This study examines the effective differences between the Realistic Mathematics Education (RME) approach and conventional learning in improving mathematics outcomes for integer multiplication and division. A quasi-experimental design with a Post-test Control Group Design was employed, involving 32 third-grade students from SDN 4 Wameo Baubau, divided into experimental (RME) and control (conventional) groups. Data was collected through a validated 10-item essay test and analyzed using an Independent Sample T-Test with SPSS. Key findings: (1) The experimental group’s average score (89.38) was significantly higher than the control group’s (55.94); (2) Hypothesis testing yielded t-value = -9.223 and sig. 0.000 (p < 0.05), confirming RME’s superiority. These results advocate for RME as an alternative approach to elementary mathematics learning, particularly for context-oriented topics.