## Abstract

The authors assessed the associations of ethnicity and socioeconomic status (SES) with body size judgments in Black and White young adults. Self-perceived and ideal body size judgments were measured using the Stunkard nine-figure scale (higher value = larger body) at the year 7 examination (1992–1993) of the Coronary Artery Risk Development in Young Adults (CARDIA) Study. In sex-specific adjusted multiple regression models, the difference between self-perceived and ideal body size judgments was larger for Whites than for Blacks: 0.74 vs. 0.57 for White men vs. Black men (p < 0.05) and 1.48 vs. 0.96 for White women vs. Black women (p < 0.0001). This ethnic difference was evident in all body mass index-stratified adjusted models (all p's < 0.05). In ethnicity/sex-specific adjusted models, lower education was associated with a smaller difference between self-perceived and ideal body size for all groups except White women (p's for trend: White women, 0.57; Black women, <0.0001; White men, 0.0007; Black men, 0.016). Judgments of self-perceived body size differed by ethnicity but not by SES, and judgments of ideal body size differed by SES but not by ethnicity. Learning to make medically accurate judgments of healthy body size may increase the motivation to lose weight in some persons.

Excess body weight is an important risk factor for cardiovascular disease (1). Prevalences of overweight and obesity vary according to Black/White ethnicity and socioeconomic status (SES) (2–5). The prevalence of obesity is higher among Blacks than among Whites, especially in women (2), and higher among persons of lower SES (3, 4). One possible explanation for the observed ethnic and socioeconomic differences in body mass index (BMI) is body size satisfaction. Body size satisfaction is related to weight loss attempts—persons who are more satisfied with their body size are generally less likely to attempt weight loss (6–10). A number of studies have found that Blacks are more satisfied with their body size than Whites (11–14) and that people of lower SES tend to be more satisfied with their body size than people of higher SES (9, 15). This pattern, in which the ethnic and SES groups with the highest BMIs also have the highest levels of body size satisfaction, suggests that high levels of body size satisfaction may be an important barrier to the efficacy of interventions designed to reduce weight among Blacks and low-SES persons.

A number of questions remain about the relation between body size satisfaction, Black/White ethnicity, and SES. First, it is unknown whether Black/White ethnicity and SES are independently related to body size satisfaction. We do not know of any study that has examined the relation between SES and body size satisfaction among persons from the United States. It is possible that ethnic differences in body size satisfaction may be partly due to ethnic differences in SES. Another limitation of previous work on ethnic differences in body size satisfaction is that many of the investigators did not adjust their results for BMI or adjusted for self-reported BMI rather than measured BMI. Ethnic differences in the correlation between measured and self-reported BMI may drive apparent differences in body size satisfaction.

In the current study, we addressed these limitations by measuring the associations between Black/White ethnicity and SES and body size satisfaction, adjusting for measured BMI as well as other factors. We used the discrepancy between self-perceived body size and ideal body size (body size discrepancy) as a measure of body size satisfaction. Our hypothesis was that, independently of SES, Blacks would have less body size discrepancy than Whites, indicating greater body size satisfaction, and that participants with lower SES would have lower body size discrepancy scores than participants with higher SES, also indicating greater body size satisfaction. We further hypothesized that these relations would be independent of BMI.

## MATERIALS AND METHODS

### Study population

The Coronary Artery Risk Development in Young Adults (CARDIA) Study is a multicenter, longitudinal study of the evolution of cardiovascular disease risk factors in Black and White young adults. Participants were aged 18–30 years when they were recruited in 1985–1986 from four US cities: Birmingham, Alabama; Chicago, Illinois; Minneapolis, Minnesota; and Oakland, California (16). The CARDIA sample, comprising 5,115 participants at baseline, was designed to include approximately equal numbers of subjects by ethnicity (Black/White), sex, age (18–24 years/25–30 years), and education (high school or less/more than high school). Data from the year 7 examination (1992–1993) were used in the current study.

### Exclusions

Of the 4,086 participants who attended the CARDIA year 7 examination, participants were excluded from the current analysis if they were missing data on the body size judgment variables (n = 67), missing BMI measurements at year 7 (n = 86), missing data on years of education at year 7 (n = 7) or were pregnant (n = 13). This resulted in a final sample size of 3,913.

### Measurements

Body size judgments were obtained using the Stunkard figure rating scale (see figure 1), which consists of a nine-figure scale of numbered silhouettes that increase gradually in size from very thin (a value of 1) to very obese (a value of 9) (17). Two body size perception variables were included in the current study. “Self-perceived body size” is the number of the figure selected by participants in response to the prompt “Choose the figure that reflects how you think you look.” “Ideal body size” is the number of the figure chosen in response to the prompt “Choose your ideal figure.” This scale has good test-retest reliability and adequate validity (10, 13). Following the methods of a number of other investigators (13, 18–20), we defined body size satisfaction as the difference between self-perceived body size and ideal body size. A body size discrepancy variable was created for each participant by subtracting the number of the figure selected as the ideal body size from the number of the figure selected as the self-perceived body size. A high body size discrepancy value signifies low satisfaction with body size, and a low value signifies greater satisfaction with body size.

FIGURE 1.

The nine-figure body size scale of Stunkard et al. (17). Reprinted with permission from The Genetics of Neurological and Psychiatric Disorders (17). Copyright 1983, Raven Press.

FIGURE 1.

The nine-figure body size scale of Stunkard et al. (17). Reprinted with permission from The Genetics of Neurological and Psychiatric Disorders (17). Copyright 1983, Raven Press.

Height and weight were measured while participants were wearing no shoes and light clothing. BMI was calculated as the ratio of weight (kg) to standing height (m) squared (kg/m2). Age, ethnicity, years of education, sex, and marital status were assessed by self-report. Participants reported their annual family income by selecting one of the following eight categories: <$5,000;$5,000–$11,999;$12,000–$15,999;$16,000–$24,999;$25,000–$34,999;$35,000–$49,999;$50,000–$74,999; and ≥$75,000.

SES is commonly measured using occupation, income, or education (21). In the current study, we used educational level and income as indicators of SES.

### Statistical analysis

Participant characteristics and the proportion of participants at each educational level and in each BMI stratum at the year 7 examination were calculated by ethnicity/sex group. We tested for ethnic differences within each sex group using t tests for continuous variables and chi-squared tests for categorical variables.

We calculated the proportion of participants in each ethnicity/sex group who selected each figure size for the self-perceived and ideal body size variables and the proportion of participants at each value of the body size discrepancy score. The distribution data are presented because most responses across groups were clustered at one or two levels of each variable. The mean values for self-perceived body size judgment for White women, Black women, White men, and Black men were 4.4, 4.6, 4.7, and 4.5, respectively. Corresponding means for ideal body size judgment were 3.1, 3.5, 3.9, and 4.0.

The association of Black/White ethnicity with body size judgment was evaluated using sex-specific linear regression models stratified by BMI group, with adjustment for years of education, age, and marital status. Separate models were created for each of the three body size variables: self-perceived body size, ideal body size, and body size discrepancy. Three BMI strata were used: normal weight (BMI 18.5–<25 kg/m2), overweight (BMI 25–<30 kg/m2), and obese (BMI ≥30 kg/m2). Because of small numbers of underweight participants (n = 71), persons with a BMI of <18.5 kg/m2 were excluded from BMI-stratified analyses. In addition to the stratified models, we used sex-specific linear regression models to examine the effect of Black/White ethnicity with adjustment for continuous BMI, years of education, age, and marital status. All of the analyses described above were repeated with adjustment for income level instead of education. Regression coefficients were also calculated for assessment of the relation between BMI and both body size variables.

The association of SES with body size judgment was evaluated using three sets of ethnicity/sex-specific linear regression models, with ideal body size judgment, self-perceived body size judgment, and body size discrepancy included as dependent variables and with adjustment for continuous BMI, age, and marital status. In the first set of models, years of education was included as a continuous independent variable, and tests for trend were performed. Mean values for five levels of education are also presented. A second set of analyses was performed in which years of education was replaced with level of income.

## RESULTS

Descriptive statistics are shown in table 1. Black participants were less likely to be married, were slightly younger, had fewer years of education, and had higher BMIs than White participants.

TABLE 1.

Demographic characteristics of participants, Coronary Artery Risk Development in Young Adults (CARDIA) Study, 1992–1993†

 Women Men Whites (n = 1,041) Blacks (n = 1,080) Whites (n = 988) Blacks (n = 804) Mean or no. % Mean or no. % Mean or no. % Mean or no. % Mean age (years) 32.6 31.6*** 32.5 31.4*** % married 54 33*** 50 37*** Mean years of education 15.4 13.9*** 15.3 13.5* Mean BMI‡,§ 25.03 29.0*** 26.0 27.0*** BMI group Underweight (BMI <18.5) 28 2.7 25 2.3 8 0.8 10 1.2 Normal weight (BMI 18.5–24.9) 632 60.7 376 34.8 437 44.3 310 38.5 Overweight (BMI 25–29.9) 213 20.5 265 24.5 406 41.1 296 37.0 Obese (BMI ≥30) 168 16.1 414 38.3*** 137 13.9 188 23.3*** Education (years) <12 29 2.8 66 6.1 39 4.0 83 10.3 12 182 17.5 300 27.8 166 16.8 263 32.7 >12–<16 221 21.2 455 42.1 242 24.5 283 35.2 16 305 29.3 183 16.9 278 28.1 112 13.9 >16 304 29.2 76 7.0*** 263 26.6 63 7.8*** Annual income <$16,000 130 12.6 314 29.7 99 10 218 27.6$16,000–$34,999 291 28.2 375 35.4 253 25.7 273 34.6$35,000–$49,999 213 20.7 186 17.6 211 21.4 148 18.8$50,000–$74,999 205 19.9 133 12.6 224 22.7 94 11.9 ≥$75,000 192 18.6 50 4.7*** 199 20.2 56 7.1***
 Women Men Whites (n = 1,041) Blacks (n = 1,080) Whites (n = 988) Blacks (n = 804) Mean or no. % Mean or no. % Mean or no. % Mean or no. % Mean age (years) 32.6 31.6*** 32.5 31.4*** % married 54 33*** 50 37*** Mean years of education 15.4 13.9*** 15.3 13.5* Mean BMI‡,§ 25.03 29.0*** 26.0 27.0*** BMI group Underweight (BMI <18.5) 28 2.7 25 2.3 8 0.8 10 1.2 Normal weight (BMI 18.5–24.9) 632 60.7 376 34.8 437 44.3 310 38.5 Overweight (BMI 25–29.9) 213 20.5 265 24.5 406 41.1 296 37.0 Obese (BMI ≥30) 168 16.1 414 38.3*** 137 13.9 188 23.3*** Education (years) <12 29 2.8 66 6.1 39 4.0 83 10.3 12 182 17.5 300 27.8 166 16.8 263 32.7 >12–<16 221 21.2 455 42.1 242 24.5 283 35.2 16 305 29.3 183 16.9 278 28.1 112 13.9 >16 304 29.2 76 7.0*** 263 26.6 63 7.8*** Annual income <$16,000 130 12.6 314 29.7 99 10 218 27.6$16,000–$34,999 291 28.2 375 35.4 253 25.7 273 34.6$35,000–$49,999 213 20.7 186 17.6 211 21.4 148 18.8$50,000–$74,999 205 19.9 133 12.6 224 22.7 94 11.9 ≥$75,000 192 18.6 50 4.7*** 199 20.2 56 7.1***
*

p < 0.05;

**

p < 0.01;

***

p < 0.001.

Ethnic differences within each sex group were tested with t tests and chi-squared tests for continuous and categorical variables, respectively.

BMI, body mass index.

§

Weight (kg)/height (m)2.

Overall, Black men were somewhat more likely than White men to select smaller figures for the self-perceived body size variable, whereas Black women were more likely than White women to select larger figures for the self-perceived body size variable (table 2). The ideal body size variable showed less variation than the self-perceived body size variable; a majority of men selected from only three figures (figure numbers 3–5), and a majority of women selected from only two figures (figure number 3 or 4). Nonetheless, in both sex groups, Black participants tended to select larger ideal figures than White participants. Finally, in both sex groups, Black participants had lower body size discrepancy scores than White participants (table 2), suggesting that they were more satisfied with their body size (0.74 vs. 0.57 for White men vs. Black men (p < 0.05) and 1.48 vs. 0.96 for White women vs. Black women (p < 0.0001)).

TABLE 2.

Distribution of body size judgment scores† across ethnicity/sex groups, Coronary Artery Risk Development in Young Adults (CARDIA) Study, 1992–1993‡

 Women Men Whites Blacks Whites Blacks Self-perceived body size (%) 1 or 2 (very thin) 7.3 9.7 7.8 10.7 3 14.0 13.5 13.8 17.8 4 35.5 22.0 16.4 18.8 5 25.8 28.3 34.7 28.4 6 10.9 15.6 19.2 15.9 7–9 (very large) 6.5 11.0** 8.2 8.4* Perceived ideal body size (%) 1 or 2 (very thin) 15.0 11.0 2.5 5.7 3 59.0 44.1 29.7 26.1 4 24.0 34.4 45.1 35.7 5 1.7 8.1 21.5 28.8 6–9 (very large) 0.4 2.4** 1.1 3.7** Body size discrepancy (self-perceived − ideal) Too thin/just right (≤0) 19.7 30.5 34.0 52.7 A bit too fat (1) 42.8 30.3 34.8 23.5 Much too fat (>1) 37.5 39.2** 31.2 23.8**
 Women Men Whites Blacks Whites Blacks Self-perceived body size (%) 1 or 2 (very thin) 7.3 9.7 7.8 10.7 3 14.0 13.5 13.8 17.8 4 35.5 22.0 16.4 18.8 5 25.8 28.3 34.7 28.4 6 10.9 15.6 19.2 15.9 7–9 (very large) 6.5 11.0** 8.2 8.4* Perceived ideal body size (%) 1 or 2 (very thin) 15.0 11.0 2.5 5.7 3 59.0 44.1 29.7 26.1 4 24.0 34.4 45.1 35.7 5 1.7 8.1 21.5 28.8 6–9 (very large) 0.4 2.4** 1.1 3.7** Body size discrepancy (self-perceived − ideal) Too thin/just right (≤0) 19.7 30.5 34.0 52.7 A bit too fat (1) 42.8 30.3 34.8 23.5 Much too fat (>1) 37.5 39.2** 31.2 23.8**
*

p < 0.01;

**

p < 0.0001.

Determined according to the nine-figure body size scale of Stunkard et al. (17).

Ethnic differences in distributions of body size judgment scores within sex groups were tested with chi-squared tests.

Table 3 shows that, independently of BMI, Black/White ethnicity was significantly associated with judgments of self-perceived body size and body size discrepancy. After adjustment for years of education, age, and marital status, in BMI-stratified models both male and female Black participants chose a smaller figure than White participants to represent their current (self-perceived) body size. In contrast, in every BMI stratum, except among normal-weight women, there was no statistically significant association between ethnicity and participants' judgments of ideal body size after adjustment for education, age, and marital status. Table 3 further indicates that in every BMI stratum, except among normal-weight men, Black participants perceived themselves as being closer to their ideal body size than did White participants (they had lower body size discrepancy scores). The mean values and significance values were virtually identical when education was replaced with income in the models (data not shown).

TABLE 3.

Adjusted mean values for body size perception† by ethnicity and body mass index, Coronary Artery Risk Development in Young Adults (CARDIA) Study, 1992–1993‡

 Women Men Whites Blacks Whites Blacks Self-perceived body size Normal weight (BMI§,¶ 18.5–24.9) 3.77 3.36**** 3.74 3.35**** Overweight (BMI 25–29.9) 5.04 4.64**** 5.16 4.80**** Obese (BMI ≥30) 6.3 5.96** 6.47 5.90**** Perceived ideal body size Normal weight (BMI 18.5–24.9) 2.93 3.04* 3.81 3.74 Overweight (BMI 25–29.9) 3.34 3.33 3.9 3.85 Obese (BMI ≥30) 3.81 3.93 4.43 4.34 Perceived body size discrepancy (current − ideal) Normal weight (BMI 18.5–24.9) 0.84 0.32**** −0.07 −0.39*** Overweight (BMI 25–29.9) 1.7 1.31**** 1.26 0.94**** Obese (BMI ≥30) 2.5 2.03**** 2.04 1.56***
 Women Men Whites Blacks Whites Blacks Self-perceived body size Normal weight (BMI§,¶ 18.5–24.9) 3.77 3.36**** 3.74 3.35**** Overweight (BMI 25–29.9) 5.04 4.64**** 5.16 4.80**** Obese (BMI ≥30) 6.3 5.96** 6.47 5.90**** Perceived ideal body size Normal weight (BMI 18.5–24.9) 2.93 3.04* 3.81 3.74 Overweight (BMI 25–29.9) 3.34 3.33 3.9 3.85 Obese (BMI ≥30) 3.81 3.93 4.43 4.34 Perceived body size discrepancy (current − ideal) Normal weight (BMI 18.5–24.9) 0.84 0.32**** −0.07 −0.39*** Overweight (BMI 25–29.9) 1.7 1.31**** 1.26 0.94**** Obese (BMI ≥30) 2.5 2.03**** 2.04 1.56***
*

p < 0.05;

**

p < 0.01;

***

p < 0.001;

****

p < 0.0001.

Determined according to the nine-figure body size scale of Stunkard et al. (17).

Stratified models were linear regression models adjusted for years of education, age, and marital status.

§

BMI, body mass index.

Weight (kg)/height (m)2.

The estimated regression coefficients for continuous BMI in the model predicting self-perceived body size, adjusting for ethnicity, education, age, and marital status, were 0.18 for women (p < 0.0001) and 0.22 for men (p < 0.0001). Corresponding coefficients for the relation between continuous BMI and perceived ideal body size were 0.06 (p < 0.0001) for women and 0.05 (p < 0.0001) for men.

Multivariate analyses adjusting for BMI, age, and marital status found that neither SES variable was associated with the judgment of self-perceived body size for any ethnicity/sex group (table 4). However, both education and income were negatively associated with ideal body size for all ethnicity/sex groups except White women. At all levels of education and for almost all levels of income, White women had the lowest mean ideal body size of any group. Table 4 shows that, with the exception of White men, body size discrepancy was larger with increasing education across all groups. For body size discrepancy, Blacks showed the same patterns across income and education, but Whites did not. For White women, income was more strongly related to ideal body size than was education, while the reverse was true for White men.

TABLE 4.

Adjusted mean values for body image* by education and income, Coronary Artery Risk Development in Young Adults (CARDIA) Study, 1992–1993

 Educational level Annual income Less than high school High school graduation Some college College graduation Postcollege p trend† <$16,000$16,000–$34,999$35,000–$49,999$50,000–$74,999 ≥$75,000 p trend‡ Self-perceived body image White women 4.54 4.32 4.44 4.41 4.42 0.30 4.34 4.39 4.51 4.44 4.4 0.43 Black women 4.68 4.61 4.6 4.62 4.78 0.34 4.6 4.65 4.67 4.66 4.56 0.40 White men 4.54 4.77 4.65 4.66 4.74 0.82 4.68 4.77 4.71 4.66 4.63 0.23 Black men 4.45 4.54 4.38 4.54 4.49 0.74 4.52 4.5 4.46 4.34 4.45 0.55 Ideal body image White women 3.21 3.08 3.16 3.14 3.13 0.58 3.19 3.06 3.16 3.18 3.15 0.32 Black women 3.9 3.56 3.42 3.36 3.31 <0.0001 3.6 3.54 3.29 3.33 3.16 <0.0001 White men 4.21 3.98 3.95 3.86 3.76 <0.0001 4.07 3.88 3.88 3.97 3.72 0.02 Black men 4.11 4.08 3.94 3.82 3.81 0.003 4.17 4 3.86 3.77 3.9 0.006 Body image discrepancy White women 1.34 1.24 1.28 1.27 1.29 0.57 1.19 1.26 1.29 1.32 1.32 0.13 Black women 0.77 1.06 1.19 1.26 1.47 <0.0001 1 1.1 1.38 1.33 1.4 <0.0001 White men 0.33 0.79 0.71 0.8 0.98 0.0007 0.6 0.89 0.83 0.69 0.91 0.49 Black men 0.33 0.45 0.44 0.72 0.67 0.016 0.35 0.49 0.59 0.57 0.55 0.05
 Educational level Annual income Less than high school High school graduation Some college College graduation Postcollege p trend† <$16,000$16,000–$34,999$35,000–$49,999$50,000–$74,999 ≥$75,000 p trend‡ Self-perceived body image White women 4.54 4.32 4.44 4.41 4.42 0.30 4.34 4.39 4.51 4.44 4.4 0.43 Black women 4.68 4.61 4.6 4.62 4.78 0.34 4.6 4.65 4.67 4.66 4.56 0.40 White men 4.54 4.77 4.65 4.66 4.74 0.82 4.68 4.77 4.71 4.66 4.63 0.23 Black men 4.45 4.54 4.38 4.54 4.49 0.74 4.52 4.5 4.46 4.34 4.45 0.55 Ideal body image White women 3.21 3.08 3.16 3.14 3.13 0.58 3.19 3.06 3.16 3.18 3.15 0.32 Black women 3.9 3.56 3.42 3.36 3.31 <0.0001 3.6 3.54 3.29 3.33 3.16 <0.0001 White men 4.21 3.98 3.95 3.86 3.76 <0.0001 4.07 3.88 3.88 3.97 3.72 0.02 Black men 4.11 4.08 3.94 3.82 3.81 0.003 4.17 4 3.86 3.77 3.9 0.006 Body image discrepancy White women 1.34 1.24 1.28 1.27 1.29 0.57 1.19 1.26 1.29 1.32 1.32 0.13 Black women 0.77 1.06 1.19 1.26 1.47 <0.0001 1 1.1 1.38 1.33 1.4 <0.0001 White men 0.33 0.79 0.71 0.8 0.98 0.0007 0.6 0.89 0.83 0.69 0.91 0.49 Black men 0.33 0.45 0.44 0.72 0.67 0.016 0.35 0.49 0.59 0.57 0.55 0.05
*

Determined according to the nine-figure body size scale of Stunkard et al. (17).

Ethnicity/sex-specific linear regression models adjusting for body mass index, age, and marital status, with educational level included as a continuous variable.

Ethnicity/sex-specific linear regression models adjusting for body mass index, age, and marital status, with income level included as a continuous variable.

## DISCUSSION

Consistent with previous studies, in the current study we found that Blacks were more satisfied with their body size (i.e., they had less body size discrepancy) than were Whites. To our knowledge, this study is the first to show that the Black/White difference in body size satisfaction is independent of SES and BMI, as well as age and marital status. We also confirmed prior research indicating that lower-SES persons are more satisfied with their body size than are higher-SES persons. In addition, we believe this study is the first to demonstrate the relation between SES and body size satisfaction in a US sample. The effect of SES on body size satisfaction was somewhat stronger in Blacks than in Whites. SES was unrelated to body size satisfaction in White women (body size discrepancy was very high in White women at all levels of both income and education). In White men, educational level was strongly associated with body size discrepancy, but income level was not. In Black men and women, both education and income were highly positively related to body size discrepancy.

Ethnicity was related to body size discrepancy through its influence on self-perceived body size judgments. Ethnicity had little influence on ideal body size judgments. At the same BMIs, Black participants selected a smaller figure to represent their current body size than did White participants. This difference was evident across BMI strata, with adjustment for SES, age, and marital status. Other studies using the Stunkard figure rating scale also found that at the same (self-reported) BMIs, Black participants chose smaller figures to represent their current body size than did White participants (22, 23). Unlike previous investigators, we used measured BMI rather than self-reported BMI, and therefore we were able to confirm that ethnic differences in self-perceived body size assessments are independent of actual BMI.

The average BMI for each Stunkard figure selected for self-perceived body size was higher for Blacks than for Whites, but the correlations between the figures selected for self-perceived body size and BMI were almost identical in Blacks and Whites (White women: r = 0.83; Black women: r = 0.82; White men: r = 0.74; Black men: r = 0.73). This indicates that the scale mapped onto the BMI distribution of both groups in the same way. In the US population at large and in the CARDIA sample in particular, the BMI distribution of Blacks is shifted upwards relative to that of Whites. One explanation for current findings is that Blacks interpret the Stunkard figure scale as representing Black body sizes and Whites interpret the scale as representing White body sizes; that is, people judge their own body size relative to that of other people in their ethnic group. Because US Blacks, especially women, are heavier on average than US Whites (2), ethnicity-specific reference points for evaluation of body size would be larger for Blacks than for Whites. Therefore, at a given BMI, relative judgments of self-perceived body size will be smaller for Blacks than for Whites.

In contrast to prior studies, we found very little difference between the two ethnic groups in judgments of ideal body size. Across BMI strata, no difference was observed between Black men and White men; among women, a statistically significant difference was observed only in the normal-weight group (Black women selected a slightly higher ideal body size than White women), but the difference was quite small. Other studies have found that Black women select a larger ideal body size than White women (22, 24–26). One possible explanation for the difference between our findings and those of previous studies is that investigators in many of the previous studies did not adjust for BMI and/or SES. In our study, both variables were highly positively associated with perceived ideal body size. Thus, it is possible that Black/White differences found in previous studies were confounded with BMI and/or SES. Further research is necessary to disentangle the effects of Black/White ethnicity, SES, and BMI on judgments of ideal body size.

While ideal body size was not correlated with ethnicity, it was highly associated with SES. Across groups, with the exception of White women, persons with higher SES selected a smaller ideal body size. Thus, SES differences in body size discrepancy were due to differences in perceived ideal body size rather than differences in self-perceived body size. This finding is consistent with the hypothesis that the relation between SES and obesity is mediated in part by socioeconomic differences in body size perception (27). It further suggests that there are distinct mechanisms underlying ethnic and socioeconomic differences in body size satisfaction. While Black/White ethnic differences appear to be driven by judgments of self-perceived body size, socioeconomic differences appear to be driven by judgments of ideal body size.

White women were unique among ethnicity/sex groups in the current study in that their perception of ideal body size, and thus their body size discrepancy, was unrelated to their SES (both education and income). In fact, at every level of education and income, White women selected a lower mean ideal body size than did any other ethnicity/sex group at any education or income level. Other studies have also failed to find a relation between ideal body size and SES in White women (28). However, SES has been shown to be related to other body judgment measures in White women. For example, higher-SES White women have lower thresholds for defining overweight body sizes than do lower-SES White women, and they report greater dietary restraint (15, 28–30). Our finding that White women at all socioeconomic levels have an extremely low ideal body size may be related to the fact that White women are more likely than Black women, Black men, and White men to link their general self-esteem to their body size (12, 13). Further research is necessary to explore determinants of ideal body size judgment that are unique to White women.

### Implications for obesity interventions

Both of the factors that we found to be associated with higher body size satisfaction, Black ethnicity and low SES, are also associated with higher BMI in the United States (2–4). Higher body size satisfaction may indirectly lead to higher BMI by decreasing motivation to maintain or lose weight. The current study suggests that body image and body satisfaction are complex constructs that are related to the social environment in systematic ways. Further research is necessary to better understand these relations, particularly why SES is related to ideals and ethnicity is related to self-perception. A better understanding of how distinct social factors influence body size satisfaction would suggest new avenues for intervention.

The current findings suggest that lay people and medical professionals have different conceptions of “satisfactory” body size. This difference is greater among low-SES groups and Blacks, who are more likely to be overweight. Lay conceptions are influenced by factors that are not relevant to medical definitions. Medically speaking, healthy body weight is not dependent on ethnicity or SES. Weight control might be facilitated if lay conceptions of body weight more closely approximated medical conceptions. Of course, health-care providers should inform their patients when they are overweight. In addition, providers should instruct their patients in the body sizes that are associated with medical weight categories. It is important that information about healthy and unhealthy weights be taught in a perceptual format as well as a verbal one. This could be achieved by simply showing patients a figure scale like the Stunkard scale and pointing out which figures are underweight, normal weight, overweight, and obese. Perceptual knowledge about healthy body size is necessary to counteract the natural tendency for body size norms to shift as overweight becomes the modal body size.

### Limitations

A possible limitation of this study is that the Stunkard figure rating scale was developed for White populations and therefore may not be appropriate for use with Black persons. To our knowledge, the Stunkard figure rating scale has not been formally validated in Black populations. However, Patt et al. (31) found that the Stunkard scale performed similarly to scales developed specifically for Black women. For self-perceived body size ratings, they found similar average BMI values associated with each figure across four different nine-figure scales, and each scale correlated with BMI. As we mentioned above, in the CARDIA data set used in this study, the correlations of the self-perceived body size variable with BMI were equivalent across ethnic groups. This suggests that the scale was similarly related to BMI in both groups.

Another limitation is the cross-sectional nature of this study. In future research, investigators should address the longitudinal relation between body size perceptions and weight change.

### Conclusions

In summary, we believe this study is the first to demonstrate that Black/White ethnicity and SES are independently associated with body size satisfaction. This study provides evidence that Black/White differences in body size satisfaction are due to differences in judgments of self-perceived body size, whereas socioeconomic differences in body size satisfaction are due to differences in ideal body size judgments. This research suggests that interventions targeting body size norms may increase the motivation of overweight persons to lose weight.

### Abbreviations

Abbreviations
• BMI

body mass index

• CARDIA

Coronary Artery Risk Development in Young Adults

• SES

socioeconomic status

This research was supported by contracts N01-HC-95095, N01-HC-48047, N01-HC-48048, and N01-HC-48049 from the National Heart, Lung, and Blood Institute.

Conflict of interest: none declared.

## References

1.
Wilson
PW
D'Agostino
RB
Sullivan
L
, et al.  .
Overweight and obesity as determinants of cardiovascular risk: the Framingham experience
Arch Intern Med
,
2002
, vol.
162
(pg.
1867
-
72
)
2.
Flegal
KM
Carroll
MD
Ogden
CL
, et al.  .
Prevalence and trends in obesity among US adults, 1999 –2000
JAMA
,
2002
, vol.
288
(pg.
1723
-
7
)
3.
Wardle
J
Waller
J
Jarvis
MJ
Sex differences in the association of socioeconomic status with obesity
Am J Public Health
,
2002
, vol.
92
(pg.
1299
-
304
)
4.
Paeratakul
S
Lovejoy
JC
Ryan
DH
, et al.  .
The relation of gender, race and socioeconomic status to obesity and obesity comorbidities in a sample of US adults
Int J Obes Relat Metab Disord
,
2002
, vol.
26
(pg.
1205
-
10
)
5.
Drewnowski
A
Specter
SE
Poverty and obesity: the role of energy density and energy costs
Am J Clin Nutr
,
2004
, vol.
79
(pg.
6
-
16
)
6.
Lee
RE
Harris
KJ
Catley
D
, et al.  .
Factors associated with BMI, weight perceptions and trying to lose weight in African-American smokers
J Natl Med Assoc
,
2005
, vol.
97
(pg.
53
-
61
)
7.
Jaeger
B
Ruggiero
GM
Edlund
B
, et al.  .
Body dissatisfaction and its interrelations with other risk factors for bulimia nervosa in 12 countries
Psychother Psychosom
,
2002
, vol.
71
(pg.
54
-
61
)
8.
Riley
NM
Bild
DE
Cooper
L
, et al.  .
Relation of self-image to body size and weight loss attempts in Black women: The CARDIA Study
Am J Epidemiol
,
1998
, vol.
148
(pg.
1062
-
8
)
9.
Wardle
J
Waller
J
Rapoport
L
Body dissatisfaction and binge eating in obese women: the role of restraint and depression
Obes Res
,
2001
, vol.
9
(pg.
778
-
87
)
10.
Thompson
JK
Altabe
MN
Psychometric qualities of the figure rating scale
Int J Eat Disord
,
1991
, vol.
10
(pg.
615
-
19
)
11.
Altabe
M
Ethnicity and body image: quantitative and qualitative analysis
Int J Eat Disord
,
1998
, vol.
23
(pg.
153
-
9
)
12.
Baturka
N
Hornsby
PP
Schorling
JB
Clinical implications of body image among rural African-American women
J Gen Intern Med
,
2000
, vol.
15
(pg.
235
-
41
)
13.
Smith
DE
Thompson
JK
Raczynski
JM
, et al.  .
Body image among men and women in a biracial cohort: The CARDIA Study
Int J Eat Disord
,
1999
, vol.
25
(pg.
71
-
82
)
14.
Wilfley
DE
Schreiber
GB
Pike
KM
, et al.  .
Eating disturbance and body image: a comparison of a community sample of adult black and white women
Int J Eat Disord
,
1996
, vol.
20
(pg.
377
-
87
)
15.
McLaren
L
Kuh
D
Women's body dissatisfaction, social class, and social mobility
Soc Sci Med
,
2004
, vol.
58
(pg.
1575
-
84
)
16.
Friedman
GD
Cutter
GR
Donahue
RP
, et al.  .
CARDIA: study design, recruitment, and some characteristics of the examined subjects
J Clin Epidemiol
,
1988
, vol.
41
(pg.
1105
-
16
)
17.
Stunkard
AJ
Sorensen
T
Schulsinger
F
Kety
S
Rowland
L
Sidman
R
, et al.  .
Use of the Danish Adoption Register for the study of obesity and thinness
The genetics of neurological and psychiatric disorders
,
1983
New York, NY
Raven Press
(pg.
115
-
20
)
18.
Sanchez-Johnsen
LA
Fitzgibbon
ML
Martinovich
Z
, et al.  .
Ethnic differences in correlates of obesity between Latin-American and black women
Obes Res
,
2004
, vol.
12
(pg.
652
-
60
)
19.
Snooks
MK
Hall
SK
Wildes
JE
, et al.  .
Relationship of body size, body image, and self-esteem in African American, European American, and Mexican American middle-class women
Health Care Women Int
,
2002
, vol.
23
(pg.
460
-
6
)
20.
Flynn
KJ
Fitzgibbon
M
Body images and obesity risk among black females: a review of the literature
Ann Behav Med
,
1998
, vol.
20
(pg.
13
-
24
)
21.
Krieger
N
Williams
DR
Moss
NE
Measuring social class in US public health research: concepts, methodologies, and guidelines
Annu Rev Public Health
,
1997
, vol.
18
(pg.
341
-
78
)
22.
Becker
DM
Yanek
LR
Koffman
DM
, et al.  .
Body image preferences among urban African Americans and whites from low income communities
Ethn Dis
,
1999
, vol.
9
(pg.
377
-
86
)
23.
Bhuiyan
AR
Gustat
J
Srinivasan
SR
, et al.  .
Differences in body shape representations among young adults from a biracial (Black-White), semirural community: The Bogalusa Heart Study
Am J Epidemiol
,
2003
, vol.
158
(pg.
792
-
7
)
24.
Parnell
K
Sargent
R
Thompson
SH
, et al.  .
Black and white adolescent females' perceptions of ideal body size
J Sch Health
,
1996
, vol.
66
(pg.
112
-
18
)
25.
Powell
Kahn
AS
Racial differences in women's desires to be thin
Int J Eat Disord
,
1995
, vol.
17
(pg.
191
-
5
)
26.
Stevens
J
Knapp
RG
Keil
JE
, et al.  .
Changes in body weight and girths in black and white adults studied over a 25 year interval
Int J Obes
,
1991
, vol.
15
(pg.
803
-
8
)
27.
Sobal
J
Stunkard
AJ
Socioeconomic status and obesity: a review of the literature
Psychol Bull
,
1989
, vol.
105
(pg.
260
-
75
)
28.
Wardle
J
Griffith
J
Socioeconomic status and weight control practices in British adults
J Epidemiol Community Health
,
2001
, vol.
55
(pg.
185
-
90
)
29.
Wardle
J
Robb
KA
Johnson
F
, et al.  .
Socioeconomic variation in attitudes to eating and weight in female adolescents
Health Psychol
,
2004
, vol.
23
(pg.
275
-
82
)
30.
Jeffery
RW
French
SA
Socioeconomic status and weight control practices among 20- to 45-year-old women
Am J Public Health
,
1996
, vol.
86
(pg.
1005
-
10
)
31.
Patt
MR
Lane
AE
Finney
CP
, et al.  .
Body image assessment: comparison of figure rating scales among urban Black women
Ethn Dis
,
2002
, vol.
12
(pg.
54
-
62
)