Department of Health Policy, Management and Behavior, University at Albany, SUNY, One University Place, Room 183, Rensselaer, NY 12144. E-mail: email@example.com
Objective:To assess familial links in fat stereotypes and predictors of stereotypes among girls and their parents.
Research Methods and Procedures:Fat stereotypes were assessed using a questionnaire developed for this study. Participants indicated the extent to which they agreed with nine statements about thin people (e.g., thin people are smart) and the same statements about fat people (e.g., fat people are smart). Predictors of fat stereotypes that were examined include weight status (BMI; girls and parents), education (parents), income (parents), self-investment in physical appearance (parents), maladaptive eating attitudes (girls), and parenting practices and peer interactions focused on body shape and weight loss (girls).
Results:Girls and parents exhibited fat stereotypes. Fathers who were more educated and had a higher family income were more likely to endorse fat stereotypes, as were mothers and fathers with a high investment in their physical appearance. Although no associations were found between girls’ and parents’ fat stereotypes, girls were more likely to endorse fat stereotypes when interactions with parents and peers focused on body shape and weight loss. Girls were also more likely to endorse fat stereotypes when they reported higher levels of maladaptive eating attitudes. No associations were found between weight status and fat stereotypes.
Discussion:Although there was no association between girls’ and parents’ fat stereotypes, girls were more likely to express negative attitudes about obesity and obese persons when parenting practices and interactions with peers promoted a lean body type, suggesting that there may be an implicit link between the lean ideal and fat stereotypes.
Idealization of the lean physique combined with epidemic rates of obesity (1, 2) and widespread concern about body weight and dieting (3, 4, 5, 6) have created a social environment that is extremely hostile toward obesity and obese persons. Negative attitudes toward obesity, or fat stereotypes, are consistently reported in the literature (7). Relative to lean individuals, overweight individuals are described as lazy, sloppy, lacking in self-discipline, overindulgent, and having poor personal hygiene (7). These negative attitudes have translated into discrimination toward obese individuals in a number of settings, including the workplace, medical and health care settings, educational settings, and housing (7, 8).
Children also exhibit negative attitudes toward obese persons. Specifically, research shows that children are reluctant to play with overweight peers (9, 10, 11) and are more likely to assign negative adjectives such as lonely, lazy, sad, stupid, ugly, and dirty to an overweight child than to an average weight or lean child (12, 13, 14, 15). Negative attitudes toward obesity and fat stereotypes are present among children as young as 3 (11, 14), and they increase with age (16). Finally, recent research indicates that, although one in three children is overweight or obese (2) and rates of obesity have doubled among children since the 1960s (2, 17), stigmatization of obese children has increased during this time period (18). As a result of negative attitudes toward obesity, overweight children are frequently victimized in the school setting (19, 20, 21, 22) and have been shown to exhibit low self-esteem (23, 24, 25).
The beliefs that thin is good and fat is bad are broadly disseminated in newspapers, magazines, videos, television, and the music industry (26, 27, 28, 29, 30, 31). Although these negative views are pervasive, not all individuals display negative attitudes toward obesity. Therefore, it is pertinent to examine characteristics, such as educational background, weight status, and investment in physical appearance, that may influence an individual's likelihood of endorsing fat stereotypes. Research to date has not extensively examined factors that may increase the likelihood of endorsing fat stereotypes among children or adults. With respect to children, given that fat stereotypes are present as young as age 3 and increase with age, it is important to examine the familial and peer environments as well as individual characteristics that are associated with the likelihood of developing negative attitudes toward obesity. Parents may foster the development of fat stereotypes among their children as a result of their own fat stereotypes and their parenting styles that promote a lean body weight. Likewise, peers may foster the development of fat stereotypes among children as a result of social interactions specific to body shape and weight.
Therefore, using a sample of 9-year-old girls and their mothers and fathers, this study 1) examines the presence of fat stereotypes among girls and their parents; 2) assesses associations between parents’ fat stereotypes and their weight status, years of education, income, and self investment in physical appearance; (3) examines links between girls’ and parents’ fat stereotypes; and (4) assesses individual characteristics and interactions with peers and parents as predictors of girls’ fat stereotypes. In addition, this study examines links between parents’ inclination to provide socially desirable answers and their reported fat stereotypes.
Research Methods and Procedures
Participants included 178 9-year-old girls and their mothers and fathers who were enrolled in a longitudinal observational study of girls’ health and development from 5 to 15 years of age. Families were recruited for participation in the longitudinal study using flyers and newspaper advertisements. In addition, families with age-eligible female children within a five-county radius received letters inviting them to participate in the study. Eligibility criteria for girls’ (and parents’) enrollment in the project included living with their biological parents, the absence of severe food allergies or chronic medical problems affecting food intake, and the absence of dietary restrictions involving animal products. Families were not recruited based on weight status. Only data obtained when girls were 9 years old were used in this study.
Girls and their parents visited the laboratory during the summer. Girls were individually interviewed by trained interviewers, and parents completed a series of self-report questionnaires. Trained research assistants measured girls’, mothers’, and fathers’ height and weight. The study was approved by the Institutional Review Board, and written consent was obtained from mothers and fathers. In addition, mothers provided written consent for their daughter's participation.
Fat stereotypes were the dependent variables of interest and were measured for girls, mothers, and fathers. Predictors of fat stereotypes that were examined differed for girls and parents and included weight status (girls and parents); maternal, paternal, and peer influence on eating and weight-related concerns (girls); maladaptive eating attitudes (girls); self-investment in physical appearance (parents); and social desirability (parents).
The Fat Stereotypes Questionnaire was developed for this study. Nine statements were generated, and each statement was asked for both thin people and fat people, resulting in a total of 18 items. Characteristics and attributes that were assessed included happiness, intelligence, attractiveness, laziness, and friendships, in addition to the general belief that it is good to be fat/thin and it is bad to be fat/thin. Girls and their parents responded to each statement using a four-point scale (really disagree, sort of disagree, sort of agree, really agree).
Whereas the questionnaires were largely identical for girls and parents, a number of small adjustments were made to make the scale developmentally appropriate for girls. One item differed for girls and parents (i.e., “It is hard for fat people to get to know people” for parents; “It is hard for fat people to make friends” for girls). In addition, although the response scale was the same for girls and parents, girls were asked to respond in a two-step process. That is, girls were asked whether they agreed or disagreed with the statement. (They could also simply say yes or no to the statement.) Based on their response, they were asked whether they really or sort of agreed/disagreed. In addition, as previously noted, girls were individually interviewed, and parents completed the scale independently. Finally, all interviewers were asked to indicate on a five-point scale (1 = didn't understand to 5 = clearly understood) the extent to which girls seemed to understand the questionnaire. Ninety-nine percent of girls received a score of 3 or greater, indicating that they did not have difficulty interpreting the scale.
Discrepancy scores for each attribute were calculated (fat people attribute − thin people attribute); scores above zero indicated that the attribute was more strongly endorsed for thin people, and scores below zero indicated that the attribute was more strongly endorsed for fat people. This method was adopted because, although individuals may be reluctant to openly agree with the statement “fat people are lazy,” they may be more likely to agree with the statement for fat people than the same statement for thin people. In addition, a total fat stereotypes score was calculated by taking the mean of the discrepancy scores. Before calculating the mean score, the negative attributes were reverse coded so that all items could be interpreted in the same direction. Consequently, higher mean fat stereotype scores indicated a greater endorsement of negative characteristics for fat people and positive characteristics for thin people (i.e., higher fat stereotypes). Internal consistency coefficients for the combined fat stereotypes scores were α = 0.71 (girls), α = 0.65 (mothers), and α = 0.69 (fathers).
Girls’ and parents’ height (to the nearest 0.5 cm) and weight (to the nearest 0.1 kg) were measured three times by a trained research assistant, and average height and weight were used to calculate their BMI (kilograms divided by height squared). Girls’ BMI values were converted to age- and sex-specific BMI percentiles using the Centers for Disease Control 2000 growth charts, and girls were classified as overweight if their BMI percentile was ≥85 and obese if their BMI percentile was ≥95 (32). Parents were classified as overweight and obese based on BMI cut-offs of 25 and 30, respectively (33).
Perceptions of Parental Concern about Weight
Girls’ perceptions of parental concern about weight were assessed using a questionnaire developed for the larger longitudinal project. The scale was designed to reflect a Guttman scale (34) and included items that increase successively in risk inferred. Girls were asked whether 1) their mothers have ever tried to lose weight; 2) their mothers make sure that they don't eat too much so that they don't get fat; 3) it is important to their mothers that they (i.e., girls) are thin; 4) their mothers encourage them to lose weight; and 5) their mothers criticize them about their weight. The same questions were asked with reference to fathers. Scores (yes = 1, no = 0) were summed across the five items to form a total perception of parental concern about weight score for each parent. The percentage of girls endorsing items 1 to 5 was 65%, 40%, 16%, 11%, and 4% for perceived maternal concern and 41%, 32%, 12%, 10%, and 3% for perceived paternal concern, respectively. As with Guttman scales in general, the internal consistency coefficient is inappropriate in this instance because it is not expected that girls will respond similarly across all items. For example, girls who perceive a moderate level of parental concern may endorse 2 of 5 items and negate the remaining items, resulting in a low level of internal consistency.
Peer Influence on Weight Concerns and Eating Attitudes
The interaction subscale of the Inventory of Peer Influence on Eating Concerns created by Oliver and Thelen (35) was used to assess the frequency with which girls interact with their female friends about weight or body issues (e.g., “girls and I compare the size and shape of our bodies”). The scale includes five items and adopts a five-point response scale (1 = never to 5 = a lot). Using a sample of 164 third and fifth graders, Oliver and Thelen (35) reported an internal consistency coefficient of α = 0.80 and found that girls who interacted more with their peers about body issues had a higher drive for thinness (r = 0.42, p < 0.01). The internal consistency coefficient for the current sample was α = 0.79.
Maladaptive Eating Attitudes
The Children's Attitude Test (chEAT)1 (36) was used to examine girls’ maladaptive eating attitudes. The chEAT is a 26-item questionnaire designed to measure dieting, weight concerns, and food preoccupation using a nine-point Likert response scale, with response options ranging from never to always. Higher scores indicate more negative, or maladaptive, attitudes. Two items pertaining to purging were not included in the version administered in this study. An overall score ranging from 0 to 72 was created using the criteria outlined by Maloney et al. (36). The reliability and validity of the chEAT is well established (36, 37). The internal consistency coefficient for the current study was α = 0.74.
Self-Investment in Physical Appearance
The Appearance Schemas Inventory (ASI) was used to examine mothers’ and fathers’ beliefs about the importance of physical appearance (38). The self-investment (five items) and appearance stereotyping (three items) subscales were used in this study. Participants responded to statements using a five-point scale, and scores were averaged for each subscale. Higher scores for the self-investment scale denote a stronger belief that appearance is influential and central to the participant's self-concept. Higher scores for total appearance stereotyping indicate a greater assumption that social goodness/badness is linked to attractive/unattractive appearance. Previous research using a sample of 274 college women supports the internal consistency (α = 0.84) and concurrent validity of the ASI; higher scores on the ASI were associated with less favorable and less satisfying evaluations of appearance, more frequent body image dysphoria, and a stronger cognitive-behavioral investment in appearance and in one's physical ideals. The internal consistency coefficients for the current study were α = 0.75 and α = 0.73 for mothers’ and fathers’ self-investment, respectively, and α = 0.71 for both mothers’ and fathers’ appearance stereotyping.
The Social Desirability Scale (39) was used to examine the inclination of mothers and fathers to respond to questions in a socially desirable manner (i.e., responses that present the individual in a socially positive light). The 33-item true—false scale includes 18 positive behaviors that most people do not exhibit (e.g., “never hesitate to go out of my way to help someone in trouble”) and 15 negative behaviors that most people do exhibit (e.g., “I have never intensely disliked anyone”). The negative items were reverse coded, and all items were summed to give a total social desirability score, with higher scores indicating a stronger bias. Previous research supports the reliability and validity of the Social Desirability Scale (40, 41). In this study, the internal consistency coefficient was α = 0.79 for mothers and α = 0.81 for fathers.
A two-tailed t test was performed on the discrepancy score for each attribute and for the total fat stereotypes score for girls, mothers, and fathers to determine whether participants exhibited fat stereotypes (i.e., whether the scores were significantly different from zero). A nonsignificant p value indicates that the attribute was equally likely to be attributed to fat and thin people. A significant p value indicates that the attribute was more likely to be attributed to thin (positive mean discrepancy score) or fat (negative mean discrepancy score) people. Similarly, a significant p value for the total fat stereotypes score indicates that, in general, participants were significantly more likely to assign positive attributes to thin people and negative attributes to fat people if the sample mean was greater than zero, or vice versa if the sample mean was less than zero (see Figure 1).
The next series of analyses examined predictors of fat stereotypes for girls, mothers, and fathers. To reduce the number of analyses performed and, therefore, reduce the likelihood of a type I error, only the total fat stereotypes scores were assessed. Spearman rank correlation analysis was used to examine 1) individual and socioeconomic characteristics associated with parents’ fat stereotypes, including parents’ BMI, years of education, tendency to report socially desirable answers, and investment in physical appearance (see Table 1); 2) associations between girls’ and parents’ fat stereotypes, controlling for parents’ social desirability bias (see Table 1); and 3) associations between girls’ fat stereotypes and (a) individual characteristics including their weight status and maladaptive eating patterns and (b) sources of socialization including peer interactions surrounding body shape and weight, perceived parental concern about weight, and parental investment in physical appearance (see Table 2). Finally, significant (p < 0.05) and marginally significant (p < 0.10) correlates of girls’ fat stereotypes were entered into a stepwise regression analysis, using forward selection, to determine the relative importance of each predictor. To examine whether parental and peer influence on girls’ fat stereotypes differed for girls who were overweight and nonoverweight (i.e., BMI was a moderator), the significant correlates of these constructs and their interactions with girls’ BMI were entered into a multiple regression analysis predicting girls’ fat stereotypes.
Table 1. . Correlates of mothers’ and fathers’ fat stereotypes and links between girls’ and parents’ fat stereotypes
Families were non-Hispanic white; mothers and fathers were generally well-educated, completing a mean of 14.6 ± 2.2 and 14.7 ± 2.6 years of education, respectively; and the median combined family income was >$50, 000. In terms of weight status, approximately one in three girls was classified as overweight, and 14% were obese. In addition, 60% of mothers and 82% of fathers were classified as overweight, and 28% of mothers and 31% of fathers were classified as obese.
Results from the discrepancy score analysis indicated that the girls, mothers, and fathers adopted fat stereotypes (see Figure 1). Specifically, the following characteristics were significantly more likely to be attributed to thin people than fat people: having lots of friends (girls), happy (girls, mothers, fathers), smart (girls), good looking (girls, mothers, fathers), and the general statement that it is good to be thin (girls, mothers, fathers). Similarly, the following characteristics were significantly more likely to be attributed to fat people than thin people: difficulty making friends or getting to know people (girls, mothers, fathers), lazy (girls, mothers, fathers), having few friends (girls, mothers), and the general statement that it is bad to be fat (girls, mothers, fathers). Finally, the total fat stereotype scores were significantly greater than zero for girls (0.54 ± 0.04, p < 0.05), mothers (0.43 ± 0.03, p < 0.05), and fathers (0.40 ± 0.03, p < 0.05), indicating a general tendency to attribute positive characteristics to thin people and negative characteristics to overweight people.
As shown in Table 1, no links were identified between mothers’ and fathers’ BMI and their fat stereotypes. That is, parents who were more overweight were as likely to endorse fat stereotypes as leaner parents. Among fathers, higher levels of education and higher family income were associated with a higher likelihood of endorsing fat stereotypes. For mothers and fathers, higher self-investment in physical appearance and higher appearance stereotyping (i.e., the outer appearance reflects inner qualities of the individual) were associated with higher fat stereotypes, and a greater tendency to respond in a socially desirable manner was associated with a lower likelihood of endorsing fat stereotypes. No associations were identified between girls’ and mothers’, girls’ and fathers’, or mothers’ and fathers’ (r = 0.05, p = not significant) fat stereotypes. Results for these analyses did not change after controlling for mothers’ and fathers’ tendency to respond in a socially desirable manner.
Although no links were identified between girls’ and parents’ fat stereotypes, girls’ eating attitudes and interactions with parents and peers were identified as significant correlates of their fat stereotypes. As shown in the first column of Table 2, girls who reported more maladaptive eating attitudes were more likely to endorse fat stereotypes. In addition, girls who reported a higher level of peer interactions centered on body weight and shape and girls who perceived that their mothers placed a high emphasis on their being thin (e.g., encouraging them to lose weight, restricting their access to food to promote weight loss, and/or criticizing them about their weight) reported higher fat stereotypes. No association, however, was found between parents’ investment in their physical appearance and girls’ fat stereotypes. Furthermore, similar to parents, no links were identified between girls’ BMI and their fat stereotypes (r = −0.04, p = not significant).
Correlations among the various predictors of girls’ fat stereotypes are also presented in Table 2. The two sources of social influence that were significantly associated with girls’ fat stereotypes (i.e., peer interactions and perceived maternal concern about weight) were themselves correlated, and each was significantly correlated with girls’ BMI. That is, girls who were more overweight were more likely to report interacting with peers about body weight and shape and were more likely to perceive that their mothers were concerned about their weight. However, the previously identified associations between girls’ fat stereotypes and peer and maternal influence cannot be an artifact of girls’ weight status, because there was no association between girls’ BMI and their fat stereotypes.
When the correlates of girls’ stereotypes were collectively examined in a stepwise regression analysis, results showed that perceived maternal influence on weight concerns (β = 0.24, p < 0.01) and girls’ maladaptive eating attitudes (β = 0.21, p < 0.01) significantly and independently predicted girls’ fat stereotypes (R2 = 0.12, p < 0.0001). The final analyses examined whether girls’ BMI moderated relationships between the predictor variables and girls’ fat stereotypes. Results indicated that the associations between girls’ fat stereotypes and girls’ eating attitudes and parent and peer concern about weight did not differ for overweight and nonoverweight girls.
Although research consistently illustrates that children and adults endorse fat stereotypes, little is known about correlates or predictors of such stereotypes. This study examined the presence of fat stereotypes among 9-year-old girls and their parents and individual and psychosocial predictors of fat stereotypes. Results showed that girls and parents endorsed fat stereotypes and that stereotypical attitudes did not differ as a function of girls’ and parents’ weight status. With respect to predictors of stereotypical attitudes, fathers who were more educated and had a higher family income, mothers and fathers who were more invested in their appearance, and girls with more maladaptive eating attitudes (i.e., reporting a strong desire to be thin and restrictive dieting practices) reported higher fat stereotypes. Whereas no links were identified between girls’ and parents’ fat stereotypes, girls whose mothers, fathers, and friends encouraged them to lose weight and be lean were more likely to endorse fat stereotypes.
Results from this study support a large body of research indicating that children and adults adopt fat stereotypes (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18) and that overweight and nonoverweight individuals are equally likely to report stereotypical attitudes (11, 16, 42, 43). The omnipresent nature of fat stereotypes is concerning because it is likely to be the driving factor behind discrimination (7, 8) and victimization (19, 20, 21, 22) experienced by obese persons. It is surprising that overweight individuals are as likely to endorse fat stereotypes as lean individuals. Endorsing fat stereotypes is likely to place overweight or obese individuals at increased risk of low self-esteem, depression, and unhealthy dieting practices. Indeed research shows that, relative to their nonoverweight counterparts, obese children and adults report low self-esteem and perceived self-competences (25, 44, 45), high levels of depression (44, 46, 47), and extreme dieting practices such as skipping meals, purging and binging, and using laxatives (48, 49, 50). In addition to reducing quality of life, each of these factors may increase the likelihood of further weight gain (48).
The most noteworthy finding of this study is that characteristics that have been previously linked with an increased likelihood of endorsing the thin ideal were associated with higher fat stereotypes among girls and parents in this study. Research shows that non-Hispanic whites and more affluent and educated groups are more likely to support the thin ideal and exhibit heightened concern about weight (51, 52, 53, 54). In this study, higher income and education (fathers), higher investment in physical appearance (mothers and fathers), and maladaptive eating attitudes and concern about weight (girls) were linked with higher fat stereotypes. Similarly, girls were more likely to endorse fat stereotypes when their mothers and fathers encouraged them to lose weight and when interactions with peers centered around weight and body shape, suggesting that girls may learn to adopt negative attitudes about obesity and obese people as a result of interactions with parents and peers that promote being lean. The lack of an association between parents’ and girls’ fat stereotypes may reflect the possibility that parents are mindful of not expressing stereotypical attitudes in the presence of their children. Overall, these results suggest that there may be an intrinsic link between the promotion of the lean body type and the endorsement of fat stereotypes. Given the current epidemic of obesity, encouraging people to be lean (within reason) is generally appropriate. In doing so, however, care needs to be given not to portray individuals who are overweight negatively. To counteract this possibility, parents may need to make clear to children that body size impacts on an individual's health and not their personal qualities. In addition, parents should avoid using the general label of “fat as bad,” because children may interpret this as “fat people are bad.”
There are a number of strengths in the methodology of this study. First, a forced choice format was not adopted to assess fat stereotypes. An example of a forced choice format is requiring participants to assign a specific attribute such as laziness to either an overweight individual or a nonoverweight individual. The majority of research on anti-fat attitudes has adopted a forced choice format. However, this methodology has been criticized (55, 56) because it is likely to overestimate the presence of negative attitudes toward obese individuals. As stated by Jarvie et al. (56), forced choice scenarios guarantee that a cooperative participant will stereotype. Second, results in this study cannot be attributed to a social desirability bias. People are becoming increasingly aware of the inappropriateness of fat stereotypes. Consequently, they are less likely to be truthful on questionnaires assessing stereotyped attitudes. The questionnaire used in this study was designed to minimize social desirability effects. In addition, social desirability bias was measured and controlled for in the analyses.
In terms of limitations of this study, the directionality of effects cannot be determined because of the cross-sectional nature of the data. However, it is unlikely that fat stereotypes lead to higher levels of education, higher income, and weight-related interactions with parents. The fat stereotypes questionnaire used in this study examined only a limited number of attributes used to stereotype overweight individuals. Future research should include measures of fat stereotypes that assess a broader range of stereotypical attitudes. Finally, the results of this study can be generalized to only white, relatively well-educated families. It is likely that the results may differ for ethnic and economic groups that are less likely to value the thin physique.
In summary, this study supports previous research by identifying fat stereotypes in girls and their parents and extends research to date by examining predictors of fat stereotypes. Results suggest that promoting a lean body type may be implicated in the emergence of negative attitudes toward obesity and obese persons. Future research can build on the results of this study by using more diverse samples, particularly ethnic minority samples, and by examining in more detail the links between promoting a lean ideal and fat stereotypes. These are important topics to pursue because fat stereotypes have negative implications on children's and adults’ mental health, and discriminatory practices emanating from fat stereotypes have broad-reaching negative implications on the accessibility of health care, educational opportunities, employment opportunities, and the social treatment of overweight individuals.
This research was supported by National Institutes of Health Grant HD-32973.