Recidivism offenses are those for which the client was reconvicted during a set followup period after release. Two followup periods were analyzed: an approximate 6-month period immediately following release from residential care and a full 3-year period following release. Results of the analysis of recidivism 'during parole' (i.e., the 6 months after release) indicate that 39 percent of the 694 youths assigned to parole in 1982 reoffended during their parole. Only 25 percent of the 291 youths directly discharged in 1982 reoffended during the 6-month period after release. Youths with extensive prior criminal records, males, older youths, and parolees were most likely to reoffend during the 6-month followup period. Analysis of the 3-year postrelease recidivism indicates that 63 percent of the youths reoffended. Older youths, youths with extensive prior criminal records, males, and parolees were also most likely to reoffend during the 3-year period. These findings suggest that consideration should be given to placing offenders in higher security settings based on the factors predictive of recidivism. Additional study is required to assess the effectiveness of parole, since parole had little positive impact on recidivism. 11 tables and 5 figures. (Author summary modified)