We assessed the quality of center child care relationships with adults and peers for 414 children (ages 14 to 54 months). Classrooms were classified by ratio and group size provisions of the Federal Interagency Day Care Requirements (FIDCR) and by the Early Childhood and Infant and Toddler Environmental Rating Scales. Children cared for in classrooms meeting FIDCR ratios were more likely to be in classrooms rated as good or very good in caregiving and activities. Children in classrooms rated as good or very good in caregiving were more likely to be securely attached to teachers. Securely attached children were more competent with peers. Children cared for in classrooms meeting FIDCR group size were more likely to be in classrooms rated higher in activities. Children in classrooms rated high in activities were likely to orient to both adults and peers. Children with social orientations to adults and peers were more competent with peers.

Over half of child care center teaching staff (n=149) and a third of directors (n=71) interviewed in 1996 had left their centers by 2000. The demographic and professional profiles of those who left and stayed at their centers were similar. Among those who left, only half continued to work in child care. Highly-trained teaching staff were more likely to leave their jobs if they earned lower wages, worked in a climate with less stability of highly-trained co-workers, and worked with a greater percentage of teaching staff who did not have a bachelor's degree. Directors were more likely to leave if they earned lower wages. The study extends previous research by revealing the links among the characteristics and stability of the teaching staff as a whole and the retention of highlytrained teachers. It also underscores the multi-faceted benefits resulting from paying higher wages to all staff.

We assessed the quality of center child care relationships with adults and peers for 414 children (ages 14 to 54 months). Classrooms were classified by ratio and group size provisions of the Federal Interagency Day Care Requirements (FIDCR) and by the Early Childhood and Infant and Toddler Environmental Rating Scales. Children cared for in classrooms meeting FIDCR ratios were more likely to be in classrooms rated as good or very good in caregiving and activities. Children in classrooms rated as good or very good in caregiving were more likely to be securely attached to teachers. Securely attached children were more competent with peers. Children cared for in classrooms meeting FIDCR group size were more likely to be in classrooms rated higher in activities. Children in classrooms rated high in activities were likely to orient to both adults and peers. Children with social orientations to adults and peers were more competent with peers.

Data from a nationally representative survey of child care centers and a 5-site, observational study of centers were used to examine the quality of care provided to children from low-income families. Comparisons were made to a national sample of centers; among Head Start, public school-sponsored, and other community-based subsidized centers; and among centers that served families from differing socioeconomic groups. The quality of care in centers that served predominantly low-income children was adequate, but highly variable, with structural indices exhibiting higher quality than observations of global quality and of staff-child interactions. When compared to Head Start and public school-sponsored centers, the community-based centers had smaller groups and fewer children per teacher for preschoolers, but also had less well educated and compensated staff. Centers that predominantly served children from upper-income families provided the highest quality of care across multiple indices, and those that predominantly served children from middle-income families almost uniformly provided the poorest quality of care. The centers that served children from low-income families did not differ significantly in quality from the upper-income centers on most indices. However, the teachers in these programs were observed to be less sensitive and more harsh than teachers in the centers that served more advantaged families. The implications of the findings for research and policy are discussed.

Data from a nationally representative survey of child care centers and a 5‐site, observational study of centers were used to examine the quality of care provided to children from low‐income families. Comparisons were made to a national sample of centers; among Head Start, public school‐sponsored, and other community‐based subsidized centers; and among centers that served families from differing socioeconomic groups. The quality of care in centers that served predominantly low‐income children was adequate, but highly variable, with structural indices exhibiting higher quality than observations of global quality and of staff‐child interactions. When compared to Head Start and public school‐sponsored centers, the community‐based centers had smaller groups and fewer children per teacher for preschoolers, but also had less well educated and compensated staff. Centers that predominantly served children from upper‐income families provided the highest quality of care across multiple indices, and those that predominantly served children from middle‐income families almost uniformly provided the poorest quality of care. The centers that served children from low‐income families did not differ significantly in quality from the upper‐income centers on most indices. However, the teachers in these programs were observed to be less sensitive and more harsh than teachers in the centers that served more advantaged families. The implications of the findings for research and policy are discussed.