This is a (competing continuation) request for an ADAMHA Research Scientist Award (RSA). The broad objectives include research, career enhancement, and mentoring goals. The research objective is to estimate theoretically informed models that specify (1) the psychosocial antecedents and consequences of patterns of drug abuse and other deviant adaptations to stress and the constructs that moderate these relationships, for a cohort over the life span between early adolescence and the fourth decade of life (mid 30s), and (2) continuities and discontinuities across generations (the cohort and their adolescent children) of patterns of drug abuse and other deviant adaptations, and their correlates, observed at a comparable developmental stage (early adolescence) for the two generations. Data for the first set of models are from household interviews with a cohort of subjects in their mid 305 who were interviewed during their mid-to-late 20s (N=6,089) and, prior to that, during the seventh grade. The models specify direct and indirect effects of earlier life stages on later stages and specify the circumstances that moderate the nature of the effects. Deviant patterns include violence, self-destructive behaviors, functional psychiatric disorders, a range of criminal activities, and violation of normative prescriptions regarding the adoption and performance of conventional social roles, as well as patterns of drug abuse. These patterns are variously modeled as antecedents or consequences of each other. Other antecedents and consequences of drug abuse and other deviant adaptations include attitudes and behavioral responses of significant others in various social- relational contexts, chronic role strain, ascribed social identities, modes of coping, life events, personal and interpersonal resources, peer associations, self-feelings, and values. Data for the second set of models are from household interviews with the children of the cohort who have already reached the age the parent was when first interviewed in early adolescence or at the time they reach that age in the course of the study. The models will account for continuities and discontinuities between the generations during early adolescence by specifying mediating influences, and common antecedent experiences. The models will be estimated using structural among other multivariate techniques. Research skills will be enhanced by studying new methods of multivariate analysis and by collaboration with staffs of other longitudinal projects. Mentoring activities include guidance of graduate students who are employed on the project and in seminars in the substance and methods of research on substance abuse.
