The Role of Esteem-Relevance and Perceived Control in Determining the Effects of Physically Attractive Models in Advertising on Female and Male Adolescents

ABSTRACT - Previous work investigating the influence of physically attractive models on consumers has focused on one type of consumer response, cognitive-behavioral, and put less emphasis on affective responses. Some researchers argue that the two types of responses may occur simultaneously and in opposite directions, thus possibly explaining contradictory findings in previous literature. We investigate the constructs of esteem-relevance and perceived control over one’s physical attractiveness in an attempt to explain the contradictory results. The results of an experiment show perceived control did help explain differences in changes in affective responses across sexes, but did not explain changes in cognitive-behavioral responses.

Citation:

Mary C. Martin and James W. Gentry (1998) ,"The Role of Esteem-Relevance and Perceived Control in Determining the Effects of Physically Attractive Models in Advertising on Female and Male Adolescents", in E - European Advances in Consumer Research Volume 3, eds. Basil G. Englis and Anna Olofsson, Provo, UT : Association for Consumer Research, Pages: 82-89.

European Advances in Consumer Research Volume 3, 1998      Pages 82-89

THE ROLE OF ESTEEM-RELEVANCE AND PERCEIVED CONTROL IN DETERMINING THE EFFECTS OF PHYSICALLY ATTRACTIVE MODELS IN ADVERTISING ON FEMALE AND MALE ADOLESCENTS

Mary C. Martin, University of North Carolina at Charlotte, U.S.A.

James W. Gentry, University of Nebraska-Lincoln, U.S.A.

ABSTRACT -

Previous work investigating the influence of physically attractive models on consumers has focused on one type of consumer response, cognitive-behavioral, and put less emphasis on affective responses. Some researchers argue that the two types of responses may occur simultaneously and in opposite directions, thus possibly explaining contradictory findings in previous literature. We investigate the constructs of esteem-relevance and perceived control over one’s physical attractiveness in an attempt to explain the contradictory results. The results of an experiment show perceived control did help explain differences in changes in affective responses across sexes, but did not explain changes in cognitive-behavioral responses.

INTRODUCTION

Festinger’s (1954) social comparison theory has been revitalized in recent years to investigate several phenomena, including cnsumers’ comparisons of material possessions (Richins 1992) and consumers’ sensitivity to social comparison information (Bearden and Rose 1990). More recently, the theory has been used in an attempt to explain how highly attractive models in advertising may affect females and/or males (Martin and Gentry 1997; Martin and Kennedy 1993, 1994a, b; Richins 1991). The basic premise of these studies has been that consumers compare their physical attractiveness to that of models in ads and these comparisons may result in negative effects on their self-perceptions. The failure on the part of many to compare favorably has resulted in criticism of the promotion of an emaciated standard of "beauty" (Stephens, Hill, and Hanson 1994). These studies and others in the psychology literature (e.g., Thornton and Moore 1993), however, have not been able to develop a consistent set of findings to help us understand better the effects that advertising models may have on consumers. Further, these studies have focused on one type of consumer response (i.e., a cognitive-behavioral response) which hampers a full understanding of the complex reactions to social comparisons to advertising models.

The purpose of this study is to investigate the roles of "esteem-relevance" and "perceived control" in the social comparison process (Major, Testa, and Bylsma 1991). Specifically, these variables are viewed as key determinants of the consequences of comparisons to advertising models. Their incorporation into this stream of research may help us to explain the inconsistent findings in the literature and to understand more fully the potential consequences of comparisons to advertising models. The results of an experiment are presented which assess the effects of comparisons to advertising models by adolescent girls and boys.

DETERMINANTS OF THE CONSEQUENCES OF COMPARISONS

The Primary Appraisal Process

The notion of "esteem-relevance" is part of a primary appraisal process where one assesses his/her level of similarity to a comparison target and the self-relevance, or importance, of the dimension under evaluation. Comparisons with others who are perceived as similar on dimensions that are important are considered "esteem-threatening" because these comparisons have the most psychological impact on a person. On the other hand, comparisons with others who are perceived as dissimilar on dimensions that are unimportant are considered "esteem-irrelevant" and have little or no psychological impact on a person. The notion that comparisons on dimensions that are important or relevant to the self is supported by work by Tesser (1988) and his colleagues (Tesser, Millar, and Moore 1988).

The Secondary Appraisal Process

If a social comparison is considered esteem-threatening, a secondary appraisal process occurs where "assessment of his/her coping resources and options and his/her subsequent responses" occur (Major et al. 1991, p. 245). This secondary appraisal process is influenced by a person’s level of "perceived control" over the dimension being compared. The degree of perceived control one feels he/she has over the dimension being compared is determined by his/her 1) beliefs about the stability of that dimension, and 2) beliefs about his/her personal ability to maintain or change his/her status on that dimension. The importance of control in social comparison processes has been noted by several prior theorists (Crosby 1976; Singer 1981; Wills 1981). However, not until recently has its role been acknowledged and integrated into the literature on social comparison processes. For example, Testa and Major (1990) found that persons who believe that they ave little control over their subsequent outcomes experience greater negative affect (e.g., depression and hostility).

Major et al. (1991) propose further that subsequent responses to comparisons may be categorized as affective (e.g., self-esteem, anger, mood), cognitive (e.g., self-perceptions, attributions), or behavioral (e.g., self-improvement efforts). The distinction between the types of responses is necessary as these authors and others (e.g., Nadler and Fisher 1986; Wheeler and Miyake 1992) suggest that temporary changes in these responses may not be congruent or that only one type of response may occur. For example, comparisons to others who are better off on the dimension being compared ("upward" comparisons) may result in temporary negative affect (e.g., a lowering of self-esteem) and temporary positive cognitions (e.g., a raising of self-perceptions) simultaneously. As Wheeler and Miyake (1992, p. 770) write, "An upward comparison may be useful even if it results in immediate negative affect; one could feel simultaneously unhappy about an upward comparison and be motivated to do better or to be informed about how to do better."

One’s perceived control over his/her physical appearance will determine the type and level of response. Wheeler and Miyake (1992, p. 770) make a "seemingly reasonable assumption that in real life feelings of control are mixed," characterized by thoughts such as "I think I can control my status on this dimension, but then maybe I can’t." In this case, incongruent affective and cognitive-behavioral responses are likely to occur.

When perceptions of control are not mixed (i.e., one feels in control or one does not feel in control), only one type of response is likely. For example, when one feels in control of his/her physical appearance, comparisons to attractive-others are likely to result in increased cognitive-behavioral responses. On the other hand, when one does not feel in control of his/her physical appearance, comparisons to attractive-others are likely to result in decreased affective responses. The notion that one type of response may occur (e.g., a cognitive-behavioral response) while another may not (e.g., an affective response) has received support in the literature (e.g., Kenrick et al. 1993; Richins 1991; Thornton and Moore 1993). However, these studies did not incorporate the role of perceived control. The process is depicted in the Figure.

FIGURE

FRAMEWORK FOR DETERMINING THE CONSEQUENCES OF COMPARISONS

This process that Major et al. (1991) propose is very similar to Weiner’s (1985) attributional theory of achievement motivation and emotion. As Weiner (1985) suggests, physical appearance can be considered a cause of social acceptance or rejection. For example, one may attribute social rejection to his/her lack of physical attractiveness. Weiner (1985) proposes that perceived causes of success and failure have three properties: locus, stability, and controllability. Attributing events to specific causes will result in a set of unique affective and/or cognitive responses depending on one’s perception of locus, stability, and controllability. These unique responses occur as a result of a "primary appraisal" process and a subsequent causal attribution process. In the primary appraisal, an outcome is evaluated and a general positive or negative reaction occurs based on the perceived success or failure of the outcome (e.g., I am unattractive compared to the model in the ad so I feel sad). The subsequent attribution process involves seeking of a causal ascription (e.g., my lack of unattractiveness is due to my lack of effort or factors beyond my control such as genetics, so I feel angry). In fact, Major et al. (1991) allude to attribution processes playing a role in social comparison processes.

SOCIAL COMPARISONS TO ADVERTISING MODELS

The Primary Appraisal Process: Perceived Similarity and the Importance of Being Physically Attractive

As a first step in applying this framework within the context of consumers comparing their physical appearance to that of models in ads, the extent to which consumers perceive themselves similar to models and the extent to which they place importance on being physically attractive must be addressed (i.e., the primary appraisal process). Though these issues have not been addressed directly, research has provided evidence which suggests that females are likely to perceive comparisons to models in ads as esteem-threatening (i.e., they perceive themselves as similar to advertising models and they place great importance on being physically attractive).

First, research has found that females, including pre-adolescents, adolescents and college students, do compare their physical appearance to that of models in ads (Martin and Kennedy 1994a; Richins 1991). This would imply a level of perceived similarity (versus dissimilarity) with respect to physical attractiveness or "surrounding dimensions," those dimensions not being directly compared (Wood 1989) as a minimal level of similarity should be perceived for the comparison to take place at all.

Second, the domain of physical appearance has consistently been found to be a significant factor in determining self-esteem because the its importance is unable to be discounted by children and adolescents, particularly females (Harter 1986, 1993; Rosenberg 1986). The importance of being physically attractive is learned from a very early age and is reinforced throughout adolescence, particularly for females (Dion 1973, 1977; Harter 1993; Langlois and Stephan 1981; Langlois et. al. 1987). For example, girls learn that behavior, evaluations by others, and rewards are contingent upon this attribute (Dion 1977; Langlois and Stephan 1981).

The notion that girls perceive themselves as similar to advertising models and place great importance on being physically attractive can be inferred from perusing magazines targeted at girls such as Seventeen, Teen, YM, and Sassy. These magazines abound with contests and editorial material urging their readers to attempt to look like, to be like, or to even become a model (e.g., Teen magazine’s annual model search contest). Many advertisements in these magazines are for products which are supposed to help girls improve their physical appearance. In this respect, girls are given the sense that being a model is possible and desirable, hence perhaps increasing their perceived similarity to them, and their belief that being physically attractive is important is reinforced and/or strengthened.

Boys and men, on the other hand, do not face the same types of messages and probably do not have the same aspirations. Fischer and Halpenny (1993), for example, point out that it is unclear what may constitute an "ideal image" of physical attractiveness for men. The authors found that the idealized images of men in advertising do, in fact, differ from that of women and that men’s reactions to these images are different. In addition, boys face different norms for physical attractiveness and, at the same time, go through different pubertal processes than girls. For example, according to Franzoi (1995), females tend to view their bodies as "objects" where their physical beauty determines how they and others judge their overall value. Males, on the other hand, tend to view their bodies as "process" where power and function are more important criteria for evaluating their physical self. Consistent with this notion, Lerner, Orlos, and Knapp (1976) found that female adolescents’ self-concepts derived primarily from body attractiveness while male adolescents’ self-concepts were more strongly related to perceptions of physical instrumental effectiveness. This difference in body orientation results in a more culturally salient ideal of attractiveness for females (Franzoi 1995; Rozin and Fallon 1988) and a greater likelihood that females will be negatively impacted by the feminine ideal than men will be by the masculine ideal (Franzoi 1995).

In summary, it is likely tat girls perceive themselves as more similar to same-sex advertising models and place greater importance on being physically attractive than boys. Thus, hypothesis one is suggested.

H1: For girls, comparisons to models in ads are esteem-threatening while, for boys, comparisons to models in ads are esteem-irrelevant because:

a) girls perceive themselves as more similar to same-sex models in ads than boys.

b) girls place more importance on being physically attractive than boys.

The Secondary Appraisal Process: Stability of and Personal Control over Physical Appearance

If social comparisons to advertising models are perceived as esteem-threatening as a result of the primary appraisal process, a secondary appraisal process begins where girls may respond positively or negatively (or both) to those comparisons, depending on the extent to which they believe their physical appearance is stable and the extent to which they believe they can personally control or alter their physical appearance. Though Major et al. (1991) and Weiner (1985) suggest that physical appearance is a dimension which generally will be considered relatively stable and unlikely to change, we contend that in the context of advertising, this perception may not hold true, particularly for adolescents. In fact, what much of advertising suggests, either implicitly or explicitly, is that we can change our appearance (if we consume the advertised product). For example, Ogletree et al. (1990) found that the majority of commercials aired during Saturday morning cartoon programming for appearance-enhancement products was targeted towards girls. In addition, adolescence is a period when initial self-concepts are being formed, making the impact of social comparison information particularly important. For example, in explaining a lack of evidence for a lowering of self-perceptions after female college students were exposed to ads with highly attractive models, Richins (1991, p. 74) wrote:

The self-concept includes one’s beliefs about the level of one’s physical attractiveness, probably developed by examining oneself and comparing one’s appearance with others and learning the reactions of others to one’s appearance. Even children see ads with very attractive models, and this information is perhaps incorporated in forming a self-perception of attractiveness. By late adolescence, however, the sight of extremely attractive models is "old news" and unlikely to provide new information that might influence self-perception.

Further, as children grow older, they increasingly focus their comparisons on attributes they regard as personally important, such as physical appearance (Bers and Rodin 1984). Thus, the following hypotheses address girls’ perceptions of control and their subsequent responses to comparisons to models in ads. In addition, given that boys are likely to view comparisons to models in ads as esteem-irrelevant, their lack of subsequent responses is hypothesized.

H2: Young girls will exhibit higher levels of perceived control of their physical appearance than older girls.

H3: Sex and perceived control of physical appearance will interact to influence girls’ and boys’ affective and cognitive-behavioral responses to advertising:

a) For girls, comparisons to models in ads when feelings of perceived control are low will temporarily lower their affective responses.

b) For girls, comparisons to models in ads when feelings of perceived control are mixed will temporarily lower their affective responses and temporarily raise their cognitive-behavioral responses.

c) For girls, comparisons to models in ads when feelings of perceived control are high will temporarily raise their cognitive-behavioral responses.

d) For boys, comparisons to models in ads will have no effect on their affective or cognitive-behavioral responses.

METHODOLOGY

Overview of the Experiment

To test the proposed temporary effects, the experiment was a 2 (ad exposure) X 3 (level of perceived control) X 2 (sex) between-subjects design. The first factor represents the manipulation of the types of ads to which a subject was exposed. Either a subject was exposed to a series of ads with same-sex models or to a series of ads with no models. Given that social comparison did not take in those subjects who were exposed to ads with no models, a comparison of the mean responses of these two groups represents assessment of whether a response was temporarily "lowered" or "raised." The second factor, perceived control, was measured rather than manipulated. Subjects were divided into three groups (low, moderate or "mixed," and high) based on their perceived control scores. The final factor is categorical, representing the sex of the subject.

In a study to "learn about how people respond to advertising," subjects were exposed to a series of three ads. Prior to ad exposure, subjects completed the perceived control items. Following exposure to a series of ads, subjects gave ratings of self-esteem, mood, self-perceptions of physical attractiveness, and their intentions to make future efforts to improve their physical appearance. Multiple-item scales were summed to form overall measures of the variables.

Subjects

Female (n=226) and male (n=190) adolescents in grades five (n=143), seven (n=157), and nine (n=116) from a public school system in the Midwest participated in the study (total sample size of 416). The public school system is located in a county in which 98 percent of the population is white and the median family income is $31,144. As an incentive to participate, the subjects participated in a drawing for two prizes of $50.00 each. In addition, a $500.00 donation was made to the public school system. Parental permission was obtained for all subjects.

Subjects were administered the questionnaire at the schools, during an hour of class time with classroom teachers administering the questionnaires. The assignment to treatments was randomized by giving each classroom a random assortment of the two types of questionnaires. To facilitate understanding, the teachers administered the questionnaire orally by reading each survey question aloud and allowing appropriate time for the subjects to mark their responses. Though potential source effects may exist, teachers administered the uestionnaire because this is believed to have minimized any source effects as compared to having an unfamiliar authority figure collect the data. Further, this method of data collection was most convenient for the participating schools.

Advertising Stimuli

Four sets of three ads were chosen from magazines targeted at adolescent and young adult females and males: Teen, Seventeen, YM, Sassy, GQ, and Esquire. For girls, ads for skin cleanser, make-up, and jeans were included. For boys, ads for sunglasses, jeans, and exercise machines were included. For each sex, one set of three ads included same-sex models and the other set did not include models. Full-color copies were made of the ads and inserted into the survey booklets.

Measure of Perceived Similarity to Advertising Models

Similarity to advertising models was assessed with respect to age (one item), attitudes (one item), background (one item), values (one item), and physical appearance (three items) on seven-point scales (adapted from McCroskey, Richmond, and Daly 1975). For example, an item assessing perceived similarity with respect to physical appearance asked subjects to rate the model in an ad in comparison to themselves on a scale from "has a physical appearance unlike my physical appearance" to "has a physical appearance like my physical appearance." Given that subjects were exposed to three ads each, the scores for perceived similarity to each model were summed with respect to each dimension. This was done only for those subjects exposed to ads with models.

Measure of Importance of Physical Appearance

The importance of being physically attractive was assessed with two items adapted from the Harter (1985) scale. Subjects were asked to respond to "I think that how I look is important to how I feel about myself as a person" and "I think that it is important to be good looking in order to feel good about myself" on a seven-point scale ranging from "not at all true for me" to "very true for me."

Measure of Perceived Control

A measure of perceived control was used which assessed the subjects’ perceptions of the stability and alterability of their weight given that body figure perceptions (versus perceptions of individual body parts) constitute the central determinants of physical appearance (e.g., Rodin, Silberstein, and Striegel-Moore 1984). Two seven-point scale items were summed to derive an overall score of perceived control. Subjects were asked to respond to "my weight is very stable" (stability) and "by increasing the amount I exercise, I can lose weight" (personal control over change) on a scale from "not at all true for me" to "very true for me." The item assessing personal control over change was based on an item in the Dieting Beliefs Scale, a scale intended to assess the expectancy that one can affect or control one’s own weight (Stotland and Zuroff 1990). High perceived control is reflected in subjects’ perceptions that their weight is not stable and that they can personally control changes in their weight. The items were summed and then perceived control was categorized into three levels: low (scores from one to four), moderate/mixed (scores from five to ten), and high (scores from eleven to fourteen).

Affective Dependent Variables

Self-Esteem. Self-esteem is both an enduring, stable component of personality ("trait" self-esteem) as well as a variable subject to situational influences ("state" self-esteem). Given the nature of this study, it was necessary to employ self-esteem measures specific to the situatin of immediate interest (Thornton and Moore 1993). Thus, the appearance subscale of the State Self-Esteem Scale was used (Heatherton and Polivy 1991). This instrument is designed to measure momentary changes in self-esteem, with the appearance subscale most sensitive to manipulations that make physical appearance salient. For example, subjects were asked to respond to "I am pleased with my appearance right now" on a scale from "not at all true for me" to "very true for me." Six seven-point items were used. The reliability coefficient was .85.

Mood. Mood has been found to play a role in comparison processes. For example, Kenrick et al. (1993) found that subjects exposed to same sex photos of highly attractive persons showed lowered mood. Brown and Mankowski (1993) found that subjects with both high and low self-esteem evaluated themselves favorably when in a positive mood. However, when in a negative mood, subjects with low self-esteem lowered their self-evaluations. Four seven-point items assessed the subjects’ mood, from "not at all for me" to "very true for me" (The Mood Short Form, Peterson and Sauber 1983). For example, subjects were asked to respond to "As I answer these questions I feel very happy." The reliability coefficient was .72.

Cognitive-Behavioral Dependent Variables

Self-Perceptions of Physical Attractiveness. Three items measured self-perceptions of physical attractiveness. Subjects were asked to rate themselves in relation to members of their age group and sex on seven-point scales from "very unattractive, ugly" to "very attractive, beautiful (handsome)," from "very fit, in shape" to "very unfit, out of shape," and from "very thin" to "very fat" (adapted from Grubb et al. 1993). The reliability coefficient was .82.

Efforts to Improve Physical Appearance. A four-item scale assessed the subjects’ intentions to make future efforts to improve their physical appearance. For example, subjects were asked to complete a seven-point scale from "not at all true for me" to "very true for me" in response to "I plan to exercise in the future to lose weight." The reliability coefficient was .67.

TABLE 1

GIRLS' AND BOYS' PERCEIVED SIMILARITY TO ADVERTISING MODELS AND THE IMPORTANCE PLACED ON BEING PHYSICALLY ATTRACTIVE

RESULTS

Perceived Similarity and the Importance of Being Physically Attractive

To assess Hypothesis 1 which addressed the primary appraisal process, t-tests were conducted to assess whether girls perceive themselves as more similar to advertising models than boys with respect to age, attitudes, background, values, and physical appearance and whether girls place more importance on being physically attractive than boys. The results are presented in Table 1.

Girls perceived themselves as more similar to same-sex models than boys with respect to age (t₂₀₆=4.8, p<.01), values (t₂₀₆=2.8, p<.01), and physical appearance (t₁₅₄=2.2, p<.01). Girls, however, did not perceive themselves as more similar to same-sex models than boys with respect to attitudes (t₁₉₈=1.1, p>.05) or background (t₂₀₇=1.0, p>.05). The girls did place more importance on being physically attractive than boys (t₃₆₄=1.7, p<.05). Therefore, Hypothesis 1 was supported. Girls perceived themselves as more similar to same-sex models on most dimensions and they placed more importance on being physically attractive than boys.

Perceived Control in Younger versus Older Girls

To assess Hypothesis 2 which proposed that younger girls will exhibit higher levels of perceived control with respect to their physical appearance than older girls, a one-way analysis ofvariance was conducted with grade as the factor. Fifth-grade girls’ level of perceived control (mean=8.9, n=73) were compared to seventh-graders’ (mean=8.3, n=91) and ninth-graders’ (mean=8.8, n=59). There were no differences in perceived control across grades (F_{2, 220}=1.40, p>.05). Therefore, Hypothesis 2 was not supported. Had we included younger females, we might have found greater perceived control. Given pubertal development is beginning around fifth grade (e.g., onset of menarche) as was indicated by measures of pubertal development, the fifth-grade girls may have experienced lessened perceived control.

Effects of Comparisons on Affective and Cognitive-Behavioral Responses

To assess Hypothesis 3 which addressed the secondary appraisal process, three-way analyses of variance were conducted for each dependent variable. It was hypothesized that affective and cognitive-behavioral responses to models in ads will vary across levels of perceived control and that this will vary according to sex. Support for each hypothesis would be demonstrated by a significant three-way interaction and subsequent significance differences between those who saw ads with models and those who saw ads with no models across levels of perceived control for each sex. Thus, significant three-way interactions were followed by two-way analyses of variance conducted for each sex. If a significant interaction was found between ad exposure and level of control within each sex, simple main effects tests for ad exposure type within levels of perceived control were conducted (Keppel 1991). The regression approach to partitioning variance was used to control for nonorthogonality due to unequal sample sizes (Keppel 1991).

Self-Esteem. The three-way interaction was significant when self-esteem served as the dependent variable (F_{2, 384}=4.3, p<.05). Two-way analyses of variance were then conducted for each sex. For females, the ad exposure X level of control interaction was not significant (F_{2, 212}=.88, p>.05). For males, the ad exposure X level of control interaction was significant (F_{2, 172}=3.9, p<.05). Subsequent simple main effects tests indicated a significant ad exposure within high control effect. That is, for those boys who perceive a high level of control over their physical appearance, their self-esteem was significantly lower when exposed to ads with same-sex models (mean=21.9) than when exposed to ads with no models (mean=32.2). Thus, Hypothesis 3 was not supported.

Mood. The three-way interaction was significant when mood served as the dependent variable (F_{2, 389}=4.6, p<.05). Two-way analyses of variance were then conducted for each sex. For females, the ad exposure X level of control interaction was significant (F_{2, 214}=3.1, p<.05). Subsequent simple main effects tests indicated a significant ad exposure within high control effect. That is, for those girls who perceive a high level of control over their physical appearance, their mood was significantly lower when exposed to ads with same-sex models (mean=18.4) than when exposed to ads with no models (mean=21.8). For males, the ad exposure X level of control interaction was also significant (F_{2, 175}=8.4, p<.01). Subsequent simple main effects tests indicated a significant ad exposure within low control effect and a significant ad exposure within high control effect. That is, for those boys who perceive a low level of control over their physical appearance, their mood was significantly lower when exposed to ads with same-sex models (mean=16.1) than when exposed to ads with no models (mean=24.3). Similarly, for those boys who perceive a high level of control over their physical appearance, their mood was significantly lower when exposed to ads with same-sex models (mean=15.3) than when exposed to ads with no models (mean=22.1). Thus, Hypothesis 3 was not supported. The results for affective responses are presented in Table 2.

TABLE 2

EFFECTS OF COMPARISONS ON GIRLS' AND BOYS' AFFECTIVE RESPONSES

Self-Perceptions of Physical Attractiveness. The three-way interaction when self-perceptionsof physical attractiveness served as the dependent variable was not significant (F_{2, 376}=1.5, p>.05). Self-perceptions were relatively constant across age, sex, and level of perceived control and Hypothesis 3, therefore, was not supported.

Efforts to Improve Physical Appearance. The three-way interaction when efforts to improve physical appearance served as the dependent variable was not significant (F_{2, 379}=1.6, p>.05). Efforts to improve were relatively constant across age, sex, and level of perceived control and Hypothesis 3, therefore, was not supported.

DISCUSSION

The results of this study suggest that females, according to the Major et al. (1991) framework, perceive comparisons to models in ads as esteem-threatening while boys perceive them as less esteem-threatening or esteem-irrelevant. Girls perceived greater similarity to the models on most dimensions measured and they placed greater importance on being physically attractive than boys and Hypothesis 1, thus, was supported. Hypothesis 2, however, was not supported. Younger girls did not perceive themselves as being in more control of their physical appearance than older girls.

The results of this study also suggest that perceived control over one’s physical attractiveness does moderate adolescents’ affective reactions to ads, but not in the direction proposed by the Major et al. (1991) framework. In addition, no influence on cognitive-behavioral responses was found. Thus, Hypothesis 3 was not supported. This hypothesis suggested that those with low perceived control over their physical attractiveness would suffer decreased self-esteem and lowered mood responses; this was supported only with boys and mood as the dependent variable. However, those boys with high perceived control had lowered mood and self-esteem and those girls with high perceived control had lowered mood after seeing same-sex models in ads. It may be that lower perceptions of control tend to make the adolescents immune to any negative affective responses to the ads. On the other hand, the ads may serve as a reminder (threat) to those with high perceived control so that they perceive that they must put forth effort to compete with the attractive models. The presence of attractive female models may not be required to lower self-esteem since there may have been a learned association between the stimuli (products advertised) and the use of attractive models.

Clearly, more research should be conducted to investigate the Major et al. (1991) framework and its implications for advertisers. This study, for example, focused on the stability and ability to control one’s weight; the effect of one’s perception of the stability and/or changeability of other aspects of physical attractiveness, including facial appearance, for example, should be investigated.

The reasons for differing perceptions of perceived control over one’s physical appearance should also be investigated. As we suggested earlier, perhaps the pubertal development that a girl or boy is experiencing influences her/his perceptions of control. Physical changes, such as the onset of menarche or the changing of one’s voice, certainly would be regarded as beyond one’s control.

The pervasiveness and differing nature of messages targeted to girls and boys likely have an effect on their perceptions of control as well. Girls, in particular, face advertising messages which emphasize the need to improve one’s appearance. But, how does an adolescent female cope with a changing body (i.e., gaining weight as result of puberty) at the same time while she is facing messages which emphasize improving one’s body (i.e., losing weight)? Given this question, a concern arises as to the extent and manner in which adolescents should be educated about advertiser tactics. The Persuasion Knowledge Model (KM) proposed by Friestad and Wright (1994) provides a useful starting point for addressing this concern. According to Friestad and Wright (1994, p. 19), an effective program for educating children and adolescents about the tactics of advertisers would include "a full analysis of preexisting persuasion knowledge and of the knowledge acquisition process." Thus, educating adolescents about how to cope with attractiveness-related messages in advertising would first entail an analysis of what they already know with respect to attractiveness-related messages (e.g., photographs of models are air-brushed) and perceptions of control (e.g., I can lose weight through exercising) and what social comparison strategies or motives they use (e.g., self-evaluation, self-improvement, or self-enhancement). Then, they must "figure out (or be coached on) what types of cognitive or emotional actions they might perform when they notice a particular tactic being used" (Friestad and Wright 1994, p. 19). In this context, therefore, educational efforts would involve teaching adolescents realistic perceptions of control, realistic comparison standards, and how to cope with attractiveness-related messages in terms of their cognitive-behavioral and affective responses in light of those realistic perceptions of control and comparison standards. Successful educational efforts would result in fewer adolescents experiencing negative reactions to highly attractive models in ads.

REFERENCES

Bearden, William O. and Randall L. Rose (1990), "Attention to Social Comparison Information: An Individual Difference Factor Affecting Consumer Conformity," Journal of Consumer Research, 16 (March), 461-471.

Bers, Susan A. and Judith Rodin (1984), "Social Comparison Jealousy: A Developmental and Motivational Study," Journal of Personality and Social Psychology, 47, 766-779.

Brown, Jonathon D. and Tracie A. Mankowski (1993), "Self-Esteem, Mood, and Self-Evaluation: Changes in Mood and the Way You See You," Journal of Personality and Social Psychology, 64 (3), 421-430.

Crosby, F. (1976), "A Model of Egoistical Relative Deprivation," Psychological Review, 83, 84-113.

Dion, Karen (1973), "Young Children’s Stereotyping of Facial Attractiveness," Developmental Psychology, 9 (2), 183-188.

Dion, Karen (1977), "The Incentive Value of Physical Attractiveness for Young Children," Personality and Social Psychology Bulletin, 3, 67-70.

Fischer, Eileen and Kate Halpenny (1993), "The Nature and Influence of Idealized Images of Men in Advertising," in Gender and Consumer Behavior, Proceedings of the Second Conference, ed. Janeen Arnold Costa, Salt Lake City, UT: University of Utah Printing Service, 196.

Franzoi, Stephen L. (1995), "The Body-As-Object Versus The Body-As-Process: Gender Differences and Gender Considerations," Sex Roles, 33 (5/6), 417-437.

Friestad, Marian and Peter Wright (1994), "The Persuasion Knowledge Model: How People Cope with Persuasion Attempts," Journal of Consumer Research, 21 (June), 1-31.

Grubb, Henry J., Marie I. Sellers, and Karen Waligroski (1993), "Factors Related to Depression and Eating Disorders: Self-Esteem, Body Image, and Attractiveness," Psychological Reports, 72, 1003-1010.

Harter, Susan (1986), "Processes Underlying the Construction, Maintenance, and Enhancement of the Self-Concept in Children," in Psychological Perspectives on the Self, Vol. 3, eds. J. Suls and A.G. Greenwald, Hillsdale, NJ: Lawrence Erlbaum Associates,137-181.

Harter, Susan (1993) "Visions of Self: Beyond the Me in the Mirror," in Developmental Perspectives on Motivation, Nebraska Symposium on Motivation, Vol. 40, ed. Janis E. Jacobs, Lincoln, NE: University of Nebraska Press, 99-144.

Heatherton, Todd F. and Janet Polivy (1991), "Development and Validation of a Scale for Measuring State Self-Esteem," Journal of Personality and Social Psychology, 60 (6), 895-910.

Kenrick, Douglas T., Daniel R. Montello, Sara E. Gutierres, and Melanie R. Trost (1993), "Effects of Physical Attractiveness on Affect and Perceptual Judgments: When Social Comparison Overrides Social Reinforcement," Personality and Social Psychology Bulletin, 19 (2), 195-199.

Keppel, Geoffrey (1991), Design and Analysis: A Researcher’s Handbook, Third Edition, Englewood Cliffs, NJ: Prentice Hall.

Langlois, Judith H., Lori A. Roggman, Rita J. Casey, Jean M. Ritter, Loretta A. Rieser-Danner, and Vivian Y. Jenkins (1987), "Infant Preferences for Attractive Faces: Rudiments of a Stereotype?" Developmental Psychology, 23 (3), 363-369.

Langlois, Judith H. and Cookie White Stephan (1981), "Beauty and the Beast: The Role of Physical Attractiveness in the Development of Peer Relations and Social Behavior," in Developmental Social Psychology: Theory and Research, eds. S.S. Brehm, S.M. Kassin, and F.X. Gibbons, New York: Oxford University Press, 152-168.

Lerner, Richard M., James B. Orlos, and John R. Knapp (1976), "Physical Attractiveness, Physical Effectiveness, and Self-Concept in Late Adolescents," Adolescence, 11 (43), 313-326.

Martin, Mary C. and James W. Gentry (1997), "Stuck in the Model Trap: The Effects of Beautiful Models in Ads on Female Pre-Adolescents and Adolescents," Journal of Advertising, 26 (Summer), 19-33.

Martin, Mary C. and Patricia F. Kennedy (1993), "Advertising and Social Comparison: Consequences for Female Pre-Adolescents and Adolescents," Psychology and Marketing, 10 (November/December), 513-530.

Martin, Mary C. and Patricia F. Kennedy (1994a), "Social Comparison and the Beauty of Advertising Models: The Role of Motives for Comparison," in Advances in Consumer Research, (Vol. 21), eds. C.T. Allen & D. Roedder John, Provo, UT: Association for Consumer Research, 365-371.

Martin, Mary C. and Patricia F. Kennedy (1994b), "The Measurement of Social Comparison to Advertising Models: A Gender Gap Revealed," in Gender and Consumer Behavior, ed. Janeen Arnold Costa, Beverly Hills, CA: Sage Publications, 104-124.

Major, Brenda, Maria Testa, and Wayne H. Bylsma (1991), "Responses to Upward and Downward Social Comparisons: The Impact of Esteem-Relevance and Perceived Control," in Social Comparison: Contemporary Theory and Research, eds. Jerry Suls and Thomas Ashby Wills, Hillsdale, NJ: Lawrence Erlbaum Associates, 237-260.

McCroskey, James C., Virginia P. Richmond, and John A. Daly (1975), "The Development of a Measure of Perceived Homophily in Interpersonal Communication," Human Communication Research, 1, 323-332.

Ogletree, Shirley M., Sue W. Williams, Paul Raffeld, Bradley Mason, and Kris Fricke (1990), "Female Attractiveness and Eating Disorders: Do Children’s Television Commercials Play a Role?" Sex Roles, 22 (11/12), 791-797.

Peterson, Robert A. and Matthew Sauber (1983), "A Mood Scale for Survey Research," in American Marketing Association Educators’ Proceedings, eds. Patrick Murphy et al., Chicago, IL: American Marketing Association, 409-414.

Richins, Marsha L. (1991), "Social Comparison and the Idealized Images of Advertising," Journal of Consumer Research, 18 (June), 71-83.

Richins, Marsha L. (1992), "Media Images, Materialism, and What Ought to Be: The Role of Social Comparison," in Meaning, Measure, and Morality f Materialism, eds. F. Rudmin and M. Richins, Kingston, Ontario: Research Workshop on Materialism and Other Consumption Orientations, 202-206.

Rodin, Judith, Lisa Silberstein, and Ruth Striegel-Moore (1984), "Women and Weight: A Normative Discontent," in Nebraska Symposium on Motivation, Vol. 32, ed. T.B. Sonderegger, Lincoln, NE: University of Nebraska Press, 267-307.

Rosenberg, Morris (1986), "Self-Concept From Middle Childhood Through Adolescence," in Psychological Perspectives on the Self, Vol. 3, eds. J. Suls and A.G. Greenwald, Hillsdale, NJ: Lawrence Erlbaum Associates, 107-136.

Rozin, P. and A.E. 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, 342-345.

Singer, E. (1981), "Reference Groups and Social Evaluations," in Social Psychology: Sociological Perspectives, eds. M. Rosenberg and R.H. Turner, New York: Basic Books, 66-93.

Stephens, Debra Lynn, Ronald Paul Hill, and Cynthia Hanson (1994), "The Beauty Myth and Female Consumers: The Controversial Role of Advertising," The Journal of Consumer Affairs, 28 (Summer), 137-153.

Stotland, Stephen and David C. Zuroff (1990), "A New Measure of Weight Locus of Control: The Dieting Beliefs Scale," Journal of Personality Assessment, 54 (1 & 2), 191-203.

Tesser, Abraham (1988), "Toward a Self-Evaluation Maintenance Model of Social Behavior," in Advances in Experimental Social Psychology, Vol. 21, ed. L. Berkowitz, San Diego, CA: Academic Press, 181-227.

Tesser, Abraham, Murray Millar, and Janet Moore (1988), "Some Affective Consequences of Social Comparison and Reflection Processes: The Pain and Pleasure of Being Close," Journal of Personality and Social Psychology, 54 (1), 49-61.

Testa, Maria and Brenda Major (1990), "The Impact of Social Comparisons After Failure: The Moderating Effects of Perceived Control," Basic and Applied Social Psychology, 11 (2), 205-218.

Thornton, Bill and Scott Moore (1993), "Physical Attractiveness Contrast Effect: Implications for Self-Esteem and Evaluations of the Social Self," Personality and Social Psychology Bulletin, 19 (4), 474-480.

Weiner, Bernard (1985), "An Attributional Theory of Achievement Motivation and Emotion," Psychological Review, 92 (4), 548-573.

Wheeler, Ladd and Kunitate Miyake (1992), "Social Comparison in Everyday Life," Journal of Personality and Social Psychology, 62 (5), 760-773.

Wills, T.A. (1981), "Downward Comparison Principles in Social Psychology," Psychological Bulletin, 90, 245-271.

Wood, Joanne V. (1989), "Theory and Research Concerning Social Comparisons of Personal Attributes," Psychological Bulletin, 106 (2), 231-248.

Wood, Joanne V. and Kathryn L. Taylor (1991), "Serving Self-Relevant Goals Through Social Comparison," in Social Comparison: Contemporary Theory and Research, eds. Jerry Suls and Thomas Ashby Wills, Hillsdale, NJ: Lawrence Erlbaum Associates, 23-49.

----------------------------------------