Cross-lagged analysis of adolescent sensation seeking and health risk behaviors: Testing reciprocal causality and causal direction

Using multiple waves from the 2000, 2002, and 2004 NLSY Young Adult cohorts (n=1700), this paper addresses how the causal inter-relationships of depression and risk proneness (sensation seeking) influence adolescent alcohol use and sexual risk taking. Structural equation modeling with cross-lagged data will test the reciprocal causality of risk proneness and depressive symptoms and their affect on health risk behaviors over time among adolescents ages 14 to 21. This phenomenological cycle will be evaluated by applying statistical weights for each of the respective years, prior to calculating the covariance matrix for path analyses performed in AMOS. In preliminary analyses, the direct and indirect influence of depression and risk proneness on adolescent alcohol use and sexual risk taking suggest a one-way direction of causation. This research builds on existing findings from cross-sectional data, extending the model from one point in time to determine how Time 1 risk proneness propensity influences Time 2 health risk behaviors which affect Time 3 outcomes, i.e. severity index of adolescent alcohol use in the past 30 days and sexual risk taking. Implications for community-level intervention programs are discussed.