Results provide empirical support for the use of these types of comparison groups in domestic violence research. The data indicate that shelter-living abused (SLA) women differ from community-living abused (CLA) women in some ways but not in others, and that relationship distress does not account for differences found. SLA women evidenced greater frequency and severity of recent abuse than CLA women. Their past histories of abuse were similar. This difference in recent severity of abuse may speak to their current decision not to access shelter services. Both CLA and SLA women showed high levels of depression and anxiety with SLA women reporting lower self-esteem. Overall, what distinguished SLA from CLA women was recent abuse severity and endorsement of aggression as a family problem-solving technique. The findings point to important differences between these groups underscoring the usefulness of including appropriate comparison groups. Intimate partner violence is a disturbing social problem. Abused women show high rates of depression and anxiety. However, a significant weakness of the literature is the lack of appropriate groups. In this study, shelter-living, abused women were compared to community-living, abused women, martially distressed women, and non-distressed non-abused women, on relationship characteristics, emotional status, and family problem-solving and coping. Tables, references