Abstract This paper analyses differences in academic achievement associated with attendance at publicly funded private schools and public schools across 11 European countries. Using eight waves of PISA data (2000–2022), we apply OLS, IPW and IPWRA estimators to account for observed heterogeneity and mitigate selection bias. The results reveal pronounced cross‐country variation. Publicly funded private schools are associated with higher achievement in a limited set of countries—most notably Belgium, Spain and Ireland—while associations are either not systematic or remain modest in countries such as Denmark, the United Kingdom or Sweden and negative in Luxembourg. To interpret this heterogeneity, the paper develops a comparative institutional typology of publicly funded private schooling systems, distinguishing configurations according to governance arrangements, regulatory constraints and degrees of school autonomy. Additional analyses show that associations tend to be stronger among students from middle‐ and high‐socio‐economic backgrounds, though there are important cross‐national differences. A broad set of robustness checks supports the stability of the results, and a decomposition offers further insight. We also explore how the COVID‐19 pandemic affected learning outcomes across school sectors, revealing substantial cross‐country heterogeneity and no uniform evidence that publicly funded private schools experienced smaller learning losses than public schools. Overall, the results indicate that sector performance follows institutional configurations rather than ownership categories, highlighting the contextual nature of publicly funded private schooling across Europe.
Building similarity graph...
Analyzing shared references across papers
Loading...
Priya Maurya
Léonard Moulin
Denis Fougère
British Educational Research Journal
Centre National de la Recherche Scientifique
Université Paris-Est Créteil
Center for Economic and Policy Research
Building similarity graph...
Analyzing shared references across papers
Loading...
Maurya et al. (Mon,) studied this question.
www.synapsesocial.com/papers/69df2c50e4eeef8a2a6b15e9 — DOI: https://doi.org/10.1002/berj.70161