The implications of age of onset for delinquency risk. II: Longitudinal data.

The role of age of onset in the level of involvement in delinquent behavior as marked by seriousness and chronicity of involvement continues to draw extensive attention from researchers. This issue bears on some of the key causal contentions about the dynamism of involvement and the validity of a developmental model of antisocial behavior risk. Five waves of the National Youth Survey were utilized here to determine if, among a nationally representative sample, there was evidence of onset age influence on later involvement. Results suggest that early onset (before age 12) relates to higher rates of more serious acts over a longer period of time for boys and girls. Overall, the results suggest support for early onset spurring on later involvement, but the contribution is small once psychosocial predictors are considered. Onset age seems most important in understanding involvement in serious crime over several years. Involvement is explained best by peer variables for males and school and family variables for females. Onset age is explained by a wider range of variables than involvement and there is greater similarity of the psychosocial variables that explain onset for both genders. The interaction of involvement and predictors was noted, suggesting a dynamic model of risk. Implications for prediction and prevention are discussed.