The purpose of this study was to determine the risk factors for injury that could be studied in both the Canadian Violence Against Women Survey and the National Violence Against Women survey in the United States; and to determine which risk factors predicted injury occurrence. Both surveys were telephone surveys. It was hypothesized that experiencing violence before the marriage or common-law union, having a partner that was drinking, having children witness the violence, experiencing more than one violent episode by the same partner, fearing one’s life was in danger, and experiencing high levels of emotional abuse were related to an increased risk of injuries. Results were presented on bivariate and multivariate findings, model fit across the data sets, and statistical comparisons of findings across the data sets. There were many similarities across the two data sources in terms of the magnitudes of associations between the risk factors and injury occurrence. There were no differences in the magnitudes of associations between injury risk and the variables assessing onset of violence, children witnessing, perpetrator drinking, other abuse by the same perpetrator, and low or moderate levels of emotional abuse. There were differences in the magnitudes of associations between injury risk and the variables assessing feared injury or death and high levels of emotional abuse. Fearing injury or death was more predictive of injury risk in the Canadian survey than in the United States survey. Experiencing high levels of emotional abuse was more predictive of injury risk in the United States survey than in the Canadian survey. Both the Canadian and the United States data suggest factors that likely increase a woman’s risk of injury from intimate partner violence. Alcohol use by the partner and having been victimized more than once by the same partner increased a woman’s risk of injury in both data sets. 5 tables, 14 references