the influence of fashion on mediated beauty in white and black female adolescents.
decrlo
article_for_assignment.pdf
COMMUNICATION RESEARCH • June 2002David et al. • Body Image, Race, and Fashion Models PRABU DAVID GLENDA MORRISON MELISSA A. JOHNSON FELECIA ROSS
Body Image, Race, and Fashion Models Social Distance and Social Identification in Third-Person Effects
The perceived effects of advertising on body-image factors were tested in both Black and White college-age women. After seeing magazine ads that portrayed either Black fashion models or White fashion models, respondents rated per- ceived effects of these ads on body-image factors. The effects were rated on self, on other Black women on campus, and other White women on campus. When projecting perceived effects on others—of the same race or a different race— both Blacks and Whites indicated that media effects would be maximal when the race of the model matched the race of the respondent. However, when rating perceived effects on self, whereas Blacks identified strongly with Black models, there was no significant difference in the way Whites identified with fashion models of either race. The results are examined within the framework of social distance and social identification.
Biases in social judgment about perceived negative media effects on self and others has drawn considerable attention from communication scholars, par- ticularly those who are interested in the connection between such “third- person perceptions” and the willingness to restrict media content (e.g., Cohen, Mutz, Price, & Gunther, 1988; Davison, 1983; Duck, Terry, & Hogg, 1995; Eveland, Nathanson, Detenber, & McLeod, 1999; Gunther & Hwa, 1996; Perloff, 1989; Salwen, 1998; Salwen & Driscoll, 1997). For instance, Shah, Faber, and Youn (1999) found a significant relationship between third-person perceptions and the willingness among respondents to censor advertising of controversial services, such as gambling. Other researchers have found support for controls on violent or misogynic rap music (McLeod, Eveland, & Nathanson, 1997), pornography (Gunther, 1995; Rojas, Shah, &
270
COMMUNICATION RESEARCH, Vol. 29 No. 3, June 2002 270-294 © 2002 Sage Publications
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
Faber, 1996), television violence (Rojas et al., 1996), and political communica- tion (Rucinski & Salmon, 1990; Salwen, 1998). Although the link between third- person perceptions and behavioral outcomes, such as support for or against policies, has not been clearly established, the underlying theoretical issues merit attention.
One of the interesting aspects of the third-person phenomenon is the social-distance corollary. As the social distance between a respondent and a comparison group increases, there is a widening of the third-person gap. In the typical demonstration, when college students are asked to estimate the perceived effect of media on self, followed by perceived effects on other stu- dents on campus (a group similar to the respondents), and finally perceived effects on other people in the state (a group that is socially farther away and less similar to the student participants), perceived effects on other people in the state has been observed to be greater than perceived effects on students on campus (e.g., Cohen et al., 1988; Gunther, 1991).
Perloff (1993) explained the social distance phenomenon on the basis of psychological distance. Psychological distance is generally conceived as degree of similarity or dissimilarity between the respondent and the compar- ison group on critical dimensions relevant to the media effect. Some of the dimensions of social and psychological distance have been examined and results are mixed with some interesting but predictable patterns (Brosius & Engel, 1996; Cohen et al., 1988; Eveland et al., 1999; McLeod et al., 1997).
Brosius and Engel (1996) varied psychological distance by using two targets—friends and acquaintances as the close group and people in general as the distant group—and found that perceived effects were greater for the general group. Duck and Mullin (1995) varied comparison groups on two dimensions, namely vagueness and closeness. While closeness was operationalized as friends versus others, vagueness was varied from the “average person” to a specific person, such as a named celebrity. Their find- ings suggest that perceived effects were the greatest when the comparison group was a vague, distant other.
In a couple of studies, however, the social-distance effect has been con- founded by other more salient attributes such as likelihood of exposure to the message and severity of the message. In these studies (McLeod et al., 1997; Eveland et al., 1999) in which misogynist rap and death-metal lyrics were used as stimuli, students rated a specific intended target audience as more vulnerable than the more general group. For example, in the earlier study (McLeod et al., 1997), youth from New York or Los Angeles were perceived to be more vulnerable to “gangsta” rap lyrics than the average person. Intu- itively this is a powerful finding because it is logical to expect youth in New
271
David et al. • Body Image, Race, and Fashion Models
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
York and Los Angeles to be more vulnerable to gangster rap because of the gang-related reputations of these cities.
Eveland and colleagues conducted a similar two-part study (Eveland et al., 1999). In the first part, groups were “distanced” by age and education. Although there was partial support for the social-distance corollary, again respondents appeared to build into their answers common-sense judgments about who might be exposed to non-mainstream media such as violent and misogynist music. In several tests, betas for perceived media exposure were significantly larger than betas for perceived social distance, overturning the traditional social distance corollary. However, the present study examines fashion advertising, which is ubiquitous in U.S. mass media and hence not dependent on selective exposure.
In addition to perceived exposure considerations, the aforementioned study also dealt with media that feature violent or hateful messages, ranging from violent action films to death-metal music.
If the media stimuli are socially undesirable at some level, such as porno- graphic material or misogynistic lyrics, perceived exposure to media is likely to be highly relevant. Even before respondents focus on the negative per- ceived effect on others, it is likely that they engage in a subtler form of third- person bias by distancing themselves and people close to them from exposure to socially undesirable media content.
In short, it appears the effects of social distance are contingent on general- ity or specificity of the target group, similarity or dissimilarity between the respondent and the target group, the comparison group’s perceived likelihood of exposure to the target stimuli, and real-world heuristics.
Although a number of social-distance attributes have already been exam- ined, to our knowledge third-person differences based on race or ethnicity have not been studied. Moreover, as a variable, race offers a promising avenue to examine psychological mechanisms underlying in-group versus out-group differences based on social identification. Perloff (1993) noted that social dis- tance could be conceptualized in at least two ways: along a continuum of “just like me” to “not at all like me” or in terms of the homogeneity and size of the group. Race appears to fit both these continua. Moreover, as media continue to be more segmented and personalized, media researchers will have to find appropriate ways to study media perceptions among in-groups and out- groups. The third-person framework provides a viable platform for testing differences based on social identification.
Hence, in this article, we extend previous work on the social-distance com- ponent of third-person effects by introducing race as a marker of social or psy- chological distance. Using social identity as a theoretical foundation, we focus on how Black and White women perceive effects of advertising on body-image
272
COMMUNICATION RESEARCH • June 2002
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
factors. The findings are discussed within the context of our research on the role of the media in shaping body-image ideals in young women.
Media Portrayals and the Ideal Body Image
The ultra-thin model and the idealized images of beauty presented as the norm in advertising and other media content have been criticized by scholars, media critics, and popular-media writers. Researchers in public health have found that the mismatch between the ideal body image portrayed in the media and one’s actual body image leads to body-image dissatisfaction (Richins, 1991; Stice, Schupak-Neuberg, Shaw, & Stein, 1994; Stormer & Thompson, 1996). In extreme cases, body-image dissatisfaction or physique anxiety hurts an individual’s self-esteem or self-concept and could even lead to behavioral consequences such as eating disorders (e.g., Thompson, 1995).
Although the relationship between media use and body-image percep- tions is tenuous, there is an emerging body of research documenting signifi- cant relationships. For example, Harrison (2000) found a significant relation- ship between bulimia and exposure to television shows with overweight characters. In the same study, a significant relationship was found between thin-ideal magazine content and increased anorexia. Others have found sig- nificant relationships between selective media use and drive for thinness (Botta, 1999, 2000; Harrison & Cantor, 1997) and between media exposure and body-image elasticity (Myers & Biocca, 1992). Given these findings, we set out to examine self-reported perceptions of media effects on body image.
Media Effects on Body-Image Factors
For the purposes of this study, media influences are presented as a cluster of effects with varying levels of undesirability, collectively referred to as body-image factors. If body-image dissatisfaction can be expressed as a mis- match between ideal and actual body-image perceptions, the role of media can be examined by assessing the extent to which they are responsible for cre- ating and triggering these body-image ideals. Body-image dissatisfaction itself is not inherently insidious as much as are the negative outcomes that could stem from the dissatisfaction. In some cases, body-image dissatisfac- tion could be channeled into a desirable outcome. For example, the Nike com- mercial of the “Just Do It” genre that exhorts viewers to exercise is in some ways equivalent to a public-service announcement that seeks a positive out- come. But the focus of this study is media portrayal of thin, idealized models in fashion magazines.
273
David et al. • Body Image, Race, and Fashion Models
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
As just noted, one behavioral outcome measure of body-image media effects on women is an eating disorder such as anorexia or bulimia. However, in our framework, we explore a range of effects, including the effect on percep- tion of ideal body weight (a perceptual outcome) and effect on perception of one’s self-esteem (an attitudinal or psychological outcome). In other words, the dependent variables in our framework extend hierarchically from the less-desirable media effect that affects perceptions of ideal body weight, to the more undesirable effect on the respondent’s psychological state, to the most undesirable behavioral outcome that is manifested as an eating disor- der. This hierarchy of body-image media effects is referred to as body-image factors in this study. These factors have been examined in the past and the results suggest that as the degree of undesirability of the outcome increases, the third-person gap widens (David & Johnson, 1998).
Body-Image Dissatisfaction, Eating Disorders, and Race
Conventional wisdom and a growing body of empirical evidence suggest that Black women are less vulnerable to body-image dissatisfaction and are more proud of their bodies than Anglo-Americans and Hispanics (Altabe, 1998; Fitzgibbon et al., 1998; Story, French, Resnick, & Blum, 1995). Studies show that Blacks are less concerned about weight than are Whites (White, Hudson, & Campbell, 1985), exhibit fewer weight-reduction behaviors such as dieting and exercising (Kann & Ross, 1996), and have lower incidences of eating dis- orders than Whites (Gray, Ford, & Kelly, 1987). Ironically, however, African Americans exhibit these positive body-image attributes despite having higher body-mass index scores and higher fat composition than Whites (Akan & Grilo, 1995; Ellis, Abrams, & Wong, 1997; Field, Colditz, & Peterson, 1997; French et al., 1997). In general, health researchers have been more con- cerned about obesity in African Americans than eating disorders.
Harris (1995) observed that the varying degrees of emphasis on thinness among Blacks and Whites could be attributed to culturally ingrained norms, with thinness as a strongly internalized ideal among White women, but less of an ideal among Black women. Some argue that the individualistic nature of some White cultures emphasizes individual control—including control of body— rather than dependence on the family, which is more common to some Black cultures.
Despite the racial and ethnic differences on body image, a number of vari- ables associated with eating disorders are common among women of all races. For instance, regardless of race, eating disorders have some common mark- ers, such as perfectionist personality traits and dysfunctional family environ- ments. One study found that Black girls who were teased about their weight
274
COMMUNICATION RESEARCH • June 2002
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
had a significantly higher drive for thinness than those who were not teased (Striegel-Moore, Schreiber, Pike, Wilfley, & Rodin, 1995), suggesting that Blacks are not immune to the social pressures of ideal body image. Silber (1986) argued that as Blacks make strides in the workplace, there is growing pressure to conform to the norms of the dominant culture, namely the Euro- centric ideals of beauty and attractiveness.
After taking into consideration the differing body-image values and norms between races, we expected some differences in third-person percep- tions about body image. At least two underlying processes are at play when projecting to others of a different race—one, the traditional third-person effect that widens with social distance; and two, the confounding of social dis- tance by race because of differences in perceived vulnerabilities of the two races, which are driven by cultural stereotypes and conventional wisdom. The findings by Eveland et al. (1999) would suggest that Whites would be more vulnerable to body-image media effects because it is generally con- strued that White women rather than Black women are the intended targets of ads that emphasize extreme thinness.
Social Distance and Social Identification
Dissimilarity based on race can be examined from the standpoint of categori- zation processes that are described in social-identity theory. Social-identity theory holds that our self-concept is based in part on the way we perceive oth- ers and how we interact with them (Hogg & Abrams, 1999; Serino, 1998; Tajfel, 1978, 1981; Tajfel & Turner, 1986; Turner & Onorato, 1999). Self- concept depends not only on personality traits of an individual but also on how an individual relates to others within a group. Because part of a person’s self-concept is based on group membership, social comparisons between groups produce pressures to earn collective self-esteem for the in-group rela- tive to out-groups. Self-categorization theory is a companion theory that specifies the social-cognition aspects and contrasts with others.
Using social identity and social comparison, some researchers have focused on racial and ethnic identity (Liebkind, 1992; Pelham & Hetts, 1999) because of its usefulness in studying stereotypes, prejudice, and other behav- iors in interracial settings. Prior research in this area suggests that an indi- vidual’s beliefs about personal and social identities play a key role in group- relevant judgments (Pelham & Hetts, 1999).
Studies in social-identity theory and social categorization theory do not specifically deal with the concept of social distance. However, the intersection between social distance and racial identity and how it affects intergroup per- ception has been explored. One study on race and social distance found that
275
David et al. • Body Image, Race, and Fashion Models
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
Blacks rated Whites to be less acceptable than Whites rated Blacks in the three domains studied: classmate or coworker, next-door neighbor, and inti- mate contact (Hraba, Radloff, & Ray-Gray, 1999).
Social-identity scholars might explain this by looking at the cultural norms of the respondents, honing in on the individualistic nature of some cul- tural groups and the collectivist cultures of others. Collective cultures put more emphasis on the welfare of the group than the welfare of the individual. For instance, studies have found that there is a slight tendency for individu- alists to be less discriminating than collectivists in in-group favoritism (Morales, López-Sáez, & Vega, 1998).
Race, Social Distance, and Third-Person Effects
Our research objective was to formulate social distance using race as a marker and to examine how third-person media effects are perceived for in-groups and out-groups. If an ethnic or racial group is perceived to be socially more distant, the social-distance corollary of the third-person effect should hold up. The bias against out-groups or a bias in favor of the in-group can be interpreted as a self-serving motivational bias that enhances the posi- tions of in-group members. This bias in favor of the in-group can be tied to social identification.
In previous research on third-person effects, social distance has been stud- ied primarily as distance between the respondent and the comparison group. Another component of social distance is the distance between the source of the media message and the respondent. In essence, social identity extends in two directions: from the source of the message to the respondent (Tyler & Cook, 1984) and from the respondent to the comparison group. Therefore, in this article we attempt to extend previous research by manipulating two com- ponents of social distance that are critical to eliciting the third-person effect: similarity between the source and respondent, and similarity between the respondent and the comparison group. Similarity and dissimilarity were operationalized on the basis of race, which was an intentional design artifact.
Hypotheses and Research Questions
The purpose of this study was to examine the effects of how body-image por- trayals in fashion advertising are perceived by Black and White college-age women. In addition to perceived effects on self, both groups of students esti- mated perceived effects on other Black students and other White students. Both groups responded to two sets of magazine ads, one with only White fash- ion models and another set with only Black fashion models.
276
COMMUNICATION RESEARCH • June 2002
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
The first hypothesis captures the basic third-person perception that oth- ers are more vulnerable to media effects than self.
Hypothesis 1: Perceived effect on others will be greater than perceived effect on self.
The next hypothesis is derived from findings reported by David and John- son (1998) that the third-person gap widens as the media outcome becomes more undesirable. In other words, the third-person difference for the eating- disorder outcome would be greater than the third-person difference for the self-esteem outcome, which in turn will be greater than the third-person gap for the ideal-body-weight outcome.
Hypothesis 2: The difference between perceived effect on self and others will widen with an increase in social undesirability of the outcome.
The next hypothesis is essentially a test of social identification, which sug- gests that respondents are more likely to identify with a similar source who is part of their in-group than someone who is dissimilar and part of the out- group.
Hypothesis 3: Perceived effect on self will be greater when the race of the respondent matches the race of the model in the ad than in the mis- matched condition.
In general, one would expect a widening of the third-person gap with an increase in social distance and dissimilarity between self and the comparison group. However, because of cultural stereotypes and the different body- image norms among Blacks and Whites, which were discussed earlier, a dif- ferential effect was predicted on the basis of race of the respondent.
Hypothesis 4: For Black college women, perceived effect of media on other White women (dissimilar to self) will be greater than perceived effect on other Black women (similar to self).
Hypothesis 5: For White college women, perceived effect on other White women (similar to self) will be greater than perceived effect on other Black women (dissimilar to self).
In addition to these hypotheses, the critical research question for the study was to examine the interaction among race of the fashion model, the race of the respondent, and the race of the comparison group.
277
David et al. • Body Image, Race, and Fashion Models
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
Research Question 1: What are the interactions among the race of the respondent, the race of the comparison group, and the race of the fash- ion model on body-image factors?
Method
Scales
Two psychological constructs, self-esteem (Rosenberg, 1965) and social- physique anxiety (Hart, Leary, & Rajeski, 1989), were measured using 10- item scales. Third-person perceptions were measured using a 11-point scale, where 0 = no influence and 10 = considerable influence.
Stimulus
Ten ads were chosen from major fashion magazines, such as Vogue, Cosmo- politan, Essence, and Ebony. The authors chose the ads with help from gradu- ate students. Half of these ads featured thin Black models and the other half featured thin White models. These models were chosen on the basis of thin- ness, attractiveness, skin tone, hair type, and sexiness. The Black models var- ied from dark to lighter skin tones, with short natural hair to straightened hair. The skin tones of the White models varied from extremely fair to well tanned, with hair color ranging from brunette to light blonde.
We tried to match the ads featuring the Black and White models on a num- ber of dimensions, including style of dress, extent of skin exposed, activity or pose, and the mood conveyed by the ad. To minimize the role of familiarity, ads with well-known supermodels were excluded.
Participants and Study Design
Eighty female undergraduate students participated in the study for extra credit. One half of the participants were Black, and the other half were White. In the absence of a random between-subjects experimental manipulation, it would be accurate to characterize this study as a quasi-experiment, with a 2 (Respondent’s race: Black, White) × 2 (Order: perceived effect on self first, perceived effect on self last) × 2 (Race of the fashion models: Black, White) × 3 (Social distance: self, other Black women, other White women) mixed design. Respondent’s race was the only between-subjects factor. Race of the fashion models and social distance were both within-subject factors.
278
COMMUNICATION RESEARCH • June 2002
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
Procedure
When students arrived, they were given a consent form and the self-esteem and physique-anxiety scales, which were stapled together as a booklet. After filling out the scales, students evaluated the ads. Each ad was evaluated on three 10-point scales, which tapped attractiveness of the model, thinness of the model, and the similarity between the model’s body type and the respon- dent’s body type.
The ads were evaluated in two sets, with five ads in each set. One set of ads featured only Black models, and the other set featured only White models. After evaluating all the ads in a set, respondents rated the perceived effect of these ads on one’s own body image, the perceived effect on other Black (or White) women on campus, and the perceived effect on other White (or Black) women on campus.
Perceived effect of advertising was obtained for three body-image factors: ideal body weight, self-esteem, and likelihood of developing an eating disor- der. Perceived effects of advertising on all the body-image factors were mea- sured using an 11-point scale (0 = no influence and 10 = considerable influ- ence). The question was phrased using the following format: How much influence do you think these ads have on your perception of ideal body weight? This question was varied by changing the comparison group to “other Black women on campus” and “other White women on campus,” and repeated for “effect on self-esteem” and the “likelihood that that these ads could lead to an eating disorder.”
One half of the students rated the ads with Black models first, and the other half rated the ads with White models first. Order of evaluation of the third-person perceptions was counterbalanced through the use of four ver- sions of the third-person questionnaire (two versions for the Black students and two versions for the White students). In one version, the first assessment was on self, which was followed by other Black/White women (same race as the respondent), which was followed by other White/Black women (of a differ- ent race). Essentially, in one version the questions began with perceived effect on self and extended to those who were most dissimilar. In the other version, the questions began with perceived effects on those who are most dis- similar and moved inward toward perceived effects on self. This was a conser- vative protocol, given that previous explorations of the impact of question order on third-person perceptions have found minimal differences (David & Johnson, 1998; Gunther, 1991, 1995; Price & Tewksbury, 1996; Tiedge, Silverblatt, Havice, & Rosenfeld, 1991).
279
David et al. • Body Image, Race, and Fashion Models
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
At the end of the experiment, before debriefing, students answered some media-use questions and provided personal data including race, height, and weight.
Results
Manipulation Check
Black students found Black models more attractive (M = 6.0, SD = 2.5) than White models (M = 3.9, SD = 2.4). They rated their body types to be more simi- lar (M = 3.3, SD = 2.4) to the body types of the Black models than to the body types of the White models (M = 2.1, SD = 2.3). Also, Black students rated the White models as thinner (M = 7.7, SD = 2.0) than the Black models (M = 6.7, SD = 1.8). White students, on the other hand, found Black models to be slightly more attractive (M = 6.9, SD = 2.3) than White models (M = 6.1, SD = 2.3). In line with the ratings provided by the Black students, White students also found the White models to be thinner (M = 8.0, SD = 1.4) than the Black models (M = 7.5, SD = 1.5). For similarity of body type, White students did not see any differences based on the race of the model. All significant differences were at p < .05 and were tested using paired t tests (see Table 1).
Although the significant differences in similarity and attractiveness by the race of the models for Black respondents is indicative of the intended social identification manipulation, there was a reversal among White stu- dents, who found Black models more attractive and saw no difference in simi- larity based on the race of the model. However, the finding that both Blacks and Whites found the White models to be thinner than Black models does raise the question of the validity of our manipulation. Despite our finding of null difference in the pretest, in which the set of ads with White models and
280
COMMUNICATION RESEARCH • June 2002
Table 1 Mean Ratings of Attractiveness, Thinness, and Similarity to Fashion Models by Race of the Students
Black Respondents White Respondents
Black White Black White Models Models Models Models
M SD M SD M SD M SD
Attractiveness 6.0 2.5 3.9 2.4 6.9 2.3 6.1 2.3 Thinness 6.7 1.8 7.7 2.0 7.5 1.5 8.0 1.4 Similarity 3.3 2.4 2.1 2.3 4.0 2.3 3.7 2.3
Note. Responses were measured on a 10-point scale. Higher scores mean higher attractiveness, thinness, and similarity.
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
the set with Black models were administered between subjects, some signifi- cant differences for thinness were observed in this study. However, because the focus of our study is on relative differences between assessments rather than the absolute value of the perceived media effect, we continued with the analysis. Also, we have tried to control for this difference by using the thin- ness evaluation as a covariate in the analysis of variance.
Next, we compared the physique anxiety and self-esteem scores by the race of the respondent. Reliabilities were high for both scales, with Cronbach’s alpha = .93 for physique anxiety and Cronbach’s alpha = .84 for self-esteem. The physique-anxiety score and self-esteem score for Black par- ticipants were not significantly different from the corresponding scores for White participants. These results suggest that both groups were comparable in terms of the their physique anxiety and self-esteem scores.
Analysis of Variance
Given the complexity of the design, each of the body-image factors, namely ideal body weight, self-esteem, and eating disorders, were analyzed sepa- rately. We used a 2 (Black respondent, White respondent) × 2 (Order: per- ceived effect on self first, perceived effect on self last) × 2 (Black fashion mod- els, White fashion models) × 3 (perceived effect on self, perceived effect on others of a similar race, perceived effect on others of a different race) mixed- factor design. In addition, physique anxiety, self-esteem, thinness rating of Black models and thinness rating of White models were entered as covariates. In preliminary analysis, when self-esteem was entered as a covariate, it was not statistically significant. In subsequent analysis, self-esteem was not included as a covariate. A summary of the means is presented in Table 2.
For the ideal-body-weight outcome, main effects for physique anxiety, F(1, 73) = 9.98, MSE = 11.43, p < .005, ω2 = .06, and thinness rating of Black mod- els, F(1, 73) = 10.26, MSE = 11.43, p <.005, ω2 = .06, were significant. In addi- tion, the expected three-way interaction, Race of Model × Comparison Group × Race of Respondent, F(2, 146) = 30.16, MSE = 3.01, p < .001, ω2 = .09, was significant.
For the self-esteem outcome, main effects for physique anxiety, F(1, 73) = 7.33, MSE = 9.56, p < .01, ω2 = .03, and thinness rating of Black models, F(1, 73) = 10.72, MSE = 9.56, p <.01, ω2 = .05, were significant. The critical three-way interaction captured by Race of Model × Comparison Group × Race of Respondent, F(2, 146) = 22.63, MSE = 3.41, p < .001, ω2 = .08, was also significant.
For the eating-disorder outcome, the main effect for physique anxiety, F(1, 73) = 4.87, MSE = 13.59, p < .05, ω2 = .03, and the thinness rating of Black
281
David et al. • Body Image, Race, and Fashion Models
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
models, F(1, 73) = 4.11, MSE = 13.59, p < .05, ω2 = .03, were significant. The targeted three-way interaction among Race of Model × Comparison Group × Race of Respondent, F(2, 146) = 22.28, MSE = 2.25, p < .001, ω2 = .05, was significant.
In within-subjects designs, multiple observations are obtained from the same subject and the independence assumption is violated, which in turn inflates the F value. The Greenhouse-Geisser epsilon and the Huynh-Feldt epsilon are corrections to the F value to account for the violations of com- pound symmetry and sphericity. After applying these corrections, the critical three-way interaction was significant at p < .001.
In summary, physique anxiety, which was entered as a covariate, had a sig- nificant effect on the ratings provided by the respondents. Also, the thinness rating of the Black models was significant. The order in which the third- person ratings were obtained was not significant. The key finding was the expected three-way interaction—Race of the Model × Comparison Group × Race of Respondent—which was significant for each of the three body-image factors after controlling for the difference in perceived thinness between Black and White models. To explain this interaction, we examined the rela- tionships among these three factors through a series of mean comparisons.
282
COMMUNICATION RESEARCH • June 2002
Table 2 Perceived Effects of Advertising on Self and Others as a Function of the Race of the Respondent and the Race of the Fashion Model
Black Respondents White Respondents
Black White Black White Models Models Models Models
M SD M SD M SD M SD
Body Weight Self 5.5 3.1 3.6 3.0 5.8 3.1 5.9 2.4 Other Blacks 6.1 2.5 3.7 2.4 6.3 2.1 5.0 1.9 Other Whites 4.3 3.4 9.1 1.8 6.7 2.0 7.5 1.5
Self-Esteem Self 3.9 2.7 2.5 2.5 4.5 2.7 4.5 2.5 Other Blacks 5.3 2.1 3.3 2.4 5.8 2.3 4.8 2.0 Other Whites 3.9 3.5 8.7 1.3 5.9 2.1 6.8 2.0
Eating Disorder Self 2.1 3.0 1.6 2.7 2.2 2.6 2.5 2.7 Other Blacks 3.5 2.6 2.8 2.2 5.2 2.1 4.0 2.4 Other Whites 4.2 3.5 8.6 1.6 5.4 2.1 6.2 2.1
Note. Means are ratings on a 11-point scale from 0 (no influence) to 10 (considerable influence).
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
Third-Person Effects When Race of Model Matched Race of the Respondent
First, perceived effect on others of the same race was significantly greater than perceived effect on self when there was a match between the race of the model and the race of the respondents. Among Whites, the perceived effect on self and on other White respondents was significantly different for the body image outcome (Self M = 5.9, Other Whites M = 7.5, Mean difference = 1.6, t(39) = 4.69, p < .001), the self-esteem outcome (Self M = 4.5, Other Whites M = 6.8, Mean difference = 2.3, t(39) = 7.58, p < .001), and the eating-disorder out- come (Self M = 2.5, Other Whites M = 6.2, Mean difference = 3.7, t(39) = 8.82, p < .001). Also, as the undesirability of the outcome increased, the third- person gap widened significantly among the White respondents (see Figure 1).
Among Blacks, the third-person effect was significant for only two of the three outcomes, and there was no systematic widening of the gap with increasing undesirability. The perceived effect on self and other Black stu- dents was not significant for the body-weight outcome (Self M = 5.5, Other Blacks M = 6.1, Mean difference = 0.6), but was significant for the self-esteem outcome (Self M = 3.9, Other Blacks M = 5.3, Mean difference = 1.4, t(39) = 2.90, p <. 005) and the eating-disorder outcome (Self M = 2.1, Other Blacks M = 3.5, Mean difference = 1.4, t(39) = 3.15, p < .005).
Baseline Effects on Self Among Blacks and Whites
When the race of the model matched the race of the respondent, there were no statistically significant differences in perceived effect on self between Black and White respondents for all three outcomes, namely perceived effect on ideal body weight, effect on self-esteem, and likelihood to lead to an eating
283
David et al. • Body Image, Race, and Fashion Models
Figure 1. Perceived Effect of Same-Race Fashion Models on Self and Other Women of the Same Race for Three Body-Image Factors
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
disorder. The absence of a significant difference between Blacks and Whites on perceived effect on self runs contrary to the conventional wisdom that Black women consider themselves less vulnerable to body-image media effects than White women. The implications of this finding will be taken up in the discussion section.
Race of the Model and Perceived Effect on Self
For the Black respondents, perceived effect on self was significantly different when the race of the model was different, suggesting strong social identifica- tion. When exposed to Black models, Blacks showed significantly higher effect on self (body weight M = 5.5, self-esteem M = 3.9, eating disorder M = 2.1) than when exposed to White models (body weight M = 3.6, self-esteem M = 2.5, eating disorder M = 1.6). All paired t tests were significantly different at p < .05. In contrast, among White respondents there was no significant effect of the race of the model on the perceived effect on self. Possible reasons for the absence of social identification among the White participants are addressed in the discussion.
Race of the Model and Perceived Effect on Others of the Same Race
The purpose of this line of analysis was to examine how participants project social identification onto others of their race. Whereas the Black respondents continued to indicate strong social identification for other Black women (their in-group), the White respondents also indicated strong social identifi- cation for their in-group, which was a reversal of the pattern that Whites pro- vided for themselves. Black respondents stated that other Black women would be affected more when the race of the model was Black (body weight M = 6.1, self-esteem M = 5.3, eating disorder M = 3.5) than when the race of the model was White (body weight M = 3.7, self-esteem M = 3.3, eating disor- der M = 2.8). By the same token, Whites indicated that other Whites would be influenced more by White models (body weight M = 7.5, self-esteem M = 6.8, eating disorder M = 6.2) than they would by Black models (body weight M = 6.7, self-esteem M = 5.9, eating disorder M = 5.4). All tests were significant at p < .05.
Race of the Model and Perceived Effect on Others of a Different Race
Finally, we examined the perceived effect on other students of a different race (out-group) as a function of the race of the model. When projecting to other
284
COMMUNICATION RESEARCH • June 2002
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
students of a different race, both Blacks and Whites rated the out-groups as showing strong social identification. However, there was a marked asymme- try in terms of how Blacks and Whites saw social identification by the out- group. Black participants’ estimation of social identification by the White students was significantly higher than the White participants’ estimation of social identification by other Whites and Blacks.
Among Black participants, perceived effects on other White students after exposure to White models were notably higher (body weight M = 9.1, self- esteem M = 8.7, eating disorder M = 8.6) compared to perceived effects after exposure to Black models (body weight M = 4.3, self-esteem M = 3.9, eating disorder M = 4.2). Among White respondents, although a similar pattern was evident, the size of the projected effects of social identification was smaller, with perceived effects on other Black students after exposure to Black models (body weight M = 6.3, self-esteem M = 5.8, eating disorder M = 5.2) being higher than effects estimated after exposure to White models (body weight M = 5.0, self-esteem M = 4.8, eating-disorder M = 4.0). All relevant within group mean differences were tested using paired t tests and were significant at p < .01 (see Figure 2).
In the discussion, the results are examined within the framework of social-identification theory and the cultural and racial stereotypes associ- ated with body-image factors.
Discussion
The purpose of this study was to examine third-person perceptions related to body-image factors by manipulating two key variables on the basis of race: social distance between the respondent and the comparison group, and social distance between the respondent and the media source. In addition, we tried to replicate the results reported in a study by David and Johnson (1998), wherein the third-person effect widened with the undesirability of the body-image outcome. To study these issues, we examined five hypotheses and some research questions. The discussion is grouped into three sections. We will start with the findings related to the third-person effect. Then we will take up the results associated with social distance before discussing the interaction between social distance and race.
Third-Person Effects
To examine the basic third-person effect within this quasi-experimental design, the perceived effect on others was examined without any confounding in terms of the race of the model or the race of the comparison group. In other
285
David et al. • Body Image, Race, and Fashion Models
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
words, the third-person results discussed in this section are based on per- ceived effect of media on self and others of the same race following exposure to fashion models also of the same race as the respondent.
In support of Hypothesis 1, there was strong evidence of third-person effects across the board, which indicates that respondents perceived media effects on others to be stronger than effect on self. As predicted in Hypo- the- sis 2, the third-person gap widened with increasing undesirability of the out- come. This trend was particularly strong among White respondents but was not as pronounced among Black respondents. Among Whites, the third- per- son gap increased from 1.57 for body weight to 2.33 for self-esteem to 3.67 for eating disorders. Among Blacks, the gap increased from 0.68 for body weight to 1.35 for self-esteem to 1.43 for eating disorders. As discussed later, the dif- ference in magnitudes also suggests that Black and White students’ beliefs about the power of media differ.
The dynamics of the widening of the third-person gap with outcome undesirability was similar to the findings reported by David and Johnson (1998). As the undesirability of the outcome increased, the White partici- pants trimmed down the effect on self but did not make comparable adjust- ments for others. Another way to interpret the perceived differential effects between self and others is that while participants made adjustments for the likelihood of occurrence on self for the more serious outcomes, which have a lower base rate of occurrence, they failed to make similar adjustments when extrapolating to others.
One explanation for the lack of clear support for Hypothesis 2 among Blacks could be due to the differences in the interpretation of undesirability of the outcomes. First, based on previous findings, there is evidence that ideal
286
COMMUNICATION RESEARCH • June 2002
Figure 2. Perceived Effect on Self and Others as a Function of the Race of the Respon- dent, Race of the Fashion Models, and Race of the Comparison Group for the Self-Esteem Outcome
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
body weight is not treated with as much emphasis among Black women as it is among White women. Hence, the absence of a third-person effect for body weight among Blacks is not surprising. Furthermore, the absence of the wid- ening of the third-person gap with undesirability can be attributed to the low incidence of clinical eating disorders among Blacks compared to Whites.
Social Distance and Third-Person Effect
When social distance was narrowly defined based on race, the traditionally robust effects of social distance were nullified or reversed. It appears that the participants’ perceptions of in-group and out-group norms took precedence over normative social distance defined along racial lines.
Race and the Perceptions of Body-Image Media Effects on Self
Conventional wisdom, cultural differences, and some findings in the litera- ture (e.g., Akan & Grilo, 1995; Altabe, 1998) suggest that Black women are less vulnerable to body-image effects compared to Whites. Although this study does not provide irrefutable evidence to the contrary, the findings from this study suggest that there were no statistically significant differences between Blacks and Whites in terms of perceived media effects when the race of the model matched the race of the respondent.
The absence of a significant difference between Blacks and Whites on per- ceived effects on self deserves attention. Social commentators and media observers have noted that as Black women make progress in their careers and achieve higher status, they feel the same pressure that Whites do to con- form to the ideal body image that dominates the overall culture. It is also pos- sible that the effect of media images on self was heightened for the Black par- ticipants in this study because we matched the race of the media model with the race of the respondent. It is plausible that even if Blacks can generally ignore the ideals of body image presented in the mainstream media predomi- nantly through White models, they could be vulnerable to body-image pres- sures presented in the Black media in the form of Black fashion models. This highlights the need for social responsibility among ethnic media as well as general-market media in body-image portrayals.
Social Identification Hypothesis
We had predicted in Hypothesis 3 that the perceived effect on self would be greater when the race of the model matched the race of the respondent. This was true for the Black participants but not for Whites. Across all three
287
David et al. • Body Image, Race, and Fashion Models
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
outcomes, among White participants there were no statistically significant differences on perceived effect on self on the basis of the race of the model.
There are at least three explanations for the absence of a main effect for race of the model on White respondents. One is the social desirability bias not to appear racist. This supports the motivational explanation of third-person effect, because not being racist is an esteem-preserver or enhancer. The other interpretation is that the fashion industry selects models that fit common standards of thinness and attractiveness and the White students did not see a significant difference based on the race of the model. Studies show, for instance, that Black models in advertisements tend to be light-skinned (Entman & Book, 2000). Nonetheless, White students perceived the race of the model to have a significant effect on other White women and other Black women. In short, although White respondents’ own ratings suggest that they saw themselves as color blind, they did not project the same pattern to other in-group (other Whites) and out-group (other Blacks) women.
A third possibility is that Blacks in the entertainment industry—includ- ing fashion—may be more accepted than in other social roles. Even if White students perceived Black models as different from them, the fashion industry is a safe milieu in which to be different. White college students may have per- ceived racial difference in fashion models as exotic—an ideal in the fashion industry. Merelman (1995) would argue that this is an example of the success of Black cultural projection among Whites, who are accustomed to seeing Blacks in glamorous roles in the entertainment industry. Our finding that White students found Black models to be more attractive is one indicator of this.
Unlike White respondents, Black respondents showed strong in-group social identification, which is expected among ethnic or minority groups in a pluralistic culture. The differences in social identification between Blacks and Whites led to other interesting effects when respondents were asked to project the perceived effect of race of the model to other Whites and Blacks.
Social Identification and Projecting to Others
Using the race of the fashion model as a cue for social identification, we tested the validity of hypotheses presented in Hypothesis 4 and Hypothesis 5. The findings related to Hypothesis 4 and Hypothesis 5 varied considerably based on the race of the model and the race of the respondent, indicating strong interactions with social distance and social identification.
Among Blacks, perceived effects on other White women (out-group) was greater than perceived effects on other Black women (in-group) only after exposure to the White fashion models. After exposure to Black fashion
288
COMMUNICATION RESEARCH • June 2002
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
models, Blacks rated perceived effects on other Black women (in-group) higher than perceived effects on other White women (out-group). This offers only partial support for Hypothesis 4 because the social distance corollary was held only for Black fashion models.
Among Whites, perceived negative effect of media on other White women (in-group) was greater than perceived effect on other Black women (out- group) when the race of model was White. Essentially, this was a reversal of the social-distance corollary, which was predicted in Hypothesis 5. But when the race of the model was Black, there were no statistically significant differ- ences in terms of how the White respondents perceived effects on in-group and out-group.
In summary, when projecting to others, both Blacks and Whites perceived significant social identification by other women. This effect was significant for others of the same race and others of a different race. When projecting media effect on others, the key difference between Black and White respon- dents was the magnitude of the effect. This mismatch is best captured by the ratings provided by Blacks for the Whites (out-group), which ranged from 8.6 to 9 on a 10-point scale, compared to ratings of 6.2 to 7.5 provided by Whites for their in-group.
In this article we have focused primarily on social identification as the mechanism to explain the perceived differences associated with the race of the model. An alternate explanation for the perceived immunity that African Americans expressed toward White fashion models might be rooted in their suspicions of the general-market media in the United States, where histori- cally they are accustomed to negative portrayals (e.g., Greenberg & Brand, 1994; Wilson & Gutiérrez, 1995). Finally, one cannot ignore the role of racial and cultural differences when interpreting these results. The conventional belief that Blacks are less vulnerable than Whites to body image factors is a powerful heuristic, which certainly could have played an important role in these judgments about others.
Directions for Future Research
Black students’ manifestation of a stronger social identity may be explained by the more collectivist leanings of minority cultures. In addition, because social comparisons between groups produce pressures to achieve collective self-esteem (e.g., Hogg & Mullin, 1999), African-American respondents may have exhibited a greater desire to preserve Black group self-esteem. This is a topic that deserves further scrutiny. One of the other areas of social-identity theory not previously discussed is the natural tension between an individ- ual’s desire to assimilate and to differentiate (e.g., Brewer & Pickett, 1999).
289
David et al. • Body Image, Race, and Fashion Models
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
At one level, the basic third-person perception is an example of differentia- tion, whereas the strong in-group versus out-group pattern exhibited by Blacks could be an indicator of assimilation. In the future, it would be benefi- cial to examine the differences within and between groups in terms of how they relate to differentiation and affiliation.
The results point to a need to explore the interplay between social identity and social-desirability aspects of the third-person. Social psychologists Tajfel and Turner (1979) claimed that definition of the “other” stems from the need to maintain a positive social identity. How Black and White college students conceive their own identities and introduce “otherness” into their vision of media effects on themselves and others is an area to pursue. In addition, how cultural projections via mass media might mediate these identities and ver- sions of otherness is a compelling area of study.
One of the limitations of our study is the thinness manipulation because there was a significant difference in the perceived thinness of the Black and White models. Although we tried to address this by introducing thinness rat- ings of the models as covariates in the analysis, it is possible that this differ- ence could have skewed the results. Clearly this study should be replicated with stimuli that are more rigorously balanced on all the critical attributes. Finally, this study is a quasi-experiment with only 80 participants, which has the typical limitations of any experiment. To improve ecological validity, it would be useful to replicate the study with a larger survey sample.
Conclusions
The findings from this study confirm some of the popular stereotypes about perceived effects of glamour advertising on body-image factors in Black and White college-age women. There were a number of significant findings. Both Blacks and Whites exhibited strong third-person effects when there was no confounding by race. Also, the third-person gap widened with increasing undesirability of the outcome, particularly among Whites. When projecting media effects on others, both Blacks and Whites indicated that others would identify more closely with same race or in-group models than with different race or out-group models.
When estimating perceived effects on self, however, Blacks and Whites dif- fered slightly. Whereas Whites reported that the race of the model does not make a significant difference on perceived effect on self, Blacks identified closely with Black models. Moreover, contrary to the widely held perception that Blacks are less vulnerable to body-image effects compared to Whites, there were no significant differences between the two groups on perceived effect on self when the race of the models matched the race of the
290
COMMUNICATION RESEARCH • June 2002
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
respondents. In the face of corroborating evidence (e.g., Botta, 2000), it seems plausible that the gap between Blacks and Whites on body-image vulnerabil- ities might be closing.
We hope the findings from this study add to the discussion on the role of advertising in creating norms of ideal beauty. Obviously, more systematic research programs are needed to accurately assess negative body-image con- sequences that could result from the idealized beauty presented in the media.
References
Akan, G. E., & Grilo, C. M. (1995). Sociocultural influences on eating attitudes and behaviors, body image, and psychological functioning: A comparison of African American, Asian American, and Caucasian college women. International Journal of Eating Disorders, 18(2), 181-187.
Altabe, M. (1998). Ethnicity and body image: Quantitative and qualitative analysis. International Journal of Eating Disorders, 23(2), 153-159.
Botta, R. A. (1999). Television images and adolescent girls’ body image distur- bance. Journal of Communication, 49(2), 22-41.
Botta, R. A. (2000). The mirror of television: A comparison of black and white adolescents’ body image. Journal of Communication, 5(3), 144-159.
Brewer, M. B., & Pickett, C. L. (1999). Distinctiveness motives as a source of the social self. In T. R. Tyler, R. M. Kramer, & O. P. John (Eds.), The psychol- ogy of the social self (pp. 71-87). Mahwah, NJ: Erlbaum.
Brosius, H., & Engel, D. (1996). The causes of third-person effects: Unrealistic optimism, impersonal impact, or generalized negative attitudes toward media influence? International Journal of Public Opinion Research, 8(2), 142-162.
Cohen, J., Mutz, D., Price, V., & Gunther, A. (1988). Perceived impact of defa- mation: An experiment on third-person effects. Public Opinion Quarterly, 52(2), 161-173.
David, P., & Johnson, M.A. (1998). The role of self in third-person effects about body image. Journal of Communication, 48(4), 37-58.
Davison, W. P. (1983). The third-person effect in communication. Public Opin- ion Quarterly, 47(1), 1-15.
Duck, J. M., & Mullin, B. (1995). The perceived impact of the mass media: Reconsidering the third person effect. European Journal of Social Psy- chology, 25, 77-93.
Duck, J. M., Terry, D. J., & Hogg, M. A. (1995). The perceived influence of AIDS advertising: Third-person effects in the context of positive media content. Basic and Applied Social Psychology, 17(3), 305-325.
Entman, R. M., & Book, C. (2000). Light makes right: Skin color and racial hierarchy in television advertising. In R. Andersen & L. Strate (Eds.), Crit- ical studies in media commercialism (p. 214-224). New York: Oxford.
Ellis, K. J., Abrams, S. A., & Wong, W. W. (1997). Body composition of a young, multiethnic female population. American Journal of Clinical Nutrition, 65(3), 724-731.
291
David et al. • Body Image, Race, and Fashion Models
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
Eveland, W. P., Jr., Nathanson, A. I., Detenber, B. H., & McLeod, D. M. (1999). Rethinking the social distance corollary: Perceived likelihood of exposure and the third-person perception. Communication Research, 26(3), 275-302.
Field, A. E., Colditz, G. A., & Peterson, K. E. (1997). Racial/ethnic and gender differences in concern with weight and in bulimic behaviors among adoles- cents. Obesity Research, 5(5), 447-454.
Fitzgibbon, M. L., Spring, B., Avellone, M.E., Blackman, L.R., Pingitore, R., Stolley, M.R. (1998). Correlates of binge eating in Hispanic, Black, and White women. International Journal of Eating Disorders, 24(1), 43-52.
French, S. A., Story, M., Neumark-Sztainer, D., Downes, B., Resnick, M., & Blum, R. (1997). Ethnic differences in psychosocial and health behavior correlates of dieting, purging, and binge eating in a population-based sam- ple of adolescent females. International Journal of Eating Disorders, 22(3), 315-322.
Gray, J. J., Ford, K., & Kelly, L. M. (1987). The prevalence of bulimia in a Black college population. International Journal of Eating Disorders, 6, 733-740.
Greenberg, B. S., & Brand, J. E. (1994). Minorities and the mass media: 1970s to 1990s. In J. Bryant & D. Zillman (Eds.), Media effects: Advances in the- ory and research (pp. 273-314). Norwood, NJ: Erlbaum.
Gunther, A. C. (1991). What we think others think: Cause and consequences in the third-person effect. Communication Research, 18(3), 355-372.
Gunther, A. C. (1995). Overrating the X-rating: The third-person perception and support for censorship of pornography. Journal of Communication, 45(1), 27-38.
Gunther, A. C., & Hwa, A. P. (1996). Public perceptions of television influence and opinions about censorship in Singapore. International Journal of Pub- lic Opinion Research, 8(3), 248-265.
Harris, S. M. (1995). Family, self, and sociocultural contributions to body-image attitudes of African American women. Psychology of Women Quarterly, 19, 129-145.
Harrison, K. (2000). The body electric: Thin-ideal media and eating disorders in adolescents. Journal of Communication, 50(3) 119-143.
Harrison, K., & Cantor, J. (1997). The relationship between media consump- tion and eating disorders. Journal of Communication, 47(1), 40-67.
Hart, E. A., Leary, M. R., & Rajeski, W. J. (1989). The measurement of social physique anxiety. Journal of Sport Psychology, 11, 94-104.
Hogg, M. A., & Abrams, D. (1999). Social identity and social cognition: Histori- cal background and current trends. In D. Abrams & M. A. Hogg (Eds.), Social identity and social cognition (pp. 1-25). Oxford, UK: Blackwell.
Hogg, M. A., & Mullin, B. A. (1999). Joining groups to reduce uncertainty: Subjective uncertainty reduction and group identification. In D. Abrams & M. A. Hogg (Eds.), Social identity and social cognition (pp. 248-279). Oxford, UK: Blackwell.
Hraba, J., Radloff, T., & Ray-Gray, P. (1999). A comparison of black and white social distance. Journal of Social Psychology, 139(4), 536-539.
Kann, L., & Ross, J. G. (1996). Youth risk behavior surveillance—United States, 1995. Centers for Disease Control. Retrieved from http://www.cdc. gov/epo/mmwr/preview/mmwwrnum/00043812.num
292
COMMUNICATION RESEARCH • June 2002
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
Liebkind, K. (1992). Ethnic identity—Challenging the boundaries of social psychology. In G. M. Breakwell (Ed.), Social psychology of identity and the self concept (pp. 147-185). London: Surrey University Press.
McLeod, D. M., Eveland, W. P., Jr., & Nathanson, A. I. (1997). Support for cen- sorship of violent and misogynic rap lyrics: An analysis of the third-person effect. Communication Research, 24(2), 153-174.
Merelman, R. M. (1995). Representing Black culture: Racial conflict and cul- tural politics in the United States. New York: Routledge.
Morales, J. F., López-Sáez, M., & Vega, L. (1998). Discrimination and beliefs on discrimination in individualists and collectivists. In S. Worchel, J. F. Morales, D. Páez, & J. Deschamps (Eds.), Social identity: International per- spectives (pp. 199-210). London: Sage.
Myers, P. N., & Biocca, F. (1992). The elastic body image: The effect of televi- sion advertising and programming on body image distortions in young women. Journal of Communication, 42(3), 108-133.
Pelham, B. W., & Hetts, J. J. (1999). Implicit and explicit personal and social identity: Toward a more complete understanding of the social self. In T. R. Tyler, R. M. Kramer, & O. P. John (Eds.), The psychology of the social self (pp. 115-143). Mahwah, NJ: Erlbaum.
Perloff, R. M. (1989). Ego involvement and the third person effect of televised news coverage. Communication Research, 16(2), 236-262.
Perloff, R. M. (1993). Third-person effect research 1983-1992: A review and synthesis. International Journal of Public Opinion Research, 5(2), 167-184.
Price, V., & Tewksbury, D. (1996). Measuring the third-person effect of news: The impact of question order, contrast and knowledge. International Jour- nal of Public Opinion Research, 8(2), 120-141.
Richins, M. L. (1991). Social comparison and the idealized images of advertis- ing. Journal of Consumer Research, 18(1), 71-83.
Rojas, H., Shah, D. V., & Faber, R. J. (1996). For the good of others: Censorship and the third-person effect. International Journal of Public Opinion Research, 8(2), 163-186.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.
Rucinski, D., & Salmon, C. T. (1990). The “other” as the vulnerable voter: A study of the third-person effect in the 1988 U.S. presidential campaign. International Journal of Public Opinion Research, 2(4), 345-368.
Salwen, M. B. (1998). Perceptions of media influence and support for censor- ship: Third-person effect in the 1996 presidential election. Communica- tion Research, 25(3), 259-285.
Salwen, M. B., & Driscoll, P. D. (1997). Consequences of third-person percep- tion in support of press restrictions in the O. J. Simpson trial. Journal of Communication, 47(2), 60-78.
Shah, D. V., Faber, R. J., & Youn, S. (1999). Susceptibility and severity: Percep- tual dimensions underlying the third-person effect. Communication Research, 26(2), 240-267.
Serino, C. (1998). The personal-social interplay: Social-cognitive prospects on identity and self-others comparison. In S. Worchel, J. F. Morales, D. Páez, &
293
David et al. • Body Image, Race, and Fashion Models
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
J. Deschamps, Social identity: International perspectives (pp. 24-43). Lon- don: Sage.
Silber, T. (1986). Anorexia nervosa in Blacks and Hispanics. International Journal of Eating Disorders, 5, 121-128.
Stice, E., Schupak-Neuberg, E., Shaw, H. E., & Stein, R. I. (1994). Relation of media exposure to eating disorder symptomatology: An examination of mediating mechanisms. Journal of Abnormal Psychology, 103(4), 836-840.
Stormer, S. M., & Thompson, J. K. (1996). Explanations of body image distur- bance: A test of maturational status, negative verbal commentary, social comparison, and sociocultural hypotheses. International Journal of Eating Disorders, 19(2), 193-202.
Story, M., French, S. A., Resnick, M. D., & Blum, R. W. (1995). Ethnic/racial and socioeconomic differences in dieting behaviors and body image per- ceptions in adolescents. International Journal of Eating Disorders, 18(2), 173-179.
Striegel-Moore, R. H., Schreiber, G. B., Pike, K. M., Wilfley, D. E., & Rodin, J. (1995). Drive for thinness in Black and White preadolescent girls. Interna- tional Journal of Eating Disorders, 18(1), 59-69.
Tajfel, H. (1978). Differentiation between social groups: Studies in the social psychology of intergroup relations. London: Academic Press.
Tajfel, H. (1981). Human groups and social categories: Studies in social psy- chology. Cambridge, UK: Cambridge University Press.
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33-47). Monterey, CA: Brooks-Cole.
Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W. G. Austin (Eds.), Psychology of intergroup relations (2nd ed., pp. 7-24). Chicago: Nelson Hall.
Thompson, J. K. (1995). Assessment of body image. In D. B. Allison (Ed.), Handbook of assessment methods for eating behaviors and weight-related problems: Measures, theory, and research (pp. 119-144). Thousand Oaks, CA: Sage.
Tiedge, J. T., Silverblatt, A., Havice, M. J., & Rosenfeld, R. (1991). Discrepancy between perceived first-person and perceived third-person mass media effects. Journalism Quarterly, 68(1/2), 141-154.
Turner, J. C., & Onorato, R. S. (1999). Social identity, personality, and the self-concept: A self-categorization perspective. In T. R. Tyler, R. M. Kramer, & O. P. John (Eds.), The psychology of the social self (pp. 11-46). Mahwah, NJ: Erlbaum.
Tyler, T. R., & Cook, F. L. (1984). The mass media and judgments of risk: Dis- tinguishing impact on personal and societal level judgments. Journal of Personality and Social Psychology, 47(4), 693-708.
White, W. C., Hudson, L., & Campbell, S. (1985). Bulimarexia and Black women: A brief report. Psychotherapy, 22, 449-450.
Wilson, C. C., II, & Gutiérrez, F. (1995). Race, multiculturalism and the media: From mass to class communication. Thousand Oaks, CA: Sage.
294
COMMUNICATION RESEARCH • June 2002
at UNIV OF NC PEMBROKE on September 19, 2016crx.sagepub.comDownloaded from
