Abstract
Time-trends in the prevalence of adolescent sexual initiation exhibit periods of increase, decrease and equilibrium. We attempted to explain, in mechanistic terms, how such dynamics arise by developing an Agent-Based Model. The model assumes that sexual initiation diffuses socially both within and across cohorts. The model behavior matched, qualitatively, the empirical time-trends. The impact of two intervention strategies suggested that the age at which an intervention is implemented effected system behavior as did the choice of which specific subpopulation was targeted. Suggestions for how computational models might be used to explore research questions in developmental science were discussed.
ACKNOWLEDGMENTS
This work was funded, in part, by a seed grant from The Institute for Social and Economic Research in Policy at Columbia University.