<![CDATA[Reflective thinking and self-efficacy are two crucial constructs in education, particularly in mathematics education, and both have been reported to share a positive relationship across various contexts. However, empirical findings regarding the strength of this relationship remain inconsistent. This study presents a meta-analysis to investigate the relationship between reflective thinking and self-efficacy by addressing four research questions: (1) What is the overall strength of the relationship between reflective thinking and self-efficacy? (2) Is there significant heterogeneity among the studies? (3) Do the year of study, country, participants, educational level, and sample size serve as moderators? (4) How can the findings of this meta-analysis be integrated into mathematics learning practices? A total of 28 articles meeting the inclusion criteria were synthesized from the Scopus, ScienceDirect, and ERIC databases, following the PRISMA search protocol. The analysis involved calculating effect sizes, estimating a random-effects model, testing for heterogeneity, analyzing moderator variables, and examining publication bias. The results revealed a positive and statistically significant relationship between reflective thinking and self-efficacy (t(27) = 5.56, p < 0.001). The heterogeneity analysis indicated substantial variation among the included studies. Moderator analysis indicated that participant characteristics were a significant source of this variability, with pre-service mathematics teachers representing one of the key contributing groups. These findings highlight the importance of integrating reflective activities into mathematics learning in ways that are responsive to learner characteristics across educational levels. The results also underscore the need for future research to examine theory-driven pedagogical moderators to further clarify the mechanisms linking reflective thinking and self-efficacy.]]>
