The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 transformed the focus of welfare policy in the United States. Instead of the entitlement to cash assistance embodied in the Aid to Families with Dependent Children, Temporary Assistance to Needy Families (TANF) encourages economic self-sufficiency through employment. After five years, evidence is accumulating as to the policy's successes and failures. A key feature of this research involved the employment outcomes and well-being of welfare leavers, most of them single women and children who exit from welfare programs. Such dramatic policy changes provide evaluators with expansive opportunities for policy-relevant research, but they also raise questions about the role of evaluation in informing policy and about the information most valuable for this purpose. A key challenge facing evaluators is that stakeholders often have diverse needs and expectations. As a result, policy is often best informed by an array of evaluations that use different methods. Recognizing the difficulties involved, this issue contributes to the welfare policy debate by reporting on a multisite project examining families who leave welfare. The studies used multiple sources of data, employed multiple approaches to analysis, and included multiple value-based outcomes. The goals of this issue are to use this planned mixture of multiple methods to provide a richer understanding of life after TANF and to illustrate how to incorporate multiple or mixed methods in the social policy arena.