Student Scores in Public and Private Schools: Evidence from PISA 2009

Abstract

This paper examines critically the presumption that, other things equal, private schooling offers higher quality education than public schooling. We apply multilevel regression on the 2009 PISA to estimate the differential effect of public and private schooling on student scores in Australia. We control for observable and unobservable influences, at school and student levels. We find that public-private schooling quality estimated differences are not statistically significant, but Catholic schools perform better than both. Differences by sector in the level of resourcing, plays a minor role. Student socioeconomic status differences and resulting selection, drive the observed better private schooling scores outcomes.

Keywords: PISA, government schools, school quality, multilevel modelling

JEL Classification: I24, I21, I28