because our measure of multiple forms of violence included violence by all kinds of perpetrators, the characteristics of violence reported by men and women are likely to be different, and gender-based comparisons of the risk of reporting symptoms of psychological ill health may be misleading.