INTRODUCTION Stereotypes in Perception of Ideal Body Size for Different EthnicitiesWestern societies place a large amount of pressure on their women. They are expected to act, speak, walk and look a certain way. Emphasis is especially placed on a woman’s looks, and expectations of conformity to the ideal image are becoming increasingly higher. Everywhere women turn they see what is considered ideal: in movies, on TV shows, in magazines, on runways, on billboards. There is no escape from the media saturation of the idea that “thin is in”. This has been so ever since a model aptly named Twiggy appeared on the scenes in the late 1960’s. Twiggy led a wave of supermodels and actresses whose thin frames and size 0 bodies became highly desirable, but which was usually unattainable for most women. Eating disorders, such as anorexia nervosa and bulimia, have been linked to the constant media saturation that thinness is the ideal body image (Johnson & Petrie, 1995).It is common knowledge that White women suffer from more eating disorders than do Black women (Crago, Shisslak & Estes, 1996). Studies have shown that White women also report significantly lower body satisfaction, more disordered eating behaviour and a higher drive for thinness on a Body Esteem Scale than Black women (Henriques, Calhoun & Cann, 1996). Research conducted by Bissell (2002) showed that in 80% of cases where White women reported a greater desire than Black women to look like models, the models were thin. Furthermore in the other 20% of cases where Black women had the greater desire to look like the model shown, the model was overweight. This is indicates that not only are White women more dissatisfied with their bodies than Black women, they also seek to be thinner. Black women on the other hand appear to be more satisfied with their bodies and if given the choice would rather be overweight than underweight (Bissell, 2002). It is also interesting to observe what males find attractive in terms of women’s body size and weight. Investigations by Demarest and Allen (2000) showed that men of different races did not differ significantly in what body size they found attractive. While all the women in this study believed that men would find thin bodies most attractive, there was a significant disparity in the level of thinness Black women and White women would have expected the men to find attractive. The Black women estimated closest to what men chose as ideal, whereas the White women believed that men would find extremely thin women to be most attractive. Bissell’s (2000) research asked male respondents to rate pictures of female models of varying sizes both on attractiveness and their willingness to date someone looking like the model. They were also shown a chart and asked to circle the body size closest to their perception of the ideal body shape for women. It was found that while White males chose a thinner body as more ideal, it was not a significant disparity from the ideal body chosen by Black males. White males were more likely to rate overweight women as most unattractive, and were less likely than Black males to date an overweight woman.While many investigations exist into the disparities of body satisfaction of White and Black women, no literature has been found discussing whether women of each race holds a different body ideal for their own race than for the other. Furthermore, no investigations exist that show whether males view the ideal body size White and Black women differently from each other. No research has been conducted to unearth stereotypes in body ideals for White and Black women. We therefore examined whether or not stereotypes exist as to what body size is considered acceptable for White and Black women on a whole, by both males and females. We predicted, based on previous research, that all participants will rate overweight White women as significantly more dissatisfied with their bodies than Black women of the same size.
METHOD MethodParticipants The sample consisted of 69 Loyola Undergraduate students, recruited using convenience sampling. They came from all four standings: Freshman, Sophomore, Junior and Senior. The participants were both male (n=27) and female (n=41) and not discriminated against based on race, and thus a wide variety of races participated. There were 47 White participants, 11 Blacks and 10 who fell into other categories (such as Latino or Asian). Participation was on a purely volunteer basis, although some participants received course credit where offered by their professors. The participants were recruited through the Psychology participant pool, from on-campus organisations and from the investigators own classes. The results of one White male were removed from the study, with his permission, because he felt that he was not in a position to determine whether or not the women were satisfied with their bodies. He responded ‘neutral’ for every slide shown. MaterialsThe test packet contained consent forms, a demographic question sheet and a body-satisfaction survey. Pictures of women’s bodies (Stunkard, Sorenson & Schulsinger, 1983) of different sizes and colored by the investigators were used in conjunction with the body-satisfaction survey. The pictures were not included in the test packet given to the participant as they were shown as slides controlled by the investigators. The participants used pencils provided by the investigators to answer the questions. The demographic question sheet was compiled by the investigators (See Appendix A). It contained questions regarding age, race, sex, class standing, nationality and sexual preference. The body-satisfaction survey was done using a 5-point Likert Scale answering the following question for each of the eighteen slides shown: “I think that this woman is satisfied with her body”. The answers ranged from “Strongly Agree” to “Strongly Disagree”. (See Appendix B)The pictures of women’s bodies were duplicated, and one set colored peach to represent White women, and the other set colored brown to represent Black women. (See Figure 1)Design & Procedure The study was a within subjects factorial design. The participants were subjected to each and every combination of the levels of the independent variables and there were equal numbers of participants in each cell (69 per cell). The independent variables were race of the woman, as represented by the color of the drawing of the woman; and the size of the woman. For the I.V. of race, the levels were brown and peach, to represent Brown and White women respectively For the I.V. of the size of the woman, the levels were the pictures, each of which was assigned a number, as seen in Figure 1. Race Size of woman – as assigned in Figure 1 1 2 3 4 5 6 7 8 9Peach 69 69 69 69 69 69 69 69 69Brown 69 69 69 69 69 69 69 69 69 The dependent variable was the response on the Satisfaction survey. This D.V. would then be used to determine whether or not stereotypes exist by examining whether people gave significantly different satisfaction ratings for Black and White women of the same size. The Sampling method was a convenience method and as a result the participants were randomized as they were not put into certain time slots by the investigators, but rather chose times themselves. The participants acted a control against themselves, and being that they were shown all the slides in one sitting no temporal effects were expected. The study was conducted in Loyola University classrooms. The participants were tested in groups, one group every 45 minutes. The participants were seated and given pencils with which to answer the questions. They were told that they were participating in a study on body ideals – no specifics were given. They were given two consent forms: read, signed and dated each, and kept one copy for themselves and returned the other to the investigator. The first part of the test packet, the demographic survey, was handed out. Participants were given 5 minutes to fill out the survey. The survey was then collected. Participants received the second part of the test packet, the body-satisfaction survey. They were presented images on a projector screen of 18 women, 9 White and 9 Black. These images were shown individually and in a previously chosen random order (this order remained the same for all test groups). The participants were given 10 seconds to make their response on the body-satisfaction survey. This survey was then collected. Following this, the participants were debriefed. They were told that the study was an investigation into stereotypes held by people towards ideal body types for Black and White women. Any questions were answered. The investigators also signed participation slips for those receiving extra credit.
RESULTSResultsOur hypothesis was that heavier Black woman would be rated as more satisfied than White women of equal size by all participants. To test this we analyzed the mean satisfaction rating for the White and Black women that made up each pair using a Paired Samples T-test on SPSS software. We analyzed all data for p<.05, two tailed. These analyses showed that there were significant disparities in the mean satisfaction ratings of five pairs. These were women 3, 4, 5, 7 and 8 (see listing in Figure 1). For Pair 3 the Black woman was rated significantly more dissatisfied than the White woman (t (67) = 2.921, p = .009). For Pair 4 the Black woman was rated significantly more dissatisfied than the White woman (t (66) = 3.278, p = .026). For pair 5 the White woman was rated significantly more dissatisfied than the Black woman (t (67) = 5.097, p = .001). Pair 7 also had the White woman rated as more dissatisfied than the Black woman (t (66) = -3.647, p = .006). Likewise for Pair 8 (t (67) = -4.254, p = .004). (See results in Table below). Pairs 4 and 5 support the hypothesis that heavy Black women would be rated as more satisfied than White women of the same size. Mean Std. Deviation Black White Black White t df pPair 1 4.57 4.54 .68 .69 -.497 6467676667 .621Pair 2 4.47 4.47 .72 .80 .000 67 1.000Pair 3 4.38 4.10 .81 .79 2.921 67 .005Pair 4 3.54 3.85 .75 .61 3.278 66 .003Pair 5 2.79 2.79 .89 .84 5.097 67 .000Pair 6 3.03 2.82 .86 .86 1.524 67 .132Pair 7 2.63 3.15 .95 1.22 -3.647 66 .001Pair 8 3.28 3.90 1.06 1.02 -4.254 67 .000Pair 9 3.84 4.01 1.03 1.11 1.841 67 .070In terms of descriptives, woman 1 was rated as most dissatisfied of the White women (M = 4.54, SD = .69) and Figure 6 was rated as the least dissatisfied of the White women (M = 2.82, SD = .86). Woman 1 was also rated as most dissatisfied of the Black women (M = 4.57, SD = .68). It should be noted that there was no significant difference in the means of the two colors for woman 1. Woman 5 was rated as the least dissatisfied Black woman (M = 2.79, .89). (See results in table above).
DISCUSSION Discussion Two of the heavier Black women (4 & 5) were rated significantly less dissatisfied with their bodies than the White women of the same size. This was congruent with our hypothesis that heavier Black women are perceived as more satisfied with their bodies than White women of the same size, and indicates that stereotypes do exist as to what size is acceptable for Black women and White women. However, the ratings for woman 3 contradicted our hypothesis as the White woman was rated significantly less dissatisfied than the Black woman. There was also no significant difference in dissatisfaction rating for the two heaviest sets of women (1 & 2), which also contradicted our hypothesis. Although this was not part of our hypothesis, we also examined the significant difference in ratings for the smaller women. Due to previous studies that showed that White women were more likely to want to be skinny than Black women (Henriques, Calhoun and Cann, 1996) and that Black women would choose to be overweight rather than underweight (Bissell, 2002), we were surprised to see that for women 7 and 8, both of who were at the skinnier end of the scale, the Black women were rated as less dissatisfied with their bodies. This is also inconsistent with the images we see in the media that show White women constantly trying to lose weight, or being presented with very skinny supermodels and actresses (versus fuller models and actresses such as Tyra Banks and Queen Latifah). We would have expected that the White women at this end of the scale would have been rated as less dissatisfied with the Black women. We also noticed that at either extremity of the scale (very obese or very skinny) there were no significant differences in the satisfaction ratings for the different races of the pair. Race was not an issue but both were rated as very highly dissatisfied. This leads us to believe that while stereotypes may exist in terms of what weight is acceptable for White women and Black women, there is no difference at extreme weights. Neither too obese nor too skinny is considered to be satisfactory.There were several limitations to the present study that should be accounted for if further investigation into this idea is undertaken. First, and possibly most important, was that when the slides were projected onto the screen the colors were not as distinctive as planned. The peach drawings looked White and the brown drawing looked light brown. Upon questioning some participants after debriefing, there were mixed responses to their viewing of the slides. Some participants said they did not notice there was a color difference, some said that they noticed the difference but did no attribute it to different races, and some said that they both noticed the colors and attributed it to different races. By the time we realized that the colors were not having their full effect we had already collected a large amount of data and due to time constraints were reluctant to fix the problem and have to throw out all our data. In the future it would be necessary to ensure that the colors show up properly and that difference is notable, quite possibly by doing pilot tests.Secondly there were several issues with the sample. The sample size was fairly small (n = 68). If the sample size was larger we may have had more power over the date. Furthermore the sample was made up of college students from a predominantly White university, and we feel the results may not be generalizable to the public at large. Unfortunately we could not account for and possible effects that the participants’ race may have had on their ratings, since there was a large disparity in the number of White and other races, specifically Blacks. We also could not account for any possible effects that the sex of the participants may have had on their ratings as the ratio between men and women in the sample was very uneven. Future studies should try to look at these aspects, and would need fairly large and equal sample sizes for each to do so. We also did not specifically take age into account, although being that we sampled college undergraduates we assumed that the range would have been between 18 and 24 for the most. We had several participants who were much older. There were several in their 30’s and one woman who was in her late 40’s. Although we did not examine whether or not age would have an impact on this stereotype it is possible, and therefore these few outliers could have had an effect on the results. Future studies should either examine a wider variety of ages or have a requirement as to the range. Thirdly, being that the pictures were shown in a single preselected random order there may have been order effects, which would have affected the rating. For example, the Black woman 1 was shown right after the White woman 8. Was the woman 1 rated more harshly than woman 8 because of the size of her figure or because of the comparison of a very skinny figure with a very heavy figure? This contrast effect could have been avoided by showing different groups the pictures in one of several preselected random orders. The investigators themselves may also have had a subconscious effect on the participants’ response. The investigators were two females, one Black and one White, and one male of White-Asian decent. They each conducted the experiment, sometimes alone, sometimes in combination with one or both of the other investigators. It is possible that the race and sex of the investigator, or combination of investigators may have had an effect on the response of the participant. Furthermore, several of the participants were friends of the investigators and may have responded differently than other participants as such. Although the investigators tried to minimize nuisance variables in the room they still occurred, particularly when participants who were friends came in together. They tended to joke around, and this may have caused them not to the answer the questions with the seriousness the study deserved or may have acted as a distraction to the other participants. Other disturbances which may have had an effect occurred when people arrived after the study had already commenced and when the equipment had technical difficulties. In conclusion, we can infer that stereotypes do exist as to ideal body types for Black and White women, although at very extreme body types race is not an issue. Knowledge of this could lead to a greater awareness of one’s potential to hold stereotypes and therefore maybe reduce pressure placed on women to look a certain way, as they will see that what is expected of them is not necessarily expected from others. Future research could focus on how participant characteristics such as age, sex, race and sexuality affect these stereotypes. Future research should also attempt to examine whether stereotypes exist for more races than just Black and White, such as Asian or Latino. It would also be interesting to attempt to determine if biracial women suffer from competing expectations for their body.
REFERENCES ReferencesBissell, K.L. (2002). I want to be thin like you: gender and race as predictors of cultural expectations for thinness and attractiveness in women. New Photographer, 57, 4-12.Crago, M., Shisslak, C.M., & Estes, L.S. (1997). Eating disturbances among American minority groups: A review. The International Journal of Eating Disorders, 19, 239-248.Demarest, J., & Allen, R. (2000). Body Image: Gender, Ethnic and Age Differences. The Journal of Social Psychology, 140, 465. Garner, D.M. (1997). The 1997 Body Image Survey Results. Psychology Today. Retrieved December 1, 2002, from www.psychologytoday.com/htdocs/prod/contents/PTO/PTO-19970201-000023.htmlHenriques, G.R., Calhoun, L.G., & Cann, A. (1996). Ethnic differences in women’s body satisfaction: an experimental investigation. The Journal of Social Psychology, 136, 689-697.Johnson, C., & Petrie, T. (1995) The relationship of gender discrepancy of eating attitudes and behaviours. Sex Roles, 33, 405-416. Nelson, R. (2002). Ethnicity does not influence body size, preference or tolerance. Obesity, Fitness & Wellness Week, 5-6.Stunkard, A.J., Sorenson, T., & Schulsinger, F. (1983). Use of the Danish adoption register for the study of obesity and thinness. Genetics of Neurological and Psychiatric Disorders, 115- 120.
APPENDIX A Appendix ADemographic SurveyCode ______________Instructions: Please answer the following questions to the best of your ability. Raise your hand when finished, and a researcher will collect your sheet.Sexć Maleć Female Age _________ Nationality ______________________________Race (check all that apply)ć African American ć Asian Americanć Caucasianć Hispanicć OtherCurrent Class Standing ć Freshmanć Sophomore Major(s) _______________________________ć Juniorć Seniorć Other Sexual Orientationć Heterosexualć Homosexualć Bisexual
APPENDIX B Appendix BSatisfaction SurveyCode _______________Instructions: For each slide presented, circle the answer that corresponds to the way you would rate the statement “I think this woman is satisfied with her body.” Strongly Agree Agree Neutral Disagree Strongly Disagree 1. SA A N D SD 2. SA A N D SD 3. SA A N D SD 4. SA A N D SD 5. SA A N D SD 6. SA A N D SD 7. SA A N D SD 8. SA A N D SD 9. SA A N D SD10. SA A N D SD11. SA A N D SD12. SA A N D SD13. SA A N D SD14. SA A N D SD15. SA A N D SD16. SA A N D SD17. SA A N D SD18. SA A N D SD
Table 1 - Stunkard Body Shape Scale |
