Objective: To examine the concept of the “good enough” body size acceptability across a wide range of ages and weight status.
Research Methods and Procedures: Subjects were 303 children, 427 adolescents, 261 young adults, and 326 middle-age adults who selected acceptable body sizes from an array of drawings representing their own age and gender. They also selected body sizes representing their own actual and ideal size.
Results: A large majority (87%) of subjects considered their own body size socially acceptable. This finding applied to both genders in all age groups and to underweight, normal weight, and overweight subjects. Even among obese subjects, 48% considered their own body size socially acceptable. For the large percentage of subjects who reported a discrepancy between their actual and ideal body sizes, most considered their own body size acceptable. This finding also applied to both genders in all age groups and to underweight, normal weight, and overweight subjects.
Discussion: Most male and female subjects across a wide range of ages and status considered their own body size to be within the range of socially acceptable body sizes even though, for many, it did not match their ideal. The implications of expanding body size research to include the conceptual framework of body size acceptability is discussed in terms of contributing to a paradigm of positive psychology.
Studies in the literature on body size frequently assume that the difference between a person's ideal and actual size provides a valid measure of body size dissatisfaction (1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12). This assumption merges two separate concepts, those of body size difference and of body size dissatisfaction. It masks the possibility that one can find one's own body size to be acceptable despite the fact that it falls short of the ideal. In other areas of life, such as wealth, intelligence, achievement, and athletic ability, the acceptability of one's current status can coexist with the knowledge that it is not ideal.
Using the difference between one's actual and ideal body size as a measure of dissatisfaction highlights displeasure with one's own body and fuels cultural pressures toward negative self-evaluation (e.g., (2, 10, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24)) while neglecting the context of social acceptability. To avoid this bias toward negativity that occurs when these concepts are merged, the present study contextualizes body size evaluations in the realm of what might be acceptable. It examines the hypothesis that, across a wide range of ages and weight status, people consider their own body size to be “good enough,” within the range of socially acceptable body sizes even though it may not match their ideal.
Research Methods and Procedures
Subjects were selected from four age groups: 303 children (185 girls, 118 boys), 427 adolescents students (241 girls, 186 boys), 261 young adults (173 women, 88 men), and 326 middle-age adults (208 women, 118 men). Their demographic characteristics are presented in Table 1.
Table 1. Demographic characteristics of the total subject pool and subject age groups
Total pool (n = 1317)
Children (n = 303)
Adolescents (n = 427)
Young adults (n = 261)
Middle-age adults (n = 326)
Socioeconomic status (SES) was estimated by use of the Hollingshead–Redlich Index of class (61). Class I represents high SES, II and III middle SES, and IV and V lower SES. 39% (119) ofelementary students did not provide SES information; the actual percentage of SES I elementary students is probably much lower.
All subjects were volunteers. The children were in the fourth and fifth grades. The adolescents were high school students. The students attended five elementary and three high schools, and participated after written parental consent had been obtained. The young adults were students at a large university and a majority (200) were enrolled in an introductory psychology course. The remaining young adults were approached at their place of work or recreation. Criteria for the middle-age adults were that they be older than age 26 and not enrolled in a college or university. The majority were approached at their place of work or worship; some were asked to solicit participation from acquaintances (using the snowball method). The study was approved by the University Institutional Review Board.
Line drawings of male and female figures were used in the rating of body size (Figure 1). The figures portrayed four specified age ranges: children (ages 6 to 10 years), young adults (ages 16 to 25), middle-age adults (ages 35 to 45), and older adults (ages 55–65). Each age group array depicted sets of both male and female figures. Each set displayed body sizes numbered from one to nine, ranging in size from very thin to very fat. Arrays for each age range were presented on a different page. The age range was specified at the top of each page, e.g., children (ages 6 to 10), with male and female figures in the same age range appearing on the same page. The line drawings of children, young adult, and middle-age adults were modifications of those used by Collins (2). The line drawings of older adults were modifications of those used by Sorensen et al. (25). The modifications are described in Rand and Wright (26).
Subjects were shown the array of nine body sizes representing their own age and gender and asked to select 1) the number of the body size that most resembled how they currently looked and 2) that they would most like to resemble. They were then given male and female arrays of all four age ranges and told to circle the numbers of all body sizes that they considered “socially acceptable” in each array. The specific instructions were, “When you see people at school, at the mall, or on the beach, what body sizes do you think look okay? Circle the number under all body sizes you think are acceptable in addition to the size you like best.”
In this exploratory study, “socially acceptable” was considered a face valid measure, operationalized as body sizes that looked “okay” to the subject. The factor of setting was defined as broadly as possible by asking subjects to imagine figures in a variety of settings. (Note: Selection of these choices was between line drawings and not between “actual” and “ideal” body sizes.)
Line drawings are a standard instrument used to assess body sizes. Validity and reliability of outline drawings for assessing current, desired, and ideal body sizes for adults and children are considered good (2, 26, 27, 28, 29, 30, 31). For this study, two reliabilities were assessed, based on a 1- to 3-week test-retest of 187 subjects (59 children, 71 adolescents, and 57 middle-age adults) and using ratings in arrays matching the subject's own age group. One measure of reliability evaluated subjects’ ratings of all sizes that they rated as being socially acceptable. This reliability, measured as the mean test-retest agreement, was 93 ± 7% for children, 92 ± 6% for adolescents, and 92 ± 8% for middle-age adults. The second measure of reliability assessed the consistency of subjects’ inclusion of their own body size as one of those evaluated as socially acceptable. This reliability, measured as the test-retest agreement, was 75% for children, 79% for adolescents, and 73% for adults.
Questionnaire and Instructions
Instructions, rating forms, and the demographic data questionnaire were read aloud to children and adolescents, who also completed practice arrays to check for task comprehension. Young adults and middle-age adults were given written instructions concerning the questionnaire and rating materials.
Validity of Height and Weight Data
Self-report of height and weight are considered to be of adequate validity for epidemiological studies, although there is a consistent bias toward reporting values in a socially desirable direction (32, 33). The self-reports of height and weight of adolescents, young adults, and middle-age adults were considered acceptable for purposes of the present study. Height and weight data from children contained too many improbable and/or missing values and were excluded from analyses.
Actual size was the body size selected by the subject that most resembled how he or she currently looked. Ideal size was the body size selected by the subject that he or she would most like to resemble. Body size acceptability occurred when a subject's actual size matched a body size included in his or her ratings of acceptable body sizes in the array representing his or her age and gender. Body size discrepancy occurred when a subject's actual size differed from his/her ideal size (4). Three categories were created to measure the magnitude of the discrepancy, as follows: 0 (actual/ideal size difference = 0), 1 (actual/ideal size difference = 1), and 2 (actual/ideal size difference ≥ 2). (Note: There were insufficient numbers of subjects with larger discrepancies for additional categories).
Data were analyzed using SAS 6.04 (34) and are reported as mean ± S.E. The Bonferroni correction was used to adjust significance levels within each group of tests.
A large majority (87%) of subjects in the total subject pool considered their actual body size to be within the range of socially acceptable body sizes (Table 2.) This finding held true for both females and males. This finding also was true for subjects when each of the four age groups was examined separately and for males and females in each age group. There were significant age group differences (p < 0.042), with a smaller percentage of middle-age adults than children, adolescents, or young adults considering their actual body size socially acceptable. Gender differences were not significant.
Table 2. Body size acceptability, by total subject pool, age group, and gender
Body size acceptability × age group: χ2 = 8.211, 3 df, p < 0.042. The only significant paired age group analysis was that between young adults and middle-age adults (χ2 = 7.237, 1 df, p < 0.007; Bonferonni correction, p = 0.05/6 = 0.008).
Body size acceptability × gender, χ2 = 6.162, 1 df, p < 0.013. Differences within individual age groups were not significant.
Some subjects rated their actual body size as being the same as their ideal (discrepancy = 0), whereas others rated their actual body size as being different (discrepancy = 1 or discrepancy ≥ 2). These three subsets were then compared. Subjects with a discrepancy of 0 or 1 were much more likely than subjects with a discrepancy ≥ 2 to consider their body size socially acceptable (p < 0.0001) (Table 3). Similar results were found when data were analyzed separately for each age group.
Table 3. Body size acceptability and magnitude of body size discrepancy, by total subject pool and age group
Body size acceptability × magnitude of discrepancy, χ2 = 330.836, 2 df, p < 0.0001. Significant body size acceptability × magnitude of discrepancy relationships were obtained within each individual age group analysis (p values < 0.0001; Bonferroni correction, p = 0.05/4 = 0.01).
A few subjects in each age group selected the same figure to represent their actual and desired body size (discrepancy = 0) but did not consider their actual body size within the range of socially acceptable body sizes. This unusual assessment may have reflected the evaluation of subjects who were satisfied with their own size but simultaneously considered it too thin or heavy to be socially acceptable. Anonymous questionnaires prevented exploration of this apparent paradox.
Further analyses were conducted to determine if weight status influenced ratings of body size acceptability. Adolescent, young adult, and middle-age adult subjects were grouped into one of four weight categories. (Recall that children were excluded from analyses involving weight.) Weight categories were based on recent guidelines for body mass index (BMI, height in meters divided by weight in squared kilograms) (35): underweight (BMI < 18.5), normal weight (18.5 ≤ BMI ≤ 24.9), overweight (25.0 ≤ BMI ≤ 29.9), and obese (BMI ≥ 30.0).
A large majority of three weight groups (underweight, normal weight, and overweight) considered their actual body size to be within the range of socially acceptable body sizes. Even in the obese group, almost half (48%) of the subjects considered their actual body size to be within this range (Table 4). More moderate but similar findings were obtained when each of the three age groups was examined separately (Table 5). Significant differences were found among weight categories, both for the total subject pool (p < 0.001) and for each age group (p values < 0.002). More subjects in the normal weight category judged their actual size to be socially acceptable than other weight categories. The obese category had the fewest subjects judging their size to be acceptable. The only significant gender difference occurred in the overweight category (p < 0.02); more men (n = 99, 92%) than women (n = 96, 79%) rated their actual size within the range of socially acceptable sizes (χ2 = 6.446, 1 df, p = 0.011).
Table 4. Body size acceptability by weight category for the total subject pool*,†
Body size acceptability
Children were excluded from analyses of weight category.
Weight category × body size acceptability; χ2 = 118.689, 3 df, p < 0.0001. All comparisons between pairs of weight categories were significant (p values < 0.003; Bonferroni correction, p = 0.05/6 = 0.008) except for underweight vs. overweight (not significant).
Table 5. Body size acceptability by age group and weight category*
Children were excluded from analyses of weight categories.
Significant body size acceptability × weight category relationships were obtained within each individual age group analysis (p values < 0.002; Bonferroni correction, p = 0.05/3 = 0.017).
The three magnitude of discrepancy subsets were compared by weight category (Table 6). For discrepancy = 0 and discrepancy = 1, a majority of subjects in all weight categories considered their body size to be socially acceptable. For discrepancy ≥ 2, approximately half of the normal and overweight and a third of the underweight and obese considered their body size socially acceptable. The relationship between the magnitude of discrepancy and body size acceptability was significant for each weight category (p values < 0.005). Small cell numbers prevented separate age group or gender analyses.
Table 6. Body size acceptability and body size discrepancy, by weight category, for the total subject pool*
Children were excluded from analyses of weight category.
Significant body size acceptability × magnitude of discrepancy relationships were obtained within each individual weight category analysis (p values < 0.005; Bonferroni correction, p = 0.05/4 = 0.012).
One might speculate that individuals who endorse a greater number of body sizes as being socially acceptable (i.e., those with a wider “latitude of acceptance”) might be more likely to evaluate their actual size as being acceptable. To address this question, a point bi-serial correlation was calculated. It was found that the latitude of acceptance was related to subjects’ body size acceptability (point bi-serial correlation = 0.22, p < 0.001). However, the relationship between latitude of acceptance and subjects’ body size acceptability did not account for the age group and weight category differences that were found in this study.
Children's latitude of acceptance was significantly less than that of all other age groups (26), yet the percentage rating their body size as socially acceptable was not significantly different from other age groups. In fact, the only significant difference in body size acceptability occurred between young adults and middle-age adults, groups that did not differ in their latitude of acceptance. The latitude of acceptance also differed by weight category (F[3,1004] = 3.20, p < 0.03). Underweight subjects’ latitude of acceptance was significantly less than that of all other weight categories. However, obese subjects were the least likely to rate their body size socially acceptable.
The results of this study confirmed the hypothesis that, across a wide range of ages and weights, most people consider their own body size within the range of acceptable body sizes, even though it may not match their ideal. This was true for the large majority (81% to 95%) of all subjects, males and females, in all age groups (children, adolescents, young adults, and middle-age adults). This was found for the large majority (79% to 94%) of underweight, normal weight, and overweight subjects within each age group. Even among obese subjects, almost half (48%) considered their body size socially acceptable. When examining subjects whose own size was discrepant from their ideal, a majority of all subjects with a discrepancy = l (91%) considered their own size to be socially acceptable, as did almost half (47%) of those with a discrepancy ≥ 2. These results clarify the seemingly contradictory findings that people can be both dissatisfied with their own body size and relatively satisfied with their own appearance (40, 41). The findings are only contradictory if the complexity of attitudes toward body size in different contexts is ignored.
The obese weight group stands out as having far fewer subjects who judged their actual size within the range of socially acceptable body sizes compared to subjects in other weight groups. These obese subjects undoubtedly were aware of and may have shared the widespread prejudice against obesity (29, 36, 37, 38, 39). It is remarkable that so many considered their actual size within the range of socially acceptable sizes despite the prejudice against obesity. This result may be explained in part by the age of the obese subjects in the present study. Most (75%) were middle-age. It is this age group that has been found to be most accepting of variations in body size (26).
Socially acceptable was defined in this study as a body size that “looks okay” at school, at the mall, or on the beach. Evaluations were made on arrays of body sizes that were portrayed in apparel intended to maximize critical appraisal (swimsuits). It is possible that the results might change if subjects were asked to evaluate body sizes in specific situational contexts. Further refinement is needed to clarify whether the latitude of acceptability is the same when applied to self as well as to others and if and how settings and clothing influence this evaluation. In future studies it would be advisable to obtain the actual height and weight of subjects rather than the self-reports used in the present study. In addition, it would be desirable to include larger samples of underweight and obese subjects, as well as subjects representing a greater racial and ethnic diversity.
When the study of body size dissatisfaction is defined by actual/ideal discrepancy, the self is compared to an ideal. This is a stringent comparison that automatically limits the social context: There is only one ideal. This comparison maximizes the likelihood of finding a difference between self and ideal. It is conducive to the erroneous conclusion that this body size difference is the same as body size dissatisfaction. In contrast, the concept of body size acceptability compares the self both to the ideal and to a range of deviations from the ideal that might be acceptable. This comparison assesses how an individual sees him- or herself within a broader social context, shifting the focus from the ideal to the acceptable or “good enough” body size. It maximizes the likelihood of inclusion of one's own body size as a normal, acceptable variation.
Why has the literature emphasized body size dissatisfaction, particularly among women, rather than body size acceptability? Undoubtedly, serious concerns about eating disorders, involving pervasive body size dissatisfaction is one factor (9). Although the prevalence of clinically severe eating disorders is low, dieting and moderate preoccupations with food and eating are common (42, 43). Research has also taught us that people tend to give more emphasis to negative than positive attributes and feelings (44), thus pushing dissatisfaction aspects to the forefront.
The focus on body size dissatisfaction also may have been encouraged through the publication of research results that emphasize group differences. Significant group differences have been documented in regard to gender, age, and race or ethnicity (1, 2, 4, 20, 24, 45, 46, 47, 48, 49, 50, 51). By highlighting these differences, findings of similarities that may also reflect reality have been ignored (52). Moreover, once a research area has been established, further research becomes legitimized (53).
Our culture has distorted what people really look like in favor of idealized images, especially for women (9, 54, 55, 56, 57, 58, 59). Studies investigating body size dissatisfaction may have the inadvertent consequence of perpetuating a restricted view of what is acceptable. As researchers, we have the opportunity to expand the literature and broaden the discussion to include the study of body size acceptability. As stressed by Worell and Etaugh (60), “Reality is created, in part, by the scientific process.” By reorienting interest toward body size acceptability, research will contribute to the renewed paradigm of a positive psychology.
We thank P. K. Young Laboratory School and the elementary and high schools in Alachua County, Florida, that took part in this study for their participation.