The influence of theoretically prominent family processes on late adolescent college student identity development was the focus of this study. The primary purpose was to test a model of adolescent identity development. This model proposed that family health variables would predict identity development, and that attachment and separation-individuation would each make unique and additive contributions to identity development. The second purpose was to identify instruments which discretely measured the family processes. The third purpose was to better understand family influences by measuring the processes of exploration and commitment. Participants were 150 male and 150 female college students, between the ages of 18 and 23, and from intact families. Questionnaires completed measured family functioning and identity development. Family functioning measures covered three domains (family health, attachment to parents, and separation-individuation) which formed the set of independent variables. The identity measures (ego identity status and identity process) comprised the dependent variables. A hierarchical regression design was employed where family health variables were entered first, followed by attachment variables, then separation-individuation variables. The results indicated mixed support of the model. First, the proposed model was statistically supported for females as all domains predicted identity achievement and diffusion. For males, only family health predicted …

The emerging consensus among investigators seems to be that bulimia is a multidetermined disorder with a number of contributing factors, including biological components, sociocultural factor, personality, and family characteristics (Garfinkel & Garner, 1982). An etiological model was examined in this study integrating two important theoretical perspectives in the bulimia literature: the stress-coping perspective (Cattanach & Rodin, 1988) and the family systems perspective (Minuchin et al., 1978). Five latent variables: Family Characteristics, Coping Resources, Psychological Disturbance, Environmental Stressors, and Bulimia were represented by twelve measured variables. Structural Equation Modeling analysis allowed for the simultaneous examination of the hypothesized interrelationships between model variables. Findings confirmed a direct impact of psychological disturbances on bulimic symptoms. Hypothesized indirect relationships of family characteristics, coping resources and environmental stressors to bulimia were confirmed. Treatment implications as well as directions for future research were discussed.