The growing demand for quality education in the globalization era requires students to develop strong mathematical conceptual understanding alongside essential 21st-century competencies. However, international assessments such as PISA indicate that students’ conceptual mastery in several ASEAN countries remains relatively low, partly due to instructional approaches that are less contextual and culturally relevant. In response, ethnomathematics-based learning has been proposed as an alternative model that integrates local cultural contexts into mathematics instruction to enhance meaningful understanding. This study aims to quantitatively synthesize empirical evidence regarding the effectiveness of ethnomathematics-based learning models in improving students’ mathematical conceptual understanding and to examine the influence of moderator variables. A meta-analysis was conducted on 26 eligible studies published between 2018 and 2025, selected through purposive sampling from the Google Scholar database. Effect sizes were calculated using Hedges’ g under a random-effects model, followed by outlier detection, heterogeneity testing, moderator analysis, and publication bias assessment. The results reveal a statistically significant and very high pooled effect size (g = 1.21, p < 0.0001) after outlier removal, indicating strong effectiveness compared to conventional instruction. Despite substantial heterogeneity, moderator analyses highlight variations across educational levels, mathematical topics, cultural contexts, and learning models. Overall, ethnomathematics-based learning demonstrates robust empirical support as an effective approach to strengthening students’ mathematical conceptual understanding.
Copyrights © 2026