A Foster Care Research Agenda for the '90S

Abstract

In a recent review of the nation's foster care system, the General Accounting Office found that the population in out-of-home care grew from 276,000 in 1985 to 429,000 in 1991, or 55% in six short years. Other observers, like the Children's Defense Fund, have reviewed what trend data are available and have concluded that the foster care population in the United States will continue to grow [Children's Defense Fund 1992]. A recent study showed that foster care caseloads have more than doubled over the past ten years in five of the largest states [Goerge et al. 1994a]. Given the size of the population of children in foster care and the negative emotional and developmental impact that parent-child separation is thought to entail, policymakers, planners, and practitioners would be well-served by research that identifies those factors that account for the growing demand for foster care and those factors that minimize placement's impact on child and family well-being.This article poses a research agenda that would strengthen what we know about the role foster care plays in the lives of troubled families and the place child welfare occupies as part of the social safety net. We start with a recommendation that new research be shaped in a way that both builds upon past efforts and yet is contemporary in the concerns it addresses. We are mindful of the observation that child welfare research has tended to be noncumulative [Kadushin 1978]. Investigators have too often ignored the findings of previous research, thus adding to the fragmentation of the knowledge base. Nevertheless, Kadushin and Martin [1988] have documented much of the research in their chapters on family foster care and institutional services in their widely used textbook. In addition, Kamerman and Kahn's recent review of social services in the United States builds a broad, contemporary framework that provides the necessary social policy context for the development of child welfare services [Kamerman and Kahn 1990]. These works, if properly integrated, can help create the coherence that foster care research has missed.Of course, a coherent frame of reference is of limited value if the research questions posed by researchers are not themselves contemporary. The middle section of this article concentrates on two sets of issues that seem particularly relevant at this juncture. The first is the need for and availability of data. Since P.L. 96-272 was passed in 1980, critics of the child welfare system have often lamented the reality that administrative data systems have failed to yield reliable information about the experiences of children. In our view, these criticisms have been somewhat shortsighted. Moreover, the new Adoption and Foster Care Analysis and Reporting System (AFCARS) and Statewide Automated Child Welfare Information Systems (SACWIS) requirements should substantially increase the availability of program data. If researchers hope to have an impact on foster care and the development of child welfare services, these data must be used to their fullest advantage. To this end, we describe recent efforts to expand the research utility of administrative data systems. Second, we review research questions that seem to occupy a place close to the center of most child welfare debates. In some cases, past research offers a sound basis for follow-up work; in other cases, relatively new research themes have emerged as policy and practice have shifted to meet the demands of today's families.Finally, the last section of this article deals with the statistical methods used to study foster care. The discussion hinges on the observation that although most foster care research is concerned with fundamental questions about change in the life course of children and parents, very little attention has been given to how change is typically analyzed. Yet without closer attention to the measurement of change, it is unlikely that researchers will contribute all that they can to the process by which services to children and families are improved. …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 71 citations exist upstream.