The study group consisted of 145 female parolees who were released from the custody of the Massachusetts Department of Corrections to parole supervision in 1989, 1990, and 1991 in the Greater Boston area; they had completed or were terminated from their term of parole. The quantitative research design used retrospective information obtained through a systematic review of official State records and a brief questionnaire that was completed by the parole officers who supervised the subject. The quantitative data were supplemented by qualitative information collected through formal interviews with the parole supervisors and by information collected at informal meetings with several female inmates who had prior experience with both community social service programs and parole supervision. The qualitative information was used to inform the interpretation of the quantitative data. Recidivism, the dependent variable, was measured in two ways: the number of recidivistic arraignments during the period between a subject's release from parole and the end of the 1-year follow-up period and the number of days to the first recidivism. Independent variables were grouped into four domains: demographics, needs, characteristics of parole, and criminal history. The use of services and type of service used were also used as intervening independent variables. Five of the service-use variables were found to produce an effect on the recidivism of the women in the study: starting an outpatient substance abuse treatment, a multiservice center, a residential treatment center, completing an employment program, or an outpatient substance abuse treatment program. There was a decrease in recidivism from beginning community programs (other than outpatient substance abuse), which is magnified for women who complete the programs. 9 tables, 14 notes, and 19 references