This research was conducted with the purpose of meta-analysis of the effectiveness of game-based interventions on improving and promoting mathlearning among Iranian students. The design was based on meta-analysis and socio-statistical methodology, including all studies and researches conducted in the field of game effectiveness on mathlearning. In the stage of collecting research documents, 1248articles and 80 these were collected based on input and output criteria, and after screening, the effectiveness of‌48game-based intervention methods was examined inthe form of 39research documents. In order to calculate the effect size, CMA2 software was used. Examining the assumptions based on the homogeneity/heterogeneity of the effect size and publication bias was indicative of the heterogeneity of the effect size and the lack of publication bias. The average random effect size of game-based interventions on math learning was equal to 0.589, which was within the confidence range, indicating a strong and significant effect of different forms of games on math learning among students. Moderation analysis showed that the effect of games on math learning among second grade students and students with learning disabilities is greater than others and that computer, group and physical/movement or behavioral games have a greater effect on math learning. The more the number of math-oriented game sessions is increased, the more intense math learning is. Different forms and forms of games, especially among students with learning disorders, have an effective role and a considerable function in promoting learning and academic progress in mathematics, and improve the performance and calculation skills of students