PREDICTING DESISTANCE IN A HIGH‐RISK SAMPLE: EXAMINING THE INFLUENCE OF INDIVIDUAL AND CONTEXTUAL FACTORS

Delinquent behavior typically emerges in early adolescence, peaks in mid-adolescence, and decreases in early adulthood (Tremblay et al., 2004). The age crime-curve has been observed across crime categories, countries, and historical periods (Hirschi & Gottfredson, 1983). Further, childhood antisocial behavior strongly predicts later antisocial behavior, yet not all children who display early behavioral problems continue to exhibit such behavior into adulthood (Moffitt, Caspi, Dickson, Silva, & Stanton, 1996). Thus, adolescence becomes a critical period for determining whether a youth will persist in or desist from criminal behavior (Loeber & Farrington, 2012). Desistance is the process that results in the termination of antisocial behavior and criminal activity (Laub & Sampson, 2001). As noted above, desistance typically takes place during late adolescence and early adulthood, irrespective of antisocial behavior onset (Loeber, Farrington, Stouthamer-Loeber, & White, 2008). Desistance is likely to result from a range of complex developmental, psychological, and sociological processes (Laub & Sampson, 2001). Although few theories exist to specifically explain desistance, both ecological and developmental-ecological perspectives have been applied to examining antisocial behavior in children and adolescents (Belsky, 1980; Bronfenbrenner, 1979;