Multivariate meta-analysis of self-regulated learning and academic performance in higher education: Moderators, mediators, and methodological insights

Self-regulated learning Academic performance Higher education Meta-analysis Educational interventions Moderator analysis

Authors

Downloads

This meta-analysis examines the relationship between self-regulated learning (SRL) and academic performance in higher education, quantifying the overall effect size and exploring factors contributing to variability. A systematic review using Google Scholar and Scopus identified 62 studies (14 articles, N=6,991 participants) published between 2014 and 2024, involving undergraduate, medical, and EFL learners. Using a random-effects model, the analysis revealed a pooled moderate positive effect of SRL (Zr=0.239, 95% CI: 0.204–0.274, p<0.001) on academic performance. Subgroup analyses revealed higher effect sizes among medical students (Zr=0.326) compared to undergraduate (Zr=0.228) and EFL learners (Zr=0.284). Path analysis studies yielded larger effect sizes (Zr=0.312) compared to correlational designs (Zr=0.239), highlighting SRL’s mediating role. No significant publication bias was detected (p=0.484, Egger’s test). Practically, these findings suggest educators should design interventions tailored to academic contexts, incorporating explicit goal-setting, structured self-monitoring, and reflective practices. Future research should address methodological gaps by adopting longitudinal designs, diversifying samples, and standardizing SRL measurement frameworks.