The study defines a recidivist as any ex-offender who was returned to a State or Federal correctional facility for a period of 30 days or more during the followup period. Monitoring of recidivism data since 1971 has revealed a systematic reduction of recidivism rates from 1971 to 1978, from 25 percent to 16 percent. However, recidivism rates of the 1979 and 1980 release populations increased to 26 percent. Individuals who had experienced furlough prior to release had significantly lower rates of recidivism than those with no furlough experience. The data also showed that participation in prerelease programs prior to community release reduced recidivism rates. Finally, individuals released directly from medium- or minimum-security institutions, including prerelease centers and halfway houses, had significantly lower rates than offenders released directly from maximum-security institutions. All these relationships held even when selection factors were controlled and remained constant when followup periods were extended from 1 to 2 years and then to 5 years. The graduated social reintroduction of the offender, accomplished through a series of moves through institutions with descending levels of security and size, was associated with reductions in recidivism. Despite the recent increase in recidivism, participation in reintegration programs has remained associated with lower rates of recidivism. Statistical tables and a bibliography of 28 research publications are appended.