PATTERNS OF VIOLENT JUVENILE RECIDIVISM (PENNSYLVANIA)

The problem of the serious juvenile offender has emerged as an important policy concern. Some of the issues identified by prior researchers which are reexamined in the present study of violent juveniles are those of offense specialization, career escalation, and the relationship between age at onset and chronicity. These analyses of longitudinal data have two objectives: (1) to examine differences in the criminal careers of violent and nonviolent delinquents, and (2) to identify those members of a birth cohort who are chronically violent. The data are from the Philadelphia 1958 Cohort Study and include the police records of juveniles from their first police contact to the age of eighteen. The descriptions of offenses from the police investigation reports were coded and scored for seriousness using the techniques developed by Sellin and Wolfgang in The Measurement of Delinquency. For purposes of the present analysis a violent offense was defined as any offense involving injury or according to the Uniform Crime Reports\u27 classification system as homicide, rape, robbery, or aggravated assault. A violent offender was defined as a juvenile who had committed at least one violent offense. Significant differences were found between violent and nonviolent delinquents. For example, violent delinquents had an earlier age at onset than did other delinquents. The results also indicate that a large proportion of violent juveniles are nonwhite, low SES males. The chronically violent juveniles with three or more violent offenses were responsible for the majority of all offenses by delinquents in the birth cohort and for those offenses involving serious injury. Furthermore, the chronically violent delinquents tended to commit many nonviolent offenses before their first violent crime. As was found in prior research of nonviolent delinquents, the criminal careers of violent juveniles were characterized by cafeteria-style delinquency and can be modelled by a Markov chain. There was also little evidence of escalation of offense seriousness. Consequently, it was impossible to predict violent recidivism based on the prior record. Nonetheless, the findings of the present study indicate that the violent juvenile offender is a chronic offender who will probably commit many nonviolent and violent offenses