Carrying the Weight of Self-Derogation?Disordered Eating Practices as Social Deviance in Young Adults

ABSTRACT

This study asks if self-derogation in early adolescence increases the odds of adopting disordered eating practices (DEPs) in young adulthood. Using a racially/ethnically diverse random sample of 1,209 youth, our (longitudinal) data show that both boys and girls who disliked themselves at age 13 were more likely to adopt disordered eating practices at age 22. Following Kaplan's theory of deviance, we view disordered eating as a form of internally directed deviance that youth may adopt in an effort to “correct” for negative self-feelings. We find that girls/women adopt DEPs more than boys/men do, although this is not true for African Americans. Family socioeconomic status fails to predict DEPs. Exploratory analyses aimed at understanding connections between self-derogation and DEPs show that perceptions about weight matter. In particular, women who are more anxious about their weight are more apt to engage in DEPs (compared to less anxious women). Our study contributes to debates about theories of deviance and theories of gender, relative to explaining disordered eating practices.

Acknowledgments

This work is supported by Grant 5 R01 DA 10772 from The National Institute on Drug Abuse to R. Jay Turner.

Notes

See Schatzki (2001) and Martin (2003) for recent conceptions in social theory of practices. Martin focuses on practices associated with gender.

Although we do not address sexual orientation, some research shows that gay men are more concerned with their bodies than straight men (Carlat et al. 1997; French et al. 1998).

Increases in problematic eating may be among the negative consequences of the acculturation process experienced by ethnic minorities (Hsu 1990; Bowen et al. 1991; Cachelin et al. 2000). In fact, acculturative stress has been found to moderate the relationship between body dissatisfaction and bulimia for ethnic minority women (Perez et al. 2002). Because ethnic minorities have been understudied, particularly in large random samples, the question of ethnic variations in eating disorders is uncertain. Since self-destructive eating behaviors vary between and within different ethnic groups (Striegel-Moore & Franko 2002), we explore their patterns in our large, random sample of youth/young adults (see Hypothesis 2).

Accurate estimates of the prevalence of eating disorders are difficult to ascertain due to a lack of large-scale epidemiological studies (Levine & Smolak 2002).

Because they are complex and multidimensional, feelings about one's body have been difficult to operationalize (Hsu 1990), and no uniform definition is accepted. Body assessments are usually defined as they relate to other constructs such as weight satisfaction or weight anxiety, size perception accuracy, body satisfaction, appearance satisfaction, and body dysphoria (Thompson et al. 1999). J. Thompson (1990:1) argues that the construct of “body image” has been used to describe “phenomena that vary widely in their specific characteristics, [and definitions used] are very idiosyncratic to the particular type of body image under investigation.”

The parent study on which this paper is based consisted of five waves of data collection conducted between 1990 and 2003. The present paper restricts analyses to the two waves of data collected in 1992/1993 and 2002/2003. To avoid confusion, we henceforth refer to these waves of data collection as Time 1 and Time 2. A full description of the study design is provided elsewhere (Turner and Gil 2002). A full description of the context and design associated with the other three waves of the research has been reported elsewhere, and a summary of the findings has been published by Vega and Gil (1998). A detailed summary of the fourth wave of data collection has also been published (Turner and Gil 2002).

Detailed analyses provided assurance that participants were highly representative of the population from which they were drawn (Vega and Gil 1998).

Comparisons of persons interviewed with the random sample drawn from the original study population reveals no statistically significant differences on a wide array of early adolescent behaviors and family characteristics as well as mental health issues and practices.

Disordered eating practices were only measured at Time 2.

^a Group mean significantly different from men respondents.

^b Group mean significantly different from African American respondents.

^c Group mean significantly different from Cuban American respondents.

^d Group mean significantly different from Non-Cuban Hispanic American respondents.

Note: Ethnic contrasts employed Scheffe post hoc multiple comparisons significance test (p < .05).

*p < .05;

**p < .01;

***p < .001.

Notes: Unstandardized regression coefficients. Standard Errors in Parentheses.

^† p < .10;

*p < .05;

**p < .01;

***p < .001.

It is important to note the substantial increases in standard errors in the coefficients from equation 5 to equation 6. This is due, in large part, to operational confounding between perceptions of overweight and weight anxiety (r=.437).