The projects were funded by the Department of Health, Education, and Welfare from 1974 through 1977 to test alternative strategies for dealing with child abuse and neglect. Data were collected on 1,724 clients who entered the projects' caseloads for treatment from January, 1975, through November, 1976. Information on client characteristics, services received, and outcomes were recorded. Impact measures included reincidence of abuse or neglect while in treatment and reduced propensity for future maltreatment by the time services are terminated. Results showed that seriousness of the assault that brought a case into treatment had a much stronger relationship with reincidence than type of treatment or service model. Clients who both physically abuse and neglect their children, emotional maltreators, and clients with difficult household indicators were less likely to improve on the functioning indicators used in the study. Clients in treatment for at least 6 months and clients receiving lay services were most likely to show improved functioning by the end of treatment. The lay service model had the strongest effect of the service models studied. The lay service model also had the greatest effect on propensity for future abuse or neglect. Given that about 30 percent of the clients had severe reincidence while in treatment, the initial intervention strategies of the projects must be questioned. Findings suggested that earlier intervention is needed and that child abuse and neglect programs currently cannot expect to have much more than a 40 percent to 50 percent success rate. Findings also showed that the most effective services tended to be those which were least expensive. Recommendations for future research, footnotes, tables, a list of 23 references, and extensive appendixes presenting data collection instruments, additional results, and a list of major evaluation report and papers are included. For related evaluations, see NCJ 73114-15, 73117-23, and 73090.