Effects of Single-Sex Schooling in High School: A Comparison of Analysis of Covariance and Propensity Score Methods

• In Event: National Databases, Science and Math Student Achievements, and a Methodological Assist

Abstract

In analyzing effects of single-sex schooling, researchers often have to rely on observational studies based on small non-representative samples. Here, we compare the effects of single-sex education in high school on Grade 12 and post-secondary outcomes using a subsample of the longitudinal High School and Beyond study (2379 students attending Catholic schools, 29 girls’ schools, 22 boys’ schools, 33 coeducational schools), the strongest dataset available. Focusing on achievement-related, motivational and social outcomes, we contrast ANCOVA and propensity score adjustment methods. When background and base-year covariates are controlled, uncorrected apparent differences between single-sex and coeducational schools disappear. The pattern of results is similar with both adjustment methods. Advantages of propensity score methods compared to typically used ANCOVA approaches are discussed.

Authors

• Benjamin Nagengast, Tuebingen University

• Herbert W. Marsh, University of Western Sydney

• Kit-Tai Hau, Chinese University of Hong Kong