MIXED METHODS FOR POLICY AND PRACTICE 455 “If You Work in This Country You Should Not be Poor, and Your Kids Should be Doing Better”: Bringing Mixed Methods and Theory in Psychological Anthropology to Improve Research in Policy and Practice Thomas S. Weisner Abstract New Hope (NH) was a successful poverty reduction program that offered a positive social contract to working-poor adults. If you worked full time, you were eligible to receive income supplements, childcare vouchers, health care benefits, a community service job, and client respect. NH did reduce poverty and increase income and earnings for some participants, and improved outcomes for some children. But in spite of relatively generous benefits, NH was only selectively effective. Only those not working when NH began and those with few barriers to work were positively affected by the program through achieving more work hours, poverty reduc- tion, and income gains. Boys in program families benefited, girls did not. Take-up of NH benefits was typically partial and episodic; for instance, some parents would not use childcare programs for young children. Ethno- graphic evidence was essential for understanding these sometimes-surprising program impacts and their policy and practice implications, and was effectively combined with an experimental, random-assignment research design. Psychological anthropology can bring its traditions of integrating qualitative and quantitative methods and its focus on experience, context, and meaning to understanding and improving policies and practices within a scientific frame of the committed, fair witness. [mixed methods, policy and practice, family, poverty, adolescence] There is wide support in the United States for helping children and parents in poor families, but there is much less consensus on how to do that, and who should be given support. In this article, I describe the New Hope (NH) work-support program to reduce poverty. A wide range of qualitative and quantitative evidence was collected by the NH study team. Parents’ cultural models of work and parenting goals and practices were important for understanding responses to the program, as has been shown in other psychological anthropological studies of poverty (Strauss 1992, 2002). The NH study followed a group of children ages one to ten for eight years, from 1995 to 2004, during and after the NH program itself was in operation (1995–98). Children were nine to 19 at the final eight-year assessment. Some NH program impacts persisted into the adolescent years—an important result given the repeated finding that generally it is difficult to sustain positive effects of support programs. Yet NH was selectively successful: it assisted some kinds of parents and children, and not others, and those it assisted were helped in different ways. The program benefits of NH ETHOS, Vol. 39, Issue 4, pp. 455–476, ISSN 0091-2131 online ISSN 1548-1352. C  2011 by the American Anthropological Association. All rights reserved. DOI: 10.1111/j.1548-1352.2011.01208.x Journal of the Society for Psychological Anthropology