Four recent in-depth studies found new evidence documenting the relationship between domestic violence and welfare. The studies range in methodological rigor, but reviewed together, they can assist in a better understanding of the role domestic violence plays in a poor women's ability to become economically self-sufficient. The Passaic County Study of AFDC (Aid to Families with Dependent Children) Recipients in a Welfare-to-Work Program was a research effort conducted by the Passaic County Board of Social Services on a sample of 846 women on AFDC in one northern New Jersey county participating in a mandatory pre-employment assessment and training program between 1995 and 1997. In Harm's Way?, Domestic Violence, AFDC Receipt and Welfare Reform in Massachusetts reported on a random sample of 734 women in the Massachusetts welfare caseload, surveyed between January and June 1996. This study sampled the State's entire AFDC caseload for both current and past prevalence of domestic violence. The Worcester Family Research Project was a 5 year study of 436 women, most welfare recipients and homeless, conducted between 1992 and 1995. The Effects of Violence on Women's Employment was a random survey of 824 women in one low-income neighborhood of Chicago, Illinois between 1994 and 1995. The four studies found large and consistently high percentages of women on AFDC were currently abused by partners. It also documented that the majority of women on welfare were past victims of domestic violence. Key findings included: (1) high percentages of abused women reported arguments about child support, visitation, child custody, police visits to their homes, and interference from intimate partners with education, training, and work, and (2) many women did not suffer from physical and mental health problems, but a significant percentage did. Women on welfare and abused suffered from depression, other mental health problems, posttraumatic stress disorder, drug and alcohol abuse, and physical health problems at higher rates than non-abused counterparts and the general female population. The new data supported the hypothesis that many women on welfare who do not comply with work or training requirements while receiving assistance would be prevented from doing so by the direct behavior of an abusive partner, or by the indirect effects of the abuse on their health and well being. The studies clearly demonstrated that domestic violence is a factor in many welfare recipients' lives. The following recommendations were made regarding policy implications: identify each battered woman's needs, provide time to use domestic violence services, revise paternity and child support policies, envision new programs for battered women, and involve men.