Meta-analysis reveals substantial gains in academic achievement among students using AI-supported education, highlighting adaptive tutoring as a particularly effective intervention.
This meta-analysis examined the effect of AI-supported education on academic achievement in Korea. A total of 24 experimental and quasi-experimental studies published in KCI-indexed journals from 2017 to February 2026 were analyzed. Hedges’ g was calculated from post-test means, standard deviations, and sample sizes, and effects were pooled using a random-effects model. The overall effect was positive and significant, Hedges’ g = 0.89, 95% CI [0.56, 1.22], p < .001. AI intervention type showed the clearest moderator pattern, with adaptive tutoring showing the largest point estimate. Education level, subject area, study design, and publication year were not significant moderators. The findings suggest a positive effect of AI-supported education, but should be interpreted cautiously because of high heterogeneity, small subgroup sizes, and possible publication bias.
No takes yet. Share an insight, caveat, or question.
KWAK et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: