Teacher homework involvement and student learning outcomes: A three-level meta-analysis

A three-level meta-analysis examined the association between teacher homework involvement and student learning outcomes and tested substantive and study-level moderators. Teacher homework involvement encompassed homework quality, autonomy support, content support, feedback quality, feedback quantity, and combined support, whereas learning outcomes encompassed motivational and affective, behavioral and self-regulatory, and academic outcomes. The analysis included 506 effect sizes from 29 independent studies involving 35,702 students. Teacher homework involvement had an overall small-to-moderate positive association with student learning outcomes ( r = 0.246, p < .001), which remained stable under study-clustered CR2 inference and leave-one-study-out analyses. This association varied by involvement dimension, outcome type, geographical region, publication type, and publication year, but not by grade level or subject matter. Omnibus differences across involvement dimensions and outcome types remained statistically significant under study-clustered inference. The most consistent evidence indicated stronger associations for homework quality, feedback quality, and autonomy support than for feedback quantity. Across outcomes, autonomous motivation showed the largest association, suggesting that teacher homework involvement may be particularly relevant to whether students experience homework as personally meaningful, valued, and self-endorsed. Associations were also stronger in studies conducted in Asia than in those conducted in the United States or Europe. Overall, the findings highlight the potential value of well-designed assignments, high-quality feedback, and autonomy-supportive guidance.