[Truncated abstract] Eating disorders affect a significant proportion of adolescent and young adult women and a smaller proportion of children, men, and older individuals. They are associated with a range of physical, psychological, and social consequences that can have a profound and lasting impact on affected individuals. Eating disorder symptoms (e.g., marked weight and shape concern, strict dieting, binge eating, purging) are also associated with physiological and psychosocial morbidity, and are reported by up to half of adolescent girls and one-third of adolescent boys. If eating disorders are to be effectively prevented or treated, it is imperative that risk and maintaining factors for the conditions are identified. ... This thesis aimed to identify the variables and models that may best account for the development and maintenance of eating disorder symptoms, through two broad studies and six sub-studies. Study 1 focused on identifying predictors of cognitive (i.e., elevated weight and shape concern) and behavioural (i.e., binge eating) eating disorder symptoms in pre- to early-adolescent children followed over time. Prospective tests of the dual-pathway (Stice, 2001) and cognitive-behavioural (Fairburn, 2002; Fairburn, Cooper, & Shafran, 2003) models of eating pathology were also conducted, and a distinction was made between weight and shape concern and weight and shape over-evaluation. Low selfesteem, perceived media pressure to be thin, weight and shape over-evaluation, and maternal concern about child weight prospectively predicted increases in child weight and shape concern over time (Studies 1b and 1c), and weight and shape concern was the most robust predictor of increases in dietary restraint (Studies 1b and 1d). Dietary restraint and affect-related eating prospectively predicted binge eating onset (Studies 1a and 1d), and the dual-pathway and cognitive-behavioural models were both able to account for the development of binge eating over a two year period (Study 1d). Evidence was also provided for the relative superiority of the most recent, enhanced cognitivebehavioural model (Fairburn et al., 2003) over other available theoretical accounts. Study 2 focused more specifically on the role of mood intolerance in predicting and maintaining eating pathology in adults. In Study 2a, a new measure of mood intolerance was developed, revised, and evaluated. In Study 2b, the role of mood intolerance in cross-sectionally predicting binge eating and purging was examined within the framework of the enhanced cognitive-behavioural model. The importance of mood intolerance in accounting for eating disorder symptoms was confirmed, and additional support for the cognitive-behavioural model was provided. Collectively, the six empirical studies have provided new data regarding the relative importance of different variables in the development and maintenance of different eating disorder symptoms. They also provide initial insight into the relative validity of alternate theoretical models in this area. The results suggest that the most recent, enhanced cognitive-behavioural model may provide the best account of how eating disorder symptoms develop and are maintained, providing that the mood intolerance component of the model is specified.

The purpose of this investigation was to assess the association between body mass index as well as race/ethnicity and established correlates of disordered eating including drive for thinness, body dissatisfaction, bulimia, dietary restraint and social physique anxiety in male and female college students. Difference between actual, desirable and perceived body weight was also assessed. ANOVA suggested that as actual body mass index increased, in general, attitudes and behavior toward weight, body shape and eating increased. A two-tailed t-test suggested that males and females differed significantly with regards to attitudes and behavior toward weight body shape and eating. ANOVA indicated that little to no association existed between race/ethnicity and established correlates of disordered eating.