<p>The recent decade has witnessed a growing body of studies on the relationship between boredom and academic achievement. While most studies reported negative boredom-achievement relations, the magnitude of this correlation remains inconclusive, varying across different studies. A few studies have attempted to synthesize the findings on this relationship and to explore potential moderators but they still drew inconsistent conclusions. To address the aforementioned issue, the present study attempts to conduct a three-level meta-analysis of 109 publications, including 147 independent studies with a total of 131,446 students, and yielding 259 effect sizes. This meta-analysis was conducted from December 2023 to November 2024. The results of the three-level meta-analysis showed a small to medium negative overall correlation between boredom and academic achievement (<i>r</i> = −.190, 95% confidence interval (CI) [−.207, −.172], <i>p</i> &lt; .001). In addition, univariate and multivariate moderator analyses revealed that the effect sizes were found to be stronger when (a) the samples were drawn from China; (b) the subject domain was language; (c) a domain-specific boredom scale (e.g., the Foreign Language Learning Boredom Scale) was used; (d) academic achievement was measured by course grades. In contrast, factors such as educational level, learning context, temporal specificity of boredom, the direction of relationship, publication type, publication year, and study quality did not demonstrate any significant and unique moderating effect. The results of the funnel plot, Egger’s regression test, and the fail-safe N test indicated the absence of publication bias. Based on the findings, some pedagogical implications and suggestions for future studies are provided.</p>