Heterogeneity in the Effects of Formative Assessment on Self-Regulated Learning: A Meta-Analysis

• In Event: Cognition, Emotion, Motivation, and Creativity: Evidence for Holistic Learning Design

Abstract

This meta-analysis investigates the effects of formative assessment (FA) on students’ self-regulated learning (SRL), focusing on when, for whom, and under what condition FA works. We synthesized 999 effect sizes from 138 studies involving over 23,000 students. While FA significantly improves SRL, substantial heterogeneity was observed. To explain this variation, we conducted confirmatory moderator analyses using correlated and hierarchical effects meta-regression, followed by exploratory multimodel inference. Two moderators—type of outcome measure and number of feedback sources—emerged as consistently significant and robust across sensitivity analyses. Descriptive patterns across other moderators suggest consistent trends. The results can inform policymakers, professional development designers, and educators on what works and what does not, for whom, and under what conditions when implementing FA.

Authors

• Elie Chingyen Yu, University at Albany - SUNY

• Heidi L. Andrade, University at Albany - SUNY

• YangHyun Kim, City University of New York

• Wim Van den Noortgate, KU Leuven

• Mariola Moeyaert, University at Albany - SUNY

• Angela M. Lui, CUNY - School of Professional Studies

• Diana Akhmedjanova, National Research University "Higher School of Economics"

• Dylan R. Wiliam, UCL Institute of Education