Week-12

Consumer Reactions to Attractive Service Providers: Approach or Avoid?

LISA C. WAN ROBERT S. WYER JR

Attractive service providers are often assumed to elicit favorable consumer reac- tions and to increase purchase intentions. However, this may not always be true. A pilot study and five field and laboratory experiments show that when a self- presentation concern is made salient, consumers react less positively to highly attractive providers than to less attractive ones. This concern can be influenced by chronic social anxiety or can be aroused by unrelated experiences that consumers have before being exposed to the service interactions. In addition, it can be activated by the type of product being sold, that is, a product that is likely to cause embarrass- ment. Thus the attractiveness of a service provider can have either positive or negative effects on consumers’ reactions to a consumption experience and their con- sequent purchase intentions, depending on the type of product under consideration. These effects occur when the service provider is both of the opposite sex and the same sex. However, self-presentation concerns when an opposite-sex provider is at- tractive are driven by sexual motives, whereas these self-presentation concerns when a same-sex target is attractive are stimulated by social comparison processes.

Keywords: attractive service providers, self-presentation concern, social anxiety,

embarrassing consumption, social comparison processes

INTRODUCTION

Consumers often interact more effectively with physi-cally attractive service providers, resulting in greater consumer satisfaction and stronger purchase intentions

(e.g., Ahearne, Gruen, and Jarvis 1999). Indeed, consumers sometimes evaluate a product more favorably when it has simply been touched by an attractive service provider (Argo, Dahl, and Morales 2008). The assumption that physically attractive salespersons are particularly effective is pervasive.

The present research qualifies this assumption, however. We find that attractive service providers can sometimes lead consumers to have self-presentation concerns (that is, concerns about their ability to make a good impression on others; Leary et al. 1994). Research on impression manage- ment (Leary and Kowalski 1990) suggests that people are generally motivated to make a favorable impression on at- tractive others and that their inability to accomplish can create anxiety (Leary 1986; Schlenker and Leary 1982). This suggests that when a provider’s physical attractive- ness elicits self-presentation concerns, it might have a det- rimental effect on their purchase decisions. We predicted that when consumers’ self-presentation concerns are heightened, they often avoid interacting with physically at- tractive providers and so the providers are relatively inef- fective. As we discuss presently, the processes that underlie this effect are likely to depend on whether the

Lisa C. Wan is assistant professor of School of Hotel and Tourism

Management, the Chinese University of Hong Kong, Shatin, N.T., Hong

Kong ([email protected]). Robert S. Wyer Jr. is visiting professor

of marketing, the Chinese University of Hong Kong, Shatin, N.T., Hong

Kong ([email protected]). The authors gratefully acknowledge the helpful

input of the editor, associate editor, and three reviewers. The authors also

gratefully acknowledge Anthony Cheng, Kendra Tang, and a group of stu-

dent helpers from the Chinese University of Hong Kong for their assis-

tance in data collection. In addition, the authors would like to thank 228IN

Toy Store and KINNETEC Company for their help in data collection for

field experiments. This research was supported in part by Grants GRF

493113, 452813, and 14502514 from the Research Grants Council, Hong

Kong. Supplemental materials are included in the online-only version of

this article.

Darren Dahl served as editor, and Andrea Morales served as associate

editor for this article.

Advance Access publication August 27, 2015

VC The Author 2015. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc.

All rights reserved. For permissions, please e-mail: [email protected] � Vol. 42 � 2015 DOI: 10.1093/jcr/ucv044

578

consumer and provider are the same or the opposite sex. That is, the self-presentation concerns that are stimulated by exposure to an attractive opposite-sex provider are sex- ually driven, whereas the concerns activated by an attrac- tive same-sex target are a result of social comparison processes. Nevertheless, their effects on provider effective- ness are similar.

A provider’s physical attractiveness is obviously not suf- ficient to elicit concerns about the impression one forms; other conditions are necessary as well. Self-presentation concerns can be influenced by chronic individual differ- ences in social anxiety (Arkin, Appelman, and Burger 1980). However, they can also be activated by unrelated experiences that occur prior to a social interaction and by features of the interaction itself. In the latter regard, the consumption of some products (e.g., condoms, weight-loss pills) is likely to cause embarrassment (see Dahl, Manchanda, and Argo 2001). These factors are likely to combine with a salesperson’s physical attractiveness to ac- tivate consumers’ concerns about their self-presentation, leading them to avoid interacting with the salesperson and decreasing their likelihood of making a purchase.

To our knowledge, our research is the first to distinguish between conditions in which a service provider’s physical attractiveness increases the favorableness of consumers’ reactions to a service encounter and the conditions in which it decreases the favorableness of these reactions. We investigated these conditions in six studies. An observa- tional study provided initial evidence that individuals’ so- cial anxiety influences their self-presentation concerns and their reactions to attractive opposite-sex service providers in an actual consumption setting. Four laboratory studies confirmed the implications of this study, showing that con- sumers’ avoidance of an attractive opposite-sex provider depends on their self-presentation concerns and that this is true regardless of whether these concerns are chronic (study 1), induced in an unrelated situation (study 2), or ac- tivated by characteristics of the consumption situation it- self (studies 3 and 4). A fifth experiment demonstrated similar effects when the attractive service provider was of the same sex rather than the opposite sex.

CONCEPTUAL BACKGROUND

The Effects of Physical Attractiveness

The positive impact of physical attractiveness on social interactions is well established (e.g., Berscheid and Walster 1974; Dion, Berscheid, and Walster 1972; Eagly et al. 1991). The “beautiful is good” stereotype is perva- sive. Thus physically attractive people are perceived to have more desirable personalities, more favorable attrib- utes, and more satisfying life experiences. People think that attractive individuals are more persuasive, have more influence on others, and are better liked (Sigall and

Aronson 1969). They also perceive attractive individuals to be more intellectually competent, popular, and sociable (Eagly et al. 1991). They therefore find interacting with at- tractive individuals to be an aesthetically pleasant experi- ence (Dion et al. 1972).

Consumer research appears to confirm implications of these findings. Studies of service marketing have found that attractive service providers are often better liked and produce greater consumer satisfaction and purchase inten- tions (e.g., Ahearne et al. 1999; DeShields, Kara, and Kaynak 1996; Koernig and Page 2002; Reingen and Kernan 1993). An attractive salesperson can sometimes in- crease evaluations of a product by simply touching it (Argo et al. 2008). The impact of a provider’s attractiveness de- creases over time and repeated exposure (Ahearne et al. 1999). Nevertheless, past research perpetuates the view that attractive service providers increase the aesthetic pleasantness of a consumption experience and have a posi- tive effect on first-time purchases.

However, little if any previous research has investigated the conditions in which the physical attractiveness of a ser- vice provider might decrease the favorableness of con- sumers’ reactions to a service encounter. A field study by Dabbs and Stokes (1975) is nonetheless suggestive. They found that both male and female pedestrians walked further away from an attractive woman than from an unattractive one. This was apparently because they were nervous about approaching an attractive person. These findings suggest that individuals often have self-presentation concerns when they anticipate interacting with an attractive service pro- vider and consequently may avoid doing so. Moreover, these avoidance reactions may be evident regardless of whether the provider is of the same or opposite sex. However, the factors that mediate these reactions may dif- fer. That is, self-presentation concerns about interacting with an attractive opposite-sex provider are likely to stem from anxiety about creating a poor impression on a sexu- ally appealing other. In contrast, the self-presentation con- cerns activated by an attractive same-sex provider may elicit social comparison processes that create jealousy and dislike of the person who gives rise to these feelings. The next section elaborates these differences.

Self-Presentation and Opposite-Sex Providers

The impressions that others form of a person have impli- cations for how the person is treated (Goffman 1959). Consequently, people attempt to control the impressions they convey (Leary and Kowalski 1990). Impression man- agement has two components. First, impression motivation concerns the desire to make a favorable impression on others and occurs when people are concerned about how the others will evaluate them (Leary 1996; Leary and Kowalski 1990). The second, impression construction,

WAN AND WYER 579

concerns the strategies that people use to influence others’ impressions of them. We consider each component in turn.

Impression Motivation. Individuals’ motivations to pre- sent themselves in a favorable light when they encounter an attractive person of the opposite sex may be influenced in part by the person’s physical attractiveness. When a per- son is not particularly attractive, individuals may not al- ways think about the impression they make. When the person is attractive, however, self-presentation concerns are more likely to come into play (Schlenker 1980). As noted earlier, physically attractive persons are attributed generally favorable personal qualities (Berscheid and Walster 1974) and are preferred as dating partners (Walster, Aronson, and Abrahams 1966). These persons can therefore spontaneously activate self-presentation con- cerns that are sexually driven.

Evidence of this motivation in opposite-sex interactions is suggested in a study by Snyder, Tanke, and Berscheid (1977). They found that when male participants believed that the woman with whom they were talking in a tele- phone conversation was physically attractive, they commu- nicated in a warmer and more personable way that elicited the qualities they expected the woman to have. In a quite different situation, Dong and Wyer (2014) found that when individuals conversed with an opposite-sex partner that they believed to be of average attractiveness, they were motivated to form an impression of the partner and paid particular attention to things the partner said during the conversation. When they believed their partner to be physi- cally attractive, however, they were more concerned about creating a good impression on the other rather than forming an impression of the other. Thus they were more attentive to things they personally said.

Impression Construction. When individuals want to make a good impression on someone and feel able to do so, they may be motivated to interact with this person. The strategies they use to attain this objective can involve self- descriptions, nonverbal behaviors, and often deception. Thus, for example, female participants eat less snack food in the presence of an attractive male in order to enhance their apparent femininity (Mori, Chaiken, and Pliner 1987). Both male and female participants are more apt to lie about their personal attributes to get a date if the prospect is physically attractive than if (s)he is less so (Tyler and Feldman 2004). Moreover, males display luxury posses- sions to impress attractive women (Griskevicius et al. 2007).

Situations can arise, however, when individuals are mo- tivated to make a favorable impression on someone but be- lieve they are unable to attain this objective. In this case, they could expect that interacting with the person will be aversive and might try to avoid doing so. Evidence of this is suggested in a study by Bernstein et al. (1983). They in- vited male participants to participate in a movie-rating

exercise and offered them a choice of two TV monitors to watch. Two chairs were in front of each monitor about 3 inches apart, and either a highly attractive or relatively un- attractive woman was seated in front of one of the moni- tors. Some participants were told that different movies would be shown on the two monitors, whereas others were told that the same movie would be shown on both. When the woman was highly attractive, they were more likely to sit beside her when different movies were being shown on the monitors than when the same movie was shown. This was apparently because they had an excuse for affiliating in the former condition (i.e., their preference for one movie over the other). If this excuse was unavailable, the men’s fear about being negatively evaluated led them to avoid the attractive female.

Although purchasing a product or service can give con- sumers a clear excuse to affiliate with an attractive sales- person of the opposite sex, they may nevertheless avoid doing so if their self-presentation concerns are activated. As noted earlier, however, another’s physical attractive- ness is unlikely to be a sufficient condition for activating concerns about one’s self-presentation. Both individual differences and situational factors can also come into play. Some individuals have chronic social anxiety and may generally be apprehensive about interacting with an attractive other regardless of other considerations (Wenzel and Emerson 2009). In other cases, a recent experience that has decreased individuals’ feelings of self-efficacy might influence their perception that they can interact effectively with others. Moreover, situational factors that lead individuals to anticipate feeling embar- rassed can increase their apprehension about their ability to create a good impression independently of the chronic anxiety they experience.

The latter possibility is of particular relevance in con- sumer situations. Feelings of embarrassment can result from public actions that observers might consider foolish or inappropriate (Dahl et al. 2001). These actions threaten the positive public image that people try to convey and can impair the effectiveness of their social interactions (Brown 1970; Miller 1987; Schlenker and Leary 1982). Because embarrassment is aversive, people are likely to avoid situa- tions that perpetuate these feelings (Apsler 1975).

In consumer situations, self-presentation concerns are likely to be activated by the anticipation of purchasing a product or service that is typically used privately or calls attention to undesirable personal attributes (e.g., con- doms, breast pads, weight-loss pills). Such purchases are particularly likely to elicit embarrassment when the sales- person is attractive and of the opposite sex. Therefore, people may be especially motivated to avoid interacting with such a salesperson. (For evidence that fear of embar- rassment decreases affiliation, see Buunk and Hoorens 1992; Dong, Huang, and Wyer 2013; Sarnoff and Zimbardo 1961.)

580 JOURNAL OF CONSUMER RESEARCH

Self-Presentation Concerns in Same-Sex Interactions

The conditions described in the previous section are likely to be evident in service situations in which the salesperson or service provider is an attractive person of the opposite sex. However, self-presentation concerns could also induce avoidance of interactions with attractive same-sex providers as well, albeit for different reasons. Attractive members of the same sex are generally regarded as potential threats to a person’s status and self-esteem in social interactions (Agthe, Spörrle, and Maner 2011; Maner et al. 2009). In this case, self-presentation concerns are more likely to be driven by a social comparison pro- cess. For one thing, people are often more attentive to at- tractive members of the same sex because they pose threats to their social status (Maner et al. 2009). Thus they try to impress members of the same sex by buying luxury prod- ucts in intrasex competitions (Hudders et al. 2014; Wang and Griskevicius 2014). However, comparisons with an os- tensibly superior person can lead to jealousy and can de- crease self-esteem (Tesser and Collins 1988; see also Lockwood and Kunda 1997). These reactions are particu- larly likely when the individual to whom one compares oneself is of the same sex. Attractive same-sex individuals stimulate upward comparison and consequently threaten self-esteem (Agthe et al. 2011; Kenrick et al. 1993; Wood 1989). Women often compare their appearance to that of attractive female peers (Jones and Buckingham 2005) and models (Dittmar and Howard 2004; Richins 1991). These comparisons can elicit negative affect and decrease their satisfaction with their own physical attractiveness. Thus al- though people are in a positive mood after being exposed to attractive opposite-sex faces, they are in a negative mood after viewing attractive same-sex faces. In organiza- tional contexts, both men and women react negatively to attractive job candidates of the same sex, being less likely to work with them and giving them poor performance eval- uations (Agthe et al. 2011).

Note that the different effects of physical attractiveness on reactions to same-sex and opposite-sex others may be reflected in liking for these persons (Sigall and Aronson 1969; Sigall and Landy 1973). Although consumers may be motivated to avoid interacting with attractive opposite- sex salespersons when they experience self-presentation concerns, this avoidance does not necessarily decrease their liking for them. (As noted earlier, male participants may approach an attractive woman when they have an ex- cuse for affiliating; Bernstein et al.1983.) However, if at- tractive same-sex providers stimulate upward comparisons, they are likely to feel inferior and experience envy (Agthe, Spörrle, and Maner 2010; Maner et al. 2009). Therefore, in contrast to their reactions to attractive opposite-sex others, consumers are likely to decrease their liking for an attrac- tive same-sex provider.

Summary

The conceptualization we have outlined in the preceding discussion is summarized in Figure 1. We assume that when individuals encounter service providers of average attractiveness, their concerns about making a good impres- sion are unlikely to be activated (Dong and Wyer 2014), and so they are unlikely to think about the quality of their interaction with the person. In this case, individuals’ desire to interact with a provider will increase with the provider’s physical attractiveness, and this may be true regardless of whether the provider is of the same or the opposite sex.

When the provider is physically attractive, however, they may have concerns about their self-presentation. These concerns could be activated by (1) individuals/ chronic social anxiety, (2) recent experiences that have had a negative impact on self-efficacy, and (3) characteristics of the interaction situation itself (e.g., the type of product or service being considered). When the provider is of the opposite sex, consumers are likely to experience evaluation apprehension, and their willingness to interact may de- crease. When the provider is of the same sex, consumers are likely to engage in upward comparison and evaluate themselves negatively in relation to the provider. In this case, they may dislike the provider and be unwilling to in- teract for this reason. Thus activating self-presentation con- cern should generally decrease the willingness to interact with a highly attractive service provider, but the reason may depend on whether the provider is of the same or the opposite sex.

The research to be reported confirms implications of this conceptualization. An observational study found that consumers with chronically high social anxiety react less favorably to attractive than to unattractive opposite-sex service providers in an actual retail environment. A labora- tory replicated these findings when consumption was pub- lic but not when it was private. Study 2 demonstrated that the impact of self-presentation concerns on consumers’ reactions to an attractive opposite-sex service provider can be triggered by unrelated experiences that occur before the interaction takes place. Two other studies found that self- presentation concerns can be elicited by embarrassing con- sumption experiences in an actual shopping environment and can decrease the tendency to interact with attractive opposite-sex providers in such an environment. A final study indicated that consumers also react less favorably to attractive same-sex service providers that they encounter in a potentially embarrassing consumption situation.

PILOT STUDY

Individuals with high social anxiety have stronger self- presentation concerns than those with low anxiety (Arkin et al. 1980). To this extent, consumers with chronically high social anxiety should react less favorably to attractive

WAN AND WYER 581

than to unattractive opposite-sex service providers in an ac- tual retail environment. An observational study confirmed this hypothesis.

Method

Otaku is a group of Hong Kong Chinese who have an obsessive interest in computer technology but are socially inept in interpersonal interactions (Niu, Chiang, and Tsai 2012). Therefore, we expected these individuals to be high

in social anxiety and have high self-presentation concerns. Two additional characteristics of Otaku are their love of collecting toys and figures, and their disposition to shop alone.

Based on these assumptions, we conducted an observa- tional study at a store in a Hong Kong shopping center (Richmond Shopping Arcade) that specializes in Japanese figures, models, and gifts and is a popular place for Otaku to shop. The store that we considered had recently opened and allowed us to conduct the study on two weekdays dur- ing the first month of its operation. Because we were inter- ested in observing how consumers interact individually with salespersons, we conducted the study during periods in which the store traffic was minimal (i.e., 2 to 6 PM). Before conducting the study, we observed that about 70% of the customers who stopped and looked at the window display during this period were male. The size of the store was small (162 square feet), and potential customers could easily see the physical appearance of the salesperson through the store window.

Two female confederates with different levels of attrac- tiveness (age 22 years) were used as salespersons. The highly attractive confederate wore makeup to accentuate her attractiveness; the average-looking confederate did not. A pretest using a between-subject design (n¼42) con- firmed that one confederate (after wearing makeup) was more attractive than the other (8.28 vs. 5.29 along a scale from 1 to 10; F(1, 40)¼56.78; p < .001). Two observers were used to collect data. One observer stood outside the store and counted the number of male consumers who en- tered the store. The other stayed in a small room near the cashier behind a one-way mirror and observed the interac- tion between the salesperson and the consumers. The ob- servers recorded the total number of male consumers who stopped to take a look at the window display, the number who entered the store and took a look at the products, the number who interacted with the salesperson (e.g., asking about product information, buying the product, etc.), and the duration of time that the salesperson and consumer in- teracted. Finally, we determined the amount of consumers’ purchases from the store’s sales receipts. (Because the ma- jority of consumers were men and we were only interested in observing male consumers who interact with the sales- person alone, we excluded analyses for female consumers and male consumers who entered the store with friends.)

Results

The results of this study are consistent with predictions. A similar number of male consumers stopped to look at the window display in both high attractiveness (n¼84) and av- erage attractiveness (n¼80) conditions. However, fewer male consumers entered the store when the salesperson was attractive (n¼48) than when she was average looking (n¼65; v2¼11.11, p < .01). Ninety-eight of the 113 male

FIGURE 1

CONCEPTUAL MODEL OF CONSUMER REACTIONS TO ATTRACTIVE SERVICE PROVIDER

Opposite Sex

Provider Attractive?

Self-Presentation Concern (e.g., chronic anxiety, embarrassment, private vs. public. Etc.)

Sex of Provider?

Evaluation Apprehension Social Comparison

Interact

Yes

Yes +

Purchase Intention

+

‐‐ Same Sex

+

Provider

No

+

No

582 JOURNAL OF CONSUMER RESEARCH

customers who entered the store interacted with the sales- person. The proportion of customers who interacted with the salesperson did not depend on whether the salesperson was attractive (40.8%) or average looking (59.2%; v2 < 1). However, the duration of their interaction was shorter when the salesperson was attractive than when she was not (4.30 min vs. 6.59 min; F(1, 97)¼18.35; p < .001). Finally, fewer males made a purchase when the salesperson was attractive than when she was average looking (38 vs. 56; v2¼10.27, p < .01), and the average cost of the prod- ucts they purchased was less in the former condition than the latter (US$28.85 vs. US$41.49; F(1, 92)¼4.98; p < .05).

STUDY 1

Consumers’ avoidance of attractive salespersons in the pilot study was presumably a result of their chronic social anxiety and therefore of their concern with how they pre- sented themselves to these persons. Study 1 confirmed this effect in a laboratory experiment. As Leary and Kowalski (1990) note, individuals may attempt to manage not only the impressions that others form of them but also their im- pressions of themselves. However, we expected that the self-presentation concerns that are stimulated by an oppo- site-sex other’s physical attractiveness would only occur when consumption requires interacting with the provider and would not be evident in private consumption situation such as online shopping.

We expected that in a public consumption situation, con- sumers with low chronic social anxiety would not be con- cerned about their ability to create a favorable impression on a highly attractive consultant and would be more willing to interact with such a consultant than with a less attractive one. When consumers have high chronic anxiety, however, we expected the opposite to be true. In contrast, this differ- ence should not be evident when interacting with the con- sultant is not in person.

Method

To identify participants who were high or low in social anxiety, 320 male participants completed an unrelated con- sumer survey two weeks prior to the main study. The sur- vey included a seven item social anxiety scale (e.g., “I often feel nervous even in casual get-togethers”) adopted from Leary (1983). Participants with a high (low) social anxiety score that was 1.00 above (below) the median (a¼ .86; median¼5.43; along a scale from 1 (Strongly disagree) to 10 (Strongly agree)) were invited by phone to participate in another 15-minute study with HK$20 (ap- proximately US$3). A total of 176 men agreed to partici- pate in the study and were randomly assigned to conditions of a 2 (consumption conditions: public vs. private)�2 (consultant attractiveness: high vs. average)�2 (social

anxiety: high vs. low) between-subjects design, with 22 participants in each cell.

In the experiment, male participants imagined them- selves to be interacting with a female consultant at a new university health care service center. Pictures were used to manipulate the attractiveness level of the consultant. The consultant in each picture was smiling and wearing a white shirt but differed in attractiveness. A pretest (n¼50) using a between-subject design confirmed that one consultant was more attractive than the other (8.12 vs. 5.20 along a scale from 1 to 10; F(1, 48)¼70.43; p < .001).

In private consumption conditions, participants were told to imagine that they would use an online system to send their health information to the consultant for assess- ment and get a report with advice. In public consumption conditions, however, they were told to imagine that they would interact with the consultant to obtain health care advice.

Participants then reported their intention to make an ap- pointment with the consultant and their willingness to give health care information to the consultant along scales from 1 (Very unlikely) to 10 (Very likely). They also indicated their liking for the consultant along a scale from 1 (Not at all) to 10 (Very much) and reported their agreement that the consultant was attractive along a scale from 1 (Strongly disagree) to 10 (Strongly agree).

Results

Manipulation Check. The attractiveness manipulation was successful; the highly attractive consultant was rated more positively than the average consultant (8.03 vs. 4.92, respectively; F(1, 168)¼132.46, p < .001).

Intentions to Patronize. Participants’ intentions to pa- tronize the health care center are summarized in the top section of Table 1 as a function of consumption conditions, consultant attractiveness, and social anxiety. Analyses of these data indicated that participants were generally more likely to patronize the health center when the consultant was highly attractive than when she was average looking (6.85 vs. 6.19, respectively; F(1, 168)¼9.19, p < .01). However, this difference is qualified by an interaction of attractiveness and social anxiety (F(1, 168)¼4.62, p < .05) and a three-way interaction of these variables and consumption conditions (F(1, 168)¼6.85, p < .05).

The form of the latter interaction is consistent with ex- pectations. As shown in Table 1, participants who imag- ined communicating privately had greater intentions to patronize the health center when the consultant was highly attractive than when she was less so (7.13 vs. 6.21, respec- tively; F(1, 168)¼9.27, p < .01), and this was true regard- less of whether participants’ chronic anxiety was high (7.20 vs. 6.18, respectively) or low (7.05 vs. 6.23, respec- tively). The provider’s attractiveness had a similar effect

WAN AND WYER 583

when low-anxiety participants imagined interacting with her in person (7.55 vs. 6.09 when attractiveness was high vs. low, respectively; F(1, 168)¼27.59, p < .001). If par- ticipants had high anxiety, however, they reported less in- tention to patronize the health center when the consultant was attractive than when she was less so (5.61 vs. 6.25, respectively; F(1, 168)¼5.77, p < .05).

Intentions to Provide Information. Participants’ inten- tions to provide information, shown in the second section of Table 1, were affected similarly. Analyses of these judg- ments indicated that participants were generally more likely to give information in private than in public con- sumption conditions (7.09 vs. 6.48, respectively; F(1, 168)¼9.54, p < .01) and more likely to do so when the consultant was highly attractive than she was moderately so (7.14 vs. 6.43, respectively; F(1, 168)¼12.58, p < .01). Low-anxious participants were generally more likely to give information than high-anxious participants were (7.05 vs. 6.53, respectively; F(1, 168)¼6.93, p < .01).

However, the three-way interaction of consultant attrac- tiveness, anxiety, and consumption conditions was also sig- nificant (F(1, 168)¼3.78, p < .05). Participants reported a greater likelihood of communicating information privately to a highly attractive consultant than to a less attractive one regardless of their chronic anxiety (7.62 vs. 6.57, respec- tively; F(1, 168)¼14.55, p < .001). When they imagined communicating in person, however, this was true only when their anxiety was low (7.64 vs. 6.36, respectively; F(1, 168)¼21.34, p < .001). When participants were high in anxiety, they were less inclined to interact with the at- tractive consultant than with the less attractive one (5.68 vs. 6.23, respectively; F(1, 168)¼4.41, p < .05).

Liking for the Consultant. Although high-anxious par- ticipants reported less willingness to interact with a very at- tractive consultant than with a moderately attractive one in the public consumption, this was not true of their liking for her. Analyses of consultant evaluations, shown in the last

section of Table 1, yielded only a main effect of consultant attractiveness (F(1, 168)¼72.64, p < .001), indicating that participants expected to like the consultant more when she was very attractive (M¼7.20) than when she was moder- ately attractive (M¼5.73). This difference was similar re- gardless of whether participants were high or low in social anxiety.

Discussion

The results of study 1 show that when participants imag- ined themselves interacting privately with the health con- sultant, they generally reported greater intentions to use the service and to provide personal information when the con- sultant was highly attractive. When participants imagined interacting publicly with the consultant, however, this was only true among consumers who were chronically low in social anxiety. When participants were high in anxiety, they were less willing to use the service and provide infor- mation when the consultant was physically attractive. Thus although individuals with high social anxiety like highly attractive individuals as much or more than less attractive ones, their concern over creating a poor impression leads them to avoid interacting with these individuals.

This latter effect, which was replicated in the next three studies to be reported, distinguishes the effects of an oppo- site-sex provider’s attractiveness on liking for the provider from the effects of a same-sex provider’s attractiveness. We return to this difference in study 5.

STUDY 2

Study 1 provided evidence that consumers’ chronic so- cial anxiety generates self-presentation concerns and con- sequently decreases their willingness to interact with a highly attractive service provider. However, these concerns can also be influenced by recent experiences in which a person has been evaluated negatively and his or her

TABLE 1

CONSUMER REACTIONS TO SERVICE PROVIDERS WITH DIFFERENT ATTRACTIVENESS LEVELS:—STUDY 1

Private consumption Public consumption

Moderate attractiveness High attractiveness Moderate attractiveness High attractiveness

Intention to patronize High anxiety 6.18a (1.48) 7.20c (1.38) 6.25a (1.45) 5.61b (1.51) Low anxiety 6.23a (1.69) 7.05c (1.43) 6.09ab (1.32) 7.55c (1.34)

Intention to provide information High anxiety 6.64a (1.26) 7.55c (1.30) 6.23a (1.60) 5.68b (1.25) Low anxiety 6.50a (1.41) 7.68c (1.29) 6.36ab (1.18) 7.64c (1.22)

Liking High anxiety 5.77a (1.02) 7.05b (1.12) 5.86a (1.25) 7.18b (1.01) Low anxiety 5.59a (1.33) 7.14b (1.13) 5.68a (1.29) 7.41b (0.91)

NOTE.—Numbers in parentheses indicate standard deviations. Means with unlike subscripts are significantly different at p < .05.

584 JOURNAL OF CONSUMER RESEARCH

self-efficacy is called into question. Study 2 investigated this possibility.

Method

A total of 144 participants (74 men) were paid HK$20 (approximately US$3) to take part in a 15 minute experi- ment. They were randomly assigned to cells of a 2 (consul- tant attractiveness: high vs. average)�2 (self-presentation priming conditions: high self-presentation concern vs. low concern) between-subjects design. (Consultants were of the opposite sex in all conditions.)

Participants took part in two ostensibly unrelated experi- ments. They were told that the purpose of the first study was to understand people’s reactions to job interview situa- tions. On this pretense, they were asked to imagine a situa- tion in which they needed to have an important job interview. In high self-presentation concern conditions, they imagined that they did not have enough time to pre- pare for the interview, whereas in low concern conditions, they imagined that they had plenty of time to do so. Participants were asked to put themselves into the situation and to form as vivid an image as possible. After reading the scenario, they listed their thoughts and feelings about their anticipated performance in the interview.

After a five minute break, participants then performed an ostensibly unrelated experiment in which they reported their reactions to a new personalized health care service under conditions identical to the public consumption con- ditions of study 1. That is, they viewed a picture of an op- posite-sex consultant beside a health care counter. They were told to imagine that they would interact with the con- sultant to obtain health care advice. Pretests using a be- tween-subject design were conducted to assess consultants’ attractiveness. Twenty-six male participants evaluated the high and moderate attractiveness female pictures as 8.07 and 5.54, respectively, along a scale from 1 to 10; F(1, 24)¼55.37; p < .001). Correspondingly, 26 females rated the two male pictures as 8.08 and 5.15, respectively F(1, 24)¼33.93, p < .001).

Participants were then given a questionnaire to assess their intention to make an appointment with the consultant and their willingness to give their health care information to the consultant along scales from 1 (Very unlikely) to 10 (Very likely). They also indicated their liking for the con- sultant along a scale from 1 (Not at all) to 10 (Very much) and reported their agreement that the consultant was attrac- tive along a scale from 1 (Strongly disagree) to 10 (Strongly agree). They also indicated the expected duration of their interaction with the consultant in minutes.

Results

Manipulation Check. Two independent coders rated each response generated during the priming task. Interrater

agreement was 91%, and disagreements were resolved through discussion. As expected, participants in self-presen- tation concern conditions, relative to no-concern conditions, listed more thoughts that conveyed worry about their perfor- mance and the impression they made (e.g., “I am so worried about my interview,” “I feel nervous about my perfor- mance”) (2.21 vs. 0.49, respectively; F(1, 140)¼128.01, p < .001). In contrast, they listed relatively fewer thoughts that conveyed confidence in their performance (e.g., “I am confident in the interview as I am well prepared”) (0.48 vs. 2.03, respectively; F(1, 140)¼125.69, p < .001).

The attractiveness manipulation was also successful. Male participants evaluated the highly and moderately attractive- ness female consultants as 8.22 and 5.02, respectively (F(1, 70)¼165.02; p < .001). Correspondingly, women rated the highly attractive and moderately attractive male consultant as 8.06 and 5.11, respectively; F(1, 66)¼129.39, p < .001).

Preliminary analyses of each dependent variable as a function of priming, consultant attractiveness, and sex of subject yielded no significant effects involving the latter variable. Therefore, data are pooled over this variable in the analyses to follow.

Intentions to Patronize. Participants’ intentions to pa- tronize the health care center are shown in the first section of Table 2 as a function of priming and consultant attrac- tiveness. Analyses of these data yielded a main effect of priming (F(1, 140)¼7.52, p < .01) and an interaction of priming and attractiveness conditions (F(1, 140)¼13.87, p < .001). As shown in Table 2, participants with low self- presentation concerns were more inclined to patronize the center when the consultant was very attractive than when he or she was moderately so (7.20 vs. 6.36, respectively; F(1, 140)¼19.10, p < .001). When participants’ self- presentation concerns were high, however, they were less

TABLE 2

CONSUMER REACTIONS TO SERVICE PROVIDERS WITH DIFFERENT ATTRACTIVENESS LEVELS:—STUDY 2

Moderate attractiveness

High attractiveness

Intention to patronize High self-presentation concern 6.55a (1.13) 5.92b (1.04) Low self-presentation concern 6.36a (1.02) 7.20c (1.05)

Intention to provide information High self-presentation concern 7.12a (1.26) 6.13b (1.42) Low self-presentation concern 6.94a (1.35) 7.56c (1.33)

Interaction time (in minutes) High self-presentation concern 23.78a (11.49) 18.54b (11.54) Low self-presentation concern 23.67a (11.50) 29.98c (13.34)

Liking High self-presentation concern 6.47a (1.44) 7.10b (1.13) Low self-presentation concern 6.72a (1.21) 7.34b (1.75)

NOTE.—Numbers in parentheses indicate standard deviations. Means

with unlike subscripts are significantly different at p < .05.

WAN AND WYER 585

inclined to do so in the former case (5.92 vs. 6.55, respec- tively; F(1, 140)¼10.81, p < .001).

Intentions to Provide Information. The effects of self- presentation concerns and consultant attractiveness on inten- tions to provide information to the consultant were very similar to their effects on intentions to patronize the health center. The main effect of priming and its interaction with attractiveness conditions were significant (F(1, 140)¼8.15, p < .01 and F(1, 140)¼8.39, p < .01, respectively). As shown in the second section of Table 2, participants were more willing to give information to a highly attractive con- sultant than to a moderately attractive consultant when their self-presentation concerns were low (7.56 vs. 6.94, F(1, 140)¼8.49, p < .01), but the opposite was true when their self-presentation concerns were high (6.13 vs. 7.12, re- spectively; F(1, 140)¼19.37, p < .001).

Interaction Duration. The time that participants imag- ined interacting with the consultant was affected in much the same way. The effect of priming and its interaction with attractiveness were again significant (F(1, 140)¼8.15, p < .01 and F(1, 140)¼8.39, p < .01, respectively). As shown in the third section of Table 2, participants imagined spending more time interacting with the highly attractive consultant than with the moderately attractive one when their self-presentation concerns were low (29.98 minutes vs. 23.67 minutes, respectively; F(1, 140)¼10.06, p < .01) but were less willing to do so when their self-presentation con- cerns were high (18.54 minutes vs. 23.78 minutes, respec- tively; F(1, 140)¼6.90, p < .01).

Liking for the Consultant. Although participants with high self-presentation concerns reported less willingness to interact with a very attractive consultant than with a mod- erately attractive one, this was not true of their liking for him or her. Analyses of consultant evaluations, shown in the last section of Table 2, yielded only a main effect of consultant attractiveness (F(1, 140)¼7.27, p < .01), indi- cating that participants expected to like her more when she was very attractive (M¼7.22) than when she was moder- ately attractive (M¼6.60), and this was true regardless of whether participants had concern about their self-presenta- tion or not.

Discussion

The results of study 2 provided further evidence that when participants’ self-presentation concerns are relatively low, they have a greater intention to interact with a highly attractive service provider than with a moderately attrac- tive one. When they are concerned about the impression they create, however, the reverse is true. This difference is evident even when self-presentation concerns are activated in a situation that is ostensibly quite unrelated to the con- sumption situation in which the interaction is contem- plated. Moreover, it occurs despite the fact that persons

believe they would like the attractive consultant more. Thus although individuals who are concerned with their self-presentation have more favorable reactions to highly attractive individuals than to less attractive ones, their con- cern over creating a bad impression overrides these reac- tions and leads them to avoid interacting with the individuals.

STUDY 3

Study 2 demonstrated that self-presentation concerns can be situationally primed, carrying over from one context to another. However, these concerns can also be induced by features of the consumption situation itself. As noted earlier, the consumption of some products or services (e.g., condoms, hemorrhoid creams, medical checkups, and weight-loss services) can generate embarrassment (Dahl et al. 2001), an emotional state that threatens an individ- ual’s desired social identity in social interactions. Embarrassing consumption is likely to endanger the posi- tive self-image one is motivated to convey in social situa- tions, and this may be particularly true when others in the situation are attractive and thus are people one wants to im- press. If this is so, consumers may wish to avoid interacting with attractive providers in an embarrassing consumption situation. The next two studies indicate that this is in fact the case.

Method

Study 3 was conducted in a health care products retail setting. Consumers’ reactions to attractive (vs. unattrac- tive) providers were assessed in embarrassing and nonem- barrassing consumption situations and employed a 2 (attractive: high vs. average)�2 (consumption conditions: embarrassing vs. not embarrassing) between-subjects design.

We contacted a company that regularly organizes road shows to sell health care products at major hospitals in Hong Kong. The company has two different product lines concerning body care and foot care and was launching a new product in each line: a thermal waist belt (a product to facilitate weight loss) and a foot insole (for use in re- lieving foot pressure and pain). A pretest with 40 partici- pants showed that consumers associated a higher level of embarrassment with purchasing the thermal waist belt than with purchasing the foot insole (6.06 vs. 2.67, re- spectively, along a scale from 1 to 10; F(1, 38)¼173.33; n¼40).

According to the company’s experience, about 70% of road show consumers in hospitals (patients’ family and friends, administrative staff, nurses, and doctors) were women from 25 to 50 years of age. The company arranged to have two part-time male salespersons (age 31 and 33 years, respectively) with different levels of physical

586 JOURNAL OF CONSUMER RESEARCH

attractiveness. A pretest of their attractiveness with 40 par- ticipants using a between-subject design indicated that one of them was more attractive than the other (7.90 vs. 5.25, re- spectively, along a scale from 1 to 10; F(1, 38)¼71.07). We therefore expected that consumers would be more likely to interact with the attractive salesperson when he was selling foot insoles but would be less likely to interact with this per- son when he was selling thermal waist belts.

The road show was conducted at a major hospital in Hong Kong between 10 AM and 6 PM (excluding lunch time). The company set up two booths at the show about 5 meters apart, one of which displayed each type of prod- uct. One salesperson manned each booth, switching their position every hour.

To measure consumer purchasing decisions, two ob- servers counted the total number of people who (1) stopped and looked at the products and (2) interacted with the sales- person (asking about the product, etc.). The observers also recorded the length of the interactions between the sales- person and consumers. In addition, we determined the number of purchases that consumers made from the com- pany’s receipts. (The price of each thermal waist belt was around US$84, and the price for each pair of foot insoles was around US$80.)

Results

A total of 239 people walked by the booths; 80.3% of the consumers were women between the ages of 30 and 50. The thermal waist belt was designed for women, and a few male consumers purchased the thermal waist belts as a gift. However, we restricted our consideration to female con- sumers who were shopping alone.

Number of People Who Stopped. A total of 153 female consumers stopped to look at the products. Of these, 78 stopped to look at the nonembarrassing product, and 75 stopped to look at the embarrassing product. However, al- though more consumers stopped at the nonembarrassing products booth when the salesperson was attractive (n¼44) than when he was average looking (n¼34), fewer consumers stopped at the embarrassing product in the for- mer condition than in the latter (29 vs. 46). The difference in these proportions was significant (v2¼4.83, p < .05).

Interaction Duration. Of the 153 women who stopped at the booths, 92 interacted with the salesperson and made a purchase and 17 interacted but did not make a purchase. Of the 44 consumers who did not interact with the salesper- son, only 7 made a purchase.

When the product was not embarrassing, more con- sumers interacted with the attractive salesperson (n¼36) than with the average-looking one (n¼20) and spent rela- tively more time doing so (4.44 minutes vs. 3.75 minutes, respectively; F(1, 105)¼7.24; p < .01). When the product was embarrassing, however, consumers interacted less

often with the attractive salesperson (n¼20) than with the average-looking one (n¼33) and, if they interacted, spent relatively less time doing so (2.95 minutes vs. 4.55 minutes, respectively; F(1, 105)¼36.71; p < .001).

Purchases. As predicted, more consumers made a pur- chase of the nonembarrassing product when the salesperson was attractive (n¼34) than when he was not (n¼19), whereas fewer consumers made a purchase of the embar- rassing product when the salesperson was attractive (n¼16) than when he was not (n¼30). This difference was signifi- cant (v2¼8.50, p < .01). Analyses of the amount of money spent on purchases yielded a marginally significant interac- tion of attractiveness and embarrassment conditions (F(1, 95)¼3.79, p¼ .05). The price of the embarrassing and nonembarrassing products was similar (US$84 vs. US$80, respectively). However, consumers spent more money on purchases of the nonembarrassing product when the sales- person was attractive than when he was not (US$117.65 vs. US$101.50, respectively; F(1, 95)¼4.85; p < .05) but spent less money on purchases of the embarrassing product in the former case than in the latter (US$89.25 vs. US$103.60, re- spectively; F(1, 95)¼3.62; p¼ .06).

Discussion

Study 3 provided evidence that consumers are actually more likely to interact with a physically attractive sales- person than with an unattractive one when they are buying nonembarrassing products but are less likely to do so if they are buying embarrassing products. Although the re- sults of this field study are provocative, however, we were unable to examine the factors that mediated the impact of embarrassing consumption on consumer reactions. Furthermore, the study used different product lines (i.e., body care and foot care) to exemplify embarrassing and nonembarrassing consumption. Finally, mood was not as- sessed in either this or other studies we have reported. It is possible that seeing an attractive (vs. unattractive) pro- vider induces a positive mood and results in more favor- able consumer reactions when self-presentation concern is not activated but leads to a more negative mood and to stronger negative reactions when self-presentation con- cern exists. Study 4, a laboratory study, eliminated these ambiguities.

STUDY 4

Method

A total of 132 female participants received HK$50 (around US$6.41) for participating. They were assigned to cells of a 2 (salesperson attractiveness: high vs. aver- age)�2 (consumption conditions: embarrassing vs. not embarrassing) between-subjects design.

WAN AND WYER 587

Participants were told that a company wanted to receive feedback about a new product (thermal waist belt, the body care product used in study 3) and their sales representa- tives. Participants were run individually in a room that sim- ulated a retail setting, showing the product display, leaflets, and advertising posters (see appendix). The leaf- lets and posters contained pictures and text messages that manipulated the embarrassment level of the product. In em- barrassing consumption conditions, the materials indicated that the primary function of the waist belt was to reduce weight and contained a picture of the front of a model us- ing the waist belt (without showing her face). In the nonembarrassing consumption condition, the materials in- dicated that the waist belt was intended to relax muscles, improve circulation, and alleviate lower back pain. A pre- test confirmed that consumers’ a priori preference for these functions did not differ (5.92 vs. 5.68; respectively; F < 1).

A physically attractive male confederate (age 26 years) served as the salesperson. In the high attractiveness condition, he styled his hair and wore a T-shirt that fitted him well and flattered his shape. In the average attrac- tiveness condition, the same confederate was ungroomed and wore an oversize T-shirt and a pair of glasses to detract from his appearance. A pretest (n¼38) using a between-subject design confirmed the difference in the confederate’s attractiveness in the two conditions (8.05 vs. 5.94, respectively, along a scale from 1 (Not at- tractive at all) to 10 (Very attractive); F(1, 36)¼62.81, p < .001).

Participants in each condition had a chance to touch the product, to read the product information, and to interact with the confederate (e.g., asking for more product infor- mation, asking for help on a free trial, etc.). After doing so, they completed a short questionnaire. First, partici- pants reported their agreement with two statements, “I would buy this product,” and “I would be interested in using the product” along a scale from 1 (Very unlikely) to 10 (Very likely). Responses to these items were correlated .69 and were averaged to provide an index of purchase in- tentions. They then reported their liking for the salesper- son (“I like the salesperson”) and the extent to which they felt nervous when interacting with the salesperson (“I felt nervous during the interaction”), along scales from 1 (Not at all) to 10 (Very much). The latter item was used to measure their self-presentation concern (Schlenker and Leary 1982).

As manipulation checks, participants then reported their agreement that “The salesperson is physically attractive” and “The product is embarrassing” along scales from 1 (Strongly disagree) to 10 (Strongly agree). Finally, they re- ported their mood (a¼ .86) along three scales from 1 (Sad/ bad/depressed) to 10 (Happy/good/cheerful).

In addition, an observer watched participants’ interac- tion with the confederate through a one-way mirror and re- corded the duration of their interaction.

Results and Discussion

Our manipulations of both physical attractiveness and embarrassing consumption were successful. Participants judged the salesperson to be more attractive in the high attractiveness condition than in the average attractiveness condition (8.17 vs. 5.53, respectively; F(1, 128)¼123.67, p < .001). Furthermore, participants felt the product was more embarrassing in embarrassing consumption conditions than in nonembarrassing consumption condi- tions (6.11 vs. 4.38, respectively; F(1, 128)¼67.16, p < .001).

We expected that participants in nonembarrassing con- sumption conditions would have higher purchase intentions and would interact a longer time with the attractive sales- person than with the average-looking one. However, we expected participants in embarrassing consumption condi- tions to have a lower purchase intention and to interact for less time when the salesperson was attractive than when he was not.

Intention to Purchase. Participants’ intentions to pur- chase the product are shown in Table 3 as a function of at- tractiveness level and embarrassing consumption. The interaction of these variables was significant (F(1, 128)¼10.25, p < .01) and of the form expected. That is, participants in nonembarrassing consumption conditions reported a greater intention to purchase the product when the salesperson was attractive than when he was not (6.50 vs. 5.61, respectively; F(1, 128)¼11.92; p < .001). However, participants in embarrassing consumption condi- tions reported less intention to purchase when the salesper- son was attractive than when he was not (5.29 vs. 6.08, respectively; F(1, 128)¼9.48; p < .01).

TABLE 3

CONSUMER REACTIONS TO SERVICE PROVIDERS WITH DIFFERENT ATTRACTIVENESS LEVELS:—STUDY 4

Moderate attractiveness

High attractiveness

Intention to patronize Nonembarrassing 5.61ac (1.27) 6.50b (1.80) Embarrassing 6.08ab (1.07) 5.29c (1.77)

Interaction time, minutes Nonembarrassing 4.06a (1.32) 5.09b (1.28) Embarrassing 3.52a (1.42) 2.76c (1.30)

Liking Nonembarrassing 4.88a (1.75) 6.70b (1.78) Embarrassing 5.00a (1.54) 6.85b (1.77)

Self-presentation concern Nonembarrassing 3.67a (1.22) 3.48a (1.44) Embarrassing 3.97a (1.47) 5.30b (1.59)

Mood Nonembarrassing 5.59a (1.22) 6.65b (1.44) Embarrassing 5.56a (1.47) 6.16c (1.59)

NOTE.—Numbers in parentheses indicate standard deviations. Means

with unlike subscripts are significantly different at p < .05.

588 JOURNAL OF CONSUMER RESEARCH

Interaction Duration. The duration of participants’ in- teraction with the salesperson is summarized in the second section of Table 3. Participants were generally more likely to interact with the salesperson in nonembarrassing con- sumption conditions than in embarrassing conditions (4.58 vs. 3.14 minutes, respectively; F(1, 128)¼38.57; p < .001). However, the interaction of salesperson attrac- tiveness and embarrassment conditions was significant (F(1, 128)¼14.88, p < .001) and indicated that partici- pants in nonembarrassing consumption conditions inter- acted longer with the attractive salesperson than with the average-looking one (5.09 min vs. 4.06 min, respectively; F(1, 128)¼20.31; p < .001). In contrast, participants in embarrassing consumption conditions interacted for a shorter time with the attractive salesperson than with the average-looking one (2.76 minutes vs. 3.52 minutes, re- spectively; F (1, 128)¼11. 31; p < .01).

Liking for the Salesperson. Participants reported greater liking for the highly attractive salesperson than for the mod- erately attractive one (6.78 vs. 4.94; F(1, 128)¼37.69, p < .001) and, as in other experiments, this was true regard- less of whether participants were embarrassed (6.85 vs. 5.00) or not (6.70 vs. 4.88). In other words, embarrassing consumption did not have a negative impact on participants’ liking of the attractive consultant; it only affected their will- ingness to interact with him.

Self-Presentation Concern. The self-presentation con- cern that participants reported are shown in the third sec- tion of Table 3. Participants reported greater concern about their self-presentation in embarrassing consumption conditions than in nonembarrassing consumption condi- tions (4.64 vs. 3.58, respectively; F(1, 128)¼18.03, p < .001). However, the interaction of salesperson attrac- tiveness and embarrassment conditions was also significant (F(1, 128)¼9.20, p < .01) and indicated that participants in embarrassing consumption conditions had greater con- cern with the impression they created when the salesperson was attractive than when he was average looking (5.30 vs. 3.97, respectively; F(1, 128)¼16.26, p < .001). In nonem- barrassing consumption conditions, however, this was not true (3.48 vs. 3.67, respectively; F < 1).

Mediation. We assumed that the interactive effects of attractiveness and embarrassing consumption on purchase intentions would be mediated by their effects on self-pre- sentation concerns. Bootstrapping analyses confirmed this assumption; based on 5000 samples, the indirect path of self-presentation concerns was significant; that is, a model of attractiveness�embarrassing consumption ! self- presentation concern ! purchase intention (model 8, Hayes, 2013) was supported; 95% confidence interval (CI), from�1.899 to� .3687).

Participants had a more positive mood when the sales- person was highly attractive than when he was average

looking (6.41 vs. 5.58, respectively; F(1, 128)¼14.55, p < .001). However, additional analyses using mood as a mediator in each embarrassing and nonembarrassing con- sumption condition showed mood to have no effect (p > .10).

In summary, study 4 confirmed the implications of study 3. That is, consumers react more favorably to an attractive salesperson when they are buying nonembarrassing prod- ucts, but they react more negatively to an attractive salesper- son when they are buying embarrassing products. Moreover, these reactions influence purchase intentions. Finally, our results confirm the assumption that the effect of embarrass- ing consumption on purchase intentions was mediated by self-presentation concerns and the impact of these concerns on interaction behavior. Although feelings of anxiety are likely to reflect self-presentation concerns (Schlenker and Leary 1982), two additional items adapted from Leary et al. (1994) were used to measure this construct in study 5.

STUDY 5

The preceding study demonstrated that when a consump- tion situation is likely to be embarrassing, attractive oppo- site-sex providers can lead consumers to have self- presentation concerns and that when this occurs, it has a detrimental effect on purchase decisions. In contrast, study 5 examined consumers’ reactions to attractive providers of the same sex. We expected consumers to react more favor- ably to an attractive (vs. unattractive) same-sex provider when consumption was not embarrassing and did not acti- vate self-presentation concerns. However, we expected that attractive same-sex providers would lead consumers to have self-presentation concerns when the product being considered was potentially embarrassing and would resist interacting with these providers. In contrast to interactions with attractive opposite-sex providers, however, we ex- pected self-presentation concerns to be driven by a social comparison motive. Thus in contrast to their effect on indi- viduals’ reactions to attractive opposite-sex providers, these concerns should decrease consumers’ liking for at- tractive same-sex providers. We expected that when they anticipated interacting with a same-sex salesperson, partic- ipants in an embarrassing consumption situation would be less willing to interact with a highly attractive same-sex salesperson than with an average-looking one, and that this disposition would be accompanied by a higher social com- parison tendency, more negative emotions, and a decreased perception of personal attractiveness. In nonembarrassing consumption conditions, however, these differences were not expected.

Method

A total of 112 female undergraduate students received HK$30 (around US$3.85) for participating. They were

WAN AND WYER 589

assigned to cells of a 2 (salesperson attractiveness: high vs. average)�2 (consumption conditions: embarrassing vs. not embarrassing) between-subjects design.

Participants were told to imagine that a company wanted to receive feedback from consumers about having a ther- mal wear product and their sales representatives. On this pretense, their attention was directed to a computer screen showing a picture of a same-sex salesperson and some pic- tures of the thermal wear with product descriptions. They were asked to imagine themselves interacting with the salesperson.

A physically attractive female confederate (age 28 years) served as the salesperson in the picture. In the high attractiveness condition, she styled her hair, wore makeup, and wore a T-shirt that fitted her well and flattered her shape and looked thin. In the average attractiveness condi- tion, the same confederate did not wear makeup and wore an oversize T-shirt and a pair of glasses to detract from her appearance. A pretest using a between-subject design (n¼56) confirmed the confederate’s difference in attrac- tiveness in the two conditions (8.10 vs. 4.96, respectively, along a scale from 1 to 10; F(1, 54)¼249.77, p < .001).

Written messages were used to manipulate the embar- rassment level of the product. In embarrassing consump- tion conditions, the materials indicated that the primary function of the thermal wear was to reduce weight and keep the body slim during the winter. In the nonembarrass- ing consumption condition, the materials indicated that the thermal wear was intended to keep the body warm during the winter. A pretest confirmed that consumers’ a priori preference for these functions did not differ (6.96 vs. 6.73; respectively; F < 1).

After reading the information, participants were told that in understanding people’s responses to situations, it was of- ten helpful to know their perceptions of themselves. On this pretense, they were asked to rate themselves with re- spect to six attributes (e.g., intelligent, honest, dependable), one of which was “physically attractive.” These ratings were reported along scales from 1 (Not at all) to 10 (Very). They then responded to measures of purchase intention and liking that were identical to those employed in study 4 ex- cept that an additional two items adapted from Leary et al. (1994) were used to measure self-presentation concerns. Responses to the three items (When you thought about in- teracting with the salesperson, to what extent did you “Think about how the salesperson would evaluate you?” “Imagine how the salesperson would perceive you?” and “Feel nervous?”) were reported along a scale from 1 (Strongly disagree) to 10 (Strongly agree) and were aver- aged (a¼ .83). They also estimated the likely duration of their interaction with the salesperson in minutes.

Participants’ social comparison motivation was inferred from responses to three items adapted from Richins (1991) (e.g., “To what extent did you think about how well or how badly you look compared to the salesperson?”) along a

scale from 1 (Definitely no) to 10 (Definitely yes) and were averaged (a¼ .86). Finally, participants reported their general mood along three scales from 1 (Sad/bad/ depressed) to 10 (Happy/good/cheerful) (a¼ .78) and indi- cated their feelings of jealousy along a scale from 1 (Not at all) to 10 (Extremely).

Results and Discussion

Both physical attractiveness and embarrassing consump- tion were successfully manipulated. Participants judged the salesperson to be more attractive in the high attractiveness condition than in the average attractiveness condition (8.04 vs. 4.91, respectively; F(1, 108)¼80.26, p < .001). Furthermore, participants felt the product was more embarrassing in embarrassing consumption than in nonem- barrassing consumption conditions (6.27 vs. 3.16, respec- tively; F(1, 108)¼80.66, p < .001).

We expected that female participants in nonembarrassing consumption conditions would have a greater liking for the salesperson, interact a longer time with her, and have higher purchase intentions when she was attractive than when she was average looking. However, we expected participants in embarrassing consumption conditions to like the salesperson less, to imagine interacting with her for less time, and to have less intention to make a purchase when the salesperson was attractive than when she was not.

Intention to Purchase. Participants’ intentions to pur- chase the product are shown in Table 4 as a function of at- tractiveness level and embarrassing consumption. Participants reported a greater intention to purchase the product in nonembarrassing consumption conditions than in embarrassing consumption conditions (6.61 vs. 5.68, re- spectively; F(1, 108)¼19.80; p < .001). However, the in- teraction of salesperson attractiveness and embarrassment conditions was also significant (F(1, 108)¼10.57, p < .01) and of the form expected. That is, participants in nonem- barrassing consumption conditions reported a greater inten- tion to purchase the product when the salesperson was attractive than when she was not (6.95 vs. 6.27, respec- tively; F(1, 108)¼11.44; p < .01). In embarrassing con- sumption conditions, however, participants reported lower intentions to make a purchase when the salesperson was at- tractive than when she was not (5.34 vs. 6.02, respectively; F(1, 108)¼11.45; p < .01).

Interaction Duration. The duration of participants’ imagined interaction with the salesperson is summarized in the second section of Table 4. Participants were willing to interact with the salesperson for a longer time in nonem- barrassing consumption conditions than in embarrassing consumption conditions (5.24 vs. 4.29 minutes, respec- tively; F(1, 108)¼11.16; p < .01). However, the interac- tion of salesperson attractiveness and embarrassment conditions was also significant (F(1, 108)¼6.04, p < .05)

590 JOURNAL OF CONSUMER RESEARCH

and indicated that participants in nonembarrassing con- sumption conditions imagined interacting longer with the attractive salesperson than with the average-looking one (5.68 min vs. 4.79 min, respectively; F(1, 108)¼10.31; p < .01). In contrast, participants in embarrassing con- sumption conditions imagined interacting for a shorter time with the attractive salesperson than with the average- looking one (4.04 min vs. 4.54 min, respectively; F(1, 108)¼3.56; p¼ .06).

Liking for the Salesperson. Participants’ liking for the salesperson is summarized in the third section of Table 4. The interaction of salesperson attractiveness and embar- rassment conditions was significant (F(1, 108)¼5.86, p < .01). That is, participants in the nonembarrassing con- sumption condition liked the attractive salesperson more than the average-looking one (6.36 vs. 5.75, respectively; F(1, 108)¼6.88; p < .01). In contrast to conditions in which the salesperson was of the opposite sex, however, participants in embarrassing consumption conditions liked the attractive salesperson less than the average-looking one (5.39 vs. 5.96, respectively; F(1, 108)¼6.08; p < .05).

Self-Presentation Concern. The self-presentation con- cern that participants reported are shown in the fourth sec- tion of Table 4. Participants reported more concern when the salesperson was highly attractive than when she was average

looking (5.26 vs. 4.51, respectively; F(1, 108)¼11.11, p < .01), and more concern in embarrassing consumption conditions than in nonembarrassing conditions (5.27 vs. 4.50, respectively; F(1, 108)¼11.83, p < .01). However, the interaction of salesperson attractiveness and embarrassment conditions was also significant (F(1, 108)¼4.94, p < .05). That is, participants in embarrassing consumption conditions felt greater concern when the salesperson was highly attrac- tive than when she was average looking (5.89 vs. 4.64, re- spectively; F(1, 108)¼31.56, p < .001). In nonembarrassing consumption conditions, however, this was not true (4.62 vs. 4.37, respectively; F(1, 108)¼1.94, p > .10).

Emotional Reactions to Social Comparison. We ex- pected that a highly attractive salesperson would stimulate participants to engage in social comparison and that this would have negative consequences on both their self- perceptions of attractiveness and their reactions to the salesperson. Responses summarized in Table 4 confirm these expectations. Participants reported a greater tendency to engage in social comparison when the salesperson was highly attractive than when she was average looking (4.72 vs. 4.06, respectively; F(1, 108)¼10.02, p < .01), and in embarrassing conditions than in nonembarrassing conditions (4.69 vs. 4.10, respectively; F(1, 108)¼7.97, p < .01). However, the interaction of salesperson attractiveness and embarrassment conditions was also marginally significant (F(1, 108)¼3.65, p¼ .059). That is, participants in embar- rassing consumption conditions had a higher social compari- son tendency when the salesperson was highly attractive than when she was average looking (5.21 vs. 4.16, respec- tively; F(1, 108)¼26.6, p < .001). In nonembarrassing con- sumption conditions, however, this was not true (4.23 vs. 3.96, respectively; F(1, 108)¼2.49, p > .10).

As a result of these comparisons, participants reported their own attractiveness to be a lower rating when the sales- person was highly attractive than when she was not (5.93 vs. 6.45, respectively; F(1, 108)¼6.08, p < .05). However, the interaction of salesperson attractiveness and embarrass- ment conditions was significant (F(1, 108)¼3.83, p¼ .05) and indicated that this difference was evident in embarrass- ing consumption conditions (5.61 vs. 6.54, respectively; F(1, 108)¼20.44, p < .001) but not in nonembarrassing consumption conditions (6.25 vs. 6.36, respectively; F(1, 108)¼1.08, p > .10).

Analyses of participants’ mood as a function of experi- mental conditions yielded a marginally significant interac- tion of salesperson attractiveness and embarrassment conditions (F(1, 108)¼3.33, p¼ .07). That is, participants in embarrassing consumption conditions experienced less positive moods when the salesperson was highly attractive than when she was average looking (4.62 vs. 5.10, respec- tively; F(1, 108)¼6.10, p < .05). In nonembarrassing con- sumption conditions, however, this was not true (5.43 vs. 5.14, respectively; F(1, 108)¼2.74, p > .10).

TABLE 4

CONSUMER REACTIONS TO SERVICE PROVIDERS WITH DIFFERENT ATTRACTIVENESS LEVELS:—STUDY 5

Moderate attractiveness

High attractiveness

Intention to patronize Nonembarrassing 6.27a (1.05) 6.95b (1.17) Embarrassing 6.02a (1.04) 5.34c (1.16)

Interaction time, minutes Nonembarrassing 4.79a (1.59) 5.68b (1.22) Embarrassing 4.54a (1.32) 4.04c (1.79)

Liking Nonembarrassing 5.75ac (1.21) 6.36b (1.31) Embarrassing 5.96a (1.35) 5.39c (1.29)

Self-presentation concern Nonembarrassing 4.37a (1.05) 4.62a (1.16) Embarrassing 4.64a (1.19) 5.89b (1.34)

Social comparison Nonembarrassing 3.96a (1.10) 4.23a (1.07) Embarrassing 4.16a (1.14) 5.21b (1.10)

Self-attractiveness rating Nonembarrassing 6.36a (1.13) 6.25a (1.24) Embarrassing 6.54a (0.96) 5.61b (1.10)

Mood Nonembarrassing 5.14a (1.11) 5.43a (1.05) Embarrassing 5.10a (1.07) 4.62b (1.19)

Jealousy Nonembarrassing 2.46a (1.29) 3.21b (1.48) Embarrassing 2.71a (1.18) 4.79c (1.50)

NOTE.—Numbers in parentheses indicate standard deviations. Means

with unlike subscripts are significantly different at p < .05.

WAN AND WYER 591

Finally, participants reported feeling more jealous when the salesperson was highly attractive than when she was average looking (4.00 vs. 2.59, respectively; F(1, 108)¼ 29.78, p < .001), and more jealous in embarrassing condi- tions than in nonembarrassing conditions (3.75 vs. 2.84, respectively; F(1, 108)¼12.41, p < .01). As shown in Table 4, however, the interaction of salesperson attractive- ness and embarrassment conditions was significant (F(1, 108)¼6.09, p < .05) and indicated that the impact of physical attractiveness on jealousy was greater in embar- rassing consumption conditions (4.79 vs. 2.71, respectively; F(1, 108)¼32.37, p < .001) than in nonembarrassing condi- tions (3.21 vs. 2.46, respectively; F(1, 108)¼4.21, p < .05).

Mediation. To test the underlying process of the interac- tive effects of attractiveness and embarrassing consumption on purchase intention, mediation analyses using bootstrap- ping were performed. As in study 4, the interactive effects of attractiveness and embarrassment conditions on purchase intentions was significantly mediated by self-presentation concerns, that is, the model of attractiveness�embarrassing consumption ! self-presentation concern ! purchase in- tention (model 8, Hayes, 2013) was supported; based on 5000 bootstrapping samples, 95% CI from� .9609 to �.0432). More important, a sequential mediation analyses indicated that the effects of self-presentation concerns were driven by a social comparison process; that is, a model of at- tractiveness � embarrassing consumption ! social compar- ison ! self-presentation concern ! purchase intention was also supported (model 6, Hayes, 2013; based on 5000 boot- strapping samples, 95% CI from �.4043 to �.0583).

In summary, study 5 generalized the effects of previous experiments to same-sex interactions. That is, consumers react more favorably to an attractive same-sex salesperson than to an average-looking one when they are buying nonembarrassing products but react less favorably to an at- tractive salesperson than to an unattractive one when they are buying embarrassing products. (Note: These conclu- sions were confirmed in a preliminary study we conducted in which sufficient mediators were not assessed; for details, see the online appendix.) Moreover, these effects are medi- ated by self-presentation concerns. However, these con- cerns are driven by a social comparison motive, as reflected in an increased tendency to engage in social com- parison, a decrease in self-perceptions of attractiveness, a decrease in mood, increased feelings of jealousy, and a de- crease in liking for the salesperson.

GENERAL DISCUSSION

Consumers often react more favorably to physically at- tractive than to unattractive service providers, as evidenced by their greater satisfaction with the service and their inten- tions to purchase the products being sold (Ahearne et al. 1999; Argo et al. 2008; Reingen and Kernan 1993). To our

knowledge, the conditions in which the physical attractive- ness of a service provider can decrease as well as increase consumption behavior have not previously been examined. To this extent, our research not only identifies these condi- tions but also provides evidence of the mechanisms that underlie these effects.

When consumers are motivated to make a good impres- sion on an attractive provider, factors that make them feel apprehensive about their ability to accomplish this can lead them to anticipate that interacting with the provider will be unpleasant and to avoid doing so, decreasing their likeli- hood of purchasing the product or service at hand. These self-presentation concerns can be affected by chronic so- cial anxiety or can be aroused by unrelated experiences that consumers have before being exposed to the service interactions. Moreover, it can also be activated by the type of product being sold. More important, although consumers interact the same way with an opposite-sex attractive sales- person and a same-sex attractive salesperson, our research findings show that different processes underlie the effects in the two cases. When the service provider is of the oppo- site sex, self-presentation concerns are driven by sexual motivation. Consumer avoidance of interacting with the provider is motivated by attraction to the provider and ap- prehension about conveying a positive impression. When the provider is of the same sex, however, self-presentation concerns appear to be driven by social comparison pro- cesses, leading consumers to dislike the provider and to avoid interacting for this reason.

Our conceptualization of these effects intersects with the- ory and research in three different areas, concerning (1) the effect of the “beautiful is good stereotype,” (2) impression management motivation, and (3) the determinants and con- sequences of social anxiety and embarrassment. Each factor is a necessary condition for the adverse effect of a pro- vider’s attractiveness to occur. That is, when a provider is of only average attractiveness, individuals are likely to focus their attention on the product or service they are considering and are unlikely to be sensitive to the impression they create on the provider. When the provider is very attractive, con- sumers’ desire to make a good impression is increased (Dong and Wyer 2014), and if they feel relatively confident about their ability to create this impression, they may be more motivated to interact with the provider than they might otherwise be. However, if they are chronically anxious about their ability to create a good impression in social situ- ations, or if recent experiences have decreased their confi- dence in their self-efficacy, these concerns may generalize to the interaction they anticipate having with the service provider and lead them to avoid engaging in this interaction.

This avoidance motivation may also be activated by the type of product or service being considered, as studies 3, 4, and 5 indicate. The impact of embarrassment in the situa- tions we investigated extend previous research in this area. Previous research has been mainly concerned with how

592 JOURNAL OF CONSUMER RESEARCH

embarrassment prevents consumers from achieving posi- tive benefits, for example, purchasing condoms to have safe sex (Moore et al. 2006), having appropriate medical examination to save lives (Consedine, Krivoshekova, and Harris 2007; Consedine et al. 2011), and using in-store coupons to save money (Brumbaugh and Rosa 2009). With few exceptions (e.g., Dahl et al. 2001; Dong et al. 2013; Wan 2013), little attention has been given to the factors that influence consumers’ behavioral reactions to embar- rassing consumption situations. To this extent, our research adds to our understanding of the moderating effect of em- barrassment on consumers’ behavior in actual consumption situations and the conditions in which this effect occurs.

In this regard, our research suggests constraints on the desirability of using attractive service providers to increase the sale of products. This strategy may indeed be effective when the product being promoted is not embarrassing. It can have an adverse effect, however, when the purchasing of the product is embarrassing. That said, it is important to note that the negative impact of attractive service providers is restricted to face-to-face situations in which the motive to create a favorable impression is relatively high. The use of attractive models or celebrities in advertisements for embarrassing products could have a positive effect on on- line shopping in which social interaction is not an issue.

Cultural or dispositional variables might also moderate the relationship between attractiveness and consumer reac- tions to embarrassing consumptions. For example, people with an interdependent self-construal are more concerned with impression management than those with an indepen- dent self-construal (Lalwani, Shavitt, and Johnson 2006). If this is so, these differences should be particularly evident in embarrassing consumption situations. In contrast, if nonembarrassing consumptions do not elicit self-presenta- tion concerns, differences in self-construal might not affect purchase behavior. This possibility is worth exploring.

DATA COLLECTION INFORMATION

The first author conducted the pilot study at the Hong Kong Richmond Shopping Arcade in the winter of 2013 and study 3 at the Hong Kong Prince of Wales Hospital in the spring of 2013. The first author supervised the collection of data for studies 1, 2, 4, and 5 by three research assistants at the Chinese University of Hong Kong in 2013–14. The first author supervised the collection of data for study 5 by a re- search assistant at the Chinese University of Hong Kong in the spring of 2015. Data from all studies were analyzed by the first author and were discussed by both authors.

APPENDIX

Study 4 Advertisement posters

Embarrassing Consumption

Non-Embarrassing Consumption

WAN AND WYER 593

REFERENCES

Agthe, Maria, Matthias Spörrle, and Jon K. Maner (2010), “Don’t Hate Me Because I’m Beautiful: Anti-attractiveness Bias in Organizational Evaluation and Decision Making,” Journal of Experimental Social Psychology, 46 (6), 1151–54.

———. (2011), “Does Being Attractive Always Help? Positive and Negative Effects of Attractiveness on Social Decision Making,” Personality and Social Psychology Bulletin, 37 (8), 1042–54.

Ahearne, Michael, Thomas W. Gruen, and Cheryl Burke Jarvis (1999), “If Looks Could Sell: Moderation and Mediation of the Attractiveness Effect on Salesperson Performance,” International Journal of Research in Marketing, 16 (4), 269–84.

Apsler, Robert (1975), “Effects of Embarrassment on Behavior Toward Others,” Journal of Personality and Social Psychology, 32 (1), 145–53.

Argo, Jennifer J., Darren W. Dahl, and Andrea C. Morales (2008), “Positive Consumer Contagion: Responses to Attractive Others in a Retail Context,” Journal of Marketing Research, 45 (6), 690–701.

Arkin, Robert M., Alan J. Appelman, and Jerry M. Burger (1980), “Social Anxiety, Self-Presentation, and the Self-Serving Bias in Causal Attribution,” Journal of Personality and Social Psychology, 38 (1), 23–35.

Bernstein, William M., Blair O. Stephenson, Melvin L. Snyder, and Robert A. Wicklund (1983), “Causal Ambiguity and Heterosexual Affiliation,” Journal of Experimental Social Psychology, 19 (1), 78–92.

Berscheid, Ellen and Elaine Walster (1974), “Physical Attractiveness,” in Advances in Experimental Social Psychology, Vol. 7, ed. Leonard Berkowitz, New York: Academic Press, 157–215.

Brown, Bert R. (1970), “Face-saving Following Experimentally Induced Embarrassment,” Journal of Experimental Social Psychology, 6 (3), 255–71.

Brumbaugh, Anne M. and José Antonio Rosa (2009), “Perceived Discrimination, Cashier Metaperceptions, Embarrassment, and Confidence as Influencers of Coupon Use: An Ethnoracial-Socioeconomic Analysis,” Journal of Retailing, 85 (3), 347–62.

Buunk, Bram P. and Vera Hoorens (1992), “Social Support and Stress: The Role of Social Comparison and Social Exchange Processes,” British Journal of Clinical Psychology, 31 (4), 445–57.

Consedine, Nathan S., Yulia S. Krivoshekova, and Christine R. Harris (2007), “Bodily Embarrassment and Judgment Concern as Separable Factors in the Measurement of Medical Embarrassment: Psychometric Development and Links to Treatment-Seeking Outcomes,” British Journal of Health Psychology, 12 (3), 439–62.

Consedine, Nathan S., Inga Ladwig, Maike K. Reddig, and Elizabeth A. Broadbent (2011), “The Many Faeces of Colorectal Cancer Screening Embarrassment: Preliminary Psychometric Development and Links to Screening Outcome,” British Journal of Health Psychology, 16 (3), 559–79.

Dabbs, James M. Jr. and Neil A. Stokes III (1975), “Beauty Is Power: The Use of Space on the Sidewalk,” Sociometry, 38 (4), 551–57.

Dahl, Darren W., Rajesh V. Manchanda, and Jennifer J. Argo (2001), “Embarrassment in Consumer Purchase: The Role of

Social Presence and Purchase Familiarity,” Journal of Consumer Research, 28 (3), 473–81.

DeShields, Oscar W. Jr., Ali Kara, and Erdener Kaynak (1996), “Source Effects in Purchase Decisions: The Impact of Physical Attractiveness and Accent of Salesperson,” International Journal of Research in Marketing, 13 (1), 89–101.

Dion, Karen, Ellen Berscheid, and Elaine Walster (1972), “What Is Beautiful Is Good,” Journal of Personality and Social Psychology, 24 (3), 285–90.

Dittmar, Helga and Sarah Howard (2004), “Thin-Ideal Internalization and Social Comparison Tendency as Moderators of Media Model’s Impact on Women’s Body- Focused Anxiety,” Journal of Social and Clinical Psychology, 23 (6), 768–91.

Dong, Ping, Xun (Irene) Huang, and Robert S. Wyer Jr. (2013), “The Illusion of Saving Face: How People Symbolically Cope with Embarrassment,” Psychological Science, 24 (10), 2005–12.

Dong, Ping and Robert S. Wyer Jr. (2014), “How Time Flies: The Effects of Conversation Characteristics and Partner Attractiveness on Duration Judgments in a Social Interaction,” Journal of Experimental Social Psychology, 50 (January), 1–14.

Eagly, Alice H., Richard D. Ashmore, Mona G. Makhijani, and Laura C. Longo (1991), “What Is Beautiful Is Good, But. . .: A Meta- Analytic Review of Research on the Physical Attractiveness Stereotype,” Psychological Bulletin, 110 (1), 109–28.

Goffman, E. (1959), The Presentation of Self in Everyday Life, Garden City, NY: Doubleday.

Griskevicius, Vladas, Joshua M. Tybur, Jill M. Sundie, Robert B. Cialdini, Geoffrey F. Miller, and Douglas T. Kenrick (2007), “Blatant Benevolence and Conspicuous Consumption: When Romantic Motives Elicit Strategic Costly Signals,” Journal of Personality and Social Psychology, 93 (1), 85–102.

Hayes, Andrew F. (2013), Introduction to Mediation, Moderation, and Conditional Process Analysis, New York: Guilford.

Hudders, Liselot, Charlotte De Backer, Maryanne Fisher, and Patrick Vyncke (2014), “The Rival Wears Prada: Luxury Consumption as a Female Competition Strategy,” Evolutionary Psychology, 12 (3), 570–87.

Jones, April M. and Justin T. Buckingham (2005), “Self-Esteem as a Moderator of the Effect of Social Comparison on Women’s Body Image,” Journal of Social and Clinical Psychology, 24 (8), 1164–87.

Kenrick, Douglas T., Daniel R. Montello, Sara E. Gutierres, and Melanie R. Trost (1993), “Effects of Physical Attractiveness on Affect and Perceptual Judgments: When Social Comparison Overrides Social Reinforcement,” Personality and Social Psychology Bulletin, 19 (2), 195–99.

Koernig, Stephen K. and Albert L. Page (2002), “What If Your Dentist Looked Like Tom Cruise? Applying the Match-up Hypothesis to a Service Encounter,” Psychology & Marketing, 19 (1), 91–110.

Lalwani, Ashok K., Sharon Shavitt, and Timothy Johnson (2006), “What Is the Relation Between Cultural Orientation and Socially Desirable Responding,” Journal of Personality and Social Psychology, 90 (1), 165–78.

Leary, Mark R. (1983), “Social Anxiousness: The Construct and Its Measurement,” Journal of Personality Assessment, 47 (1), 66–75.

———. (1986), “The Impact of Interactional Impediments on Social Anxiety and Self-Presentation,” Journal of Experimental Social Psychology, 22 (2), 122–35.

594 JOURNAL OF CONSUMER RESEARCH

———. (1996), Self-Presentation: Impression Management and Interpersonal Behavior, Boulder, CO: Westview.

Leary, Mark R. and Robin M. Kowalski (1990), “Impression Management: A Literature Review and Two-Component Model,” Psychological Bulletin, 107 (1), 34–47.

Leary, Mark R., John B. Nezlek, Deborah Downs, Julie Radford- Davenport, Jeffrey Martin, and Anne McMullen (1994), “Self-Presentation in Everyday Interactions: Effects of Target Familiarity and Gender Composition,” Journal of Personality and Social Psychology, 67 (4), 664–73.

Lockwood, Penelope and Ziva Kunda (1997), “Superstars and Me: Predicting the Impact of Role Models on the Self,” Journal of Personality and Social Psychology, 73 (1), 91–103.

Maner, Jon K., Saul L. Miller, D. Aaron Rouby, and Matthew T. Gailliot (2009), “Intrasexual Vigilance: The Implicit Cognition of Romantic Rivalry,” Journal of Personality and Social Psychology, 97 (1), 74–87.

Miller, Rowland S. (1987), “Empathic Embarrassment: Situational and Personal Determinants of Reactions to the Embarrassment of Another,” Journal of Personality and Social Psychology, 53 (6), 1061–69.

Moore, Sarah G., Darren W. Dahl, Gerald J. Gorn, and Charles B. Weinberg (2006), “Coping with Condom Embarrassment,” Psychology, Health & Medicine, 11 (1), 70–79.

Mori, DeAnna, Shelly Chaiken, and Patricia Pliner (1987), “‘Eating Lightly’ and the Self-Presentation of Femininity,” Journal of Personality and Social Psychology, 53 (4), 693–702.

Niu, Han-Jen, Yung-Sung Chiang, and Hsien-Tang Tsai (2012), “An Exploratory Study of the Otaku Adolescent Consumer,” Psychology & Marketing, 29 (10), 712–25.

Reingen, Peter H. and Jerome B. Kernan (1993), “Social Perception and Interpersonal Influence: Some Consequences of the Physical Attractiveness Stereotype in a Personal Selling Setting,” Journal of Consumer Psychology, 2 (1), 25–38.

Richins, Marsha L. (1991), “Social Comparison and the Idealized Images of Advertising,” Journal of Consumer Research, 18 (1), 71–83.

Sarnoff, Irving and Philip G. Zimbardo (1961), “Anxiety, Fear, and Social Isolation,” Journal of Abnormal and Social Psychology, 62 (2), 356–63.

Schlenker, Barry R. (1980), Impression Management: The Self- Concept, Social Identity, and International Relations, Monterey, CA: Brooks/Cole.

Schlenker, Barry R. and Mark R. Leary (1982), “Social Anxiety and Self-Presentation: A Conceptualization Model,” Psychological Bulletin, 92 (3), 641–69.

Sigall, Harold and Elliot Aronson (1969), “Liking for an Evaluator as a Function of Her Physical Attractiveness and Nature of the Evaluations,” Journal of Experimental Social Psychology, 5 (1), 93–100.

Sigall, Harold and David Landy (1973), “Radiating Beauty: Effects of Having a Physically Attractive Partner on Person Perception,” Journal of Personality and Social Psychology, 28 (2), 218–24.

Snyder, Mark, Elizabeth D. Tanke, and Ellen Berscheid (1977),“Social Perception and Interpersonal Behavior: On the Self-Fulfilling Nature of Social Stereotypes,” Journal of Personality and Social Psychology, 35 (9), 656–66.

Tesser, Abraham and James E. Collins (1988), “Emotion in Social Reflection and Comparison Situations: Intuitive, Systematic and Exploratory Approaches,” Journal of Personality and Social Psychology, 55 (5), 695–709.

Tyler, James M. and Robert S. Feldman (2004), “Truth, Lies, and Self-Presentation: How Gender and Anticipated Future Interaction Relate to Deceptive Behavior,” Journal of Applied Social Psychology, 34 (12), 2602–15.

Walster, Elaine, Vera Aronson, and Darcy Abrahams (1966), “Importance of Physical Attractiveness in Dating Behavior,” Journal of Personality and Social Psychology, 4 (5), 508–16.

Wan, Lisa C. (2013), “Culture’s Impact on Consumer Complaining Responses to Embarrassing Service Failure,” Journal of Business Research, 66 (3), 298–305.

Wang, Yajin and Vladas Griskevicius (2014), “Conspicuous Consumption, Relationships, and Rivals: Women’s Luxury Products as Signals to Other Women,” Journal of Consumer Research, 40 (5), 834–54.

Wenzel, Amy and Tracy Emerson (2009), “Mate Selection in Socially Anxious and Nonanxious Individuals,” Journal of Social and Clinical Psychology, 28 (3), 341–63.

Wood, Joanne V. (1989), “Theory and Research Concerning Social Comparisons of Personal Attributes,” Psychological Bulletin, 106 (2), 231–48.

WAN AND WYER 595

Copyright of Journal of Consumer Research is the property of Oxford University Press / USA and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.