Using Propensity Score Methods to Estimate Class Size Effects on Academic Achievement in Early Grades

• In Event: Critical Dimensions in Education: A Deeper Look at Teaching Quality, Class Sizes, and Cultural Practices

Abstract

This study examines causal class size effects on children’s academic achievement in reading, mathematics, and science in grades K-2. We used a recent large-scale national probability sample of kindergartners, namely ECLS-K 2011. Data analyses incorporated propensity score methods in kindergarten, first grade and second grade. Results indicate that small class effects were not significant overall across three subjects in early grades (i.e., K-2). Nevertheless, in kindergarten, small class was associated with lower mathematics achievement. In addition, teacher effects were not significant overall, and they were not consistent across grades and subjects. Changing schools had a statistically significant and negative effect on children’s achievement in reading and mathematics in kindergarten.

Authors

• Ting Shen, Missouri University of Science & Technology

• Spyros Konstantopoulos, Michigan State University