The Dynamic Relationships between Parenting and Adolescent Delinquency: A Group-Based Model Approach

Abstract

Parenting has been found as a key factor for influencing adolescent delinquency. Past studies have shown that adolescents who expose to higher level of warmth, induction, and monitoring parenting would less likely be delinquents. However, most of these studies present the relationship between parenting and adolescent delinquency statically. That is, researchers show parenting measured at one time predicts adolescent delinquency at the same time or the other time. Researchers seldom demonstrate whether the change of parenting over time influences the change of adolescent delinquency. Theories of types of delinquents propose two types of delinquents in offender population (Patterson et al., 1992; Moffitt, 1993, 1997). Early starters show their delinquency in late childhood and persist into adulthood, while late starters show the behavior during mid-adolescence and desist after this period. Although researchers propose several predictions for distinguish types of delinquents and for explaining their developmental trajectories, parenting is the key factor among them. In predictor analysis, ineffective parenting in late childhood interacts with child's early temperament problems producing early starters. Several studies have demonstrated this argument (Fergusson et al., 2000; Nagin and Land, 1993; Wiesner and Silbereisen, 2003). In within delinquent group analysis, parenting has long been demonstrated its impact on adolescent delinquency (Lin, 2002). However, there is no research that systematically explores the relationship between parenting and delinquency over time under types of delinquents framework. In this study, we ask: How the change of parenting influences delinquent trajectories over time? Is there difference between different types of delinquents? Using 5-year panel data of adolescents in Taiwan, current study started from types of delinquents argument and provided a group-based model to distinguish groups of delinquents. Based on the delinquent groups, we presented a time-varying model for exploring the dynamic relationship between parenting and delinquent trajectories. The results showed two offender groups in the sample and correspond to theories of types of delinquents. We also found the dynamic effect of parenting over time when predicting delinquent trajectories. The parenting showed its prominent effect for late starters in the later waves, while for early starters parenting had intervention effect at every time point.