The man-to-man (m-2) program matches volunteers on a one-to-one basis with inmates and ex-inmates of Ohio's adult correctional institutions and operates on the principle of helping through friendship. The study used an ex-post-facto design entailing construction of an experimental (matched) group and two comparison groups (nonmatched and random). The experimental group included a randomly selected sample of 107 ex-offenders matched with volunteers and paroled from a State correctional institution. Members of the nonmatched group were 120 ex-offenders who met the 'estrangement' criterion of the m-2 target population. The outcome variable for the study was a criminal behavior scale in two versions: the severity index, a continuous scale based upon severity of offense as prescribed by the Ohio revised code, and a simple dichotomy of no crime/crime on parole. Comparisons were made on the basis of age, race, and other demographic variables, and on months of incarceration and prior criminal justice record. The chi-square statistic was used to examine the distribution of the groups for comparability. Analysis showed, among other things, that the m-2 match has a positive effect on white offenders but a negative effect on nonwhite offenders. However, future controlled research addressing the match relationship must be encouraged. Tables of study data, footnotes, and 12 references are provided. (Author abstract modified)