Gender identity and bulimic eating behavior

From a social-psychological perspective, the disproportionate number of women compared to men affected by bulimia nervosa implies that gender (i.e., the social construction of sex) plays an important part in the etiology of this disorder. From this perspective it was hypothesized that the risk of developing bulimia depends, in part, on the composition of a woman's gender identity. Three questionnaire studies conducted in the United States and former West Germany tested competing hypotheses on the relationship between gender identity and bulimia. Respondents were predominantly middle class and Caucasian. Study 1 was conducted in West Germany and tested the hypothesis that bulimia is related to a hyperfeminine gender identity M. Boskind-Lodahl (1976) Cinderella's stepsisters: A feminist perspective on anorexia nervosa and bulimia, Signs, Vol. 2, pp. 342–356]. Twenty-six bulimic women were compared to 26 nonbulimic women. Results from Study 1 indicated that bulimic women typically had a gender-typed identity, whereas nonbulimic controls tended to be androgynous. However, between-group differences were based on hypomasculinity rather than hyperfemininity on part of bulimic women. Study 2 and Study 3 explored the idea that masculinity positively contributes to general esteem and reduces the risk of bulimia nervosa. In Study 2, three competing theories of the optimal relationship between mental health and gender identity (congruence, androgyny, and masculinity) were related to bulimia and tested with structural equation modeling in a sample of 301 North American undergraduates. In Study 3, the findings of Study 2 were cross-validated in a sample of 464 West German high school students. The results indicate that masculinity had significant positive effects on eating behavior via esteem, whereas the effects of femininity were negligible.The research and preparation of this article were supported in part by a Fulbright scholarship, a predoctoral stipend from the University of Heidelberg, and a postdoctoral stipend from the Max Planck Institute for Human Development and Education.I thank the women who volunteered to participate in this research project as subjects. I gratefully acknowledge the support of the principals, especially Frau Ute Vater, and teachers of the Bunsen-, Feudenheim-, and Lessing-Gymnasium, who enabled Study 3.Thanks to Michael Marsiske, Ulrich Mayr, John R. Nesselroade, Kai Schnabel, Anna Stetsenko, and two anonymous reviewers for their comments on an earlier draft of this paper, and to Bettina Franzke for technical support. I also extend my thanks to Frank Faulbaum and Norbert Schwarz, without whom this research could not have been completed.