A recidivist was defined as any subject who was returned to a State or Federal correctional institution, a county house of correction, or a jail for 30 days or more during the period of followup (1 year from the date of each subject's release to the community). Data were collected for four categories of variables: (1) current offense commitment variables, (2) personal background characteristics variables, (3) criminal history variables, and (4) recidivism variables. Data were collected from the files of the Department of Corrections, the Board of Parole, and the Board of Probation. Hypothesized explanations for the increase in the recidivism rate in the 1979 cohort were a higher risk population passing through the corrections system, a policy change in the parole revocation process, a policy change in the parole releasing process, and a change in the level of participation in the reintegration model. Data analysis did not support any of these explanations, however. In addition to showing that furlough participation and successful completion of prerelease placements were strong predictors of nonrecidivism, data showed lower recidivism rates for releasees from lower security institutions and higher recidivism rates for releasees from higher security institutions. Individual variables associated with recidivism rates were marital status, education, employment history, age, type of offense, and criminal career patterns. Tabular data and 25 bibliographic entries are provided.