This article discusses the relationship between domestic violence and employment in a post-welfare reform population.
Prior research has indicated a high prevalence of domestic violence in the Aid to Families With Dependent Children population and highlighted the issues this raised for welfare reform. One study found that domestic violence among welfare recipients was associated with a general pattern of reduced stability of employment. Few welfare agencies have undertaken efforts to identify and provide services to women with domestic violence issues that might be impeding their ability to become and stay employed. A randomly selected group of Temporary Assistance to Needy Families (TANF) recipients from two central valley California counties was interviewed three times: at baseline (Round 1); 1 year later, after welfare-to-work requirements were applied (Round 2); and again 15 months later (Round 3). Measures were domestic violence, employment, varying time periods, and information about receipt of domestic violence services. The results show that the effects of domestic violence on work varied depending on the type of domestic violence. The best predictor of a lower probability of working 32 or more hours a week at the time of the Round 2 and Round 3 interviews was the estimated need for domestic violence services, which was designed to encompass factors hypothesized most likely to affect employment. Estimated need adds to the usual objective indicators of serious abuse, which are posttraumatic stress disorder as a result of adult abuse, and seeking help for domestic violence issues even if one does not meet the criteria for serious abuse. More than half of the women met the criteria for this level of abuse at some point during the 3 years of the study. Estimated need for services was associated with several concurrent and longitudinal measures of employment, but not consistently. The estimated need for domestic violence services was significant as a main effect in the 2 years after baseline, reducing the predicted probability of working 32 or more hours. 1 figure, 8 tables, 23 references