Deciphering the layers of reading achievement: A multi‐level analysis of student and school‐level predictors in top‐performing PIRLS 2021 countries

Abstract

This study investigates the student‐ and school‐level predictors of reading performance across five top‐performing countries in the Progress in International Reading Literacy Study (PIRLS) 2021. Utilising a multilevel analysis, the research examines the influence of cognitive development, socio‐cultural contexts and educational environments on reading achievement. The results reveal significant variances both between and within schools, highlighting the critical role of educational policies and practices in shaping student outcomes. Finland, known for its equity‐focused education system, demonstrates low between‐school variance, while Russia shows high between‐school variance owing to socio‐economic disparities. Key student‐level predictors include gender, digital self‐efficacy and a sense of school belonging, varying in significance across countries. At the school level, factors such as the number of computers, principal experience and academic emphasis significantly influence reading scores. The full model, incorporating both student‐ and school‐level predictors, explains a substantial portion of the variance in reading scores, especially in Hong Kong and Russia. These findings underscore the importance of a comprehensive approach to educational research, considering both individual and institutional factors. Policy implications suggest targeted interventions for countries with high between‐school variances and support for inclusive learning environments where within‐school variances are significant. Future research should explore the impact of non‐cognitive factors and emerging educational practices on reading performance to develop more effective educational strategies globally.