Juvenile justice decision-making as a longitudinal process

Findings from a longitudinal analysis are used to illustrate the importance of in-cluding processual variables in explanations of the type and severity of disposi-tions imposed within the juvenile justice system. Results show that dispositions imposed for prior offenses exert a significant impact on dispositions imposed for current offenses, even when appropriate variables are controlled. Moreover, the pattern of this relationship is one of stabilization rather than escalation: disposi-tions are likely to be repeated from offense to offense rather than becoming more severe as the criminal career develops. Previous studies of decision-making in the criminal justice system have tended to focus on the consequences of the wide-ranging discretion granted to criminal justice decision-makers. In doing so, they have relied on cross-sectional designs that examine final dispositions for only one of-fense in a person's criminal career. While appropriate for some theoretical issues, such a static orientation offers a limited view of factors that can potentially influence dispositional decisions. Indeed, it explicitly ignores the possibility that criminal justice decisions are not made in isolation from one another but occur as part of an interconnected process. Process variables can influence decision-making at two levels. The first involves the impact of pre-adjudicatory decisions on the type and severity of final dispositions imposed for a particular offense. Several studies (Bernstein et al.; Bortner; Hagan et al; LaFree) show that proces-sual variables of this type do influence final dispositions and need to be taken into account in a complete model of decision-making. The second, and more dynamic, level involves the sequence of sentences impose