A total of 9,549 offenders were released from North Carolina prisons in fiscal 1979-80. Released offenders were traced in Department of Correction (DOC) records for 29 months after being released from prison to determine the total length of all new active sentences (if any) they received for new crimes within those 29 months. Multiple regression models were fitted to the data to identify which items were associated with recidivism and to determine how well recidivism could have been predicted. The models proved to be no better or worse in predicting recidivism than simply predicting that no released offender would return to prison. The statistical association of certain variables with recidivism was analyzed. These variables included demographic characteristics; previous offenses, sentences, and convictions; prison experience; and parole supervision. Specific findings are detailed.