Background StudySeveral meta‐analysis studies have investigated the effects of mobile learning on learning performance. However, limited attention has been paid to pedagogy in mobile learning, making quantitative evidence of the effects of pedagogical approaches on learning performance in mobile learning scarce. Filling this gap can therefore help stakeholders understand which mobile pedagogical approaches might work or not under which learning conditions, hence achieving better learning experience and performance.ObjectivesTo address this gap, this study conducted a meta‐analysis and research synthesis of the effects of integrated pedagogical approaches on students' learning performance in mobile learning. Additionally, this study analysed the field of education, level of education, learning setting, sample size, and mobile device as moderating variables of the effect of pedagogical approaches.MethodsThe software Comprehensive Meta‐Analysis V.3 was used for this meta‐analysis, where Hedges' g was calculated for the effect sizes. Specifically, 70 quantitative studies (N = 5575 participants) were coded and analysed.ResultsThe results indicate that pedagogical approaches in mobile learning have a large effect on students' learning performance (g = 0.93, p < 0.001). The most effective pedagogical approach was project‐based learning (huge effect), while collaborative learning, situated learning and game‐based learning had large effects. Finally, cognitive theory of multimedia learning and inquiry‐based learning had medium effects. The results also indicate that the effect is moderated by the field of education, the level of education, the learning setting, the sample size, and the mobile device. Finally, it is found that study quality might influence the overall effect size.ConclusionsThe findings of this study can be beneficial to both researchers and practitioners as they highlight and discuss which pedagogical approaches could be more effective in mobile learning under specific conditions. Further research, on the other hand, should focus on covering additional moderators of learning performance like mobile device type and screen size which is one of the limitations of this study.