| Sean Reardon
Pennsylvania State University
Understanding the growth of achievement inequality in the early years of schooling
FINAL REPORT:
This study explores the relationship between school characteristics and race/ethnic and SES achievement gaps through a detailed analysis of student achievement data from the Early Childhood Longitudinal Study (ECLS-K). Specifically, the study provides plausibly unbiased estimates of the causal effect of Catholic schooling on children's math, reading, and science achievement trajectories from kindergarten through fifth grade. In addition, the paper estimates the extent to which the effect of Catholic schooling varies across educational markets (e.g., public school districts) and across subpopulations (e.g., low-income students) within the U.S.
The results of this study indicate no positive effects of Catholic schooling during the K-3 years. This finding is significant for educational policy, since it suggests that policies that encourage private school enrollment, such as voucher policies, are unlikely to have a positive effect on students' achievement levels, at least in the short term. The possibility remains, however, that a voucher policy could induce substantial changes in the educational market that would lead to positive effects on achievement, but we have no evidence of the extent to which that is likely.
Another aspect of this research is the development of a multi-site matching estimator to enable researchers to use multi-site data (such as data from a number of districts or schools) to estimate both the average within-site effect of an instructional policy or practice and the variation in such effects across sites. The method combines features of standard matching estimators (e.g., propensity score matching) with multi-level modeling techniques.
Back to Funded Research Grants Page |