<![CDATA[Mathematics learning outcomes remain a challenge across various educational levels, highlighting the need for instructional approaches that enhance student engagement and understanding. One of the widely implemented approaches is Game Based Learning (GBL). This study aimed to examine the overall effect of GBL on students’ mathematics learning outcomes and to identify differences in effect sizes based on game type, sample size, and educational level through a meta-analysis approach. The study employed a meta-analysis method following the PRISMA guidelines. A total of 25 primary studies published between 2019 and 2026 that met the inclusion criteria were analyzed using JASP software. Effect sizes were calculated using Hedges’ g and synthesized through a random-effects model. The results revealed that GBL had a very large positive effect on mathematics learning outcomes, with a pooled effect size of 2.184 (95% CI [1.101, 3.268]; p < 0.001). Heterogeneity analysis indicated substantial variation among studies (Q = 1129.72; I² = 98.89%), leading to moderator analyses. The findings showed that game type (F = 0.1125; p = 0.740) and sample size (F = 0.073; p = 0.930) were not significant moderators of effect size variation. In contrast, educational level was found to be a significant moderator (F = 4.563; p = 0.009), with the largest effect size observed at the higher education level (g = 10.630). Publication bias analyses using Funnel Plot, Fail-Safe N (9076), and Trim and Fill indicated that the meta-analytic findings were highly robust and minimally affected by publication bias. These findings suggest that Game Based Learning is an effective instructional approach for improving students’ mathematics learning outcomes across different educational levels.]]>
Copyrights © 2026
