The purpose of the present study was the construction of the Rape Empathy Scale (RES), designed to measure subjects' empathy toward the rape victim and the rapist in a heterosexual rape situation. The results of psychometric analyses of reliability for both a student and juror sample are presented, in addition to evidence of cross-validation on separate student and juror samples. Significant differences between male and female subjects' RES scores were found, as well as differences between scores of women who had experienced a rape situation (rape victims and rape resisters) and women with no previous exposure to rape. RES scores were predictive of both students' and jurors' ratings of defendant guilt, as well as their recommended sentences for the defendant and their attributions of responsibility for the crime. Furthermore, subjects' RES scores were predictive of their social perceptions of the rape victim and defendant, and male jurors' RES scores were negatively correlated with their reported desire to rape a woman. The results are discussed in relation to the low conviction rate for sexual assault cases and the importance of juror selection as a vehicle for increasing the number of just convictions in rape cases.