A meta-analysis of the combined effect of self-efficacy and students' mathematics interest on mathematics achievement

Purpose This study employed a meta-analysis design to systematically synthesize empirical evidence on the effect of students' mathematics interest and self-efficacy on mathematics achievement. Design/methodology/approach This study employed a meta-analysis design. Drawing on 30 peer-reviewed studies published from 2017 to 2025, the research applied strict inclusion criteria to identify relevant quantitative studies across global educational contexts. Data were extracted using a structured coding sheet, and statistical analyses were conducted in JASP, incorporating heterogeneity tests, forest plots, and publication bias assessments. Moreover, the study drew heavily on secondary data. Findings The findings revealed a moderate, statistically significant positive combined effect size for both students' mathematics interest and self-efficacy on students' mathematics achievement, with high heterogeneity (I2 > 80%) accounted for using a random-effects model. The Rosenthal Fail-Safe N test further confirmed the robustness and validity of the findings, indicating no publication bias. These results highlight the crucial role of motivational constructs, specifically interest and self-efficacy, in enhancing mathematics achievement and offer evidence-based guidance for educators and policymakers aiming to improve mathematics learning outcomes. Originality/value Most studies have employed structural equation modeling (SEM) frameworks to assess the directional and mediating effects of self-efficacy and students' mathematics interest on mathematics achievement within a unified model. Even fewer studies have drawn findings from a large pool of empirical studies through meta-analytic procedures.