A 4-year follow-up study of children in the Rural Child Care Project has shown that greatest losses in intellectual functioning for former project children occurred in the first year of public school among those scoring above 80 on their first Binet but who were retained in grade 1. IQ change during grade 1 was negatively related to attendance at a project center and unrelated to social services received by the family. IQ loss was inconsistently related to qualitative Binet performance over time. Parent awareness and use of community, school, and project resources were unrelated to length of project participation. Despite differences between project mothers, project teachers, and middle-class mothers (in intellectual functioning, educational achievement, reported educational attitudes, and control strategies), project mothers were equal to or more effective than project teachers in teaching their children to perform structured tasks. Although project teachers found parent training procedures difficult to implement, this project demonstrated the benefits of such procedures to both parents and staff. While children in the current program almost 2 years performed no better on UCLA Language Tests or Preschool Inventory than children with less than 1 year's experience, they were higher than summer Head Start children in general achievement. After discussion of these and additional findings, the document provides recommendations for improving the project within a "focussed services" model which stipulates specification of behavioral objectives and joint training of child development and homemaking staffs in the use of child-centered activities as a primary means of increasing family adequacy. (Author/MJB)