Predictors of body image dissatisfaction in elementary-age school girls

Abstract

Concerns about weight, shape, body image, dieting, and eating are evident among elementary-age school girls [Int. J. Eating Disord. 10 (1991) 199; J. Adolesc. Health 12 (1991) 307; Pediatrics 84 (1989) 482; J. Clin. Psychol. 21 (1992) 41; Addict. Behav. (2000).]. Body image dissatisfaction, found to predict the formation of eating disturbances in longitudinal studies [Eating Disord.: J. Treat. Prev. 2 (1994) 114; Int. J. Eating Disord. 18 (1995) 221.], is likely a necessary precursor of eating disturbance development [J. Abnorm. Psychol. 102 (1993) 438.]. Turning back the etiological clock, this study examines the association between body image dissatisfaction and four potential areas of influence, including peer relationships, family characteristics, personality features, and body mass index, among girls in grades three, four, and five. Each conceptual area significantly contributed to the prediction of body image dissatisfaction. The combination of statistically significant variables, including teasing, peer modeling of weight and shape concerns, daughters' perceptions of parental influence to lose or control weight, appearance comparison, sensitivity, and body mass index, accounted for 50% of the variance in general body image dissatisfaction and 45% in weight and shape dissatisfaction. Implications for prevention, treatment, and future research are discussed.

Introduction

Many children exhibit dissatisfaction with their weight, endorse the use of weight control measures, and exhibit symptoms of disordered eating Collins, 1991, Flannery-Schroeder & Chrisler, 1996, Hill & Bhatti, 1995, Hill et al., 1994, Koff & Rierdan, 1991, Maloney et al., 1989, Mellin et al., 1992, Thelen et al., 1992, Tiggemann & Pennington, 1990, Vander Wal & Thelen, 2000. Of children in grades three through six, 6.9% scored in the anorexic range on a children's inventory of eating concerns (Maloney et al., 1989), and 13.3% of children in grades five through eight met criteria for an eating disorder not otherwise specified according to prescreening and interview procedure (Childress, Brewerton, Hodges, & Jarrell, 1993). Clearly, if eating disturbances are evident among such young girls, the motivation to diet must be in place beforehand (Hill, 1993).

An important precursor of eating disturbance development is body image dissatisfaction (Leon, Fulkerson, Perry, & Cudeck, 1993). Although cross-sectional studies of children have noted associations between body image dissatisfaction and eating disorder measures Flannery-Schroeder & Chrisler, 1996, Lawrence & Thelen, 1995, longitudinal studies suggest that body image disturbance serves as an etiological risk factor for eating disorder development Attie & Brooks-Gunn, 1989, Brooks-Gunn et al., 1989, Cattarin & Thompson, 1994, Killen et al., 1996, Killen et al., 1994, Thompson et al., 1995. Both Killen et al. (1994) and Killen et al. (1996) found that girls with the greatest weight and shape concerns at baseline were more likely to develop eating disorder symptomatology over a 4-year period than girls with the fewest concerns. Together, these studies show that body image dissatisfaction precedes the development of restrictive eating practices, which then predate the development of bulimic symptoms.

Unfortunately, not much is known about variables that are associated with body image dissatisfaction (Casper & Offer, 1990) or its development (Hill, Oliver, & Rogers, 1992). The following sections highlight variables that may be associated with body image dissatisfaction in young girls.

Peer interactions, such as discussions of weight and diet or imitation of weight and diet concerns, may be associated with girls' body image dissatisfaction (Levine, Smolak, Moodey, Shuman, & Hessen, 1994). In one study, the importance that one's peers attributed to weight and eating accounted for nearly one-third of the variance in fourth- and fifth-grade girls' own weight concerns (Taylor et al., 1998). Similarly, Oliver and Thelen (1996) found that self-reported conversations about the size, shape, and appearance of one's body with one's peers was a significant predictor of a drive for thinness among girls in grades three and five. Hence, peer involvement in dieting and weight loss may predict body image dissatisfaction.

Although teasing can be benign, it typically constitutes a very negative experience for children, and appearance is the most likely target (Shapiro, Baumeister, & Kessler, 1991). Longitudinal studies Cattarin & Thompson, 1994, Thompson et al., 1995 have shown that both teasing frequency and impact are associated with the development of body image dissatisfaction among middle-school girls. Among fourth- and fifth-grade girls, Taylor et al. (1998) found that the impact, but not the frequency, of teasing about weight was significantly associated with poor self-image and weight concerns. The belief that other children would like one better if one was thinner and self-reported teasing have been shown to be important indicators of eating and body image concerns among both girls and boys in grades three through six Maloney et al., 1989, Oliver & Thelen, 1996. Hence, peer teasing is likely associated with body image dissatisfaction in children as well as adolescents. Moreover, because heavier children may be more likely to receive appearance-related teasing Cattarin & Thompson, 1994, Thompson et al., 1995, teasing may serve as a mediator or moderator between body mass index and body image dissatisfaction.

Familial pressures to control weight may also be associated with body image dissatisfaction. Although cross-sectional studies of middle-school girls have shown associations between body image dissatisfaction and self-reported family teasing (Levine, Smolak, & Hayden, 1994), findings among children are mixed. One study of mothers and their 9- and 10-year-old daughters showed that maternal disapproval of the daughter's body build and eating habits accounted for only 2% of the variance in body image dissatisfaction (Brown, Schreiber, McMahon, Crawford, & Ghee, 1995). However, in a study of parents and their fourth-grade sons and daughters, Thelen and Cormier (1995) found that daughter's body image dissatisfaction was significantly associated with parent-reported encouragement to control weight, but not after body mass index was controlled statistically. They suggest that parents may encourage overweight daughters to control their weight. Indeed, adolescent girls, who report having received encouragement from their parents to diet, tend to have had higher body mass indices than girls who do not receive such encouragement (Paxton et al., 1991). Although there may be a direct association between parental influence and body image dissatisfaction, it is also possible that parental influence will moderate or mediate the relationship between body mass index and body image dissatisfaction.

Parents may model excessive concerns about weight, shape, and diet (Striegel-Moore, Silberstein, & Rodin, 1986), which girls may observe and imitate. Children tend to acquire and retain parental food preferences, eating practices (Hill, Weaver, & Blundell, 1990), and body image ideals (Flynn & Fitzgibbon, 1996). In a study of 10-year-old daughters and their mothers, Hill et al. (1990) compared the responses of girls whose eating was highly restrained with those of girls whose eating was unrestrained. A strong relationship between the girls' restraint scores and those of their mothers was found. However, Thelen and Cormier (1995) found no significant associations between the self-reported dieting frequency or weight control measures of parents and their fourth-grade children. Perhaps, Hill et al. (1990) found a pattern due to their use of an extreme group design. To date, no study has examined the association between girls' body image dissatisfaction, which may develop prior to other eating disorder characteristics, and the weight and eating concerns of their mothers.

Girls likely differ in their degree of sensitivity to issues of appearance. Although no studies have investigated the association between social anxiety and body image dissatisfaction, the importance of this construct is suggested by two different but converging lines of research.

The first line of research involves a reconsideration of the relationship between peer teasing and eating disorders. Among college women (Thompson & Psaltis, 1988), adolescent girls Fabian & Thompson, 1989, Taylor et al., 1998, and elementary-age girls (Taylor et al., 1998), teasing effect is more strongly associated with various measures of body image dissatisfaction than teasing frequency. Hence, a girl's level of sensitivity may be more important than the actual amount of teasing received.

The anxiety disorder literature constitutes a second line of evidence. According to Atlas (1994), sensitivity to criticism includes an increased likelihood of perceiving criticism in ambiguous situations and of experiencing a negative emotive response to the criticism. Girls, who are more sensitive or socially anxious, may be more likely to perceive teasing and to experience a greater negative emotional impact (Atlas, 1994). At least, among college women, generalized anxiety and weight-related body image anxiety have been shown to be associated with body image dissatisfaction (Reed, Thompson, Brannick, & Sacco, 1991). Although the association between social anxiety and body image dissatisfaction has not been researched among elementary-age girls, the construct of social anxiety has been shown to be a reliable and valid construct among children (La Greca, Dandes, Wick, Shaw, & Stone, 1988).

Among the central features of eating disorders are pervasive concerns with appearance, deficits in public self-confidence, and obsessions with self-presentation (Striegel-Moore, Silberstein, & Rodin, 1993). A girl's evaluation of her appearance is influenced by the way others react to her, by comparing herself with peers, and by comparing herself with cultural ideals (Schonfeld, 1964). The social comparison hypothesis states that those who frequently compare their appearance to others, especially to persons deemed to be more attractive than themselves, are at an increased risk for the development of body image dissatisfaction (Thompson, 1992).

To date, evidence supporting an association between appearance comparison and body image dissatisfaction has been obtained only from college women Heinberg & Thompson, 1992a, Heinberg & Thompson, 1992b, Stormer & Thompson, 1996, Thompson et al., 1991a, Thompson et al., 1991b. However, appearance comparison is evident among elementary-age school children Feldman & Ruble, 1977, Pomerantz et al., 1995, Ruble et al., 1976. Moreover, Oliver and Thelen (1996) found that a child's tendency to compare her appearance with that of other children was associated with a drive for thinness, a construct commonly associated with appearance dissatisfaction. Hence, social comparison may also be associated with appearance dissatisfaction among children.

Among elementary-age girls, heavier children tend to have greater levels of body image dissatisfaction than their more average-weight peers Braet et al., 1997, Brown et al., 1995, Hill et al., 1994, Mendelson & White, 1982, Mendelson & White, 1985, Mendelson et al., 1996, Vander Wal & Thelen, 2000. Increased body mass index is associated with related constructs, such as drive for thinness, lower self-worth, perceived ineffectiveness, restraint of food intake, concerns about being or becoming overweight, and dieting Braet et al., 1997, Fox et al., 1994, Hill et al., 1992, Koff & Rierdan, 1991, Koff & Rierdan, 1993, Kolody & Sallis, 1995, Oliver & Thelen, 1996, Strauss et al., 1985, Striegel-Moore et al., 1995, Vander Wal & Thelen, 2000. Hence, the relationship between body mass index and body image dissatisfaction should be confirmed.

First, it is hypothesized that body image dissatisfaction will be predicted by variables from among the following four categories: (1) peer relations, i.e. peer investment in dieting and peer teasing about weight or appearance; (2) family characteristics, i.e. mother-reported parental influence, daughters' perceptions of parental influence, and mothers' body image dissatisfaction; (3) psychological features, i.e. social anxiety and physical appearance comparison; and (4) body mass index. Second, it is hypothesized that the relationship between teasing and body image dissatisfaction, as well as the relationship between direct parental influence (either mother reported or daughter's perception) and body image dissatisfaction, will be mediated or moderated by body mass index.

First, preliminary analyses were conducted to check for grade and school differences. Second, to compensate for the relatively large variables-to-subjects ratio, four sets of stepwise multiple regressions were conducted to identify the contribution of each construct area to the prediction of body image dissatisfaction. Third, blocked stepwise multiple regression analyses were used to identify predictors of body image dissatisfaction in children using only significant variables identified from each of the four categories. Finally, hypothesized tests of moderation and mediation were conducted.