Although the results of research on the effectiveness of incarceration in reducing recidivism versus those of parole are mixed, there are no indications that one form is superior to the other. Incarceration for longer periods increased the probability of future criminal activity, even when the factor of longer prison terms for poorer recidivism risks was considered. The likelihood of recidivism plays little role in decisions to retain inmates beyond their minimum sentences, because variable periods of incarceration are used by corrections officials as incentives for cooperation with the system. No treatment techniques have unequivocally demonstrated themselves capable of reducing recidivism. Detailed research on the effects of group counseling have shown that it had no effect on behavior during or after incarceration. Evaluation of a medical model program for treating addicts indicated that the program did not substantially reduce addiction or subsequent arrest rates. Assignment of smaller caseloads to parole supervisors resulted in increases in recidivism, possibly because the parole officers then had greater opportunity to identify technical violations which returned parolees to prison. Prisoners directly discharged from prison had lower return rates than those placed on parole because those in the first group were not subject to reincarceration for administrative reasons. The evidence indicates that offender characteristics influence recidivism regardless of the correctional approach used. Because none of the approaches demonstrated superior rehabilitative efficiency, decisions about punishment should be made some other basis than the anticipated rehabilitative effects of the punishment or program or their effects or recidivism. Reference notes and tabular data are included.