SEARCH

SEARCH BY CITATION

Keywords:

  • body image;
  • body size assessments;
  • ethnicity

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. References

Objective: Obesity is most common in the United States among women of ethnic minority groups (black and Hispanic). Researchers have hypothesized that these subcultures are more accepting of overweight figures. The purpose of this study was to examine body image and body size assessments in a large community sample of men and women.

Research Methods and Procedures: Participants were 801 women and 428 men: 23% Asian, 45% Hispanic, 17% black, and 15% white. The figure rating scale was used to rate: body dissatisfaction, attractive male and female shapes, acceptable female size, and perceptions of underweight to obese female figures.

Results: Controlling for age, education, and body weight, no ethnic differences were found for men. Asian women reported less body dissatisfaction than the other groups. Women were more dissatisfied with their size than men and chose thinner female figures as attractive and acceptable.

Discussion: Ethnicity, independent of age, education, and body weight, does not influence preference for female and male shapes or tolerance for obesity.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. References

Obesity is most common in the United States among ethnic minority groups (black and Hispanic), and in particular among women of lower socioeconomic status (SES) (1). It is estimated that as many as 48.5% of black women and 47.2% of Mexican American women are overweight or obese (1). At the same time, societal emphasis on the ideal woman and the pursuit of thinness make the development of disordered eating (including binge eating) a mental-health concern for ethnic minority women (2). The prevalence pattern of obesity has led researchers to hypothesize that these subcultures are more accepting of overweight figures: specifically, that black women and men have a greater acceptance of larger body sizes than do white women and men (3). Studies have shown that black (and in some cases Hispanic) women are more satisfied, or less dissatisfied, with their bodies than are white women despite higher weights (4), providing some indirect evidence for this hypothesis. Such differences in body satisfaction do not necessarily mean, however, that black or Hispanic subcultures are more tolerant of obesity in women. Black women may report less body dissatisfaction for a number of reasons; for example, they generally have high overall self-esteem (5, 6, 7, 8). Few empirical studies actually have assessed what is considered acceptable body sizes across ethnicity or examined men's preferences for female shapes. Furthermore, results that have been reported have been inconclusive.

Studies of black girls and women generally have found that although they tend to be heavier than their white counterparts (9, 10), they report less dissatisfaction with their weight (11) and perceive themselves to be thinner than they actually are (12). Rucker and Cash (11) found that black women, compared with white women, had body-size ideals that were less thin and more congruent with their current perceived size. Kemper et al. (13) reported that black adolescent girls were much more likely than white girls to be satisfied with their body size, to describe themselves as thinner than other girls, and to say that they were not overweight. In contrast, other studies have not found significant differences between black and white women on measures of body dissatisfaction (14, 15) or in the desire to not be fat (16).

Studies with Hispanic and Asian women have been fewer in number. Research with Hispanic women indicates thatalthough they may report similar (or even higher) levels of body dissatisfaction compared with white women (17, 18), they are more likely to rate themselves as attractive (19) and report more positive attitudes toward obesity (20). Findings with Asian women have been inconsistent. Although some studies have found that Asian women report less body dissatisfaction than white women (21, 22, 23), others have found similar levels of weight concern between the two groups (24, 25, 26), and an even greater fear of fat among Asians (26). One study comparing the four ethnic groups found that black and Asian women showed less body-size dissatisfaction than white and Hispanic women (19).

Limited research to date has examined ethnic differences in the body image of men. In their study of body image and realistic weight loss, Cachelin et al. (27) found no ethnic differences among men. Similarly, Mintz and Kashubeck (22) found no race differences for men on body image or body satisfaction, and Barr (24) found no differences between Asian and white adolescent boys on desired weight. In contrast, Story et al. (28) reported that compared with white adolescent boys, Asian adolescent boys were more dissatisfied and black males were less dissatisfied with their weight.

In addition to studies on body dissatisfaction, researchers have examined more direct differences in cultural acceptance of body size and shape by taking two approaches: by asking men about their preferences for female shape, and by asking men and women to rate the ideal or attractive female figure. A study comparing black and white college men found that black men chose larger ideal female silhouettes and were not as tolerant as whites of very thin figures (29). Likewise, among adolescent boys, blacks were found to prefer a significantly heavier ideal female body size than whites (30). Powell and Kahn (31) found that white men indicated less desire than black men to date a woman with a heavier than ideal body size and that white men felt they would more likely be ridiculed if they did date a woman who was larger than the ideal; at the same time, however, these researchers did not find any differences between the groups of men on their ratings of attractive female figure. Similarly, Harrison et al. (32) found that black men were less likely than white men to refuse a date with an overweight woman; however, in this study, black men and black women did not favor obesity. In a cross-cultural study, Ugandan students rated obese female figures as more attractive than did British whites, and the differences in ratings between the two groups grew wider as the figures got fatter (33). Parallel findings have been reported for women and girls. Among college students, white women rated the same figures as fatter than did black women (11). In a study of adolescents, the body size chosen by black girls as ideal for themselves and as ideal by social standards was larger than the size selected as ideal by white girls (13). Interestingly, although white girls felt that parents and significant others wanted them to lose weight or reduce their size, black girls felt that their size was considered satisfactory by parents and significant others (13).

In contrast to these findings, Altabe (19) found no differences between black, Asian, white, or Hispanic college students in terms of the cultural ideal, defined as the ideal that most men and women hold (whether they were rating their own subculture or the larger culture was not specified, though). Cachelin et al. (27), examining a community sample of black, Asian, Hispanic, and white male and female dieters, found no significant race differences in ratings of attractive female (or male) shape, and Demarest and Allen (15) likewise found no ethnic differences among men in the female shape they found most attractive. Singh (34) reported that black men and women did not differ from white men and women in perceptions of the most attractive female body shape and, moreover, overweight female figures were not rated as desirable or attractive by either group. Allison et al. (35), using a subway sample of black, white, Hispanic, and Asian men, as well as personal advertisements placed in newspapers and magazines, found no relationship between ethnicity and preference, and demonstrated that ethnicity explained only 2.1% of the variance in preference. These researchers concluded that ethnic differences in men's preferences for women's body shapes do not contribute substantially to ethnic differences in female adiposity (35).

Most published studies of body image or body-size preferences have used convenience samples of college students, have based findings on small samples of ethnic minority groups, and have not accounted for the impact of confounding variables including age, body weight, and SES. Findings from studies (11, 15, 29, 32, 33) that have utilized relatively small college samples cannot be generalized to the larger population of adults. Furthermore, there is evidence indicating that an individual's age, body weight, and SES or educational level all influence body image. Body dissatisfaction increases with age (15), young adults are more likely than other age groups to prefer thinner women (36), and younger individuals have the most distorted views of the preferences of the opposite sex (15). Adiposity and body mass index (BMI) are strong predictors of body dissatisfaction (18, 37, 38), with body image dissatisfaction increasing at higher weights (39). SES is a key factor in weight, with an inverse relationship existing between SES and obesity in Western societies (40). One study found that low-SES groups selected larger body sizes as ideal compared with middle- and high-SES groups (13). If differences in SES, body weight, and age are not controlled, it cannot be concluded that differences in body image and body-size preferences are due to ethnicity per se; any effect of ethnicity may be confounded with such other factors.

We are aware of only two studies to date that have examined ethnicity and body image while controlling for SES, body weight, and age. Miller et al. (41), controlling for age, body size, and SES, found that black men and women overall reported greater body satisfaction and less overestimation of weight than white, with Latino American men and women having scores in between the other two groups; the ethnic groups for this study, however, were very small in size and the sample consisted of college students. Smith et al. (38), after adjusting for age, BMI, and education, found that black women were more satisfied with body size than white women; this study did not include Hispanic or Asian populations. In contrast, Caldwell et al. (14), after controlling for SES, BMI, and marital status (but not age), did not find significant differences between black and white women on body dissatisfaction or discrepancies between actual and ideal weight and shape. These researchers concluded that previous findings of less body concern in black women may reflect class rather than race effects (14).

The purpose of this study was to examine body image and body-size assessments in a large sample of men and women of four ethnicities/races: black, Hispanic, Asian, and white. We utilized a community sample to represent a wide range of ages, body weights, and education levels, and we studied the effect of ethnicity by controlling for these variables. There are different definitions and measurements for the concept of SES. Because income is not necessarily a reflection of one's status or position in society (42), we utilized educational level as our measure of SES. Additionally, we included ratings specifically to measure acceptable female shapes, as well as perceptions of underweight, normal-weight, overweight, and obese female figures; few previous studies have examined subjects’ perceptions of the full thin to fat range of female figures (33). We hypothesized that: black women and men would report less body dissatisfaction than the other ethnic groups; black and Hispanic men and women, compared with Asians and whites, would accept heavier female figures and would select larger sizes as representing overweight and obese female figures (i.e., would have higher thresholds for what they consider obesity); regardless of ethnicity, women would be more dissatisfied with their size and shape than men; and women, compared with men, would select thinner female figures as attractive and acceptable.

Research Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. References

Participants

Participants (N = 1229) were 801 women and 428 men with the following ethnic breakdown: 288 Asian (189 women and 99 men), 548 Hispanic (379 women and 169 men), 208 black (132 women and 76 men), and 185 white (101 women and 84 men). Average age for the total sample was 24.3 years old (SD = 8.9, range = 18 to 83 years old), mean BMI was 24.7 (SD = 5.4, range = 14.4 to 50.3), and mean education level was some college (SD = 2.8, range from completion of grade school/no high school degree to completion of graduate or professional school).

Demographic and Weight Information

Participants reported their age, sex, current weight and height, ethnicity (Asian; black; Hispanic; white, not of Hispanic origin; American Indian; mixed/parents from two different groups), and level of education (1 = grade 6 or less; 2 = grade 7 to 12 without graduating; 3 = high school graduate or high school equivalent; 4 = some college; 5 = graduate of a 2-year college; 6 = graduate of a 4-year college; 7 = some graduate or professional school; and 8 = completed graduate or professional school). Education can be treated statistically as a categorical or continuous variable. In this study, we treated education as a continuous variable in the univariate and multivariate analyses.

BMI was calculated as weight in kilograms divided by height in square meters. Research has shown that self-reports are highly correlated with actual heights and weights and are sufficiently valid to use in epidemiological and survey studies (43, 44).

Silhouette figure ratings (45) were used to assess body dissatisfaction, attractive female and male figures, acceptable female body sizes, and perceptions of underweight, normal-weight, overweight, and obese female figures. This procedure was similar to that used by other researchers (27, 46, 47, 48). Referring to nine female and nine male silhouettes depicting body size, ranging from severe emaciation to severe obesity, 1 to 9, respectively, participants were asked to rate the following: “your current size and shape”(current); “the size and shape you would most like to be” (ideal);“the smallest size and shape you feel is realistic for you to attain” (realistic); “the female size and shape you feel is most attractive” (attractive woman); “the male size and shape you feel is most attractive” (attractive man); “the female size and shape you feel men in general find most attractive” (woman attractive to men); and “the male size and shape you feel women in general find most attractive” (man attractive to women).

Additionally, referring to the nine female silhouettes, participants were asked to rate: “the (female) figure that represents someone who is underweight” (underweight woman); “the figure that represents someone who is normal weight” (average woman); “the figure that represents someone who is overweight” (overweight woman);“the figure that represents someone who is obese” (obese woman);“the thinnest figure you think is acceptable for a woman” (thinnest woman acceptable to you); “the thinnest figure men in general would say is acceptable for a woman” (thinnest woman acceptable to men);“the heaviest figure you think is acceptable for a woman” (heaviest woman acceptable to you); and “the heaviest figure men in general would say is acceptable for a woman” (heaviest woman acceptable to men).

Body dissatisfaction was calculated for each individual as current size minus ideal size (46). This simple measure of body dissatisfaction was established by Fallon and Rozin (46, 47) and has been utilized in numerous body image studies, with demonstrated reliability and validity (13).

Procedure

Participants were recruited from colleges, community organizations, churches, and work establishments in the urban Los Angeles area. Effort was made to recruit men and women from various ethnic communities. Individuals were approached by an experimenter and asked to participate in a brief questionnaire study. Completed questionnaires were returned directly to the experimenter by being placed in a box. Informed consent was obtained, and involvement was voluntary (no compensation was provided) and anonymous (i.e., questionnaires were identified only by numbers and signed consents were kept separately).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. References

Sample Characteristics

Preliminary analyses indicated significant ethnic and gender differences on age, BMI, and educational level. Compared with men, women were significantly younger (mean age: 26.2 vs. 23.3 years, respectively, F(1,1218) = 5.6, p < 0.0001), as expected had significantly lower BMIs(mean BMI = 26.2 vs. 23.9 kg/m2, respectively, F(1,1207) = 7.2, p < 0.0001), and were significantly less educated(mean educational level = 4.7 vs. 6.5, respectively, F(1,1223) = 4.4, p < 0.0001). Among women, Asians and Hispanics were significantly younger than whites and blacks (F(3,792) = 42.9, p = 0.0001), Asians had significantly lower and blacks significantly higher BMIs than the other groups (F(3,783) = 45.7, p = 0.0001), and whites had significantly higher educational levels than the other groups (F(3,797) = 19.0, p = 0.0001). For men a similar pattern of ethnic differences existed: Asians and Hispanics were significantly younger than whites and blacks (F(3,420) = 36.8, p = 0.0001), Asians had significantly lower mean BMIs than the other groups (F(3,418) = 11.8, p = 0.0001), and Hispanics had significantly lower educational levels than the other groups (F(3,420) = 10.9, p = 0.0001; see Table 1 for mean values). Therefore, these variables were controlled by entering them as covariates in the subsequent main analyses.

Table 1.  Sample characteristics*
 AsianBlackHispanicWhite
 Men (n = 99)Women (n = 189)Men (n = 76Women (n = 132)Men (n = 169)Women (n = 379)Men (n = 84)Women (n = 101)
 Mean(SD)Mean(SD)Mean(SD)Mean(SD)Mean(SD)Mean(SD)Mean(SD)Mean(SD)
  • *

    N = 1229; education level: 1 = grade 6 or less; 2 = grades 7 to 12 without graduating; 3 = high school graduate or high school equivalent; 4 = some college; 5 = graduate of a 2-year college; 6 = graduate of a 4-year college; 7 = some graduate or professional school; and 8 = completed graduate or professional school.

Age24.1(8.1)22.0(6.6)33.2(11.5)27.5(10.3)22.2(5.5)21.1(5.3)31.0(12.0)28.6(11.2)
Body mass index23.6(4.0)20.9(3.5)27.5(6.8)27.3(6.2)26.9(4.9)24.5(5.2)26.8(4.4)23.2(4.2)
Education5.8(2.9)4.4(2.9)7.1(2.4)4.1(2.3)4.1(2.7)4.1(2.7)7.5(2.7)7.0(3.0)

Ethnic Differences in Body Image and Body Size Assessments for Women and Men

Among women, preliminary multivariate ANOVA was significant for the effect of ethnicity/race (F(45,1,977) = 4.2, p = 0.0001, Pillai Trace = 0.26). Univariate tests revealed significant race differences on most figure ratings: current, ideal, realistic, attractive woman, normal-weight woman, overweight woman, obese woman, thinnest woman acceptable to you, heaviest woman acceptable to you, and body dissatisfaction. With age, BMI, and educational level entered as covariates, overall MANCOVA was significant for the effect of race (F(45,1,911) = 1.5, p = 0.02, Pillai Trace = 0.10). Univariate tests revealed, however, few differences: white women chose a somewhat thinner female figure as attractive to men than did black women (F(3,649) = 3.8, p = 0.01); Asian women chose a somewhat larger female figure as being underweight than did black women (F(3,649) = 3.2, p = 0.02); and Asian women reported less body dissatisfaction than the other groups (F(3,649) = 2.9, p = 0.04). There were no significant race differences on the remaining variables. (Mean values for women are presented inTable 2.)

Table 2.  Figure ratings by women of different ethnic groups*
 Asian (n = 189)Black (n = 132)Hispanic (n = 379)White (n = 101)
 Mean(SD)Mean(SD)Mean(SD)Mean(SD)
  • *

    N = 801. Unadjusted mean values are shown.

Current figure3.4(1.1)4.1(1.3)3.9(1.3)3.7(1.2)
Ideal figure2.7(0.7)3.1(0.8)2.9(0.8)2.7(0.8)
Body dissatisfaction0.7(1.1)1.0(1.1)1.0(1.1)1.0(0.9)
Realistic figure2.5(0.9)3.0(0.9)2.8(0.9)2.7(0.8)
Attractive woman2.7(0.8)3.1(0.8)2.8(0.7)2.7(0.8)
Attractive man3.7(0.9)3.8(1.0)3.7(0.9)3.8(0.8)
Woman attractive to men2.7(0.8)2.8(0.9)2.7(0.8)2.6(0.7)
Man attractive to women3.7(0.9)3.6(0.9)3.5(0.9)3.6(0.9)
Underweight woman1.8(0.8)1.6(0.7)1.7(0.7)1.7(0.7)
Normal-weight woman2.9(0.6)3.0(0.7)3.0(0.6)2.8(0.7)
Overweight woman5.5(1.1)5.8(1.1)5.7(1.2)5.3(0.9)
Obese woman6.9(1.7)7.5(1.1)7.5(1.2)7.3(1.1)
Thinnest woman acceptable to you2.4(0.8)2.6(0.9)2.6(0.9)2.3(0.7)
Thinnest woman acceptable to men2.5(0.9)2.4(0.8)2.5(0.9)2.2(0.8)
Heaviest woman acceptable to you5.0(1.1)4.8(0.9)5.1(1.1)5.2(1.1)
Heaviest woman acceptable to men4.7(1.1)4.3(1.2)4.4(1.1)4.6(1.2)

For men, multivariate ANOVA was significant for the effect of ethnicity/race (F(45,1,053) = 2.1, p = 0.0001, Pillai Trace = 0.25). Univariate tests demonstrated significant race differences on the following ratings: attractive woman, normal-weight woman, overweight woman, and body dissatisfaction. With age, BMI, and educational level entered as covariates in a MANCOVA, the effect of race was no longer significant (F(45,1,011) = 0.8, p = 0.85, Pillai Trace = 0.10), and univariate tests revealed no significant differences between the ethnic groups. (Mean values for men are presented in Table 3.)

Table 3.  Figure ratings by men of different ethnic groups*
 Asian (n = 99)Black (n = 76)Hispanic (n = 169)White (n = 84)
 Mean(SD)Mean(SD)Mean(SD)Mean(SD)
  • *

    N = 428. Unadjusted mean values are shown.

Current figure4.1(1.5)4.6(1.7)4.6(1.3)4.6(1.4)
Ideal figure4.0(0.9)3.9(0.9)3.9(0.9)3.8(0.8)
Body dissatisfaction0.1(1.4)0.7(1.4)0.6(1.4)0.8(1.3)
Realistic figure3.4(1.1)3.5(1.0)3.2(1.1)3.2(1.0)
Attractive woman3.1(0.6)3.5(0.7)3.4(1.0)3.2(0.7)
Attractive man4.0(0.8)3.8(0.8)3.8(1.0)3.7(0.9)
Woman attractive to men2.8(0.9)3.0(0.9)3.0(0.8)3.0(0.7)
Man attractive to women3.9(0.9)3.5(1.0)3.6(1.0)3.7(0.9)
Underweight woman2.0(0.7)1.9(0.7)2.0(1.0)1.9(0.8)
Normal-weight woman3.3(0.6)3.2(0.7)3.3(0.7)2.9(0.7)
Overweight woman5.7(1.1)5.8(1.0)5.8(1.1)5.3(0.8)
Obese woman7.3(1.4)7.2(1.3)7.4(1.1)7.1(0.9)
Thinnest woman acceptable to you2.6(0.8)2.8(1.0)2.7(0.8)2.6(0.9)
Thinnest woman acceptable to men2.6(1.0)2.7(0.9)2.7(0.8)2.3(0.7)
Heaviest woman acceptable to you5.3(1.4)5.0(1.1)5.1(1.1)4.8(1.2)
Heaviest woman acceptable to men4.8(1.3)4.8(1.4)4.6(1.2)4.4(1.0)

In summary, when controlling for age, BMI, and educational level, for men there were no ethnic differences in perceptions of attractive and acceptable figures, and for women there were very few ethnic differences. White women did think that men preferred a thinner figure, but this difference was small.

Gender Differences in Body Dissatisfaction and Body Size Assessments

We conducted a two-factor (gender × race) MANCOVA (with age, BMI, and educational level as covariates) to examine gender differences and any interactions between race and gender on body dissatisfaction and on standards for male and female shapes. Overall MANCOVA for the effect of gender was significant (F(13,1009) = 3.5, p = 0.0001, Pillai Trace = 0.04). As expected, women reported more body dissatisfaction than men (mean for women = 0.9 vs. 0.6 for men, F(1,1021) = 23.1, p = 0.0001). While 26% (110/428) of men wanted to gain weight (i.e., become larger in size), only 9% (73/801) of women wanted to gain weight; more women (66%) than men (54%) wanted to lose weight.

Compared with men, women chose: a thinner female figure as attractive(mean = 2.8 for women vs. 3.3 for men, F(1,1021) = 7.1, p = 0.008); a thinner female figure as thinnest acceptable to you(mean = 2.5 for women vs. 2.7 for men, F(1,1021) = 6.6, p = 0.01); and a somewhat thinner female figure as heaviest acceptable to you (mean = 5.0 for women vs. 5.1 for men, F(1,1021) = 4.1, p = 0.04). Additionally, what women believed that men would rate as an attractive female figure (woman attractive to men) was thinner than what men actually rated as an attractive woman (mean = 2.7 vs. 3.3, respectively, F(1,1021) = 12.1, p = 0.0005), and what women believed that men would rate as thinnest acceptable female figure was thinner than what men actually rated (mean = 2.5 vs. 2.7, respectively, F(1,1021) = 5.6, p = 0.02). For attractive male figure, there were no significant gender differences (mean = 3.7 for women vs. 3.8 for men, F(1,1021) = 3.2, p = 0.08), and men accurately rated the male figure believed to be attractive to women (mean = 3.6 for women vs. 3.7 for men, F(1,1021) = 0.1, p = 0.7). (Mean ratings for women and men are presented in Table 4.)

Table 4.  Figure-rating comparisons between the sexes*
 Women (n = 801)Men (n = 428)
 Mean(SD)Mean(SD)
  • *

    N = 1229. Adjusted mean values are shown. Overall MANCOVA for effect of sex: F(13,1009) = 3.5, p = 0.0001, Pillai Trace = 0.04.

  • Significant sex difference at p < 0.001.

  • Significant sex difference at p < 0.01.

  • §

    Significant sex difference at p < 0.05.

Body dissatisfaction0.9(1.1)0.6(1.4)
Attractive woman2.8(0.8)3.3(0.8)
Attractive man3.7(0.9)3.8(0.9)
Woman attractive to men2.7(0.8)3.0(0.8)
Man attractive to women3.6(0.9)3.7(1.0)
Underweight woman1.7(0.7)2.0(0.8)
Normal-weight woman3.0(0.6)3.2(0.7)
Overweight woman5.6(1.1)5.7(1.0)
Obese woman7.3(1.3)7.3(1.2)
Thinnest woman acceptable to you§2.5(0.8)2.7(0.8)
Thinnest woman acceptable to men2.5(0.9)2.6(0.9)
Heaviest woman acceptable to you§5.0(1.1)5.1(1.2)
Heaviest woman acceptable to men4.5(1.2)4.7(1.2)

In terms of the interaction between gender and race, white women chose the thinnest and black men the heaviest female figure as attractive to men (F(3,1021) = 3.5, p = 0.02). However, overall MANCOVA for the interaction between gender and race was not significant (F(39,3033) = 1.3, p = 0.1, Pillai Trace = 0.05). In the total sample, the effect of race was not significant (overall F(39,2989) = 1.2, p = 0.2, Pillai Trace = 0.04).

In summary, compared with men, women reported more body dissatisfaction, chose thinner female figures as attractive and acceptable, and believed that men preferred thinner female figures than they actually did. There was agreement between the sexes on what is an attractive male figure.

Correlations between Demographic Variables and Figure Ratings

Post hoc exploratory analyses revealed several significant correlations between demographic variables and figure ratings. Among the total sample of men and women, there were significant positive correlations for age and BMI with body image and body dissatisfaction: older and heavier individuals tended to choose larger figures as their current (for age: r = 0.20, p < 0.0001; for BMI: r = 0.79, p < 0.0001), desired (for age: r = 0.12, p = 0.005; for BMI: r = 0.41, p < 0.0001), and realistic figures (for age: r = 0.14, p = 0.0003; for BMI: r = 0.56, p < 0.0001), and reported more body dissatisfaction(for age: r = 0.16, p < 0.0001; for BMI: r = 0.64, p < 0.0001). BMI and educational level were significantly correlated with several ratings of attractive and acceptable figures. Heavier individuals chose larger attractive male (r = 0.14, p = 0.0004) and attractive female figures (r = 0.25, p < 0.0001), larger normal-weight female figures (r = 0.13, p = 0.0007), and larger thinnest acceptable female figures (r = 0.22, p < 0.0001). Those with higher levels of education also chose larger attractive male (r = 0.11, p = 0.003) and attractive female figures (r = 0.10, p = 0.009), and at the same time chose thinner female figures as overweight (r =−0.10, p = 0.008).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. References

Our results do not support the hypothesis that there are ethnic differences in acceptable body size or that there is a tolerance of larger body sizes by blacks and Hispanics, which would create subcultures of women where obesity is acceptable. Most ethnic differences in body image and perceptions of attractive and acceptable female body size disappeared once differences in age, educational level, and BMI were controlled. We found no ethnic differences among the men in our sample, and very few differences among the women.

We did find that Asian women reported less body dissatisfaction(defined as current minus ideal figure ratings) compared with the other groups, which is consistent with research demonstrating less body dissatisfaction in Asian compared with white women (21, 22, 23). In our study, black women were similar to white and Hispanic women in their levels of body dissatisfaction, a finding that is not consistent with what generally has been found by most other research; other studies, however, typically have not controlled for age, SES, or BMI. White women, compared with black women, chose a somewhat thinner female figure as attractive to men; white women, therefore, may be more likely to pursue a thinner shape because they believe it to be more preferred by men. Surprisingly, Asian women chose a larger figure as representing an underweight female than did black women. There were no ethnic differences on any of the other variables for women.

Consistent with other research (15, 19, 22, 24, 27), we found no ethnic differences among the men on any variables of body image or ratings of attractive and acceptable female figures. Specifically, we found no significant ethnic group differences for preferred female body size. Black men did choose the heaviest female figure as attractive, but this difference was not significant. Although some previous studies (31, 32) have reported to have found a significant ethnic difference in romantic partner size, we found differences to no longer exist once we controlled for age, education, and BMI.

An interesting observation from our findings was that demographic characteristics not only influence a person's own body image and body dissatisfaction, but also his or her perceptions of attractive and acceptable female size. Specifically, we found that BMI was positively correlated with figure ratings, so that individuals who were heavier tended to choose larger acceptable female and attractive figures. Individuals with more education also chose larger attractive figures, but displayed a lower threshold for female obesity.

As with other research that has found gender differences in body image (15, 38, 41), we found significant differences between men and women regardless of ethnicity. Men were less dissatisfied with their size than women, were more likely to want to gain weight, and were accurate in their perception of the male figure women find attractive. Women chose thinner female figures as attractive and acceptable than did men and perceived that men prefer thinner female figures than they actually do. White women had the most distorted perception of the female figure men find attractive, a result also reported by Demarest and Allen (15). This pattern might help explain why white women are prone to developing eating disorders, such as anorexia nervosa, where drive for thinness is a core symptom (49); they (mis)perceive, even more so than other female ethnic groups, that men prefer thinner female figures.

One interpretation of our findings is that marked ethnic or cultural differences in preferences for female size or tolerance for overweight female shapes do not exist. Another interpretation is that such ethnic differences do exist, but that factors such as SES, age, and BMI are more powerful determinants or contributors to body-size perceptions (14). The issue of whether ethnic or cultural differences in preference for female shape exist is a complex one. There are many different cultural or social forces that may influence weight. For example, although perceptions of body size may be the same across ethnicities, some cultures may be less critical of deviance from the ideal or norm, creating less pressure to achieve a certain body size. Results from Kemper et al. (13), demonstrating that white girls felt encouraged by others to lose weight, whereas black girls believed that their size was satisfactory to others, support this proposal. Furthermore, body image and body-size preference are broad and complex constructs that can be defined and measured in many ways. Large-scale research investigations that use multi-faceted and cross-cultural approaches are needed to shed further light on this issue.

One limitation of our study was that it was correlational in nature and, therefore, could not address causality or rule out the possibility that ethnicity does influence body-sizepreferences. Our results suggest that the tendency of blacks and Hispanics to be overweight influences the preference or tolerance for larger sizes, because BMI was a strong determinant of body image and body-size perceptions. However, cultural acceptance of larger sizes may produce the tendency to be overweight in the first place. Longitudinal studies of children of different ethnic groups, examining body image and size preferences before a history of overweight is established, are needed. Other limitations were that we used only one measure (i.e., figure ratings) to assess body image and size perceptions, and our sample overall was fairly acculturated to U.S. society because all participants could read and write English (to fill out the questionnaires). It can be debated whether discrepancy between perceived body size and ideal body size actually constitutes a measure of body dissatisfaction; therefore, other measures of body dissatisfaction should be used in future studies. Additionally, measures of body image importance may be helpful, because dissatisfaction with one's body may assume a higher level of importance among some ethnic groups compared with others. Future studies should use multiple measures (ideally translated into participants’ native languages) to assess body image and should include large samples of ethnic minority populations that represent a wide range of acculturation levels.

In summary, this study investigated body image and perceptions of attractive, acceptable, and typical female figures, across a range of sizes from underweight to obese, in a large community sample of Asian, black, Hispanic, and white men and women. Wide ranges of age, educational level, and BMI were represented, and differences among groups on these variables were controlled. The findings suggested that ethnicity alone does not markedly influence perceptions of female body size. Future studies of body image should control for confounding effects of age, SES, and body weight.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. References