Abstarct: ABSTRACT The objective of this mexta-analysis is to combine and analyze the results of 22 primary studies that examined the impact of dynamic assessment on language skills. The analysis is guided by three main questions: What is the main effect of dynamic assessment on language skills? What variables moderate the effectiveness of dynamic assessment? To what extent the overall effect is conditioned by publication bias? Analysis revealed a substantial overall effect size of g=1.12, which indicates a large effect size according to Plonsky and Oswald's (2014) scale. baxsed on the results of the moderator analysis, it was found that dynamic assessment has the greatest impact on students' listening skills and intermediate levels of education. The Fail-safe N test of publication bias suggests that the positive results of the dynamic assessment cannot be attributed to publication bias, providing evidence of unbiased publication. The findings may have clear implications for policymakers and language teachers.