Today, hundreds of studies on the technique known as Realistic Mathematics Education (RME) have been discovered. Numerous trials yielding contradicting results. The purpose of this meta-analysis is to ascertain the overall effect of RME and to examine the moderator variables in order to explore the consequences. The ERIC database, Sage publications, Springer publications, semantic scholars, and Google scholars were used to identify empirical data. The study examined 54 effect sizes from 38 individual studies conducted in the past two decades, involving 6140 participants. The estimation procedure was according to the random effects model, and statistical calculations were performed using the CMA program. According to the research findings, the study's overall effect size was 0.97. This suggests that implementing RME has a significant favourable influence on pupils' mathematical abilities. Moderator variables analysis shows that RME implementation would be more efficient when considering sample size, intervention length, learning mix, and education level. This finding contributes to the future implementation of RME to consider those variables that are shown to moderate the studies' effect sizes. Finally, comparisons of RME across countries and limitations of the study are discussed, providing vital information as a starting point for later studies and development of RME.