This report presents an evaluation of family preservation programs in Kentucky, New Jersey, and Tennessee, and a county program in Philadelphia County, Pennsylvania. Key goals of these programs were reducing foster care placement, maintaining child safety, and improving family functioning. The statewide programs employed the Homebuilders program model, and the county program used a broader, home-based service model. Volume 1 of the report describes the study implementation, the study sites, and families in the Homebuilders model sites. Volume 2, which includes an executive summary, examines services and outcomes for both types of programs, analyzes attrition, and presents conclusions. In the evaluation, families were randomly assigned to a family preservation program or to regular service of the child welfare system. Information on parenting practices, family functioning, child well-being, and caseworker-caregiver interaction was collected through interviews. Information on family functioning was assessed at the beginning of services, at the close of Homebuilder services, and 1 year after services began. Administrative data provided information on childrens placements, reentries, and subsequent abuse/neglect allegations up to 18 months after experiment entry. Staff attitudes and characteristics were collected through questionnaires. Discussions with personnel of provider agencies provided information about services, policies, staffing, training, and the service context. Findings revealed that the experimental group received more services and more intensive services than the control group in all four states. There were no significant group differences on family-level placement rates, proportion of time in substitute care, case closings, or subsequent maltreatment. There were a few child and family functioning items in which the experimental group displayed better outcomes than the control group at the end of Homebuilders services, but these results did not occur in more than one state. There were very few differences at the 1-year follow-up. The report's discussion focuses on implications and highlights the need to rethink the functions, target group, and characteristics of services and to examine the issues of program specialization, length, and intensity. Each report section contains references or endnotes. Volume 3 of the report comprises 11 appendices, which include the screening protocol, worker safety checklist, secondary analyses for chapters in volume 2, and study instruments.(KB)