Cases were selected from the 1984 database of the Center for Juvenile Justice Training and Research, which contains information on all individuals sent to the sample placements during that calendar year. A total of 527 cases were used in the study. Each had been released for at least a year at the time of this study. Recidivism measures addressed subsequent arrests, convictions, incarcerations, rate of arrests, and number of arrests during the first 12 months. The timing of the recidivism incidents was also examined. The findings indicate that by the end of the follow-up period, 55 percent of the study sample had been arrested; 48 percent had been arrested within the first 12 months. Juveniles with more extensive arrest records were more likely to be arrested, convicted, and incarcerated after release than juveniles with fewer preplacement arrests. The younger a juvenile was at first arrest, the more likely he was to be arrested, convicted, and incarcerated after release. The older a juvenile was at release from placement, the less likely he was to recidivate. Race, chemical dependency, and family stability were not significant in the prediction of recidivism for the study sample. Poor performance in school and difficulties in adjusting to the residential placement were related to more recidivism behavior. 6 tables, 21 figures, and appended supplementary data