OBJECTIVE To assess first-year students' knowledge of, attitudes toward, and personal histories of family violence. METHODS An anonymous, self-administered, 70-item questionnaire was developed and distributed in the first six months of medical school to 390 first-year students at three New England medical schools in 1991-92. The students were tested on knowledge and asked questions about their personal histories. Attitude questions were scored on a five-point Likert scale. Attitude scales were developed by dividing questions into three content groups (education about family violence, the physician as advocate, acceptability of violent behaviors) and removing items for which inter-item coefficients with all other items were less than .20. Two-tailed t-tests were performed on continuous variables and chi-square tests on categorical variables. Ninety-five percent Cls were calculated for point estimates. RESULTS In all, 370 students (95%) responded. Of these, 139 (38%) reported personal histories of abuse. The mean knowledge score was 11.3 (of 16 questions). One-third of the students answered more than 75% of these questions correctly; 12% answered less than half of the questions correctly. The women felt more strongly than the men about the need for violence education (p < .001). The students who reported histories of abuse more strongly favored education (p < .05) and advocacy roles (p < .001) for physicians. In addition, the students who had histories of family violence were more likely to report histories of suicidal thoughts (p < .0001). CONCLUSIONS The students lacked knowledge but felt a need to learn more about family violence. Family violence curricula should be better integrated into medical education. These curricula should be sensitive to students' attitudes, given the reported high prevalence of personal histories of family violence among students.