(Sociology Of Self)Read An Article And Answer 3 Big Questions

Social Comparisons, Reflected Appraisals, and Mass Media: The Impact of Pervasive Beauty Images on Black and White Girls' Self-Concepts Author(s): Melissa A. Milkie Source: Social Psychology Quarterly, Vol. 62, No. 2, Special Issue: Qualitative Contributions to Social Psychology (Jun., 1999), pp. 190-210 Published by: American Sociological Association Stable URL: http://www.jstor.org/stable/2695857 . Accessed: 03/01/2015 17:08

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

.

American Sociological Association is collaborating with JSTOR to digitize, preserve and extend access to Social Psychology Quarterly.

http://www.jstor.org

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

Social Psychology Quarterly 1999, Vol. 62, No. 2,190-210

Social Comparisons, Reflected Appraisals, and Mass Media: The Impact of Pervasive Beauty Images

on Black and White Girls' Self-Concepts* MELISSA A. MILKIE University of Maryland

Content analyses and experimental studies often indicate strong, usually negative effects of media on the self In contrast, qualitative work suggests that individuals may exercise considerable influence in selecting, interpreting and criticizing media content. This literature, however, does not adequately consider or specify how "interpreted" media content still might affect self-concept negatively. Incorporating social compari- son and reflected appraisal processes, this study shows how media affect self-esteem indirectly, despite criticism, through beliefs about how others use and are affected by media. In-depth interviews with 60 white and minority girls, complemented by quanti- tative measures from a larger study, help to clarify how girls are affected by promi- nent images of females pervasive in media. Most girls see the images as unrealistic; many prefer to see "real" girls. White girls, despite their criticism, are still harmed by the images because they believe that others find the images important and that others in the local culture, especially boys, evaluate them on the basis of these images. Minority girls do not identify with "white" media images, nor believe that significant others are affected by them; thus their critical interpretations succeed in thwarting negative feelings. The study increases our understanding of media effects on the self- concept and suggests that researchers consider how media images may be part of social comparison and reflected appraisal processes.

Questions of whether and how media influence self-concept-both self-identities and self-evaluations-as well as their impact on beliefs, values and behaviors underlies much media research. Content analyses often show how media content departs from reality in regard to who is represented, and in how groups and situations are portrayed. Experimental studies show how content involving violence, sexual violence, or stereotypes affects self-concept, attitudes, and behavior. Yet these quantitative "strong effects" approaches have been criticized for presuming that negative messages harm a

*Direct correspondence to Melissa A. Milkie, Department of Sociology, University of Maryland, 2112 Art-Sociology Building, College Park, MD 20742 (e-mail [email protected]). This research was partially supported by NIMH grant PHS T32 MH 14588-17. I thank Christopher Botsko, Donna Eder, Semya Hakim, Bernice Pescosolido, and Sheldon Stryker for invaluable comments on earlier versions of this paper. Thanks also to the Indiana University Social Psychology Group and to three anonymous reviewers for their insightful sugges- tions.

target group, even though they never direct- ly assess this effect (in the case of content analyses), or isolate content from how peo- ple experience and interpret media in every- day life (in the case of experimental stud- ies).

Qualitative approaches to the study of media have taken these criticisms seriously and have advanced several key conceptual and methodological issues indicated decades ago by prominent scholars (Blumer 1969; Katz and Lazarsfeld 1955; Lazarsfeld, Berelson and Gaudet 1948; Mills 1963). First, researchers have shown how the social context of media use is crucial, particularly in that significant others are relevant to the way people interpret and are affected by media. Second, they have focused on peo- ple's power to select and be critical of media content, and thus to discount media mes- sages. Although recent qualitative work has increased our understanding of media processes by revealing the complexities of people's understandings of images, it has done so to the detriment of assessing how

190

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

MEDIA IMAGES AND SELF-CONCEPTS 191

media influence people. Interpretive researchers view people as powerful in rela- tion to media content, but do not examine this assumption directly.

In this study, I provide a way to bridge perspectives which argue either that media content is powerful or that people are pow- erful in interpreting media. I do so by ana- lyzing the extent to which people's power to make critical assessments of content (for example, believing that stereotyped portray- als of one's group are unrealistic or unim- portant) may prevent that content from neg- atively affecting the self. I take the case of feminine beauty images in media, and assess the relative power of critical interpretations in countering harmful effects on the self- concept, specifically on self-esteem. To clari- fy how media can affect people indirectly, I draw on basic principles in social psychology which point to the key role of others in self processes, and discuss these in terms of some unique properties of mass media. Media influence on the self can be studied in more detail through this qualitative approach, which probes further into the complexities of interpretations in examining how they matter and how they may or may not be powerful within the local context.

MEDIA: INTERPRETATIONS, CRITICISMS, AND EFFECTS ON

INDIVIDUALS

With their power to frame, define, and neglect aspects of the social world, the mass media are a principal social and cultural institution. The central position of media in everyday life ensures that symbols dis- tributed through the media become points of focus and interaction in the population. Imagery is contested and criticized, however, by groups and individuals. Ethnic, religious, age, and gender groups struggle to influence society's values, myths, symbols and informa- tion through the media (Gans 1972); individ- uals ignore, devalue, and criticize media images (Lang and Lang 1981).

Media processes are studied through multiple methods. These include quantitative work such as content analyses (e.g., Binder 1993; Cantor 1987; Pierce 1990; Tuchman 1978) and, experimental studies (e.g.,

Davidson, Yasuma, and Tower 1979; Donnerstein and Linz 1989); which conclude that a stereotyped or negative content can affect and harm individuals. For example, researchers who have done content analyses of girls' magazines suggest that unrealistic beauty images and a focus on traditional femininity may damage girls' self-image (Evans et al. 1991; McRobbie 1982; Pierce 1990). Another approach, using qualitative methods, examines meanings that people create from symbolic products (Corner, 1991; Frazer 1987; Milkie 1994; Shively 1992) but does not usually examine effects of media. Implicit in these divergent foci is a tension between assumptions regarding rela- tive power: Quantitative analyses often sug- gest that media have power to influence people, whereas interpretive work stresses individuals' power to resist ideology-to select and be critical of media.

Interpretive Approaches: Individuals Making Meaning

Interpretive approaches to media research focus on the symbolic interactionist premise that meaning is created in interac- tion with others. Researchers taking these approaches imply that effects are contingent on the meanings created and that these meanings are located not in media messages but in the reception of those messages. Such work is not concerned with "what media do to people" but with "what people do with media" (see McLeod, Kosicki, and Pan 1991:250). Although these approaches do not necessarily argue that individual inter- pretations of media preclude media influ- ence on people, the processes by which such effects may occur is unclear (Corner 1991).

Qualitative researchers analyzing media processes have emphasized at least two issues. One is the notion that cultural mean- ings are not fixed in products, but vary according to cultural, historical, and social group context (Blumer 1969). Fish (1980) argues that readers create meanings, but do so on the basis of the "interpretive commu- nity" to which they belong. Because the members of such communities have the same purposes and goals, the meanings con- structed from the texts will be shared. Those

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

192 SOCIAL PSYCHOLOGY QUARTERLY

outside such an interpretive community will construct different meanings from the same cultural product. For example, Shively (1992) found that Anglos and Native Americans interpreted Western films differ- ently: Anglos saw them as authentic and as an important part of the history of the Old West, while Native Americans interpreted them as a message about freedom and did not regard them as authentic. Similarly, Hur and Robinson (1978) showed that African- Americans found the account of black histo- ry documented in Roots to be more accurate than did whites.

A second important issue in interpretive approaches to media focuses on the under- standing that individuals are not passive recipients of media, but can be critical of media content or can reject it. Years ago, Mills (1963) observed that people do not simply internalize dominant ideological mes- sages, but sometimes resist them. He argued that people can resist by comparing alterna- tive messages in different media, comparing media messages with their own life, or com- paring these messages with the beliefs and behaviors of people in their social networks. Through this process, people may denounce or reject media messages that contradict their own experiences.

At least two kinds of critiques people make of media content should be consid- ered. First, people are sensitive and critical toward the media reflection of reality, espe- cially the portrayal of their own reference groups. When groups in the media are absent or are portrayed unrealistically, audi- ence members who share characteristics with those persons (e.g., gender or ethnicity) may be dissatisfied. Inaccurate images may affect people in the sense that they alter the "true" social definition of the group in ques- tion. This "symbolic annihilation" especially affects disadvantaged groups; they have less control over the production of media myths than men and whites do, and must struggle to project a public self-definition that is more positive (Collins, 1991; Tuchman 1978, 1979; van Zoonen 1994).

The realism of media symbols that peo- ple are sensitive to is discussed in recent qualitative studies of audiences (Currie 1997; Press 1991; Seiter et al. 1989). Press

(1991) found that middle-class women were more aware of the contradictions between their lives and those of upper-class television characters; working-class women, paradoxi- cally, viewed the upper-class media charac- ters as more realistic.

Second, people may be critical in the sense of opposing media images ideological- ly (Fiske 1990; Hall 1980). Frazer (1987:416-17) challenges content analyses which imply powerful effects of the British girls' magazine Jackie by showing that girls in her focus groups were aware of the images' non-realism and spoke about the stories in ways suggesting resistance toward traditional messages about femininity.

Qualitative researchers such as Frazer call into question the claims that media sym- bols have powerful effects such as creating or enforcing traditional feminine ideology. They assume that media are relatively pow- erless in affecting people because individuals interpret, critique, and "resist" messages, but they do not directly examine this assump- tion. Thus a compelling question regarding interpretation of imagery remains un- answered: How does criticizing portrayals of one's reference group as unrealistic or stereotyped, or opposing ideological content matter? In particular, does criticism succeed in warding off the harmful effects indicated by quantitative work such as content analy- ses?

How Media Affect the Self: Incorporating Social Psychological Principles

A unique quality of media is their public pervasiveness and people's knowledge that the images or ideas they see are also seen by many others-often millions of others. In addition, individuals believe that others are more strongly affected by media portrayals than they themselves are (Davison 1983; Perloff 1993). This belief reflects either a misperception of how others view or are influenced by media or, as Davison (1983) points out, perhaps an underestimation of the media influence on the self. Evidence for this belief-known as the "third-person effect"- includes studies showing that peo- ple believe that other people's children are affected by commercials more strongly than

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

MEDIA IMAGES AND SELF-CONCEPTS 193

their own, and that others in their communi- ties are affected by political campaigns more strongly than they are (see Davison 1983; Perloff 1993).

How does a belief that media images are powerful for others matter for individuals? The third-person effect suggests that effects of media in which the content directly influ- ences the self, attitudes, or behavior may not be the only important kind of influence. A complex, indirect effect may also occur as people account for the effects of the perva- sive imagery in media on others in their social networks, and are themselves influ- enced by perceptions of the way others see the media-distorted world. Much of what we know and understand about others outside our community-who they are, what they value, indeed, what is happening in the broader society-is filtered through the dis- torted lens of the media; and for some, this information may even come to represent the society (Altheide and Snow 1988). Indeed, media may have become a significant part of the generalized other-that is, the "society" we know-whose views we take into account in understanding and evaluating our self. Because media effects may involve how we believe others see such images, self processes involving significant others- specifically social comparisons and reflected appraisals-must be considered.

Social comparisons affect how we learn about and see ourselves in relation to indi- viduals, groups, or social categories. Are we the same or different? Better or worse? Social comparison theories suggest that we tend to compare ourselves with similar oth- ers, though we have selectivity in making such comparisons; that is, relative freedom to select the referents by whom we evaluate ourselves (Rosenberg 1986; Singer 1981). Theoretically, given freedom of comparison, people could use selectivity to escape media images that they dislike or to which they compare negatively-by ignoring or dis- counting them and by not using such images as a basis for social comparison. Yet because of the pervasiveness of media, and the way in which people believe that media affect others, it may be difficult to avoid some social comparisons with media images and felt evaluations (reflected appraisals) based

on the media-depicted world. Media images may alter ideas of what is normative or ideal or of what one thinks others believe is nor- mative or ideal, while offering an additional pervasive standard of comparison that goes beyond local cultures. If people believe that others use such images to evaluate them, they cannot simply shift away from this con- straining comparative referent.

Reflected appraisals, or how people believe that others view them, can further explain the indirect impact of media. Generally, symbolic interactionists contend that actual appraisals by others affect reflected appraisals, which in turn affect the self (Felson 1985, 1989; Felson and Reed 1986; Ichiyama 1993). At least two areas of research on reflected appraisals are impor- tant for understanding media impact.

First, research shows that people cannot easily distinguish how a particular significant other views them. Instead, reflected appraisals appear to be a proxy for the gen- eralized other-the attitudes of an entire group (Felson 1989). Given the place of media in representing society's (generalized others') norms and ideals, it is crucial to con- sider how media relate to the self via reflect- ed appraisals.

Second, it is essential to consider the relevance and importance of reflected appraisals. Rosenberg (1986) cautions researchers to examine the conditions under which reflected appraisals in fact will affect the self. Ultimately those reflected appraisals must be important to the individ- ual in order to exert influence. For example, Rosenberg argues that a racist society does not simply cause a negative impact on black children's self-esteem because black chil- dren are not necessarily aware of others' (whites') views, may not agree with them, may not find them relevant to the self, or may not care about the opinions of those others. Similarly, people might be affected (indirectly) by media images only when they believe that those important to them are so affected.

The theoretical and empirical research on social comparisons and reflected appraisals is important in considering how media criticism could be deflated. Even if someone believes that her group is por-

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

194 SOCIAL PSYCHOLOGY QUARTERLY

trayed unrealistically and does not like the portrayal, she may not know or believe that others share her criticisms. In other words, individuals see themselves through the eyes of others who they assume have been affect- ed significantly by mass media imagery. It is clear that people ignore, dislike, and belittle media portrayals, and may not wish to make a social comparison that is negative for the self. Yet the extent to which such critical assessment of media is effective, or can negate effects, may depend on the extent to which individuals know that significant oth- ers have assessed the symbols critically in the same way (see Mills 1963). Alternatively, peers may be a means of validating critical assessment if the peer network also is critical of such portrayals and if the individual knows the views of that group accurately. In such a case, an individual can act on criti- cisms, or they can be meaningful in protect- ing her self-evaluations, because she knows that the network or group devalues those images as well.

ANALYTIC STRATEGY AND RESEARCH QUESTIONS

To develop a comprehensive and tex- tured understanding of the relationship between media imagery, interpretations, and effects on the self, I use a largely qualitative approach. In this study, I move beyond pre- vious work by allowing young women's voic- es to provide insight into the nuances in interpretations and criticisms of media images, including assessments of how they believe others use the imagery. I comple- ment this analysis with a limited amount of quantitative data.

I examine the case of pervasive beauty ideals disseminated through mass media, which many suggest are harmful to young women. These images, particularly in regard to body shape, are extremely unlike "real" American women. The gap between the image and the reality has grown in recent years, as the media images have become slimmer and Americans have grown heavier (Wiseman et al. 1992). The in-depth inter- views focus on a tangible, explicit embodi- ment of idealized femininity-girls' maga- zines-which saturate their target audience

(Evans 1990). The images presented therein are also pervasive in other media such as movies and television. I address these broad research questions: How do girls interpret the female image in media, how do they cri- tique it, and how do they perceive its influ- ence? How do girls view peers' interpreta- tions of these images? How important are critical views of the imagery in protecting girls' self-esteem?

DATA AND METHOD

Data

The data come from in-depth interviews with a subsample of 60 girls, who were part of a larger survey of 210 female and 227 male grade 9 and grade 10 students. I includ- ed males in the quantitative portion of the study in order to compare data on media use, self-esteem, evaluation of appearance and other measures, but their data are not reported here. I obtained permission from the principals of two high schools to conduct the study, and from grade 9 and grade 10 teachers to gain access to their classrooms. To examine how media interpretations might vary by ethnicity and locale, I chose an ethnically diverse urban high school and an all-white rural high school in the mid- west. This choice of schools allowed compar- isons of whites' interpretations of popular national media images across localities, and permitted a comparison of ethnic groups in a single locale.

The urban high school enrolled 1,690 students. Seventy-four percent of the stu- dents were white, 22 percent African- American, and 4 percent were of other eth- nicities. The school district, with a population of close to 70,000, is located in the wealthier part of an urban county and is under a desegregation order. Seventy percent of the 421 grade 9 students at this school partici- pated. The rural high school contained 314 students, largely from working-class back- grounds; 99 percent were white, and 1 per- cent Asian. These students come from sever- al very small towns located in the district, which collectively consist of fewer than 6,000 people. Of the 174 grade 9 and grade 10 stu- dents at the rural school, 81 percent partici- pated.

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

MEDIA IMAGES AND SELF-CONCEPTS 195

Procedure

In the months preceding the study, I conducted group interviews with three small groups of young people in youth organiza- tions in order to develop an interview sched- ule for the qualitative portion of the study. The written questionnaire was also pretest- ed; the responses, questions, and comments helped to shape the final form of the ques- tionnaire.

On the day of the survey, the researcher or an assistant read instructions and handed out the questionnaires, briefly describing the study and inviting students to participate. Of those who attended class that day and had not returned the parental "non-permission" form (four at the rural school, eight at the urban school), all participated. Students took approximately 45 minutes to complete the questionnaire.

After participating in the survey, the girls were asked to sign up for an in-depth, follow-up interview focused on girls' media such as magazines, to be conducted during their free time within the next month. They were offered gift certificates worth about $3 at local restaurants in exchange for their time. Approximately two-thirds of the girls who participated in the survey signed up for the follow-up interview. From this group, I selected respondents so as to maximize diversity in ethnicity, peer group and media use.' Overall the group interviewed did not differ significantly from those who were not interviewed on quantitative measures of media use, assessment, or comparison with the media images.

Of the 60 girls interviewed, 49 were white and 11 were minorities (10 African- American and one Asian-American). About two-thirds were interviewed individually,

1 A very few girls, mainly African-Americans, read girls' magazines only rarely or were nonreaders. Even so, however, they were familiar with the maga- zines, and with the idealized images presented there- in. Fine and Kleinman (1979) argue that widespread portrayals of groups are likely to elicit some reaction from group members. It is ironic that nonviewers or nonreaders of particular products are excluded from interpretive studies. They are likely to be more criti- cal than others in their social group because they may have rejected the use of the product.

and the remainder in groups of two (and in one case three) friends. The interviews took place in a private room during the subject's study hall or lunch period, or during a class period in which a teacher had approved her absence. Respondents returned a parental informed consent form, signed their own informed consent form, and were asked for permission to audiotape the interview. Interviews were transcribed, and themes and patterns from the data were tabulated and analyzed.

Qualitative Measures

To capture the meaning of the images and their impact on girls' lives, I included multiple sets of questions in the in-depth interviews. In the first set of questions, the girls were asked about their interests, their life at school, and their plans for the future. They were asked what they thought was most important to girls in their school, as well as what they perceived as worries or problems for girls.

Next, the respondents were asked ques- tions designed to appraise their interpreta- tion and framing of the magazines and the images therein. Specifically, to clarify their understanding of the meaning and purpose of the cultural symbols, I asked them to describe the magazines as if they were talk- ing to a girl who had just moved to the United States and knew nothing about them. The respondents were asked to state their favorite parts of the magazines, and what they liked and disliked about them. Questions were asked indirectly and directly about how the images might be influential in girls' experiences.

Other questions focused on the ways in which the respondents considered the girls portrayed in the magazines to be similar to or different from girls they knew. Several questions related specifically to the link between young women's image in the media, their own self-image, and their peers' self- image. The respondents also were asked how they thought others-friends, other girls, and boys-might interpret or use the images.

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

196 SOCIAL PSYCHOLOGY QUARTERLY

Quantitative Measures

The survey contained many questions on use and assessment of media, as well as social psychological measures. One specific question was related to social comparison with media images: "When I think about how I look compared to the girls shown in girls' magazines I feel: (1) great about myself, (2) good about myself, (3) OK about myself, or (4) not very good about myself." Another question read: "How important is it for you to read girls' magazines?" The response categories were (1) not at all important, (2) a bit important, (3) somewhat important, (4) important, and (5) very important. This question was repeated for perceived importance to others: "best girl friends," "girls in your school," and "girls your age in the United States." Evaluations of appearance was measured by the follow- ing questions: "Being good-looking is impor- tant to me"; "I am good-looking"; and "I am satisfied with my appearance." Each response was measured on a five-point scale ranging from "strongly agree" (5) to "strong- ly disagree" (1). Self-esteem was measured with the 10-item Rosenberg (1986) scale.

RESULTS

Ethnic status sharply differentiated whether girls identified with the images, sup- posedly intended for and about all adoles- cent girls. This status created an important filter for social comparisons and reflected appraisal processes, and thus influenced the effectiveness of critical interpretations for shielding harm to self-concept. First I discuss how the magazines and the images they con- tain were a part of white girls' culture at both the rural and the urban school, but how black girls generally rejected the images as part of their reference group even though they occasionally read popular girls' maga- zines. Both white and black girls interpreted the images as largely unrealistic; many want- ed more normal or more "real" girls in the images.

Next I show that theoretical work on social comparisons and reflected appraisals helps to explain how critical interpretations of media do not necessarily preclude effects of media. Social comparisons with the

images were distinguished strongly by eth- nicity, with variations in the desire to emu- late the images. Even though the white girls compared themselves negatively with the images and felt poorly about themselves in relation to the ideal, they felt it difficult to opt out of such a comparison. The utility of the criticisms relates to reflected appraisals, demonstrating the limitations of "resistance" when media images are presumed to affect significant or generalized others.

Reference Groups, Media Interpretation and Criticism of Images

White girls: Peer culture and media images as role models for the reference group. Explicit in cultural products labeled Seventeen or 'Teen is the notion that such products provide images and information relevant to particular groups of people. Willis (1990) and McRobbie (1991, 1982) note that among the various media, maga- zines have a stronger ideological focus and are important in identity formation because they target such groups on the basis of their status characteristics. In interviews, the white girls clearly indicated that they regarded the images as directed toward adolescent girls. Much of the content directly advises or dis- cusses issues pertaining to females, and the girls' responses indicated their understand- ing that femininity was central in these cul- tural products.

Girls' magazines, like other media, were part of the white girls' peer culture in both schools. They helped these girls to assess how well they fit into, or were similar to, their reference group (also see Currie 1997). The magazines gave advice on, and were perceived to help with, girls' concerns about "fitting in" and being accepted by others. In the interviews, for example, many girls stated that any hypothetical girl who does not read girls' magazines does not care about others' opinions or is very independent. This com- ment implies that the information contained in these media pertains to conforming to the "norm" of adolescent femininity.

The respondents considered reading the magazines an enjoyable leisure activity: 95 percent of the white girls surveyed read them occasionally or more often; more than

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

MEDIA IMAGES AND SELF-CONCEPTS 197

half read them "always." Magazine reading as a part of peer culture, and the relative amount of interaction centered around the cultural products themselves, differed some- what in the two locales. At the rural school, where girls made slightly higher use of the imagery and evaluated it more positively (data not shown), cliques regularly discussed content during school hours and after school over the telephone. They read the magazines in the lunchroom, the hallways, and the school library, and even during class. Subscribers often shared their magazines with friends, reading them either together or to each other, and passing on copies to those who did not subscribe. Indeed, for rural white girls, a great deal of peer interaction surrounded these magazines. This is not sur- prising because in rural areas, media may be an important means of understanding the larger world and the variety of people in it, with which the rural dwellers have much less contact (Johnstone 1974; Morgan and Rothchild 1983).

Urban white girls also said they dis- cussed the magazines or particular items in the magazines with friends, but they report- ed this experience less often. Perhaps because more varied activities are available to the urban girls, magazine reading is less salient. Yet in a quantitative analysis exam- ining how often white girls read the maga- zines alone or with friends, I found no differ- ences between schools. Indeed, more white girls at the urban school than at the rural school subscribed to at least one girls' maga- zine (64 percent versus 50 percent,p < .10).

Black girls: "Maybe if there were more of us in there." Black respondents less often read mainstream girls' magazines, both indi- vidually and as a collective activity: 86 per- cent of the black girls surveyed read them at least occasionally, but only 11 percent always read them. Even though, in recent years, black models have appeared regularly in the four magazines with the greatest circulation, the magazines are perceived as largely for white girls. Most of the black girls read Ebony or Essence, aimed at black adults and black women respectively, magazines about music directed toward black youth, and hair- style publications. Thus, in contrast to white girls, these respondents largely regard main-

stream girls' magazines as something they do not want to or should not orient them- selves toward because they view the maga- zines as for and about white girls. They define the images as irrelevant to their refer- ence group for this social aspect of the self. This finding is not surprising because, as Collins (1991) and others have observed, mainstream media show an especially dis- torted image of black females-very thin, and with "whitened" hair and features. Tanya,2 in responding to how people would understand "girls" if they had only girls' magazines to look at, said:

I think this is mainly toward ... white females ... you really wouldn't see too many black people in here-so if this is all you saw, you'd be kinda scared when you saw one like me or something. (May 9, 1994; urban black girl)

Minority girls were quite critical about the realism of the images. Part of this critique was that normative adolescent femininity was portrayed as white femininity. Although ethnicity differentiated the respondents' use of these mainstream magazines, both African-American and white girls seemed to hold common perceptions about the unreali- ty of the images.

Interpretations and critical interpreta- tions of media images of females. When asked to describe the magazines to a girl who had never seen them, most of the respondents interpreted them as conveying very traditional aspects of femininity, such as appearance and romance. They mentioned fashion, makeup, styles-all related to appearance-and relationships with males. Previous analyses of these magazines (Evans et al. 1991; Peirce 1990) indicate that these reports represent accurate assessments of the content of girls' magazines. Barb's expla- nation is similar to how most girls described the magazines:

They're about how girls can do their hair, what's in fashion. They give advice on boys; sometimes they give you advice on your body and stuff like that-how to get in shape. They've got how to do your makeup right, hair-I think I already said that-

2 All names are fictitious.

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

198 SOCIAL PSYCHOLOGY QUARTERLY

what's the right jewelry. They talk a lot about stars and stuff like that, they also talk a lot about boys. (May 9,1994; urban Asian girl)

Secondarily, the respondents reported that the magazines were about girls' "prob- lems." This view is closely related to the above observation. The information present- ed about appearance and relationships with boys was interpreted by the girls as advice about problems of traditional femininity which they were experiencing or which were common to adolescent females. A minority of the girls described the magazines more broadly as about "teenagers' lives" or "everything." Only two of 60 respondents described the magazines in what might be considered feminist terms, as about girls' "being independent," although these two girls also discussed appearance as an impor- tant component of the magazines.

The great majority of the respondents, even those who seldom read the magazines, liked them as a whole or liked certain parts. The girls stated that they read them because they were interesting, entertaining, and informative. An important feature of the girls' enjoyment and interest was learning about themselves and assessing their lives and their problems in relation to their peers. Linda, a grade 9 student, explained this:

The girls will write in, and you kind of real- ize they have the same problems as you do ... you know they [other girls] kinda make you feel like you're not the only one. (May 19,1994; urban white girl)

Researchers have suggested that one reason why people are critical of media is that the media distort reality and reflect groups in distorted ways. Most of the respondents were critical in that they said media images of girls were not realistic at all, and they made negative comments about the lack of "normal" girls. In general, the respondents indicated that the feminine images in the magazines presented an unre- alistic appearance, both in the styles of cloth- ing and in the perfection of their faces, hair, and bodies in comparison with the largely imperfect local girls. A few respondents said that the girls in the magazines were some- what realistic; sometimes they referred to the pages that focused on "real" girls' prob-

lems or compared the images with the most popular or most beautiful girls in the school. The black girls were quite critical of the magazine models' physical appearance in general and tended to be critical about the lack of ethnic diversity or representation.

In discussing how the models looked, the respondents were likely to comment that they were too perfect, especially in body shape, weight, hair, facial features, and com- plexion. Sandra, a grade 10 student, dis- cussed the message sent by the magazines:

They mainly focus on models ... they make them look perfect, which nobody is. Makes everyone's expectations really high of their- self, and they don't need it. I don't think they show the true girl. You know, nobody is per- fect, and they all have their mistakes, and some of these people look like they never make a mistake. (April 21, 1994; rural white girl)

In fact, many viewed the images not merely as unrealistic, but as artificial. A girl who had recently lost a good deal of weight remarked that some models shown in the magazines have altered their "true" selves:

I think some of them might be fake. Like get contacts to change their eye color, cake on their makeup, starve themselves. Like they're really not that skinny, but they just starve themselves. (May 25, 1994; urban white girl)

Generally, the respondents disliked the fact that these pervasive media images devi- ated so much from reality. They remarked, as noted above by Sandra, that the media cre- ated an uneasy gap between image and reali- ty. Barb, while looking at the title of a girls' magazine article in front of her, observed that even the so-called "problem" bodies shown in the magazines are perfect:

Oh, if I read that "Four Weeks to a Better Body," I'd probably ... these magazines are trying to tell you "Do this and do that." Sometimes they have ... swimsuits and stuff, and what you can do if you have a problem body. If you got a big butt, big chest ... what to do. And these girls that they are showing don't have that problem. I mean you can tell they don't, and that makes me mad.... They say if you got a stick figure, wear a one-piece and ... colorful and I'm looking at the girl

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

MEDIA IMAGES AND SELF-CONCEPTS 199

and she doesn't have a stick figure. If you got big hips, if you got a big stomach-she doesn't have it-you can never understand that. (May 9,1994; urban Asian girl)

In response to open-ended questions about whether they would change anything about the magazines, particularly anything that was emphasized too much or was not included, more than one-third of the respon- dents specified that the magazines should change the feminine image to be more real- istic or "normal." Amy believes that "nor- mal" people are missing from the images:

One thing I guess would be just more nor- mal people ... not like the models, but just average. Other people that haven't really had modeling experience. (May 25, 1994; urban white girl)

Similarly, Suzanne suggests that magazine editors should be more realistic:

Probably tell them to be more realistic about the situations, and who they have in their magazines . . . people that are everyday teenagers, not just celebrities and people like that. (May 20,1994; urban white girl)

The majority of the girls indicated that the unrealistic images were a problem for themselves, and/or for some, most, or all girls. Leslie, who had discussed her "under- weight" friends' concern about eating any- thing fatty, said:

If all you ever see is all these people with the perfect teeth, the perfect complexion, with a skinny body, then they may not think that they're perfect and then they start to worry about being skinny and getting braces, wear- ing lots of makeup to cover up their face. (May 25,1994; urban white girl)

Similarly, Brittany discussed what she would tell the editors of girls' magazines:

To quit stressin' so much that you need to be so skinny for bathing suits and stuff 'cause it just makes teenagers feel bad about them- selves. (May 2,1994; rural white girl)

In sum, most of the respondents regarded media images of females, particularly those which are common in ads or fashion pages, as unrealistic. Many disliked the images for this reason, considered them harmful to

themselves or to others and advocated that media producers should alter their products to include more "real," ordinary, or "nor- mal" girls.

Media and Social Comparison Processes

Social comparison theories argue that we compare ourselves with similar and near- by others, and that social structural factors influence which referents will be chosen. Although researchers generally suggest that we are motivated and free to make compar- isons that are favorable to us (Rosenberg 1986; Singer 1981) there are limits; compar- isons that disfavor us may be unavoidable if we cannot leave a group (see Festinger 1954). This point may help to explain how cultural images of one's reference group, although rarely considered in social compar- ison research, may constitute an inescapable "group" that can have negative conse- quences as it is incorporated into local cul- ture. Insofar as one views media"others" as attractive and identifies with them as they are brought into one's peer group, they may become comparative referents, although such comparison to images is likely to have negative consequences for the self. In this section, I discuss how the mainstream female image, although most respondents view it as unrealistic and criticize it, becomes an oppressive negative referent for whites who cannot escape it easily, but not for blacks, who feel distant from it.

Emulation of media images of females verus distance from them. Both the white and the African-American respondents, but espe- cially the white girls, liked the magazines, even though they criticized the lack of real- ism of the girls pictured therein. The white respondents used the images and ideas in the magazines to assess themselves. They said fre- quently that they "felt better" or more nor- mal when reading about the problems and experiences of other girls their age. This feel- ing came from the numerous articles and advice columns that dealt with problems of relationships with boyfriends and family members, peer group pressures, and health, beauty, and fashion issues. The respondents particularly liked to assess themselves in rela- tion to their reference group by taking

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

200 SOCIAL PSYCHOLOGY QUARTERLY

quizzes that evaluated them on topics such as relationships (e.g., "How Good a Friend Are You?"). These quizzes provide scores that categorize the reader as a certain type of per- son and explain how she tends to act in situa- tions in comparison with others. Jackie explains why she reads the magazines:

I guess I like to see what ... the clothes, like what people are wearing. And like ques- tions-answers, like what people are . .. curi- ous about, and see if I'm the same, I guess. (May 23,1994; urban white girl)

The respondents often told how they learned about themselves through other girls' problems or experiences discussed in the magazines, and thus about how they compared with their peers on adolescent issues that were often troubling develop- mental concerns, such as sexuality. Many girls appreciated the magazines for helping with their problems. Yet just as the white girls sought to learn about and evaluate themselves in reference to their "media peers"' emotions, problems, behavior, and experiences, they did so with their media peers' physical appearance, though with less enthusiasm. The ordinary girls featured in the text and the atypical professional models featured in the accompanying pictures are very different types of media peers; perhaps this coexistence of "real" girls with problems (in the text) and "ideal" girls (pictured) makes the comparison to beauty images even more likely and leads to negative self- evaluations. Because of their perception that these images, however distorted, constituted others' views of adolescent femininity, the white respondents could not easily opt out of a social comparison and self-evaluation in which they were sure to fall short (also see Currie 1997; Snow 1983, p. 96,221).

Social comparison theories assume some degree of freedom in choosing comparison referents; such choices are presumed to be partly governed by motivations for self- enhancement (Singer 1981; Rosenberg 1986). Yet the girls who saw the images as unrealistic and disliked this nonrealism, and for whom comparisons would not invite self- enhancement, nonetheless made compar- isons. The great majority of white girls want- ed to look like the ideal girls featured in the

magazines, connecting such imagery to rewards such as male attention, and inevitably compared themselves with the "perfect" girl. Beth indicated why "every- body" wants to be like this:

They're so beautiful and everything and they have these really great bodies and they have the perfect hair, the perfect boyfriend, the perfect life, and they're rich and everything. (April 25,1994; rural white girl)

Similarly, Amber explained how she believed the magazines influence girls:

I think they, just girls in general kind of want to be like the girls in there. Like the models. I think they want to be like those models ... [because] they think if they're like that they're gonna get lots of guys and stuff. (April 28,1994; rural white girl)

Although the vast majority of white girls appeared to use the information and images as a reference group with which they identi- fied and whose physical appearance they emulated, four respondents stated that they did not want to change themselves to be like anybody else. For example, Beth, who com- mented about what "everybody wants," said:

Well, everybody wants to be a model but ... I just like myself as myself. I wouldn't really ... change anything about myself. (April 25, 1994; rural white girl)

The minority respondents, in sharp con- trast, did not emulate these images nor com- pare themselves as negatively with the mod- els. Even though most of the black girls occasionally read the mainstream publica- tions, they considered the images less rele- vant, belonging to "white girls"' culture and not part of a reference group toward which they oriented themselves. Strikingly, 10 of the 11 minority girls (nine black and one Asian-American) said unequivocally that they did not want to be like these girls; one mixed-race respondent (African-American and white) said that she "sometimes" did. The black girls indicated that they did not relate to the images and did not wish to emulate the rigid white beauty ideal.3

3 Many of the black girls mentioned alternative publications focusing on hairstyles for black females.

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

MEDIA IMAGES AND SELF-CONCEPTS 201

Tamika described why she generally did not read mainstream girls' magazines:

Well, I don't see a lot of black girls ... I don't see a lot of us .. . maybe if they had more, maybe I could relate to that. I don't know. 'Cause obviously we can't wear the same makeup or get our hair the same way ... things like that. So maybe if they had more. (May 11, 1994; urban black girl)

In sum, for the great majority of white girls in both locales, national media images and information about the reference group served as an additional social comparison introduced into the local context. The white girls evaluated their own behavior, prob- lems, emotions, and importantly, physical appearance in comparison with these media others. Even though they knew that the images were unrealistic, the white girls saw themselves as part of the reference group being portrayed, and compared their "prob- lems" with adolescent females' problems. They reported that they often (reluctantly) made social comparisons with the perfect physical appearance of media images because they knew that these images were what "everybody" wants. The minority respondents and a very few of the white respondents did not emulate these feminine images in media, did not bring them mean- ingfully into peer groups, and seemingly did not make social comparisons unfavorable to themselves.

Criticism, social comparisons and self- evaluations. Although the white girls liked the magazines a good deal and enjoyed find- ing out that they were "normal" on the basis of other girls' behaviors and problems, many said that they personally or that "girls" in general felt abnormal and inferior in rela- tion to the idealized feminine image. A key influence of the magazines, then, is that the great majority of white respondents said they wanted to look like the girls pictured therein, even though most saw the images as unrealistic and unattainable. These girls nec- essarily experienced relative deprivation

These publications did not contain articles, howev- er-only hairstyle photos and information. (See Parker et al. 1995 for an assessment of African- American girls' culture and appearance.)

because they could not attain the valued image promoted by the pervasive display of this unique part of the reference group. Although they generally understood that the images were unrealistic, the girls perceived that other girls in the school, and especially males, valued such an appearance. Thus it was difficult for critical appraisal of media images to become meaningful in local inter- action, a phenomenon that I discuss below.

A rural white respondent, Patty, demon- strated the complexity of being aware of nonrealism and critical of images, while simultaneously striving to achieve such an appearance. She indicated that the maga- zines influence girls "quite a bit," and explained that they did so by making girls, including herself, feel as if they wanted and needed to look like the media portrayals:

[Girls are influenced by magazines] ... by havin' to look that way. I mean by their body and stuff 'cause they're all really tall and skinny so everybody tries to be all really tall and skinny . . . I think most girls don't realize that they're bigger boned than [that] ... I'm [from a] very big-boned family ... most girls just don't realize that there's no way in the world they could look like that. (April 22, 1994; rural white girl; respondent's empha- sis)

Patty underestimated the degree of critical assessment by girls: most respondents did realize that they could not attain the look. Similarly, Alana, a tall, slim 15-year-old who had been recruited by a modeling agency, said that she and others felt as if they should measure up:

The thing with being skinny-you think sometimes "Maybe I should be skinny too." But then you sit back and you're thinking "Oh my God ... what am I thinking-think- ing that I'm going to be perfect like one of them." But there's a lot of girls, like I said, that get into that-they try to be all of that. (May 12,1994; urban white girl)

The white respondents made negative social comparisons even while they recog- nized the media distortion. They indicated that the comparisons were difficult to opt out of and made them or "girls" feel worse about themselves because the girls inevitably looked worse than the glamor-

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

202 SOCIAL PSYCHOLOGY QUARTERLY

ized, exceptional females in the media. In quantitative data from the larger group of girls surveyed (N = 210) the white girls felt significantly worse about themselves com- pared with the images (mean = 3.02 on a four-point scale, in which 4 = most negative) than the minority girls (mean = 2.34, p < .001).

Not only is it difficult for girls to opt out of this social comparison, but it is problem- atic to make appearance a less psychologi- cally central aspect of the self on which to "stake" one's self-esteem (James 1968; Rosenberg 1986). The idea that people can make aspects of the self less important for self-esteem with relative freedom downplays the lack of cultural power that disadvan- taged groups have in self-processes (Ferguson, 1980; Oyserman and Markus 1993). In addition to the actual (distorted) content of the female image, the media frame female worth in terms of appearance; and to counter this dominant cultural dis- course is quite difficult (Tlichman 1979).

The black girls' criticisms of media imagery, in contrast to the white girls', may be effective in reducing the impact of media in this case, because the black subculture as a whole is more critical of mainstream beau- ty ideals. The black girls in this study, although as concerned about appearance as the white respondents, perceived themselves as better-looking and were more satisfied with their appearance than were the white girls, and their self-esteem was higher.4 Though the black girls objectively are far- ther from the mainstream ideals of beauty in skin color, hair style, and weight (see Dawson 1988), they compared themselves

4 I regressed self-esteem on ethnicity, control fac- tors, and the comparison of self with images. The media comparison affected self-esteem (b = -3.77, p < .001). When the media comparison variable is included in the model, the relationship between minority status and self-esteem becomes nonsignifi- cant, though still positive. In other words, black girls' higher self-esteem is explained somewhat by more favorable comparisons with idealized feminine images. The negative effect of the social comparison was not reduced by including controls for how often these images were used, how much they were liked, or how important they were (i.e., criticisms). This finding further demonstrates (quantitatively) that criticism did not reduce negative impact.

more favorably with mainstream media images than did the white girls. Evidence from interviews also indicates that black girls perceived the white ideal as narrow or as less applicable to them. Eliza discussed how minority girls may strive less often to be like the images of girls shown in these maga- zines:

This is kind of a stereotype, but more of my white friends than my black friends are into [trying to be like feminine images in maga- zines]. I mean a lot of them are going on a diet or "I want that body so bad"-I don't know how anybody can be like that. (May 27,1994; urban black girl)

Lakoff and Scherr (1984) suggest that ethnic minority women, although evaluated by whites as inferior in relation to a model of white beauty which is impossible for them to achieve, currently may consider a wider range of looks as beautiful or normal within their subculture. They cite black women's magazines: Although these publications advertise hair straighteners (for a more mainstream model of beauty), they also show a wide variety of African facial fea- tures when demonstrating hair styles, as well as a range of body types, with information about how to "get bigger" as well as smaller (Lakoff and Scherr 1984). Indeed, Essence, a magazine targeted to African-American women, received a media award for its more realistic portrayal of females, including its depiction of a wider variety of body shapes (Chambers 1995).

Despite some body dissatisfaction among black women (Thomas and James 1988), this dissatisfaction usually is not mani- fested in extreme thoughts and behaviors regarding weight control.5 This is not to say, however, that black females are unaffected by externally defined everyday beauty images. Collins (1991) and hooks (1992) point out how the controlling and oppressive image of fair-skinned, long-haired "white" beauty creates pain for black women, who cannot attain this.

S Thompson (1992), however, shows that black women can suffer from eating disorders, and that this problem can stem from racist ideas about how women should look.

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

MEDIA IMAGES AND SELF-CONCEPTS 203

Although the black girls surveyed as part of the larger study (quantitative data, not shown) felt that they were significantly better-looking than white girls and were more satisfied with their looks overall (also see Parker et al. 1995; Wardle and Marsland 1990), the interviews indicated a good deal of concern and dissatisfaction about hair, which the girls felt must be altered from its natural state for a more "beautiful" look. Indeed, hair straighteners are advertised in the very publications directed to black women which have been praised for their wide variety of body shapes and facial fea- tures.

As the above analysis shows, the effects of media imagery are complex. Social psy- chological work on reference groups and social comparison processes can help to elu- cidate how the consequences of pervasive media images vary for girls of different eth- nicities. Social comparison research suggests that we compare ourselves with similar oth- ers, and although media images generally have not been examined as part of reference groups, it is likely that people shown in the media may serve this function for self- assessment (Snow 1983).

The data reported here show that a peer group depicted through media can be an important social comparison group, even though it creates negative consequences for self-evaluation. In addition, the freedom to select this comparison referent may be limit- ed. The relative freedom of choosing similar others to compare oneself to for self- enhancement purposes, as suggested in research on social comparison (see Singer 1981) is shown to be very restricted in this case. Rather, the public pervasiveness and the esteemed, glamorized position of these "perfect" female peers apparently restrict white girls' ability to ignore or downplay this comparison despite its negative conse- quences. Indeed, because the images are pre- sented in a variety of media formats such as television, magazines, film, and the Internet, the impact of the images may be much more powerful. In addition, girls have very limited power to help shape and define these public images at the institutional level of media production (Milkie 1996).

It was striking that the urban white girls' interpretations of the images and their beliefs about the place of those images in others' eyes were more similar to the rural white girls' interpretations and beliefs than to those of their black urban classmates. The black girls in this integrated school did not report that the mainstream model of female beauty was relevant to them, and thus did not generally compare themselves unfavor- ably with these media peers. As Rosenberg (1986) points out in his study of black and white children's self-esteem-blacks gener- ally made comparisons within their own group, not with whites, on aspects of self- evaluation such as physical appearance. Similarly, the black girls studied here report- ed that the media figures with whom they identified were black performers, not white or "whitened" models; the performers had a wide range of appearances, and the girls did not appear to compare themselves negative- ly with these women.

In sum, although many white girls understood that the images were unrealistic, and although they disliked the anxiety- producing gap between their own physical appearance and a media-generated ideal, they still desired such an appearance, attempted to attain it, and felt bad when they did not measure up to the media image of femininity. Some of this lack of efficacy in the white girls' critical interpretations, due to continued negative social comparisons, may be explained by reflected appraisal processes.

Reflected Appraisals: How Media Images Can Affect the Self Indirectly

Research suggests that reflected appraisals may be a proxy for the perceived attitudes of a generalized other rather than for perceived views of particular others (Felson 1989). This finding helps to explain how media images may affect us indirectly insofar as they are an important part of the generalized other; indeed, as I explain in this section, media images may represent soci- ety's views. In addition, research on reflected appraisals points to the relevance and importance of these perceived attitudes toward us, and our agreement with these

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

204 SOCIAL PSYCHOLOGY QUARTERLY

attitudes as mediators of reflected appraisals' impact on the self (Rosenberg 1986). This phenomenon explains the vari- ability in the power of critical interpreta- tions of media. For whites, beliefs about how important others-white peers-consider these images may thwart their own power to meaningfully express criticism. For blacks, this "white" generalized other is less rele- vant, and the evaluations of those who find the image desirable are less important; thus critical interpretations of these media images are more effective.

In interviews, the respondents distin- guished between the importance of appear- ance to them and to others (they almost always considered appearance more impor- tant to others), and discussed how different groups of others might use and interpret media images. Close girlfriends usually were not regarded as holding the respondents to idealized media standards, but "other" girls-those in the school or beyond-often were viewed in this light. The white girls did not believe that criticisms of images were widely shared; the black girls, however, indi- cated that their close friends were equally critical. Mary, an African-American respon- dent interviewed with a close friend, said:

It seems like sometimes we're the only two people in this entire school that don't want to hear this stuff . . Everybody else might, I'm not sure, but everybody else might be enjoying it, and we're the only [girls] being different and we don't want to hear it ... I know my friends, I know they're probably not any of them ... worried about this stuff ... in these magazines, like that magazine [refers to magazine in front of her] said "How to Kiss Better." My friends are wor- ried about stuff like how to take care of yourself and learn how to be independent. They're worried about real things that are going to help you. Like who's going to hire you just cause you can kiss good? (May 27, 1994; urban black girl)

Note Mary's distinction between her friends and "other" girls.

In the questionnaire administered to the larger group of girls, respondents were asked about the importance of media to close friends, girls in their grade at school, and girls their age in the United States. At each

level, perceptions of these groups' assess- ments of media became less knowable. Although it was likely that the girls knew their best friends' media habits and dis- cussed media with them relatively frequent- ly, it was more difficult to know how all girls in their grade at the school evaluated the imagery, and impossible to know the assess- ment of all the girls their age in the country.

In regard to the perceived importance of the imagery to other girls, a significant and decisive pattern emerges (see Table 1). As a whole, individuals perceive that these cultur- al products are more important to their best friends than to themselves, more important still to girls in their school, and even more important to female peers in the United States (all differences are significant at p < .05). The table shows comparisons between rural whites, urban whites, and urban minori- ty girls. These figures are evidence that the effect holds for all groups, but that the spe- cific pattern differs somewhat by ethnicity. White girls in both locales perceive that best friends view the cultural products as more important than do they themselves, that those in the school think them more impor- tant than do best friends, and that those in the United States, the most generalized group of others, consider them most impor- tant. This pattern generally holds for the minority girls, with one interesting excep- tion. These girls believe that their best friends, the great majority of whom are also African-American (data not shown), per- ceive the cultural products as very slightly (but not significantly) less important.

In sum, the respondents attribute rela- tively low importance to the media for them- selves, but greater importance at each level of known and unknown others. This finding complements recent work on third-person effects (e.g., Perloff 1993), in which others are perceived to be influenced more strongly by the media. The perception that others are influenced more strongly may be based par- tially on the perception that the media occu- py a more important place in others' lives than in one's own. It is unclear why media are considered more important to those at each higher level of abstraction. People, how- ever, know proportionately less of others' activities as the groups become more

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

MEDIA IMAGES AND SELF-CONCEPTS 205

Table 1. Means: Importance of Girls' Magazines to Self and Perceived Importance to Othersa b

How important How important How important How important is it for you is it for your is it for girls in is it for girls your to read girls' best girlfriends your school age in the US. magazines? to read girls' to read girls' to read girls'

magazines? magazines? magazines? All Girls (N = 210) 2.50 2.67 3.00 3.32 Rural White Girls (N = 64) 2.67 3.14 3.29 3.48 Urban White Girls (N = 112) 2.53 2.60 2.94 3.24 Urban Minority Girls (N = 34) 2.06 1.97 2.67 3.28 a Scale: (5) very important; (4) important; (3) somewhat important; (2) a bit important; (1) not at all important b For combined group, all differences (i.e., self versus best friends, best friends versus girls in the school, and girls in the school versus girls in the United States) are significant at p < .05.

unknown; this "filling in" of others' lives with the media as a public knowledge maker, is consistent with the third-person effect (see Davison 1983:9). Notably, the cohort of girls in the United States is known largely through the media. Indeed, these oth- ers who are beyond the local culture but are depicted through the media are an impor- tant part of the generalized other whose views toward our self we take (Altheide and Snow 1988).

In addition to "other girls," the white girls believed overwhelmingly that males are influenced by the unrealistic images and are uncritical of those images. The great majori- ty of white girls perceived that males evalu- ated them on the basis of females' unrealistic appearance in media imagery. Some girls indicated that boys explicitly discuss media models such as Cindy Crawford, and/or insult girls who deviate from the unrealistic standards represented by such models. Although boys rarely are exposed to these images through girls' magazines, the same models and images appear in magazines and other media formats directed to males and to general audiences. Most respondents per- ceived that males wanted this appearance in girls, even if they had not heard males talk about it. Alison described how boys may use these images:

Guys mostly look at ... the ones that are like supermodels, an' then they look at you, an' balance the scales, (and) kinda look more towards the model [laughs]. (April 5, 1994; rural white girl)

Lenore and Andrea commented:

Lenore: I think it kind of influences guys 'cause I think the guys a lot just [go for] looks and I think that makes them look more ... for looks cause they see them.

Andrea: They see them and they try to com- pare all of us to them ... I just get that per- ception. (May 13,1994; urban white girls)

The idea that males want females to be physically attractive complements abundant research on adolescent relations (e.g., Eder 1995). In this study, however, the respon- dents indicated that rather than local stan- dards, they may be evaluated by the stan- dards of the media, a public, glamorized, and unrealistic portion of the social group with which they compare unfavorably. Because the ideal-a rare extreme in women's body shape (two standard deviations from the mean in social science terms) and socially constructed "perfection" of physical fea- tures-is commonly displayed and acknowl- edged, it may seem that others whose views are important to the girls perceive the image as attainable and normative (Snow 1983).

Body image is a particularly important aspect of the effect of media culture on white girls' local culture. The girls reported that they often talked about looks and that they had friends who asked frequently if their appearance was acceptable and often commented on their weight (see Nichter and Vuckovic 1994). Several respondents made connections between the media and the girls' warped views of themselves, particular- ly in regard to body size. For example, often before they were questioned directly about

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

206 SOCIAL PSYCHOLOGY QUARTERLY

body weight, shape, or perceptions among their peers, a significant number of girls stat- ed that friends or girls they knew often talked about being overweight. The respon- dents indicated, however, that those who said these things were not actually over- weight.

Tanya, an African-American girl who was rare in having close interracial friend- ships, observed how her white friends were affected by the unrealistic media images. She pointed to reflected appraisals in citing their unhappiness with their appearance:

I think that's why some girls ... think they're too fat and try to lose weight. They look at her [points to a girl in the magazine] and they think they should look like that because they heard a guy say that she's pretty or whatever. So they feel that they should look like her and they try to go on a diet and all that stuff. So I think some people just don't know what they [others] are saying, and peo- ple hear them, and so they automatically assume they should do this and do that for other people ... I hear a lot of girls that are . . . smaller than I am talking about "I'm so fat." I say "if you're fat, then I'm obese." . . . [they say] "I'm on a diet, I can't eat." [I say] "Did you eat breakfast" [they say] "no." [I say] "If I eat lunch, are you going to eat something?" [they say] "I can't, I'm on a diet." And I say "You're already skinny! How much do you want to weigh? 100 pounds?" (May 9,1994; urban black girl)

Even when respondents were aware of peers' misperceptions of weight and did not have this misperception themselves, the media's distorted images shift the local stan- dards for what is normative. Tanya's com- ment to her friends, "If you're fat, then I'm obese," indicates how the media affect local cultures, including those in which girls are aware and critical of unrealism in cultural products.

Thus, because media images were a part of the white girls' peer culture, and because these girls perceived that significant oth- ers-other girls and especially boys in their local networks-evaluated them on the basis of media ideals that were nearly unattain- able, they were influenced regardless of how strongly they criticized the imagery. Especially important were body shape

"norms" in the media, which tended to warp average-weight and thin girls' perceptions of their weight and attractiveness, or at least made them overconcerned about weight at objectively normal, healthy, weights. Even girls who articulated the distorted nature of peers' views of attractiveness seemed to feel compelled to abide by the shifted "norm" of body shape.

The wider range of physical appearances and body shapes that the black girls seemed to accept as good-looking in themselves and others was related to a more inclusive beau- ty ideal promoted in the "black" media. In addition, the black girls were more tentative about suggesting that males evaluated them on the basis of mainstream (white) media images. Most girls indicated that some males might do so but that others would not. This belief that males (often specified as black) rejected the "whitened" image was impor- tant in reducing black girls' negative self- evaluations especially related to body size.

Several minority girls said they believed at least some black males desire women who are not extremely thin. Nadia discussed the "normal" appearance preferred by black females and males:

They're [black females] not trying to have little bodies. They want to be thick. They don't want to be fat, they just want to be thick ... A nice size behind, nice big bra ... they want to be a nice normal size. They don't want to be skinny . . . 'cause black men don't like skinny people that much. Some of 'em do, but they'd rather have a thick person ... We don't think skinny is pretty. (May 23, 1994; urban black girl)

Thus, the black girls interpreted the media differently than the white girls. In par- ticular, most perceived that the images were intended for other, white girls. They appeared to be affected less negatively by a narrow media image of female beauty. The black girls defined themselves outside the dominant culture and cultural imagery; therefore they seemed to be able to reject the images as a group. Reflected appraisals allow criticisms of nonrealism to be practi- cally effective: Because the black girls believe that African-American males also reject the narrow, "white" feminine ideal

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

MEDIA IMAGES AND SELF-CONCEPTS 207

prominent in the media and define a wider range of black feminine appearances as beautiful or "normal," they can express and act on such criticisms of mainstream beauty ideals within their subculture. The white girls' (segregated) social networks con- strained the avenues for meaningfully expressing criticism of media imagery; those of the black girls facilitated media criticism.

In sum, reflected appraisal processes are important in considering how media can affect the self indirectly, even when such images are disliked or criticized. The media are presumed to affect others; for the white girls, as noted above, an important part of the interpretations is that others use the images and find them important and attain- able, even when they themselves do not. Reflected appraisal processes indicate the importance of beliefs about how others view us; thus the usefulness of a critical interpre- tation is thwarted by imagining that others use the extremely unusual image of women that pervades the media. The black girls however, do not view the mainstream images as part of the reference group. In addition, they believe that those important to them, generally within their ethnic group, are critical of the images as well. Thus, they are protected to some degree from negative comparisons. The opinions of the main- stream white "others," who are presumed to find the image desirable and realistic, are generally not deemed important for African- American girls' view of self (see Rosenberg 1986).

CONCLUSION

This work reiterates the importance of employing qualitative approaches to under- standing people's interpretations of media, especially critical interpretations. This research also shows the utility of examining how even critically interpreted media con- tent may negatively affect individuals' self- concept, attitudes, or behavior through "mediated" social comparisons and reflected appraisals. Virtually all of the white respon- dents said that they wanted to achieve an appearance like that featured in the media, even though most regarded the images as unrealistic and many wished to see more

"real" girls in the media. The influence of the media on white girls appears to lie main- ly in publicly presenting, directly and repeat- edly, a unique part of the reference group which is glamorized. This media subgroup is oppressive in that it shifts the self- evaluations of the population of real white girls downward. They perceive, through the "media-filled" eyes of others who matter- white peers-that their own appearance is judged negatively. Ironically, girls' belief that others (particularly males) hold them to these unrealistic standards might be a bit inaccurate (see Rozin and Fallon 1988). W.I. Thomas's dictum, that if people perceive sit- uations as real, they are real in their conse- quences, appears especially apt here.

The black girls generally criticized the artificiality of females' appearance and the lack of diversity in these media; however they were more immune to unfavorable social comparisons than were the white respondents. This immunity relates to the black girls' belief that girlfriends and at least some black males are also critical of these media images and appreciate a wider array of "normal" or attractive appearances.

This study suggests that recent school policies designed to teach students to be critical of distorted media content may be helpful, but these lessons may not be enough to thwart some negative effects of media if students do not see others as critical. In addition, "alternative" media with more pos- itive images of girls (e.g., Blue Jean, New Moon) which have recently become avail- able are important, but their impact may be small since these media images are not perceived to be widely used. More positive, realistic images of females, especially in terms of body shape, should be included in mainstream girls' magazines and other media so that a distorted, narrow image does not continue to deflate white girls' self- evaluations.

This research attests to the importance of studying media influence as it actually operates in the empirical world, as Blumer (1969) suggested three decades ago. A quali- tative analysis of media captures the com- plexities of influence processes. It enables researchers to more clearly understand reac- tions to, and meanings created from, media

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

208 SOCIAL PSYCHOLOGY QUARTERLY

images themselves. It permits a rich, tex- tured assessment of how individuals under- stand others' views of the images in a media- saturated world. This study makes it clear that the images are not simply accepted or interpreted individually. Rather, the media are understood in everyday experience to be part of the collectivity of individuals' social worlds, in and beyond their local context. For self-concept researchers, this qualitative work suggests that social comparisons and reflected appraisals, which often are mea- sured quantitatively and exclude "media others," may be more complex today, when much information about others and their views comes from media; and when one "knows" the media images which others see and which supposedly influence them.

REFERENCES

Altheide, David L. and Robert P. Snow. 1988. "Toward a Theory of Mediation." Pp. 194-223 in Communication Yearbook, edit- ed by James A. Anderson. Newbury Park, CA: Sage.

Binder, Amy. 1993. "Constructing Racial Rhetoric: Media Depictions of Harm in Heavy Metal and Rap Music." American Sociological Review 58:753-67.

Blumer, Herbert. 1969. Symbolic Interaction: Perspective and Method. Englewood Cliffs, NJ: Prentice-Hall.

Cantor, Muriel G. 1987. "Popular Culture and the Portrayal of Women: Content and Control." Pp. 190-214 in Analyzing Gender: A Handbook of Social Science Research, edit- ed by B.B. Hess and M.M. Ferree. Newbury Park, CA: Sage.

Chambers, Veronica. 1995. "The Essence of Essence." New York Times Magazine, June 16,p. 24.

Collins, Patricia Hill. 1991. Black Feminist Thought: Knowledge, Consciousness, and the Politics of Empowerment. New York: Routledge.

Corner, John. 1991. "Meaning, Genre and Context: The Problematics of 'Public Knowledge' in the New Audience Studies." Pp. 267-84 in Mass Media and Society, edit- ed by J. Curran and M. Gurevitch. London: Edward Arnold.

Currie, Dawn H. 1997. "Decoding Femininity: Advertisements and Their Teenage Readers." Gender & Society 11:453-77.

Davidson, Emily S., Amy Yasuna, and Alan Tower. 1979. "The Effects of Television

Cartoons on Sex-Role Stereotyping in Young Girls." Child Development 50:597-600.

Davison, W. Phillips. 1983. "The Third Person Effect in Communication." Public Opinion Quarterly 47:1-15.

Dawson, Deborah A. 1988. "Ethnic Differences in Female Overweight: Data from the 1985 National Health Interview Survey." American Journal of Public Health 78:1326-29.

Donnerstein, Edward and Daniel Linz. 1989. "Mass Media Sexual Violence and Male Viewers: Current Theory and Research." Pp. 488-501 in Men's Lives, edited by M. Kimmel and M. Messner. New York: Macmillan.

Eder, Donna with Cathy Evans and Stephen Parker. 1995. School Talk. New Brunswick: Rutgers University Press.

Evans, Ellis D. 1990. "Adolescent Females' Utilization and Perception of Contem- porary Teen Magazines." Presented at the biennial meetings of the Society for Research on Adolescence, Atlanta.

Evans, Ellis D., Judith Rutberg, Carmela Sather, and Chari Turner. 1991. "Content Analysis of Contemporary Teen Magazines for Adolescent Females." Youth and Society 23:99-120.

Felson, Richard B. 1985. "Reflected Appraisal and the Development of Self." Social Psychology Quarterly 48:71-78. . 1989. "Parents and the Reflected

Appraisal Process: A Longitudinal Analysis." Journal of Personality and Social Psychology 56:965-71.

Felson, Richard B. and Mark Reed. 1986. "The Effect of Parents on the Self-Appraisals of Children." Social Psychology Quarterly 49:302-308.

Ferguson, Kathy E. 1980. Self, Society, and Womankind. Westport, Greenwood Press.

Festinger, Leon. 1954. "A Theory of Social Comparison Processes." Human Relations 7:117-40.

Fine, Gary Alan and Sherryl Kleinman. 1979. "Rethinking Subculture: An Interactionist Analysis." American Journal of Sociology 85:1-20.

Fish, Stanley. 1980. Is There a Text in This Class? The Authority of Interpretive Communities. Cambridge, MA: Harvard University Press.

Fiske, John. 1990. Introduction to Communication Studies. London: Routledge.

Frazer, Elizabeth. 1987. "Teenage Girls Reading Jackie." Media, Culture and Society 9:407-25.

Gans, Herbert J. 1972. "The Politics of Culture in America: A Sociological Analysis." Pp.

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

MEDIA IMAGES AND SELF-CONCEPTS 209

372-85 in Sociology of Mass Communications, edited by D. McQuail. Baltimore: Penguin.

Hall, Stuart. 1980. "Encoding/Decoding." Pp. 128-39 in Culture, Media, Language, edited by S. Hall, D. Hobson, A. Lowe, and P. Willis. London: Hutchinson.

hooks, bell. 1992. Black Looks: Race and Representation. Boston: South End Press.

Hur, Kenneth K. and John P. Robinson. 1978. "The Social Impact of Roots." Journalism Quarterly 55:19-24.

Ichiyama, Michael A. 1993. "The Reflected Appraisal Process in Small-Group Interaction." Social Psychology Quarterly 56:87-99.

James, William. 1968. "The Self." Pp. 41-50 in The Self in Social Interaction, edited by C. Gordon and K.J. Gergen. New York: Wiley.

Johnstone, John W.C. 1974. "Social Integration and Mass Media Use among Adolescents: A Case Study." Pp. 35-48 in The Uses of Mass Communications, edited by J.G. Blumler and E. Katz. Beverly Hills: Sage.

Katz, Elihu and Paul F Lazarsfeld. 1955. Personal Influence: The Part Played by People in the Flow of Mass Communications. New York: Free Press.

Lakoff, Robin Tolmach and Raquel L. Scherr. 1984. Face Value: The Politics of Beauty. Boston: Routledge.

Lang, Gladys Engel and Kurt Lang. 1981. "Mass Communications and Public Opinion Strategies for Research." Pp. 653-82 in Social Psychology: Sociological Perspectives, edited by M. Rosenberg and R.H. Turner. New York: Basic Books.

Lazarsfeld, Paul F., Bernard Berelson, and Hazel Gaudet. 1948. The People's Choice. New York: Columbia University Press.

McLeod, Jack M., Gerald M. Kosicki, and Zhongdang Pan. 1991. "On Understanding and Misunderstanding Media Effects." Pp. 235-66 in Mass Media and Society, edited by J. Curran and M. Gurevitch. London: Edward Arnold.

McRobbie, Angela. 1982. "Jackie: An Ideology of Adolescent Femininity." Pp. 263-83 in Popular Culture: Past and Present, edited by B. Waites, T. Bennett, and G. Martin. London: Open University Press. . 1991. Feminism and Youth Culture.

Boston: Unwin Hyman. Milkie, Melissa A. 1994. "A Social World

Approach to Cultural Studies: Mass Media and Gender in the Adolescent Peer Group." Journal of Contemporary Ethnography 23:354-80. . 1996. "The Limits of Audience Power:

Magazine Editors Account for the

Persistence of Unrealistic Beauty Ideals Despite Criticism." Presented at the annual meeting of the American Sociological Association, New York.

Mills, C. Wright. 1963. Power, Politics and People. New York: Oxford University Press.

Morgan, Michael and Nancy Rothchild. 1983. "Impact of the New Television Technology: Cable TV, Peers, and Sex-Role Cultivation in the Electronic Environment." Youth and Society 15:33-50.

Nichter, Mimi and Nancy Vuckovic. 1994. "Fat Talk: Body Image among Adolescent Girls." Pp. 109-31 in Many Mirrors: Body Image and Social Relations, edited by N. Sault. New Brunswick: Rutgers University Press.

Oyserman, Daphna and Hazel Markus. 1993. "The Sociocultural Self." Pp. 187-220 in Psychological Perspectives on the Self, vol. 4, edited by J. Suls. Hillsdale: Lawrence Erlbaum Associates.

Parker, Sheila, Mimi Nichter, Mark Nichter, Nancy Vuckovic, Colette Sims, and Cheryl Ritenbaugh. 1995. "Body Image and Weight Concerns among African-Amercian and White Adolescent Females: Differences that Make a Difference." Human Organization 54:103-14.

Peirce, Kate. 1990. "A Feminist Theoretical Perspective on the Socialization of Teenage Girls through Seventeen Magazine." Sex Roles 23:491-500.

Perloff, Richard M. 1993. "Third Person Effect Research 1983-1992: A Review and Synthesis." International Journal of Public Opinion 5:167-84.

Press, Andrea. 1991. Women Watching Television: Gender, Class, and Generation in American Television Experience. Philadelphia: University of Pennsylvania Press.

Rosenberg, Morris. 1986. Conceiving the Self. Melbourne, FL: Krieger.

Rozin, Paul and April Fallon. 1988. "Body Image, Attitudes to Weight, and Misperceptions of Figure Preferences of the Opposite Sex: A Comparison of Men and Women in Two Generations." Journal of Abnormal Psychology 97:342-45.

Seiter, Ellen, Hans Borchers, Gabriele Kreutzner, and Eva-Maria Warth. 1989. "'Don't Treat Us Like We're So Stupid and Naive': Towards an Ethnography of Soap Opera Viewers," Pp. 223-47 in Remote Control: Television, Audiences, and Cultural Power, edited by E. Seiter, H. Borchers, G. Kreutzner, and E. Warth. London: Routledge.

Shively, JoEllen. 1992. "Cowboys and Indians: Perceptions of Western Films among

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions

210 SOCIAL PSYCHOLOGY QUARTERLY

American Indians and Anglos." American Sociological Review 57:725-34.

Singer, Eleanor. 1981. "Reference Groups and Social Evaluations." Pp. 66-93 in Social Psychology: Sociological Perspectives, edited by M. Rosenberg and R.H. Turner. New York: Basic Books.

Snow, Robert P. 1983. Creating Media Culture. Beverly Hills: Sage.

Thomas, Veronica G. and Michelle D. James. 1988. "Body Image, Dieting Tendencies, and Sex Role Traits in Urban Black Women." Sex Roles 18:523-29.

Thompson, Becky Wangsgaard. 1992. "'A Way Outa No Way': Eating Problems among African-American, Latina, and White Women." Gender & Society 6:546-61

Tuchman, Gaye. 1978. "Introduction: The Symbolic Annihilation of Women by the Mass Media." Pp. 3-38 in Hearth and Home:

Images of Women in the Mass Media, edited by G. Tuchman, A.K. Daniels and J. Benet. New York: Oxford University Press. .1979. "Women's Depiction by the Mass

Media." Signs: Journal of Women in Culture and Society 4:528-42.

van Zoonen, Liesbet. 1994. Feminist Media Studies. London: Sage.

Wardle, Jane and Louise Marsland. 1990. "Adolescent Concerns about Weight and Eating; A Social-Developmental Perspec- tive." Journal of Psychosomatic Research 34:377-91.

Willis, Paul. 1990. Moving Culture. London: Calouste Gulbenkian Foundation.

Wiseman, Claire V., James J. Gray, James E. Mosim-ann, and Anthony H. Ahrens. 1992. "Cultural Expectations of Thinness in Women: An Update." International Journal of Eating Disorders 11:85-89.

Melissa A. Milkie is an assistant professor of sociology at the University of Maryland whose work centers on culture and social identities. Currently she is examining meanings of parenting across generations.

This content downloaded from 155.247.166.234 on Sat, 3 Jan 2015 17:08:30 PM All use subject to JSTOR Terms and Conditions