The population examined consisted of all residents released from Michigan prisons to the community between January 1, 1978 and June 30, 1978. The population was stratified according to whether persons were released to community programs or directly to parole, and a proportional random sample was drawn. The study sample consisted of 1,117 cases or 38 percent of the population. A comparison of descriptive statistics showed that persons released to community programs were less likely to be serving for a violent crime, have a prior prison commitment, juvenile history, or serious institutional misconduct than persons released directly to parole. The analysis of program operations showed that no CRP participants were considered very high risk cases. In general, findings indicate that CRP participants are subject to a more rigorous screening and monitoring process than persons released directly to parole. When compared to the latter, CRP participants commit significantly fewer felonies, their involvement in violent crime while on CRP status is almost nonexistent, and their overall felony involvement is lower than that of parolees. These observations support the conclusion that CRP participants do not present an undue risk to the community and are subject to substantial regulation. Footnotes, tabular data, and seven references are given. (Author summary modified)