Short- and Long-term Effects of Large-Scale Public Preschool Programs on Children’s Academic Achievement

• In Event: 3-044 - Evaluating Publicly-Funded Preschool Programs: Understanding Impacts, Impact Variation, and Lasting Effects

Integrative Statement

High quality early childhood education (ECE) programs have been found to produce large gains in children’s learning and development at school entry, particularly for disadvantaged children (Barnett, 2011). Long-term follow-ups have found that some ECE programs produce greater school success, higher earnings, and other positive life outcomes including decreases in crime and better health (Yoshikawa et al., 2013). Unfortunately, findings with respect to long-term outcomes have not been uniformly positive.

This study seeks to synthesize all available evidence regarding large scale public ECE program impacts on academic achievement over the last two decades. We examine the impacts on achievement from locally- and state-funded pre-K and federally-funded Head Start in the United States as well as international ECE programs for children who participated after 2000. To enable comparison across the studies we report standardized effect sizes (using standard deviation (SD) units) for each study. Tables 1 and 2 list the included studies and effect sizes.

Average effect sizes for immediate impacts (at kindergarten entry) of state and local pre-K are .29 SD for math, .24 SD for language, and .63 SD for literacy. Long-term average effect sizes of state and local pre-K programs are substantially lower: .11 SD for math and .11 SD for literacy; long-term Head Start effect sizes are zero for math and .04 for literacy. Interestingly, the average longer-term effects from international studies are .10 SD for math and .12 SD for literacy.

Several points are clear from the results. First, the vast majority of studies find that ECE programs improve cognitive abilities at entry to kindergarten in language, literacy, and mathematics. Second, these short-term effects vary considerably in size across programs. Third, estimated effect sizes vary by domain in the short-term, but not long-term where they are uniformly small. Immediate effects on literacy tend to be much larger than effects on language or math. Once into the primary grades there is little difference in magnitude between estimated impacts in literacy and in mathematics. Fourth, effect sizes tend to be much smaller shortly after entry to primary school. Fifth, some state and local pre-K programs in the United States have larger estimated effects immediately and over the long-term compared to Head Start, though this is not true of all state and local programs. Sixth, estimated effects in international studies are similar to those of the state and local programs.

Policy makers and the public should consider all the evidence when evaluating proposals to expand access to quality pre-K and other ECE programs. This review finds that large-scale public ECE programs can produce substantial gains in achievement, though these are somewhat smaller in the longer term than in the short term. However, it is encouraging that some more intensive programs are found to produce long-term effects on achievement that are remarkably close in size to those of the historical exemplars. Given the large differences in outcomes among programs, further and more rigorous investigation into the components of ECE programs that result in sustained long-term positive impacts seems warranted.

Authors

• W. Steven Barnett, NIEER/Rutgers University
  Presenting Author

• Kwanghee Jung, NIEER/Rutgers University
  Non-Presenting Author