This study sought to determine if mediation as practiced in the Brooklyn Dispute Center is a more effective method of resolving interpersonal disputes involving criminal acts than standard prosecution in the courts. A randomly selected control group was chosen from those cases which had been approved for mediation by the screening prosecutor and by the complainant (if he or she was present in the complaint room when mediation screening occurred). Complainants who had agreed to mediation but whose cases were selected as controls were informed that there was not enough room for their cases in the mediation program. The focus was on victims' rather than defendants' perceptions of the two adjudication processes and on the results they achieved. Complainants in both experimental and control groups were interviewed three times by evaluation staff. The data strongly indicate that the mediation process is perceived more positively by complainants than the court process and that going through mediation enhances complainants' perceptions of their relationships with defendants to a greater degree than does going through the court process. Still, a 4-month followup showed that mediation was no more effective than court adjudication in preventing recidivism; however, in both experimental and control groups, new interpersonal hostilities were the exception rather than the rule. Cases found to be least amenable to mediation involved disputes between intimates where there was a deep-seated pattern of serious hostilities. The less complex cases which involved problems derived from situational rather than patterned behavioral sources were more amenable to mediation. Tabular data and seven notes are provided.