Developmental Trajectories and Intentional Actions

The article turns a critical eye on problems arising from use of largesamples and complicated statistics to handle multilevel designs. I arguethat although these have a legitimate role to play in discovering causalrelations, valuable information can be lost or distorted in the processof their use. Exploratory classification and tree diagrams show howtransitional analyses can unpack effects of community, family, priormisbehavior, juvenile delinquency, and incarceration on adult criminalbehavior. The analyses showed that in the worst neighborhoods, disruptivebehavior had little discriminating power with regard to juveniledelinquency or adult crime. They showed also that good family interactionsserve as protection against crime in all types of neighborhoods. ConstructTheory, I suggest, is a way to understand the influences that seem toproduce criminal behavior.