Objective: The purpose of this study is to determine which factors best predict parental use of single versus multiple types of violence. Methodology: The study uses data from a Quebec telephone survey conducted with a representative sample of 2,469 mothers of children aged 0-17 years. The interview covered topics such as mother's attitudes and reports of violence towards children, mother's experiences of childhood violence, level of marital discord, and socioeconomic factors. A discriminant function analysis was performed using individual and family factors to predict membership in five distinct groups: (a) no violence reported; (b) reports of psychological aggression only; (c) reports of minor physical violence only; (d) reports of psychological aggression and minor physical violence; and (e) reports of psychological aggression, minor and severe physical violence. Results: One-way ANOVAs revealed that the groups differed significantly on factors such as child's age and gender, mother's attitudes, mother's experiences of childhood violence, level of marital discord, and number of children in the household. Discriminant function analysis revealed that the groups differed in two ways. The first function maximally separated the report of psychological aggression from minor physical violence only. The second function discriminated between the combined reports of all three types of violence versus the reports of minor physical violence only. Mother's attitudes and experiences of childhood violence were the two factors that best distinguished among groups and best predicted group membership. Conclusion: These results offer new insights for the prediction of mother's reports of one type versus multiple types of violence and highlight the importance of distinguishing between types and combinations of family violence in future research.