In 1981 major changes were made to the Aid to Families with Dependent Children (AFDC) program by the Omnibus Budget Reconciliation Act (OBRA). These changes redirected program resources to those welfare recipients with the least financial resources and thus affected the eligibility and benefit levels of many working welfare recipients. Numerous questions were raised about the effects of the changes—questions concerning the reasonableness of projected program savings, welfare recipients' work effort, and changes in the quality of life for families losing AFDC. This article describes an evaluation study, composed of multiple quasi-experimental research designs, that sought to address these questions. It emphasizes the kinds of issues faced by evaluators when designing research on complex and sensitive topics.
