The Effect of Realistic Mathematics Education on Students' Numeracy in Indonesia: A Meta-Analysis Study

Abstract

Numeracy is a fundamental competency in mathematics education, yet Indonesian students consistently underperform on international and national assessments. Realistic Mathematics Education (RME), grounded in Freudenthal's didactic theory, has been widely adopted in Indonesian classrooms to address this gap. However, the magnitude and consistency of RME's effect on numeracy outcomes remain unclear. This meta-analysis quantitatively synthesizes evidence on the effect of RME on students' numeracy in Indonesia and examines potential moderating variables. A systematic literature search was conducted across Google Scholar, Semantic Scholar, Crossref, and Scopus (2015-2026), yielding 2,338 initial records. After PRISMA 2020 screening, 32 quasi-experimental studies were included. Effect sizes were pooled using the DerSimonian-Laird random-effects model. Heterogeneity was assessed via Cochran's Q, I², and τ². Moderator analyses were conducted for education level, class size, and learning combination. The pooled effect size was g = 1.469, p < .001, classified as a large effect by Cohen's benchmarks. Substantial heterogeneity was detected. Subgroup analyses revealed that primary school students and junior high school students benefited more than senior high school students. Larger class sizes showed greater effect sizes than smaller classes. Standalone RME yielded the largest effect, while RME integrated with technology yielded the smallest. The funnel plot showed no severe publication bias. RME exerts a large and statistically significant effect on students' numeracy across educational levels in Indonesia. These findings provide robust empirical evidence supporting the broader implementation of RME as a core pedagogical approach to improve students’ numeracy outcomes, particularly at the primary and junior high school levels.