Body Fat Distribution and Perception of Desirable Female Body Shape by Young Black

Men and Women

Devendra Singh

(Accepted 8 October 1993)

The relation between body fat distribution as measured by waist-to-hip ratio (WHR) and perception of desirable female body shape was investigated in college-age black men and women. Subjects judged attractiveness, various personal qualities, and desirability for long-term relationships of 72 line drawings of female figures that represented three body weight categories (normal, underweight, and overweight) and four levels of WHRs (0.7, 0.8, 0.9, and 1.0). Judgments of attractiveness and desirability for long-term relationships were affected by body weight and the size of the WHR. Both male and female subjects ranked normal weight figures with 0.7 and 0.8 WHR as more attractive and desirable for long-term relationships; neither Underweight nor overweight figures, irrespective of WHR size, were assigned high ranking for these variables. These findings do not support the notion that black young men and women find overweight female figures as desirable and attractive. 0 7994 by John Wiley & Sons, lnc.

The personal and interpersonal consequences of being overweight are apparently not as negative for black women as for white women. Adolescent black females are more satisfied with their bodies and are less desirous of being thin than white adolescent females (Rosen & Gross, 1987). Adult black women have a more tolerant attitude about being fat (Rucker & Cash, 1992) and in general, obesity and body image dissatisfaction do not exert as powerful a negative influence on their lives as it does on those of white women (Thomas, 1989). Furthermore, whereas a majority of white women believe that being thin is a prerequisite to attractiveness, most black women do not. Although there are no studies demonstrating that black men and women idealize fatness, a positive stereotype of fatness has been reported for black Americans. Black men are less likely to refuse to date an overweight woman, and consider them to be attractive and sexy (Harris, Walters, & Waschull, 1991).

Devendra Singh, Ph.D., is Associate Professor of Biopsychology and Behavioral Neurosciences, Department of Psychology, University of Texas at Austin. Address correspondence and reprint requests to the author at Department of Psychoiogy, Mezes 330, University of Texas, Austin, TX 78712.

lnternational journal of Eating Disorders, Vol. 16, No. 3, 28%294 (1994) 0 1994 by John Wiley & Sons, Inc. CCC 0276-3478/94/03028946

290 Singh

However, like almost all studies on ideal female shape, these studies have explored only body weight. The female body shape is determined by both the amount of fat and its distribution. After puberty, females preferentially deposit more fat peripherally and on lower body parts (hips, buttocks, and thighs), producing a characteristic feminine or gynoid fat distribution. Males lose fat from lower body parts after puberty and start depositing fat on the upper body parts (shoulders, nape of the neck) and centrally (or intra-abdominally), thus producing the characteristically masculine or android fat dis- tribution (Bjorntorp, 1987). Although there is some population-specific variation in fat distribution, human sexual dimorphism in overall and peripheral fat distribution ap- pears to be universal. For example, Kung San of the Kalahai desert in Africa show a sex-specific body fat distribution similar to white Americans (Brown & Konner, 1987).

The fat distribution for both men and women can be quantified by the ratio of the circumference at the waist to the circumference of the hips (WHR). Typically, WHRs of 0.80 or lower are found in women and 0.85 or higher in men (National Academy of Sciences, 1991). Both white and black men and women, in general, have WHRs in their gender-specific range (Mckeigue, Shah & Marmot, 1991).

These findings suggest that both body fat distribution (body shape) and overall weight (body size) may jointly determine the perception of ideal female body shape. Previous studies with white men of various ages (age group 18-85 years old) have shown that female figures depicting a low WHR are judged as more attractive and healthier than female figures of similar body size but a higher WHR (Singh, 1993). The present study was designed to answer the following questions: (1) Would black men judge line draw- ings depicting overweight female figures as more attractive than normal and under- weight figures? (2) When presented with figures that differed in WHR as well as body weight (normal, underweight, and overweight), would black men judge figures with a low WHR as more attractive than figures with similar body weight but a high WHR? (3) Would black men and women respond to body weight and WHR similarly or would there be gender differences in judgment of desirable body shape as reported for whites (e.g., Fallon & Rozin, 1985)?

METHOD

Subjects

Eight-seven black subjects (54 men and 33 women) aged 18-23 years enrolled in a predominantly black undergraduate coIlege volunteered to participate in the study.

Stimulus Material

Stimuli consisted of 12 line drawings of female figures representing four levels of WHR (0.7, 0.8, 0.9, and 1.0) and three levels of body weight (normal, underweight, and overweight). All facial and bodily features were held constant except for WHR, which was changed by varying the line representing the waist (Singh, 1993). For testing, figures were arranged randomly and assigned identifying letters. All figures were re- produced on a single piece of white paper so the subjects could examine all 12 figures simultaneously.

Body.Fat 291

Procedure

Two black research assistants (one man, one woman) approached various professors to seek volunteers for their studies. Students who volunteered were given a typed page explaining the purpose of the study (the cover story), a page containing 12 figures, and a rating questionnaire. Subjects were asked to indicate their age, height, body weight, religious, and ethnic affiliation. Then, under the pretext of judging the relationship between body build and personality characteristics, subjects were asked to rank figures without permitting tied ranks for the following characteristics in order: good health, youthful looking, attractive, sexy, faithful, desire for children, capability for having children, ambitious and career oriented, intelligent, aggressive, interesting to talk to, kind and understanding, good companion, and sense of humor. Then subjects were asked to rank figures for desirability for a long-term relationship, such as marriage, by assuming that the figures represented women in their own ethnic group. Female sub- jects were asked to rank figures for desirability for long-term relationships on the basis of how they thought black men would rank these figures. Target figures were the same as used in previous studies on white subjects (Singh, 1993), except for this study, target figures were described as representing an international composite of female bodies.

Subjects were instructed to make an educated guess if they found it difficult to assign a ranking for any attribute but they had to assign all rankings for each attribute.

RESULTS

Parametric inferential statistics could not be used for data analyses because the rank- ing data are ordinal. Therefore, data were analyzed using a nonmetric multidimensional scaling (MDS) technique. By utilizing a special type of nonmetric MDS, multidimen- sional unfolding, 12 stimulus figures and 15 attributes were scaled into the same dimen- sional solution space. This technique enables one to examine, first, the attributes that subjects perceive to be closely related, and second, the different attributes that subjects associate with a given figure. Separate analyses for male and female subjects were performed using PROC ALSCAL in the Statistical Analysis System (SAS, 1985).

A two-dimensional solution for 12 figures and 15 attributes provided an excellent fit to the data of male [stress = 0.241; R2 (squared multiple correlation indicating the propor- tion of variance of the ranking data accounted for by the MDS solution space) = 0.9441 and female subjects (stress = 0.173; R2 = 0.971) indicating that rankings of both male and female subjects could be explained on the basis of two dimensions. Male subjects located all attributes close to each other in the same solution space except for the attrib- utes of youthfulness, faithfulness, and kindness and understanding. This solution im- plies that male subjects did not perceive youthfulness as closely related to attractiveness, companionship, or desirability for a long-term relationship. Female subjects also did not perceive youthfulness to be closely related to any of the other attributes. Both male and female subjects associated all attributes except youthfulness, faithfulness, and kindness and understanding with normal weight figure N7. Figures N8 and N9 were not grouped with these attributes by male subjects, although female subjects assigned all three nor- mal weight figures (N7, N8, and N9) to the space close to most of the attributes. None of the overweight figures for either male or female subjects were located in the solution space containing attributes of attractiveness, sexiness, companionship, and desirability for long-term relationships.

292 Singh

WHR and Attractiveness

A highly positive correlation for attractiveness rankings indicated that both male and female subjects ranked figures in a practically identical fashion (Spearman r, = .94). To determine whether subjects systematically used the WHR for attractiveness rankings, the percentage rankings of 1 (the most) and 12 (the least) assigned to each figure by male and female subjects were computed. Results clearly show that male subjects systemat- ically utilized the size of the WHR for attractiveness rankings for normal weight and underweight categories. Male subjects assigned low attractiveness rankings to over- weight figures without any consistent use of the size of the WHR. Female subjects, generally, ranked attractiveness of normal weight and overweight figures on the basis of WHR. It should be noted that neither male nor female subjects assigned higher attrac- tiveness ranking to figures with WHRs in the typical male range (0.9 and 1.0) in any body weight category.

To ascertain attributes which subjects associated with attractiveness, Spearman rank correlations were computed from average rankings assigned to 12 figures for various attributes by male and female subjects. Male subjects positively associated attractiveness with youthfulness (Y, = .58, p < .05) and healthiness (Y, = 37, p < .05). Female subjects also positively associated attractiveness with youthfulness (Y, = .71, p < .01) and health- iness (Y, = .98, p < .01) but negatively with faithfulness (r, = - .71, p < .01). Like white subjects, black subjects perceived attractive women as possessing some negative per- sonal qualities (Singh, 1994).

WHR and long-Term Relationships

To explore the question of whether the size of WHR would affect desirability for a long-term relationship, the percentage rankings of 1 and 12 assigned to each figure by male and female subjects were computed (Figure 1).

Male subjects systematically used the WHR for ranking normal weight and over- weight figures when assigning ranks for preference for long-term relationship. Figure N7 was preferred more for a long-term relationship than any other figure. When only body weight categories were considered, male subjects preferred normal figures the most and underweight figures the least. The Spearman correlation for long-term rela- tionship for male and female subjects was positive and significant (r, = .78, p < .01). There is a remarkable consensus between male and female subjects for N7, although female subjects ranked this figure somewhat higher than male subjects.

It appears that preference for a long-term relationship is largely affected by the degree of attractiveness assigned to a figure. Female subjects clearly reacted this way, as is evident from correlation between preference for a long-term relationship and attractive- ness (Y, = .97, p < .01). The correlation between attractiveness and preference for a long-term relationship for male subjects was also positive but lower (Y, = .66, p < .05). Some interesting gender differences were evident for attributes of youthfulness. Male subjects did not associate youthfulness with long-term relationships (Y, = .17, p < .05) whereas female subjects did (Y, = .75, p < .01). Likewise, whereas female subjects negatively associated preference for a long-term relationship and faithfulness (Y, = .74, p < .Ol ) , male subjects did not (Y, = .19, p > .05).

DISCUSS I 0 N In summary, results show that both male and female subjects generally assigned an

attractiveness ranking using the WHR within each body weight category. Figures with

Body Fat

- 2 50- 1

60

293

0 FEMALE RATERS I MALE RATERS -

I 1 1 1 I I I I I l l ,

4 3 a Z

LI L

2 W 0 U W a

OVERWEIGHT t NORMAL t UNDERWEIGHT

60 t

- 10

B 30 1 - E 401

similar body weight were assigned lower ranking if they had a higher WHR. Compar- ison among body weight categories shows that neither male nor female subjects asso- ciated attractiveness, sexiness, companionship, or desirability for a long-term relation- ship with overweight figures, although these figures were perceived to be desirous and capable of having children and to be kind and understanding. These findings are re- markably similar to those reported for white men and women (Singh, 1993), and do not support the belief that college-age blacks find overweight females as sexy and attractive (e.g., Harris et al., 1991). Additionally, black men and women responded to figures similarly, suggesting that both genders use similar criteria for defining a desirable body shape.

It should be pointed out that there are no published studies with American black men and women that have investigated the nature of desirable female body shape using visual stimuli as used by Fallon and Rozin (1985). The present findings suggest that when judging female figures, which differ in body size as well as shape, black and white subjects generally use similar standards to judge attractiveness of female figures. Cross- cultural surveys also indicate that standards of feminine beauty involve both body size and body shape. Brown and Konner (1987) examined 325 cultures coded by the Human Relations Area files and found that 58 cultures exhibited preferences for a certain part of the female body. For example, consistent with the report of Ford and Beach (1951) 81% of cultures were found to prefer a plump or moderately fat body but 19% of cultures showed equal abhorrence for extremely thin as well as obese women. More relevant to present findings, 90% of the societies prefer large hips and legs whereas only 10% consider slender hips and legs attractive. Thus, it is possible that those societies that are purported to admire fat women (e.g., Efik of Nigeria, Havasupai Indians in American Southwest-references in Brown & Konner, 1987) actually prefer those fat women who have a gynoid fat distribution. No data are available to support this inference. The

294 Singh

present findings, however, suggest that future studies on female attractiveness and body image satisfaction should explore both the amount of fat and the nature of its distribution.

REFERENCES

Bjorntorp, P. (1987). Fat cell distribution and metabolism. In R. J. Wurtman & J. J. Wurtman (Eds.), Human

Brown, P. J., & Konner, M. (1987). An anthropological perspective on obesity. In R. J . Wurtman & J. J.

Fallon, A. E., & Rozin, P. (1985). Sex differences in perceptions of desirable body shape. Journal of Abnormal

Ford, C. S . , & Beach, F. A. (1951). Patterns of sexual behavior. New York: Harper. Harris, M. B., Walters, L. C., & Waschull, S. (1991). Gender and ethnic differences in obesity-related behav-

McKeigue, P. M., Shah, B., & Marmot, M. G. (1991). Relation of current obesity and insulin resistance with

National Academy of Sciences. (1991). Diet and health. Washington, DC: Author. Rosen, J. C., &Gross, J. (1987). Prevalence of weight reducing and weight gaining in adolescent girls and boys.

Rucker 111, C. E., & Cash, T. F. (1992). Body images, body-size perceptions, and eating behavior among

SAS Institute, Inc. (1985). SAS user’s guide: Statistics (Version 5) . Cary, NC: Author. Singh, D. (1993). Adaptive significance of female physical attractiveness: Role of waist-to-hip ratio. Journal of

Singh, D. (1994). Is thin really beautiful and good? Relationship between physical attractiveness and waist-

Thomas, V. G. (1989). Body-image satisfaction among black women. Iournal of Social Psychology, 129, 107-112.

obesity (pp. 66-72). New York: New York Academy of Sciences.

Wurtman (Eds.), Human obesity (pp. 3946). New York: New York Academy of Sciences.

Psychology, 94, 102-105.

iors and attitudes in a college sample. Iournal of Applied Social Psychology, 22, 1545-1577.

high diabetes prevalence and cardiovascular risk in South Asians. Lancet, 337, 382-386.

Health Psychology, 6 , 131-147.

African-Americans and white college women. International journal of Eating Disorders, 12, 291-299.

Personality and Social Psychology, 65, 293-307.

to-hip ratio. Personality and Individual Differences, 16, 12S132.