1 Introduction

Prior studies of brand choice behaviors, especially those that impact on a person’s social identity, have largely focused on the approach tendencies of consumers (e.g., Chan et al. 2012; Escalas and Bettman 2003, 2005; Stokburger-Sauer et al. 2012). There is much less research on avoidance tendencies in identity-relevant brand choice. From a means-ends perspective, approach involves efforts to decrease the gap between current states and desired end states, whereas avoidance involves striving to increase the gap between current states and undesirable end states (Higgins 1998; Kardes and Cronley 2000). Thus, in a brand choice context, one might eagerly embrace a brand because one’s reference group is doing so and one wants to fit in (approach)—or vigilantly shun another brand because of risk of criticism by peers and fear of social isolation (avoidance).

Notably, in the rare instances where scholars previously have studied product and brand avoidance from a social identity perspective, they have focused on avoidance prompted by consumers’ perceptions that a target product/brand has specific negative associations (see Wooten and Reed 2004 for an exception). For example, White and Dahl (2006) found that males (but not females) are less likely to choose a steak labeled as a “ladies’ cut” rather than a “house cut,” and White and Dahl (2007) have shown that people avoid brands linked to undesirable or “dissociative” reference groups such as “jocks” or “rich girls” (see Escalas and Bettman 2005 for similar findings).

In contrast, we propose that brand avoidance behaviors can be engendered simply by social-evaluative uncertainty, that is, perceived uncertainty regarding the social acceptability of one’s consumption choices (Rudolph and Conley 2005). For our theorizing, we invoke an individual-difference construct, attention to social comparison information (ATSCI; Lennox and Wolfe 1984). ATSCI refers to a person’s degree of sensitivity to social comparison cues. The need for comparison with others is thought to be innate, since it enables a person to engage in self-evaluation as well as self-enhancement (Gibbons and Buunk 1999). It also helps a person deal with presentation of self in social settings, e.g., based on the behavioral cues of others (Calder and Burnkrant 1977). Nonetheless, the need for social comparison and interest in others’ behaviors vary across individuals, leading to the construct of ATSCI (Lennox and Wolfe 1984).

Snyder (1974) has shown that the need for social comparison information is related to self-monitoring tendencies. However, Lennox and Wolfe (1984) have highlighted major discrepancies between the conceptualization of self-monitoring posited by Snyder (1974) and his scale for measuring the construct. They have also shown that ATSCI can be distinguished from self-monitoring because of the former construct’s strong positive relationships with social anxiety, neuroticism, and fear of negative evaluation.Footnote 1 The importance of ATSCI to consumer behavior has been demonstrated by Bearden and Rose (1990), who have shown that high (vs low) ATSCI individuals believe that others are more likely to judge them by their purchases (study 1), value interpersonal considerations more in buying branded products (study 2), and conform more to peers’ preferences in making product choices (study 3). More recently, Deval et al. (2013, experiment 7) have shown that high ATSCI individuals are more likely to be influenced by social validation appeals.

The chain of logic for linking ATSCI to brand avoidance behaviors, via the construct of social-evaluative uncertainty, is as follows. A plausible explanation for the findings of Bearden and Rose (1990) is that high (vs low) ATSCI consumers have a higher need for assimilation with their peers. Furthermore, prior research suggests that in pursuing social assimilation, high ATSCI individuals will be strongly influenced by anxiety and fear of negative judgments by others (Lennox and Wolfe 1984). Consequently, we propose that high ATSCI individuals will engage in brand consumption behaviors that enable them to maintain a low profile—to blend in, and not stand out, in social settings, so as to enhance assimilation rather than identity differentiation (Berger and Heath 2007; Brewer 1991; Chan et al. 2012).

In accord with this principle, we further argue that high (vs low) ATSCI individuals will be more likely to avoid brand consumption behaviors that are distinctive or conspicuous. People engaging in such consumption behaviors are likely to stand out, attracting the attention and curiosity of others (Berlyne 1960; Mandler 1982). But the reactions of others cannot always be predicted or anticipated, and there is always a chance that those reactions could take the form of criticism rather than compliments. Because of perceived uncertainty regarding others’ reactions to attention-garnering choices and the possibility, however remote, of social disapproval, and because high ATSCI consumers crave social approval and fear social isolation, it is likely that such individuals will avoid brands and brand-related icons that might be distinctive or conspicuous.

In related research, Wooten and Reed (2004) have argued that individuals susceptible to normative influence (SNI) will seek conformity and tend to play safe in self-presentation. Hence, high SNI individuals will avoid calling attention to themselves, especially when there is a possibility of disapproval. Wooten and Reed have shown that SNI is negatively associated with individuation in self presentation and that high SNI participants respond favorably to advertising claims highlighting prevention of problems that could be socially conspicuous (e.g., a toothpaste that prevents bad breath).

The main objective of our research therefore is to contribute to the literature on brand avoidance behaviors by linking such behaviors to the construct of ATSCI. We first briefly report two preliminary studies in support of our theoretical assumptions, with data linking ATSCI to brand consciousness, brand social-evaluative uncertainty, and brand avoidance motivated by social-evaluative concerns. We then report two studies in support of our overarching theme that high (vs low) ATSCI consumers are more likely to avoid brands and brand-related icons that are distinctive or conspicuous. In study 1, we show that although high and low ATSCI individuals identify themselves with equally prestigious brands, the former avoid identifying with highly distinctive brands. In study 2, we propose and confirm that unlike low ATSCI individuals, high ATSCI consumers prefer less conspicuous brand logos even in the case of highly prestigious brands.

2 Preliminary studies: verification of theoretical assumptions

High ATSCI individuals are likely to pay close attention to branded products and brand names (Bearden and Rose 1990), thereby maintaining congruence with their reference group and avoiding social faux pas. This assumption was verified in a preliminary study (n = 218 undergraduate students) which revealed a large positive association (r = .47, p < .001) between ATSCI and brand consciousness (Nan and Heo 2007; see Appendix for measures and reliability statistics for all studies). Clearly, both theory and empirical evidence support the notion that brand choices are important to high ATSCI individuals.

A second preliminary study was conducted with participants (n = 175) recruited from an online service (Amazon Mechanical Turk). All were US residents and native speakers of English. Participants responded to measures that assessed (1) the extent to which they avoided certain brands purely on account of social-evaluative concerns, (2) perceived social-evaluative uncertainty regarding their brand choices, and (3) ATSCI. Here, as in our other studies, batteries of filler items were inserted between the various measures to clear short-term memory of previous responses. The data confirmed large positive correlations between ATSCI and brand avoidance motivated by social-evaluative concerns (r = .65, p < .001) and ATSCI and social-evaluative uncertainty regarding brand choices (r = .56, p < .001).

3 Study 1: ATSCI and brand prestige vs brand distinctiveness

It is widely recognized that on account of their symbolic meanings, brands can help consumers build as well as communicate their self-identities (e.g., Bhattacharjee et al. 2014; Chernev et al. 2011; Escalas and Bettman 2003, 2005; Levy 1959). Selectivity is integral to this process of using brands to further one’s identity goals—people identify themselves with some brands but not many others (Bhattacharya and Sen 2003). When a consumer identifies strongly with a particular brand, it can lead to a deep sense of oneness with the brand, a concept termed as consumer-brand identification (Stokburger-Sauer et al. 2012).

Prior research has established that brand prestige and brand distinctiveness are two of the important drivers of consumer-brand identification (Bhattacharya and Sen 2003; Stokburger-Sauer et al. 2012). Brand prestige refers to the status or esteem associated with a brand, whereas brand distinctiveness refers to the perceived uniqueness of a brand in relation to its competitors (Stokburger-Sauer et al. 2012). The question of interest in study 1 is how these two drivers of consumer-brand identification might vary across the ATSCI spectrum.

Irrespective of their need to assimilate or differentiate, all consumers are more likely to identify with brands of good quality and prestige as a means of self-enhancement. People uniformly have a need to affirm and maintain positive self-views, and prestigious brands can help fulfill that need (Bhattacharya and Sen 2003). Therefore, regardless of where an individual stands on ATSCI, the person is likely to identify with brands that are highly prestigious. However, for brand distinctiveness, based on our previous arguments, we anticipate a different pattern. High ATSCI individuals will be uncertain about the social acceptability of their choices and thus will prefer to keep a low profile, avoiding brands that might draw the attention of others because of their distinctiveness. High (vs low) ATSCI individuals thus are less likely to identify with brands they perceive as highly distinctive.

  1. H1:

    For identity-relevant brands, (a) brand distinctiveness will be lower among high (vs low) ATSCI individuals, and (b) brand prestige will be uniformly high, regardless of ATSCI.

3.1 Method

Participants (n = 117 undergraduate students) were provided several product categories and asked to think of brands with which they strongly identified—where they felt a “sense of oneness with the brand, the brand means something to you, the brand is like a part of you, and you feel quite attached to the brand” (Stokburger-Sauer et al. 2012). If a participant could think of one or more brands that fit this criterion, he/she wrote the brand name(s) down in the questionnaire (up to a maximum of five brands). The participant then wrote down his/her thoughts and feelings in relation to the specific brand; this elicitation procedure was included to make salient the mental representation of the brand. Subsequently, participants responded to scale items (1–7 scales; disagree/agree) for assessing brand identification, brand prestige, brand distinctiveness, and ATSCI (M = 4.51; S.D. = 1.01).Footnote 2

3.2 Results

Since our theorizing and hypothesis deal with brands that are highly relevant to self-identity, we limited the data analysis to brands that were rated at least a 6 on the 1–7 scale for brand identification, leaving us a sample of 103 cases. We first confirmed that ATSCI was not significantly related to brand identification in the sample (p > .10). Next, a general linear model (GLM; mixed model) analysis was conducted with the two brand perception measures, brand distinctiveness and brand prestige, treated as a repeated-measure dependent variable. Gender and mean-centered ATSCI scores were included as independent variables. The results showed a reliable interaction between the repeated-measure brand variable and ATSCI (F (1, 99) = 8.40, p < .01), unqualified by a three-way interaction involving gender (p > .06). Follow-up analyses confirmed a significant negative association between brand distinctiveness and ATSCI (B = −.14, p < .05) and a nonsignificant relationship between brand prestige and ATSCI (B = −.001, p > .9). As predicted in H1, brand distinctiveness (but not brand prestige) of identity-relevant brands was lower for high ATSCI participants when compared to their low ATSCI counterparts (see Fig. 1).

Fig. 1
figure 1

Study 1: brand prestige and brand distinctiveness of identity-relevant brands as a function of ATSCI (measured on 1–7 scales)

Full size image

4 Study 2: ATSCI and brand logo conspicuity

In study 2, we examine brand-related avoidance behaviors in a different context, namely preferences for brand logos in publicly consumed products. Brand logos on branded products vary considerably in the marketplace in terms of size and conspicuity (Berger and Ward 2010; Han et al. 2010). Han et al. (2010) have shown that individuals who aspire to improve their social status prefer conspicuous brand logos, especially in the case of highly prestigious luxury brands. However, an alternative perspective is possible on this issue by examining it through the lens of ATSCI. As discussed earlier, high ATSCI individuals are highly sensitive to social comparison cues and are fearful of negative evaluations by others. Such individuals are anxious to protect their current social status and are not necessarily looking to improve their social status. Viewed in terms of the regulatory focus theory of Higgins (1998), we posit that high ATSCI individuals are prevention-focused in self-regulating their behaviors for achieving social status goals, vigilantly avoiding what could be mistakes or “errors of commission.” In contrast, high need-for-status individuals are promotion-focused, eagerly approaching behaviors that might advance their social status. As such, we expect a different relationship between ATSCI and preference for conspicuous brand logos than the findings of Han et al. (2010) regarding the relationship between need for social status and logo conspicuity.

More specifically, it stands to reason that consumers in general will shun loud logos in the case of less prestigious brands, since such brand consumption behaviors would make the wrong kind of social statement to others. But, what about the case of relatively prestigious brand logos? We propose that even if a brand is highly prestigious, high ATSCI individuals will still want to keep a low profile. Loud displays of brand logos are likely to attract attention and may draw criticism from one’s reference group, especially if the particular brand has fallen out of favor with that group. Therefore, high ATSCI consumers will consistently prefer less conspicuous brand logos, regardless of the brand’s prestige, to avoid standing out among their peers—in essence, a risk-minimizing strategy in the face of social-evaluative uncertainty. In contrast, in the case of low ATSCI individuals, a prestigious brand should make the display of a highly noticeable brand logo much more acceptable. That is, for low (but not high) ATSCI individuals, brand prestige should at least partially alleviate any concerns about logo conspicuity.

  1. H2:

    In the case of highly prestigious brands, high (vs. low) ATSCI individuals will prefer less conspicuous brand logos. However, in the case of less prestigious brands, preferences for conspicuous brand logos will be uniformly low, regardless of ATSCI.

4.1 Method

Design

H2 was tested in a study (n = 391 undergraduate students) in which we manipulated brand prestige (less vs highly prestigious brands, between subjects) and presented participants with pictures of t-shirts with brand logos manipulated to be more vs (less) conspicuous.

Stimuli and manipulation checks

Separate brands were selected for males and females based on a pretest (n = 29 males and 38 females). Based on the pretest data (1–7 scales), Old Navy (less prestigious) and Polo (highly prestigious) were selected for males (M = 3.74 vs 6.38, p < .001). Old Navy (less prestigious) and Pink (highly prestigious) were selected for females (M = 3.94 vs 6. 25, p < .001). Next, t-shirts that depicted brand logos were selected for each brand based on commercially available products. A professional graphic designer then created two different images for each t-shirt, one with a small, relatively inconspicuous brand logo and the other with a very large, highly conspicuous brand logo (see Fig. 2).

Fig. 2
figure 2

Study 2 stimuli

Full size image

A second pretest (n = 40 males and 53 females) was conducted to verify the manipulation of brand logo conspicuousness. Pretest participants were shown both versions of the brand logos for either the less or the highly prestigious brands, with left-right presentation order counterbalanced. Participants then responded to items which required them to judge which of the two brand logos was (a) larger and (b) more conspicuous, with the option to judge both as equal on these criteria. A logistic regression on the data confirmed large main effects for the manipulation of brand logo on perceptions of logo size (Wald (1) = 24.7, p < .0001) and logo conspicuousness (Wald (1) = 19.9, p < .0001); the effects for gender and brand prestige were NS. Participants were much more likely to judge the large (vs small) brand logo as actually larger and more conspicuous (87 % or more of participants for the various stimuli).

Procedure

Male and female participants were assigned to separate online instruments, with the software randomly assigning participants to either the highly or less prestigious brand condition. Participants looked at the pictures of two t-shirts side by side, one with the less conspicuous brand logo and the other one with the more conspicuous brand logo. Then, they responded to a 100-point, constant-sum measure of preference between the two t-shirts. This relative preference measure was later rescaled to −100 to 100 by subtracting the points allocated to the less conspicuous brand logo from those for the more conspicuous brand logo. Participants then responded to a battery of filler questions regarding leisure activities and completed the ATSCI scale (M = 4.46; S.D. = .88).

4.2 Results

A GLM analysis was conducted on the preference data, with independent variables as ATSCI, brand prestige, and gender. The results confirmed a significant two-way interaction between brand prestige and ASTCI, (F (1, 383) = 4.26, p < .05), unqualified by a three-way interaction involving gender (p > .2). The conditional effects of ATSCI on the preference dependent variable were then probed at different levels of the moderating variable, brand prestige, with the regression approach recommended by Hayes (2013; pp. 246–254). For highly prestigious brands, the regression confirmed a significant negative relationship between ATSCI and relative preference for more (vs less) conspicuous brand logos (B = −9.28, p < .05). However, the regression for less prestigious brands showed no relationship between ATSCI and the dependent variable (B = 2.91, p > .5). These results hence support H2 (see Fig. 3).

Fig. 3
figure 3

Study 2: relative preference for more (vs less) conspicuous brand logos as a function of ATSCI and brand prestige (measured on a scale of −100 to +100)

Full size image

5 General discussion

The present research makes an important contribution to the topic of brand avoidance behaviors. As discussed earlier, prior studies of brand choice behaviors that produce social or identity-based benefits to the consumer have mainly focused on approach tendencies, wherein consumers proactively adopt brands to achieve goals such as assimilation (e.g., Escalas and Bettman 2005), individuation (e.g., Chan et al. 2012), demarcation of social status (e.g., Han et al. 2010; Levy 1959), or communal consumption (e.g., Schau et al. 2009). In the rare cases where brand avoidance behaviors have been studied, researchers have examined them in the light of undesirable associations, stereotypes, or stigmas (e.g., White and Dahl 2007).

Our work, in contrast, demonstrates that brand avoidance behaviors can also be engendered simply by social-evaluative uncertainty among high ATSCI consumers. Desiring assimilation with others, these individuals are chronically apprehensive about the reactions of their peers, a proposition verified by data from our preliminary studies. High ATSCI consumers consequently keep a low profile in their brand choices, preferring to blend in and not stand out, minimizing scrutiny and possible criticism from others (see also Wooten and Reed 2004). Thus, unlike their low ATSCI counterparts, high ATSCI consumers avoid identifying themselves with brands that are overly distinctive (study 1) and balk at conspicuous brand logos even in the case of prestigious brands (study 2).

5.1 Implications, limitations, and future research

Prior research has demonstrated a positive relationship between ATSCI and susceptibility to peer pressure in consumers’ choices (Bearden and Rose 1990). However, the present findings show that even in the absence of any explicit social influences or pressures, high ATSCI individuals avoid potentially attention-garnering brand choices such as those involving distinctive brands or conspicuous brand logos. Our findings are consistent with the arguments of Wooten and Reed (2004) about the play safe tendencies of individuals susceptible to normative influence. The findings and underlying theorizing are also congruent with the ideas of Kardes and Cronley (2000) about the relationship between avoidance tendencies in consumer behavior and the desire to minimize uncertainty in one’s beliefs.

A few limitations and possibilities for future research are worth noting. First, although our preliminary studies provide support for the theorized mechanisms, mediating processes were not explicitly tested in the main studies and will have to be verified in future research. Second, future research may be able to examine whether the brand avoidance behaviors of high ATSCI consumers are internalized to the point of being nonconscious, rather than being mindful and strategic (see Chartrand and Bargh 1999). Third, future work could clarify whether another well-established individual-difference variable, need for uniqueness (NFU; Tepper Tian et al. 2001), has an analogous relationship to approach (vs avoidance) of distinctive brands and brand icons. It could be that while high (vs low) ATSCI individuals avoid distinctive brands because they are concerned about social sanctions, high (vs low) NFU individuals seek out (approach) distinctive brands because of a greater need for individuation (Chan et al. 2012). Fourth, it would be worth investigating whether high ATSCI individuals’ avoidance of distinctive brands and brand icons is related to concerns regarding the need to justify one’s choices to others (Nowlis and Simonson 2000). Fifth, our theorizing is based on the assumption that high ATSCI individuals mainly wish to protect their existing social status and not necessarily to improve it. The emphasis on a preventive (or defensive) motivation led us to predict a different relationship between ATSCI and preference for conspicuous brand logos than the hypothesis of Han et al. (2010) about a positive association between need for social status and the desire for logo conspicuity. More research is needed to definitively distinguish ATSCI from need for status from a consumer motivation standpoint, e.g., in terms of chronic differences in regulatory foci (Higgins 1998). Finally, future research may be able to ascertain some of the descriptive correlates of ATSCI (e.g., in terms of age, gender, education, income, geography, ethnicity, etc.), thereby making the variable more useful to marketers from a targeting perspective.

In closing, we believe that a key contribution of the present research is to highlight the need for more research focusing on brand avoidance behaviors, especially when such behaviors are linked to a consumer’s social identity and motivated by the goal of reducing social-evaluative uncertainty. Our findings indicate that in making their brand choices, many consumers may be willing to sacrifice distinctiveness and individuation to reduce the possibility of disapproval by others—in essence, valuing a choice consideration such as “I can’t go wrong with brand X” over a consideration such as “I can make a real statement with brand Y.” More research on this topic may be able to produce important insights at the interface of consumers’ social identity projects and brand choice behaviors.