How Early Is Too Early to Predict Persistent Offending?

• In Event: Assessing Interventions across the Life Course

Abstract

While the identification of different patterns or trajectories of criminal offending is informative from a theoretical perspective as it summarizes “typical” criminal careers and allows for a retrospective investigation of risk factors for “group” membership, it is limited in the practical sense because these patterns of behavior are only identified post-hoc. A logical next step is to assess how early we can identify individuals who follow various criminal offending trajectories in order target these individuals for early interventions (selective interventions; Multisystemic Therapy; Borduin et al., 1990). Using data from the Rochester Youth Development Study, and an extension of trajectory analyses developed by Nagin and colleagues (in press), this research examines how well we can classify individuals in their ultimate offending trajectory prospectively from adolescence through adulthood. We then compare the ability to correctly classify individuals using prospective behavioral measures with theoretically informed risk factors from childhood. The results inform theoretical discussions regarding criminal careers and assist in the identification of individuals who would most benefit from prevention programming.

Authors

• Thomas A. Loughran, Pennsylvania State University

• Megan Bears Augustyn, The University of Texas at San Antonio

• Pilar Larroulet, University of Maryland / Universidad Catolica de Chile