Justifications for the development of special correctional pro grams and for the choice of sentencing disposition for an individual offender are frequently based on claims of greater rehabilitative efficacy. While considerable evidence exists that some types of offenders have relatively more or less likelihood of recidivism than others, there is, as yet, almost no evidence that available correctional alternatives have any impact on those likelihoods. The article reviews findings from studies of correc tion in California for five critical choices in offender process ing : (1) imprisonment or probation, (2) length of stay in prison, (3) treatment program in prison, (4) intensity of parole or probation supervision, and (5) outright discharge from prison or release on parole. The authors conclude that variations in recidivism rates among these alternatives are, for the most part, attributable to initial differences among the types of offenders processed and that the remaining differences in violation rate between programs may be accounted for by differences in inter preting an event as a violation or in officially designating it as such. No evidence was found to support claims of superior re habilitative efficacy of one correctional alternative over another .