A total of 136 male and female psychology students made judgments concerning the delinquent behaviors of a male stimulus person. Subjects were scheduled in mixed-sex groups of up to eight and shown a videotaped interview with the youth along with a 'clinic folder' containing information about the consensus (i.e., the typicality of the behavior in the youth's neighborhood) and the consistency of his behaviors (i.e., number of prior arrests). The stimulus person could be seen as reacting to environmental pressures in the neighborhood, or as acting because of internal causes. The neighborhood was seen as a reasonably stable set of environmental variables which could affect many of a person's behaviors, producing consistent behavior over time. The results showed that female observers significantly reduced the responsibility assigned to the youth (less blame and less relative responsibility) only when the behavior was described as typical in the environment and when it was highly consistent with past behaviors. The degree of consensus has less impact on the females' responses when the behavior was presented as a single event, inconsistent with prior behaviors. In these cases, females assigned greater blame and responsibility to the youth. However, males held the youth responsible as a function of the behavior's consensus only when the particular behavior was inconsistent with his past behaviors. Even though males and females employed alternative strategies for dividing the responsibility, blame was consistently perceived as correspondent with the degree of internal responsibility assigned. Consistency of these relationships across other situations, which should be studied, would require a more complex model of the attribution process. Statistical data and three references are included.