This meta-analysis study synthesizing the results of experimental and quasi experimental studies examined the effects of self-assessment interventions on student academic performance from primary education to higher education. A total of 16 studies with 46 effect sizes involving more than 7,650 participants were included in the analysis. Research synthesis showed that an overall small influence of self-assessment interventions on academic performance (g=0.37, p<0.05). Additionally, moderator analysis was used to examine moderating effects of some variables. The analysis indicated that traditional self-assessment interventions without external feedback have significantly a larger effect (g =0.47, p<0.05) than self-assessment with external feedback (g=0.28, p<0.05) on academic performance. However, effectiveness of other moderating variables (e.g. education level, assessment criteria type, self-assessment training) on academic performance were not statistically significant. The results suggest that further empirical studies are needed to reveal the moderating effects of self-assessment.