was particularly important for low SES children. Such children, placed in developmentally inappropriate kindergartens, experienced more stress (e.g., nail biting, fighting, tremors, feeling sick) and received lower grades in school than did their counterparts in developmentally appropriate programs. Another study (Holloway and Reichart-Erikson, 1988) found more positive interactions among children in child care classrooms rated as developmentally appropriate. In a more recent study of preschool programs for children from low-income families, those who attended developmentally appropriate programs as opposed to direct instruction classrooms were more successful academically as assessed by teacher grades of the extent to which they mastered basic skills in elementary school (Marcon, 1992, 1994). At least one study of preschool programs serving children from middle-class families (Hyson et al., 1990) found results similar to those for low-SES children. The middle-class children who attended developmentally appropriate programs did slightly better than those in highly academic ones on measures of academic skills and creativity. They also exhibited fewer anxious behaviors. In this study, Hyson and colleagues noted a relationship between parental beliefs and practices and the type of center their child attended.
In our review of program approaches, we found the following converging results across studies.
Both class size and staff-child ratio critically influence program quality and children’s learning and development. Class size in the model preschool programs that provide much of the research on positive outcomes for children tended to be low even compared with the recommendations of the National Association for the Education of Young Children. For example, the two best-known programs—the High/Scope Perry Preschool and the Abecedarian programs—had class sizes of 12 to 13 children with 2 teachers (Weikart et al., 1967; Ramey and Campbell, 1984). Small classes and better ratios enable teachers to provide more indi-