The 1996 survey gathered information from a representative sample consisting of approximately 6,300 women ages 18 years or over. The current study examined the relationship of violence to each of 10 potential predictor variables, controlling for the remaining variables. The 10 variables were the victim's age, country of birth, education, employment status, marital status, income, main source of income, experience of childhood physical abuse, experience of childhood sexual abuse, and experience of violence since age 15. These variables were considered with respect to their ability to predict physical violence in the last 12 months, sexual violence in the last 12 months, emotional abuse by the current male partner in the last 12 months, and multiple incidents of violence since age 15. The most striking result was that a history of violent victimization as a child or as an adult predicted future victimization, controlling for a wide range of demographic factors. Another consistent finding was that younger women had a higher risk of victimization than did older women. Finally, married women were less likely than women who were separated or who had never married to experience physical or sexual violence in the last 12 months or multiple incidents of violence from age 15. The remaining demographic variables did not consistently predict types of violence in the multivariate models. Findings provide a useful guide to identifying groups of women with an increased risk of violent victimization and are important from the perspective of prevention. Figures, tables, appended methodological information and tables, and 40 reference notes