Six hypotheses based on the feminist perspective were first tested. Among these were that the criminal justice system discriminates against women generally and that the insensitive treatment of rape victims discourages them from prosecuting. Findings showed that discrimination against female victims did not appear to be a major cause of the low conviction rate, nor did rape victims appear more likely than victims of other serious crimes to cause case attrition due to their own unwillingness to prosecute. Questioning of the victim's credibility, on the other hand, did seem to be more of a reason for case dismissal for rape cases than for other serious crimes. Corroboration also seemed to be important to the conviction rate. A multivariate analysis of the types of rape cases that did result in conviction yielded eight statistically significant variables. Witnesses and physical evidence increased the probability of conviction, while the victim's credibility was reduced if she had an arrest record. Older defendants were less likely to be convicted, while victims who were children and those who were much older were more likely to have their cases result in conviction. Conviction rates were also higher where the victim and defendant were strangers or acquaintances than where they had a closer relationship. A sodomy charge also increased the likelihood of conviction. Results support a new perspective on rape cases which emphasizes the issues of consent and credibility. Research across several jurisdictions is needed. Tables, notes, and a list of 30 references are provided. (Author abstract modified)