Effective implementation of the Sport Education Model in physical education: A meta-analysis of participant and intervention characteristics

PLOS One | https://doi.org/10.1371/journal.pone.0331228 October 16, 2025 9 / 17 [8,21,35] conducted >2 lessons per week, with one study [7] not reporting the frequency of intervention. For the dura- tion of each session, no study reported ≤30 minutes per session, 8 studies [7,8,21,35,39,41,42,44] reported 30–60 minutes per session, and 7 studies [20,36–38,40,43,45] reported ≥60 minutes per session. In terms of the experimental period, 12 studies [7,8,20,21,35,36,38,40–45] involved <18 sessions, while 3 studies [35,37,39] included ≥18 sessions. The learning outcomes assessed included cognitive indicators (such as sports ability and physical activity) in 6 studies [8,20,37,39,40,42] and non-cognitive indicators (such as motivation) in 10 studies [7,20,21,35,36,38,41,43–45]. Synthesis of the results The overall implementation effect of the Sport Education Model in physical education teaching. The assessment of the overall implementation effects of the SEM in physical education involved all 15 included studies, comprising 16 experimental groups and 15 control groups. Some studies measured multiple cognitive and non-cognitive indicators, totaling 49 (11 cognitive indicators and 38 non-cognitive indicators) (See S4 Appendix. Date used for meta- analysis for details). The heterogeneity test results indicated Q = 2200.62, P < 0.001, I² = 98%, suggesting a high degree of heterogeneity among the included studies. Therefore, a random-effects model was adopted for the evaluation. The effect

size serves as an indicator to assess the strength and relevance of the experimental effects. The results of the random- effects model showed an effect size of 0.590 (P < 0.001), indicating a moderate positive effect of SEM on students’ physical education learning (detail see Table 3). The details of all analysis results can be found in S5 Appendix: the detailed results of the meta-analysis of the overall and different moderating variables effects. Effects of the Sport Education Model on cognitive and non-cognitive dimensions in students. Table 4 shows that SEM had a small positive effect on students’ learning outcomes in the cognitive dimension (Effect size = 0.411, P < 0.042). In the non-cognitive dimension, SEM demonstrated a moderate positive effect on students’ learning outcomes (SMD = 0.641, P < 0.01). Both effects reached statistical significance. Effects of different moderating variables of the Sport Education Model on student PE learning outcomes. To explore the impact of different moderating variables on students’ physical education learning, we conducted statistical analyses on various Participant characteristics (learning stages, SEM experience, and class sizes) and Intervention characteristics (frequency, duration of each session, and experimental period). Effects of the Sport Education Model on student PE learning with different stages. This study categorized the academic stages of the included studies into elementary, secondary, and university levels. The data in Table 5 indicate

that SEM did not achieve statistical significance for physical education learning among elementary school students (P = 0.404), and therefore, SEM cannot be shown to have a significant impact on elementary students’ physical education learning. However, SEM demonstrated significant improvements for secondary (Effect size = 1.055; P = 0.003) and Table 3. The overall effect of the Sport Education Model on student PE learning. Model Effect size Standard error 95% CI Two-tailed test Heterogeneity test Lower limit Upper limit Z p Q df p I2 Random model 0.590 0.156 0.284 0.897 3.773 0.001* 2200.62 48 0.001 98% https://doi.org/10.1371/journal.pone.0331228.t003 Table 4. Effects of the Sport Education Model on cognitive and non-cognitive dimensions in students. Model Effect size Standard error 95% CI Two-tailed test Lower limit Upper limit Z p Cognitive dimension 0.411 0.202 0.015 0.807 2.034 0.042* Non-cognitive dimension 0.641 0.192 0.265 1.016 3.345 0.001* https://doi.org/10.1371/journal.pone.0331228.t004