Game-Based English-Medium Mathematics Instruction for Bilingual Elementary Learners: A Pre-Experimental Study

Abstract

Teaching mathematics through English can expand elementary learners’ academic language exposure, but it may also increase linguistic demands and reduce participation when instruction relies heavily on explanation and worksheets. This study examined whether game-based English-medium instruction improved the mathematics achievement of bilingual elementary students. A quantitative pre-experimental one-group pretest–posttest design was used with 30 fifth-grade students at a bilingual elementary school in South Sumatra, Indonesia. The intervention consisted of six 70-minute lessons on fractions delivered primarily in English through a teacher-designed game sequence, Fraction Mission, which combined board movement, visual fraction models, collaborative challenge cards, points, immediate feedback, and limited Indonesian scaffolding. Mathematics achievement was measured using a 25-item test reviewed by content and language experts. The illustrative analysis showed that the posttest mean (M = 74.33, SD = 11.41) was higher than the pretest mean (M = 62.37, SD = 8.37). A paired-samples t-test indicated a statistically significant difference, t(29) = 7.36, p < .001, with a large within-participant effect (dz = 1.34). The findings suggest that carefully scaffolded games may help bilingual elementary learners process English mathematical language while repeatedly applying concepts in meaningful and low-risk interactions. However, the one-group design does not establish causality. Future research should use comparison groups, delayed tests, and separate measures of mathematics and language development.