Eating disorders and advertising effects: an exploration.

Peterson, Robin ; Jun, Minjoon

ABSTRACT

The purpose of this manuscript is to describe a study designed to furnish insights into various relationships between advertising and anorexia nervosa and bulemia among university students. The inquiry measured self image and ideal self image and correlated these assessments to the occurrence of the eating disturbances and to measures of advertising and merchandising. The results led to a number of recommendations for marketers who choose to assist in preventing and abating eating disorders.

INTRODUCTION

Marketing and consumer researchers have devoted considerable effort in search of serviceable answers for pressing societal difficulties (Porter & Kramer, 2002; Christie, Fisher, Kozup, & Smith, 2001; Wagner & Hansen, 2002; Murphy, 1998). A significant proportion of this endeavor has applied conventional research methodologies to macro-level questions (Rentschler & Wood, 2001; Karande, Rao, & Singhapakdi, 2002). Another field of concentration for researchers has been micro-issues facing organizations with societal goals. Hibbert, Piacentini and Al Dajani (2003), for instance have examined volunteer motivation for participation in community-based food cooperatives. Carrigan and Attalla (2001) and Sen and Bhattacharya (2001) have focused on whether or not customers care about ethical behavior and socially-responsible actions to the degree that they significantly affect their purchasing behavior. In a somewhat similar study, Miles and Covin (2000) considered the relationship between environmental marketing and reputational, competitive, and financial advantages accruing to the firm.

A considerable number of research endeavors have concentrated on the effects of advertising and other promotion vehicles on phenomena that are related to social responsibility (Bloch & Richins,1992). Kreth (2000), for instance, has examined the degree to which marketers exercise social responsibility in making advertised health claims for dietary supplements (Pechmann, Zhao, Goldberg, & Reibling, 2003) have conducted research to assist in identifying effective message themes to be used in antismoking advertisements for adolescents. Peterson (2002) considered the extent and degree of favorableness of the depiction of African American Childrens' activities in television commercials. Odekerken-Schroder, De Wulf, and Hofstee (2002) have examined the extent to which gender stereotyping in advertising is related to the incidence of masculinity in the culture of various countries.

A social responsibility issue which marketing and consumer behavior researchers have largely disregarded is the widespread existence of eating disorders among young people, especially females. Research indicates that this is a major social problem (Hoskins, 2002; Johnson & Larson, 1982). Large numbers of individuals have fallen prey to anorexia nervosa and bulemia. Research estimates vary, but generally they suggest that nearly twenty percent of the individuals in the 17-20 age grouping are victims of the two eating disorders, while other age groupings possess smaller percentages (Steiger, Bruce, & Israel, 2003; Jacobson, 1985). Increasingly, larger numbers of college, high school, middle school, and even elementary school students are becoming victims (Schur, Sanders, & Steiner, 2000). The largest proportion, however, falls in the college-age female category (Schwitzer, Bergholz, & Terri, 1998). The effects on the emotional, physical, and mental health of these individuals and their families have been considerable (DiConsiglio, 2000).

Research has been undertaken in recent periods on the underlying conditions, causes, main victims, and potential cures for these disorders. Social pressure to be thin has emerged as a potential causal factor (Stice, Maxfield, & Wells, 2003). Other studies have uncovered social dependency, need for approval, and fear of rejection as influences (Hayaki, Friedman, Whisman, Delinsky, & Brownell, 2003). Significant correlations between volume of exercise, vegetarianism, low esteem, and stress, on the one hand, and the two eating disorders, on the other, have been found (McLean & Barr, 2003). In addition, studies have revealed an association between perfectionism motives and anorexia/bulemia (Sutandar-Pinnock, Woodside, Carter, Olmsted, & Kaplan, 2003). There is some evidence that body image distress is a factor (Devlin & Zhu, 2001). Eating disturbances are as common among Hispanic and African Americans as they are with Caucasians (Walcott, Pratt, & Patel, 2003). Further, numerous female collegiate athletes are subject to these disturbances (Beals, & Manore, 2002).

Exposure to outside influence appears to be a possible contributor to the two afflictions. The amount of television viewing is such a factor(Kennedy, 2000). Another is readership of health, sports, and fitness magazines, where young people are able to make social comparisons and develop what they feel are critical body image perceptions (Botta, 2003). Further, there is a demonstrated and significant relationship is between the two disturbances and exposure to advertising (Zucker, Harrell, Miner, Kathi, & Abigail, 2002).

Overall, the studies suggest that the two eating disorders are most common among the white middle and upper classes. Numerous victims are overachievers who excel in academics, sports, and other activities. Many are overactive and are heavily involved in recreational, work, and social activities. In turn, a large number have obsessive and low esteem personality traits, are subject to considerable stress, have negative images of their bodies, and are perfectionist in their appearance and performance. Many anorexics and bulemics are well-groomed, attractive, and independent, and have achieved professional success. They tend to be avid readers of health and fitness magazines and are heavily exposed to the media, especially television.

The bulk of the studies relating to eating disorders have emanated from the medical and social science literature (Botta, 2003). Researchers in consumer behavior and marketing have not addressed this problem in depth. However, it would appear that there is need for studies emanating from these disciplines. It would be useful, for instance, to examine the degree of influence on the formation and maintenance of eating disorders that is brought about through advertising. specifically, could use of emaciated models in advertising create desires for body images and in turn lead to bulemia and anorexia? Do advertisers portray ideal self physical images that are not realistic for most young women and children? Answers to questions such as these could be of value to those who seek prevention and treatment of the eating disturbances. The study set forth here was undertaken to furnish insights into these issues.

RESEARCH OBJECTIVES

The broad aim of the research was to furnish insights for advertisers, advertising agencies, governmental authorities, and other parties who have a vested interest in constraining potential advertising influences on eating disorders. Three specific objectives flow from this aim. These are:

1. To appraise the self images and ideal self images of college juniors and seniors who are experiencing anorexia nervosa and bulemia and to compare these with those who are not experiencing these disorders.

Self image is a meaningful variable for study since the "heavy" or "thin" self image held by individuals is closely related to the probability that they will contract an eating disorder (Botta, 2003).

2. To compare gauges of promotion exposure, shopping exposure, weight consciousness, and the effect of advertising on the desire to be thin between the two groups.

This objective could be useful in producing insights on the effect of advertising and merchandising efforts on the development of eating disorders.

3. Employing the self image and ideal self image measures, to suggest guidelines to advertisers and other parties who are motivated to reduce possible advertising effects on the incidence and maintenance of eating disorders.

NULL HYPOTHESES TESTED

The present research was conducted in a manner which permitted testing of the following null hypotheses:

H1: There is not a significant difference in (a) magazine advertising
 exposure, (b) television advertising exposure, and (c) clothing
 shopping exposure between (a) college students who are victims of
 anorexia and bulemia and (b) students who are not victims.

Research suggests that victims of the eating disorders tend to be heavy magazine readers, especially for health and fitness magazines (Thomsen, Weber, & Brown, 2001; Botta, 2003). Further, there is evidence that victims are heavy viewers of television and television advertisements (Kennedy, 2000). In addition, studies indicate that victims spend considerable time shopping for clothing (Leon, Carroll, Chernyk, & Finn, 1984). Thus, there is theoretical support for the alternative hypotheses.

H2: There is not a significant difference in the reinforcement by
 advertisements of the desire to be thin between the two groups.

Some studies have found that the popular media place substantial emphasis on thinness as an ideal of feminine beauty (Owen, & Laurel-Seller, 2000). Further, high risk subjects-those who are more likely to develop eating disorders-are more vulnerable to advertisements which glorify thinness than are those who are less likely to become victims (Zucker, Harrell, Miner, Kathi, & Abagail, 2001; Toro, Salamero, & Martinez, 1994).

H3: There is not a significant difference in concern with personal
 appearance between the two groups.

Past research has suggested that individuals with eating disorders are more likely than those without these disorders to be highly concerned with their personal appearance (McLean & Barr, 2003). They are more likely, for instance, to become depressed when they are exposed to videos stressing ideal personal appearance than are those without eating disorders (Heinberg & Thompson, 1995). These findings suggest a relationship between anorexia and bulemia, on the one hand, and concern with personal appearance, on the other.

H4: There is not a significant difference between the self images of
 the two groups.

Self images exert an important impact on behavior. In this regard, various studies suggest that members of the two groups possess divergent self images. Those with the eating disorders, for instance, tend to possess a higher degree of body image distress (Devlin & Zhu, 2001). Further, they tend to experience lower self esteem (McLean & Barr, 2003). In general, preadolescent girls show lower body liking and higher body dissatisfaction than boys (Roncolato, Huon & Braganza, 1998). These differences could account for some of the incidence of anorexia and bulemia.

H5: There is not a significant difference between the ideal self
 images of the two groups.

Some research suggests that those with eating disorders possess ideal self images that are based more on degree of thinness than is the case for those without these disturbances. (Thomsen, Weber, & Brown, 2001; Neumark-Sztainer, 1996). In fact, those with the disorders often have conceptions of body image that is nothing short of perfectionism (Sutander-Pinnock, Woodside, Carter, Olmsted, & Kaplan, 2003). Hence, it is possible that ideal self image may be related to anorexia and bulemia.

H6: The gaps between the self image and the ideal self images are not
 greater for those with eating disorders than those without these
 disorders.

If this hypothesis is rejected, there is reason for advertisers to reject themes, appeals, and other copy that might create gaps between the two types of self image. When consumers perceive these voids, there is a strong probability that they will experience anxiety and attitude change toward body image. Studies have revealed that the gaps between the two groups may be significant (Posavac & Posavac, 2002; Myers, & Biocca, 1992).

The hypotheses set forth above relate to the aforementioned objectives of the study. The subsequent section describes the manner in which the hypotheses were assessed.

THE RESEARCH STUDY

Cover letters and questionnaires were forwarded to a random sample of 50 marketing professors who were listed in the American Marketing Association membership roster. Each professor received thirty questionnaires and a request to ask their students to complete these and return them to the instructor. A total of 39 professors participated in the study, generating a sample size of 1,073 complete and otherwise usable questionnaires.(Some instructors did not obtain questionnaires from all thirty students. Others requested additional questionnaires so that an entire class could participate in the study). After they had secured the completed questionnaires, the instructors mailed them to the researcher. The sample was made up of 67.5% business majors and 32.5% non-business majors. As to class ranking, 33.8% were juniors, 49.1% seniors, 11.6% sophomores, and 5.5% freshmen.

The questionnaires requested that students specify (a) if they were experiencing bulemia, (b) if yes, how many purges did they carry out in an average week, (c) if they were experiencing anorexia nervosa, (d) if yes, an estimate of how many calories they consumed on an average day. Definitions of bulemia and anorexia nervosa were provided in the questionnaire.

The students were informed that they should not reveal their identities and that the questionnaires would be held in strict confidence, in order to minimize possible bias on the requested self-reports of the incidence of these disorders. The students were also requested to respond to fifty bipolar adjective scale to assess their self images and again to the same scales to assess their ideal self images. The adjective pairs were chosen and validated in a pretest study using the procedures developed by Brinton (1961) employing an initial list of 220 adjective pairs selected from Osgood, Suci, and Tannenbaum (1957). The same adjective pairs were used in a previous study of eating disorders (Peterson, 1987).

Subsequent portions of the questionnaire asked the students various questions relating to their exposure to elements of the marketing mix and their attitudes and related variables. The respondents were asked (a) how many hours per week they watched television, (b) how many magazines they read per week, as measures of advertising exposure, (c) how many times per week they shopped for clothing, as a measure of shopping exposure to items related to personal appearance, (d) if they believed that the advertisements they perceived created an image of desire to be thin, to the extent that they attempted to control their weight. This was a measure of the impact of advertisements on attitudes and behavior. and (e) the degree to which they were concerned with their personal appearance.

THE RESEARCH RESULTS

Table One presents the percentages of response to the questions asking if the students were experiencing anorexia nervosa or bulemia. It is apparent that 5.9% had bulemia, 3.6% had anorexia, and 2.1% had both. The proportions for eating disorders are approximately six percent larger than those produced by a similar study of university students in a 1987 study (Peterson, 1987), signaling a possible increase in the incidence of the disturbances since the earlier time period.

Of those students citing bulemia as a condition, the mean number of purges per week was 5.1. This is slightly higher than the results of earlier research into this eating disorder (Chiodo & Latimer, 1983; Peterson, 1987). In turn, students in the sample with anorexia reported a mean number of 917 calories consumed per average day. This figure is slightly lower than that set forth in earlier studies (Duddle, 1973). While exact caloric needs vary, depending upon factors such as weight, age, and level of physical activity, 917 is well-below the amount required to maintain physical health.

The questionnaire requested the sample members to indicate how many magazines they read in an average week. The mean for the eating disorder group was 1.7 and for the non-disorder group 1.5. The differences were not significant at the .05 level, according to a t test. Hence, null hypothesis 1A is supported by the findings.

The subjects were asked how many hours per week they watched television, on the average. Those with eating disorders produced a mean of 9.7, while those without an eating disorder generated a mean of 9.3. The differences between the two means are not significant, according to a t test at the .05 level. It appears that exposure to this advertising medium is not related to the incidence of the two disorders. The results provide support for null hypothesis 1B. The mean figures for hypotheses 1A and 1B do not support the concept that exposure to advertising may contribute to eating disorders.

The respondents were requested to specify how frequently they shopped for clothing during an average week. This was designed to measure exposure to merchandising efforts The mean for the disorder group was .6 and for the non-disorder group .4. The difference is not significant at the .05 level, according to a t test, which supports null hypothesis 1C

The subjects were asked "Do you believe that the ads you see and hear create strong images of the desirability of being thin, thereby inducing you to keep weight down?" Ninety percent of those with eating disorders answered "yes" to this question, while only sixty one percent of those without eating disorders responded in the affirmative. The difference is statistically significant at the .05 level, according to a t test. These results do not support null hypothesis 2.

It appears that those with eating disorders are not more exposed to television and magazine advertisements than those without the disorders, but that they respond in a more robust manner. These findings suggest that cutting back on the volumes of advertising would not necessarily be effective in combating bulemia and anorexia, but making the ads less suggestive of the desirability of being thin might.

The questionnaire asked the subjects how concerned they were about their personal appearance. Table Two sets forth the results. A Chi Square test for homogeneity suggests that the two samples were drawn from different populations, with a Chi Square value of 12.97, which is significant at the .05 level. Further, the weighted mean of those with eating disorders (weighing "Very concerned" as 3, "concerned" as 2, and "neutral" as 1) produces a mean of 2.7 for those with eating disorders and 2.1 for those without eating disorders. These means are significantly different at the .05 level, according to a t test. The results lead to a rejection of null hypothesis three. This suggests that anxiety regarding personal appearance, raised through advertisements, could be much more persuasive among the former than among the latter group.

Table Three presents the mean self image scores of the students with and without eating disorders on bipolar adjectives scales, where the first scale has a weight of one (very important) and the last a weight of seven (not very important). Those adjective pairs which showed significant differences between the two groups, according to a t test at the .05 level are labeled with an asterisk. Based upon the differences it is possible to construct a profile of how eating disorder victims in the sample differed from those without disorders on the various self image components. Generally, those with eating disorders perceive themselves as more intelligent, mature, grateful, compulsive, nervous, persevering, painstaking, easily jealous, neurotic, hurried, secretive, dissatisfied, and successful than those without the disorders. As there is a large number of significant differences, null hypothesis four cannot be accepted.

Table Four presents the ideal self image means for the two groups. Those with eating disorders pose their ideal self images as more assertive, intelligent, independent, friendly, persevering, thoughtful, hurried, quick, sophisticated, valuable, important, and successful than do those without an eating disorder. On the other hand, those without eating disorders were more likely to hold an ideal self image which emphasized modesty, maturity, blame others, soft hearted, adventuresomeness, not-neurotic, content, patient, and good. Since there are a large number of significant differences, null hypothesis five cannot be supported.

The results set forth above give the impression that there may be substantial differences between the components of the self image and ideal self image between the two groups. This suggests that they might respond in a different manner to advertising appeals, themes, formats, illustrations, and featured products. The differences in the self images and ideal self images also furnish opportunities for targeting consumers who do not have eating disturbances and avoiding those who do, in advertising messages. If advertisers focus their promotions on those whose self images conform to the no-disorder configuration presented in the table and avoid the disorder configuration for products and services which suggest the desirability of slimness, the probability of encouraging eating disorders may be mitigated.

The analysis included a comparison of the respondents' self image and ideal self image scores, in order to identify gaps between these two variables-identifying self image dimensions which were subject to variety or uncertainty. Consumers can be expected to respond rigorously to advertising appeals which relate to components of the self image that differ substantially from the ideal self image (Devlin & Zhu, 2001). Anxiety often arises when the self image and ideal self image vary to a large degree. One whose self image is "impatient", for instance, but whose ideal self image is "patient" may respond strongly to a promotion portraying very thin members of the same gender who are depicted as being patient. Thus, responsible marketers might choose to avoid undue emphasis on the nervousness dimension of those depicted in their promotions and instead concentrate on other components that are less likely to arouse anxiety.

Table Five presents self image and ideal self image mean scores for those with and those without eating disturbances. It also sets forth the difference between each self image and ideal self image mean score for each bipolar adjective scale. Those differences (or "D" scores) measure the gap between actual and ideal self images. Responsible advertisers are well advised to avoid themes and appeals for variables with high D scores and employing those with low D scores, in order to combat advertising-induced anxiety about slimness and weight control.

Various pairs of adjectives generated mean D scores for the eating disorder group that were significantly larger than the corresponding values for the non-eating disorder category, according to t tests at the .05 level. These were impulsive-deliberate, blame mistakes others-blame mistakes me, inflexible-adaptable to change, hardhearted-softhearted, easily jealous-not easily jealous, emotional- unemotional, impatient-patient, alert-absent minded, quick-slow, sophisticated-simple, bad-good, usual-unusual, and important-unimportant. On the other hand, the non-eating disorder group produced significantly larger D scores for boastful-modest, conceited-not conceited, painstaking-slipshod, and wasteful-thrifty. Overall, the eating disorder group had thirteen significantly larger values, while the eating disorder group generated four. These results do not support the sixth null hypothesis. The eating disorder group shows considerably more significant differences between self image dimensions and ideal self image dimensions. In accordance with the discussion presented earlier, when advertisers stress appeals in their advertisements that are related to adjective pairs with high D scores, there is a strong possibility that individuals may react in a manner that promotes eating disorders.

DISCUSSION

The purpose of this paper was to set forth the outcomes of an inquiry into eating disorders among university students. The study indicated that advertisements that carry positive images of slimness can exert a strong influence on students. Those who are victims of bulemia and anorexia possess different self image and ideal self images than those who are not victims. The study revealed various differences between self images and ideal self images which may be related to the incidence of the eating disorders, providing signals to advertisers who desire to mitigate negative influences on these eating disturbances.

This study does not, of course, prove that advertising is the sole, or even one of the more important contributors to eating disorders. Numerous other influents, such as primary and secondary reference groups, celebrities, and non-advertising messages carried in the mass media could exert a significant impact (Anonymous, 2000). Nevertheless, there is evidence that some advertising efforts have a role and should be scrutinized for possible negative effects.

There are some limitations to the study. The findings are largely based upon self reports where the subjects indicated such things as whether or not they suffer from an eating disorder and the effect of advertising on their behavior. Self reports can be biased, although assurances of anonymity in the data collection process were used to minimize possible bias. The sample of those with anorexia and bulemia was not large. However, the sample was sufficient to provide statistically significant results. Further, the subjects were students enrolled in marketing classes at various universities, and the results may not be generalizable to other populations. However, college students are among the more likely groups to become victims of the disorders, and the marketing classes attracted students from various majors within and out of the colleges of business administration, furnishing a relatively heterogeneous sample composition.

IMPLICATIONS

The study described in this paper poses implications for socially responsible advertisers. They are well-advised to avoid advertising themes, messages, and appeals which the study revealed as potentially dangerous for consumer welfare. Individuals with eating disorders appear to have anxieties which probably should not be aroused from a social responsibility perspective.

Further, there are implications for researchers. Further examination of the relationship of advertising and other promotion methods to eating disorders is merited, as a means of furnishing insights which could be useful to advertisers. In addition, the results of the study could be replicated, employing different respondents, as at the high school and younger levels, where anorexia and bulemia are relatively commonplace. Further, investigations could be carried out into the receptiveness of advertisers to modify their campaigns and individual advertisements as a means of helping in abating bulemia and anorexia.

REFERENCES

Anonymous (2000).Ads are not to blame for all the world's problems. Campaign, 28(3), 27.

Beals, K. A. & M. M. Manore (2002). Disorders of the female athlete triad among collegiate athletes. International Journal of Sport Nutrition and Exercise Metabolism, 12(3), 281-293.

Bloch, P. H. & M. L. Richins (1992). You look marvelous: the pursuit of beauty and the marketing concept. Psychology & Marketing, 9(1), 3-19.

Botta, R. A. (2003). For our health? The relationship between magazine reading and adolescents body image and eating disturbances. Sex Roles, 48(10), 389-399.

Brinton, J. H. (1961). Deriving an attitude scale from semantic differential data. Public Opinion Quarterly, 25(3), 289-295.

Carrigan, M. & A. Attalla (2001). The myth of the ethical consumer: do ethics matter in purchase behavior? The Journal of Consumer Marketing, 18(7), 560-577.

Chiodo, J. & P.R. Latimer (1983). Vomiting as a learned weight control technique. Journal of Behavior Therapy and Experimental Psychiatry, 14(1), 131-136.

Christie, J., D. Fisher, J. C. Kozup, & S. Smith (2001). The effects of bar-sponsored alcohol beverage promotions across binge and nonbinge drinkers. Journal of Public Policy & Marketing, 20(2), 240-253.

Devlin, M. J. & A. J. Zhu (2001). Body image in the balance. Journal of the American Medical Association, 286(17), 2159-2187.

DiConsiglio, J. (2000). Dying to be thin. Scholastic Choices, 16(3), 11.

Duddle, M. (1973). An increase of anorexia nervosa in a university population. British Journal of Psychiatry, 4(1), 711-712.

Hayaki, J., M.A. Friedman, M.A. Whisman, S.S. Delinsky, & K.D. Brownell (2003). Sociotropy and bulemic symptoms in clinical and nonclinical samples. International Journal of Eating Disorders, 34(1), 172-176.

Heinberg, L. J. & , J. K. Thompson (1995). Body image and televised images of thinness and attractiveness: a controlled laboratory investigation. Journal of Social and Clinical Psychology, 14(4), 325-338.

Hibbert, S., M. Piacentini, & H. Al Dajani (2003). Understanding volunteer motivation for participation in a community-based food cooperative. International Journal of Nonprofit and Voluntary Sector Marketing, 8(1), 30-42.

Hoskins, M.L. (2002). Girls' identity dilemmas: spaces defined by definitions of worth. Health Care for Women International, 23(3), 231-247.

Jacobson, B. (1985). Anorexia nervosa and bulemia: two severe eating disorders. New York: Public Affairs Committee, 1-2, 9.

Johnson, C. & R. Larson (1982). Bulemia: An analysis of moods and behavior. Psychosomatic Medicine, 44(1), 1034-1038.

Karande, K., C.P. Rao, & A. Singhapakdi (2002). Moral philosophies of marketing managers a comparison of american, australian, and malaysian cultures. European Journal of Marketing, 36(8), 768-791.

Kennedy, C. (2000). Examining television as an influence on childrens' health behaviors. Journal of Pediatric Nursing, 15(5), 272-281.

Kreth, M. L. (2000). Exploring responsibility in advertising: health claims about dietary supplements. Business Communication Quarterly, 63(6), 66-72.

Leon, G.R., K. Carroll, B. Chernyk, & S. Finn (1984). Binge eating and associated habit patterns with college student and clinically identified populations. International Journal of Eating Disorders, 11(2), 78-93.

McLean J.A. & S.I. Barr (2003). Cognitive dietary restraint is associated with eating behaviors, lifestyle practices, personality characteristics, and menstrual irregularity in college women. Appetite, 40(2), 185-192.

Miles, M. P. & J. G. Covin (2000). Environmental marketing: A source of reputational, competitive, and financial advantage. Journal of Business Ethics, 23(3), 299-311.

Murphy, P. E. (1998). Ethics in advertising review, analysis, and suggestions. Journal of Public Policy and Marketing, 17(2), 316-319.

Myers, P. N. & F. A. Biocca (1992). The elastic body image: the effect of television advertising and programming on body image distortions in young women," Journal of Communication , 42(3), 108-133.

Neumark-Sztainer, D. (1996). School-based programs for preventing eating disturbances, Journal of School Health , 66(2), 29-45.

Odekerken-Schroder, G., K. De Wulf, & N. Hofstee (2002). Is gender stereotyping in advertising more prevalent in masculine countries? A cross-cultural analysis. International Marketing Review, 19(5), 408-419.

Osgood, C. E., G. Suci, & P. H. Tannenbaum (1957). The measurement of meaning. Urbana, Ill: University of Illinois Press, 82, 84.

Owen, P. R.& E. Laurel-Seller (2000). Weight and shape ideals: thin is dangerously in. Journal of Applied Social Psychology, 30(5), 979-990.

Pechmann, C., G. Zhao, M. Goldbert, & E. Thomas Reibling (2003). What to convey in antismoking advertisements for adolescents: the use of protection motivation theory to identify effective message themes. Journal of Marketing, 67(2), 1-18.

Peterson, R. T. (1987).Anorexia and bulemia in an advertising context Journal of Business Ethics, 6(2), 495-504.

Peterson, R. T. (2002). The depiction of african american childrens' activities in television commercials: an assessment. Journal of Business Ethics, 36(4), 303-313.

Porter, M. E. & M. R. Kramer (2002). The competitive advantage of corporate philanthropy," Harvard Business Review, 80(12), 56-68.

Posavac, S. S. & H. Posavac (2002). Predictors of women's concern with body weight: The roles of perceived self media ideal discrepancies and self esteem. Eating Disorders: The Journal of Treatment and Prevention, 10(2), 153-160.

Rentschler, R. & G. Wood (2001). Cause related marketing: can the arts afford not to participate? Services Marketing Quarterly, 22(1), 57-69.

Roncolato, W., G. Huon, & C. Braganza (1998). Nutritional knowledge, food-related and body-related attitudes among preadolescent children. Australian Journal of Nutrition and Dietetics, 55(4), 195-202.

Schur, E.A., M. Sanders & H. Steiner (2000). Body dissatisfaction and eating in young children. International Journal of Eating Disorders, 27(2), 74-82.

Scwitzer, A. M. Bergholz, K. Dore, & L. Salimi Terri (1998). Eating disorders among college women-pevention, education, and treatment response. Journal of American College Health, 46(5), 202-207.

Sen, S. & C.B. Bhattacharya (2001). Does doing good always lead to doing better? Consumer reactions to corporate social responsibility,. Journal of Marketing Research, 38(2), 225-243.

Steiger, H., K. R. Bruce, & M. Israel (2003). Eating disorders in stricker. In T. George, A. Widiger, & I. B. Weiner (Eds). Handbook of Psychology. New York: John Wiley & Sons.

Stice, E., J. Maxfield, & T. Wells (2003). Adverse effects of social pressure to be thin on young women: an experimental investigation of the effects of fat talk. International Journal of Eating Disorders, 34(1), 108-117.

Sutandar-Pinnock, K., D. B> Woodside, J. C. Carter, M. P. Olmsted, & A. S. Kaplan (2003). Perfectionism in anorexia nervosa: A 6-24- month follow up study International Journal of Eating Disorders, 33(2), 225-229.

Thomsen, S.R., M.M. Weber & L.B. Brown (2001). The relationship between health and fitness magazine reading and eating-disoriented weight-loss methods among high school girls. American Journal of Health Education, 32(3), 133-138.

Toro, J., M. Salamero, & E.Martinez (1994). Assessment of sociocultural influences on the aesthetic body shape model in anorexia nervosa. Acta Psychiatrica Scandanavica, 89(3), 147-151.

Wagner, E.R. & E.N. Hansen (2002). Methodology for evaluating green advertising of forest products in the united states: A content analysis. Forest Products Journal, 52(4), 17-23.

Walcott, D. D. H. D. Pratt, & D. R. Patel (2003). Adolescents and eating disorders: gender, racial, ethic, sociocultural, and socioeconomic issues, Journal of Adolescent Research, 18(3), 223-243.

Zucker, A., N. Z. Harrel, R. Miner, K. Stewart, & J. Abigail (2000). Smoking in college women: The role of thinness pressures: media exposure, and critical consciousness. Psychology of Women Quarterly, 25(3), 233-241.

Robin Peterson, New Mexico State University Minjoon Jun, New Mexico State University

Table 1: Sample Respondents Experiencing Anorexia or Bulemia

Student Condition Number Percent of Total

Have bulemia only 63 5.9%
Have anorexia only 39 3.6
Have both bulemia and anorexia 22 2.1
Do not have either 949 88.4
Total 1,073 100.0%

Table Two: Personal Appearance Concern by the Subjects

Student Condition Very Concerned Concerned Neutral
Eating disorders 71.4% 26.2% 2.4%
Without eating disorders 43.8% 23.7% 32.5%

Table Three: Self Image Scale Means for Students With and Without
Eating Disorders

 Students with Students without
 an Eating an Eating
Adjective Pair disorder disorder

Boastful-modest 5.1 4.9
Assertive-submissive 4.3 4.2
Conceited-not conceited 5.0 5.2
Intelligent-stupid * 2.7 3.2
Clear thinking-confused 2.3 2.3
Independent-dependent * 2.1 2.7
Reliable-undependable 1.7 1.5
Mature-immature * 2.2 2.8
Impulsive-deliberate * 1.9 2.6
Hypochondriacal-not 5.8 5.5
 hypochondriacal
Nervous * -calm 3.7 4.3
Thankless-grateful * 6.1 5.6
Hostile-friendly 5.9 5.8
Blame mistakes others-blame 4.5 4.7
 mistakes me
Inflexible-adaptable to change 5.8 5.6
Hardhearted-softhearted 5.4 5.5
Unrealistic-practical 5.8 5.5
Quitting-persevering * 5.4 4.8
Painstaking*-slipshod 2.3 3.0
Thoughtful-not thoughtful 2.5 2.1
Wide interests-narrow interests 2.2 2.1
Easily jealous-not easily jealous * 2.7 3.4
Self pitying-not self pitying 4.6 4.6
Timid-adventurous 5.7 5.9
Optimistic-pessimistic 2.8 2.5
Enthusiastic-apathetic 3.1 3.0
Neurotic-not neurotic * 3.3 3.9
Fickle-loyal 5.8 5.6
Dishonest-honest 6.0 6.0
Hurried-not hurried * 2.7 3.4
Secretive-frank * 4.1 4.8
Emotional-unemotional 2.8 2.7
Dissatisfied-content * 3.0 3.6
Energetic-not energetic 3.4 3.5
Self confident-self distrustful 4.2 3.9
Suspicious-trusting 6.3 6.1
Impatient-patient 4.7 4.5
Alert-absent minded 2.5 2.5
Quick-slow 1.3 1.4
Sociable-shy 2.5 2.5
Wasteful-thrifty 5.9 6.2
Sophisticated-simple * 3.0 3.5
Graceful-awkward 1.9 1.7
Bad-good 6.8 6.9
Usual-unusual 4.1 3.8
Worthless-valuable 6.7 6.7
Important-unimportant 6.0 6.0
Unsuccessful-successful * 6.1 5.3
Sick-healthy 5.7 5.6
Weak-strong 4.7 4.6

* Statistically significant difference between items in the two columns
according to a t test at the .05 level of significance

Table Four: Ideal Self Image Scale Means for Students With and Without
Eating Disorders

 Students with Students without
 an Eating an Eating
Adjective Pair disorder disorder

Boastful-modest * 5.0 5.6
Assertive-submissive * 4.1 4.6
Conceited-not conceited 4.9 4.7
Intelligent-stupid * 2.2 2.7
Clear thinking-confused 1.7 1.7
Independent-dependent * 2.5 3.2
Reliable-undependable 2.3 2.3
Mature-immature * 2.6 2.1
Impulsive-deliberate 3.6 3.5
Hypochondriacal-not 5.6 5.5
 hypochondriacal
Nervous-calm * 3.7 3.1
Thankless-grateful * 5.5 5.0
Hostile-friendly * 6.2 5.5
 Blame mistakes others-blame
mistakes me * 5.6 5.1
Inflexible-adaptable to change 4.8 4.7
Hardhearted-softhearted * 2.9 3.5
Unrealistic-practical 5.5 5.6
Quitting-persevering * 5.8 5.1
Painstaking -slipshod 2.2 2.4
Thoughtful-not thoughtful * 2.1 2.7
Wide interests-narrow interests 3.6 3.6
Easily jealous-not easily jealous 3.4 3.5
Self pitying-not self pitying 4.7 4.5
Timid-adventurous * 5.9 5.2
Optimistic-pessimistic 3.5 3.6
Enthusiastic-apathetic 3.0 3.2
Neurotic-not neurotic * 3.0 3.9
Fickle-loyal 5.7 5.7
Dishonest-honest 5.5 5.6
Hurried-not hurried * 2.1 3.1
Secretive-frank 4.5 4.5
Emotional-unemotional 3.2 3.5
Dissatisfied-content* 3.5 4.1
Energetic-not energetic 4.3 4.1
Self confident-self distrustful 4.7 4.2
Suspicious-trusting 6.0 6.0
Impatient-patient * 3.8 4.5
Alert-absent minded * 1.9 2.7
Quick-slow * 1.5 2.0
Sociable-shy 2.3 2.6
Wasteful-thrifty 5.4 5.3
Sophisticated-simple * 2.4 3.4
Graceful-awkward 1.7 1.6
Bad-good * 5.4 6.2
Usual-unusual 3.3 3.5
Worthless-valuable * 6.5 6.0
Important-unimportant * 2.2 2.8
Unsuccessful-successful * 5.8 5.1
Sick-healthy 5.5 5.6
Weak-strong 4.9 5.1

* Statistically significant difference between items in the two columns
according to a t test at the .05 level of significance

Table Five: Self and Ideal Self Images and Gaps Between the Two

Adjective Pair Students with an eating disorder

 Self image Ideal self D Score

Boastful-modest * 5.1 5.0 .1
Assertive-submissive 4.3 4.1 .2
Conceited- not conceited * 5.0 4.9 .1
Intelligent-stupid 2.7 2.2 .5
Clear thinking- confused 2.3 1.7 .6
Independent- dependent 2.1 2.5 .4
Reliable- undependable 1.7 2.3 .6
Mature-immature 2.2 2.6 .6
Impulsive- deliberate * 1.9 3.6 1.7
Hypochondriacal-not 5.8 5.6 .2
 hypochondriacal
Nervous-calm 3.7 3.7 .0
Thankless-grateful 6.1 5.5 .6
Hostile-friendly * 5.9 6.2 .3
Blame mistakes others-blame 4.5 5.6 1.1
 mistakes me *
Inflexible-adaptable to change * 5.8 4.8 1.0
Hardhearted- softhearted * 5.4 2.9 2.5
Unrealistic-practical 5.8 5.5 .3
Quitting-persevering 5.4 5.8 .3
Painstaking-slipshod * 2.3 2.2 .1
Thoughtful- not thoughtful 2.5 2.1 .4
Wide interests- narrow interests 2.2 3.6 1.4
Easily jealous-not 2.7 3.4 .7
 easily jealous *
Self pitying- not self pitying 4.6 4.7 .1
Timid-adventurous 5.7 5.9 .2
Optimistic- pessimistic 2.8 3.5 .7
Enthusiastic-apathetic 3.1 3.0 .1
Neurotic-not neurotic 3.3 3.0 .3
Fickle-loyal 5.8 5.7 .1
Dishonest-honest 6.0 5.5 .5
Hurried-not hurried 2.7 2.1 .6
Secretive-frank 4.1 4.5 .4
Emotional- unemotional * 2.8 3.5 .7
Dissatisfied-content 3.0 3.5 .5
Energetic- not energetic 3.4 4.1 .7
Self confident- self distrustful 4.2 4.7 .5
Suspicious-trusting 6.3 6.0 .3
Impatient-patient * 4.7 3.8 .9
Alert-absent minded * 2.5 1.9 .6
Quick-slow * 2.5 1.5 1.0
Sociable-shy 2.5 2.3 .2
Wasteful-thrifty * 5.9 5.4 .5
Sophisticated- simple * 3.0 2.4 .6
Graceful-awkward 1.9 1.7 .2
Bad-good * 6.8 5.4 1.4
Usual-unusual * 4.1 3.3 .8
Worthless-valuable 6.7 6.5 .2
Important- unimportant * 6.0 2.2 3.8
Unsuccessful- successful 6.1 5.8 .3
Sick-healthy 5.7 5.5 .2
Weak-strong 4.7 4.9 .2

 Students without an eating disorder

 Self image Ideal self D Score

Boastful-modest * 4.9 5.6 .7
Assertive-submissive 4.2 4.6 .4
Conceited- not conceited * 5.2 4.7 .5
Intelligent-stupid 3.2 2.7 .5
Clear thinking- confused 2.3 1.7 .6
Independent- dependent 2.7 3.2 .5
Reliable- undependable 1.5 2.3 .8
Mature-immature 2.8 2.1 .7
Impulsive- deliberate * 2.6 3.5 .9
Hypochondriacal-not 5.5 5.5 .0
 hypochondriacal
Nervous-calm 4.3 3.1 1.2
Thankless-grateful 5.6 5.0 .6
Hostile-friendly * 5.8 5.5 .2
Blame mistakes others-blame 4.7 5.1 .4
 mistakes me *
Inflexible-adaptable to change * 5.6 4.7 .6
Hardhearted- softhearted * 5.5 3.5 2.0
Unrealistic-practical 5.5 5.6 .1
Quitting-persevering 4.8 5.1 .3
Painstaking-slipshod * 3.0 2.4 .6
Thoughtful- not thoughtful 2.1 2.7 .6
Wide interests- narrow interests 2.1 3.6 1.5
Easily jealous-not 3.4 3.5 .1
 easily jealous *
Self pitying- not self pitying 4.6 4.5 .1
Timid-adventurous 5.9 5.2 .7
Optimistic- pessimistic 2.5 3.6 1.1
Enthusiastic-apathetic 3.0 3.2 .2
Neurotic-not neurotic 3.9 3.9 .0
Fickle-loyal 5.6 5.7 .1
Dishonest-honest 6.0 5.6 .4
Hurried-not hurried 3.4 3.1 .3
Secretive-frank 4.8 4.5 .3
Emotional- unemotional * 3.5 3.5 .0
Dissatisfied-content 3.6 4.1 .5
Energetic- not energetic 3.5 4.1 .6
Self confident- self distrustful 3.9 4.2 .3
Suspicious-trusting 6.1 6.0 .1
Impatient-patient * 4.5 4.5 .0
Alert-absent minded * 2.5 2.7 .2
Quick-slow * 1.4 2.0 .6
Sociable-shy 2.5 2.6 .1
Wasteful-thrifty * 6.2 5.3 .9
Sophisticated- simple * 3.5 3.4 .1
Graceful-awkward 1.7 1.6 .1
Bad-good * 6.9 6.2 .7
Usual-unusual * 3.8 3.5 .3
Worthless-valuable 6.7 6.0 .7
Important- unimportant * 6.0 2.8 3.2
Unsuccessful- successful 5.3 5.1 .2
Sick-healthy 5.6 5.6 .0
Weak-strong 4.6 5.1 .5

* Statistically significant difference between D scores according to
at test at the .05 level of significance