver.com/smallme/products/2875259458

페인팅 스몰미 15세트 (painting smallME, 컬러링 스몰미) 페인팅스몰미, 미술교육. 창의력교육. 색칠