programs for which longitudinal studies were conducted to determine program effect appears in Table 4-1, along with program features.

We can conclude with confidence that early intervention programs can produce modest to large effects (effect sizes of 0.2 to over 1.0 standard deviation) on children’s cognitive and social development. Larger effect sizes are associated with better subsequent performance in school, particularly when the schools are of good quality (Campbell and Ramey, 1994, 1995; Lazar et al., 1982). Variation in effect size and duration is associated with the particular program features reviewed below (Ramey and Ramey, 1992).

Developmental Timing

Interventions that begin early and continue for a longer duration result in greater benefits to participants. The five major studies that demonstrated some of the largest effects of early intervention on cognitive and social development all enrolled children during infancy (the Abecedarian Project, the Brookline Early Education Project, the Milwaukee Project, Project CARE, and the Infant Health and Development Program). Data on early cognitive development for program and control children in the Abecedarian Project displayed in Figure 4-1 suggest that, without intervention, high-risk children fall substantially behind as early as the second year of life. Since no experimental design has tested for a critical period or threshold effect, however, no precise timing for intervention can be supported empirically.

Program Intensity

Programs that provide more hours of service delivery produce larger positive effects than do less intensive interventions. Within programs, children and parents who participate the most actively and regularly are the ones who show the largest developmental gains. All of the programs mentioned as effective with regard to timing also provided intensive intervention services. In addition to these, the Perry Preschool Project (Weikart et al., 1978) and the Early Training Project (Gray et al., 1982), both of which began when children were 3 or 4 years old, also provided intensive services and registered substantial program impact. Numerous examples of early intervention programs that had little or no effect on cognitive, social, or later academic performance were less intensive.1


The Utah State Early Intervention Research Institute (White, 1991), for example, found no significant effects in the 16 randomized trails of early interventions for children with developmental disabilities. None of these interventions provided full-day programs or multiple home visits per week. Similarly, a brief prenatal and postnatal program for urban teen mothers failed to affect their children’s cognitive performance or social development (Brooks-Gunn and Furstenberg, 1987). For other examples, see Ramey and Ramey (1998).

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement