The moderating effect of sport in the relationship between self-regulation, academic stress and school belonging

The main purpose of this study is to examine the explanatory and predictive relationships among academic stress, self-regulation, and school belonging, and to reveal the moderating role of sports in these relationships. The study employed a correlational survey design. The study group consisted of a total of 1007 high school students studying in secondary education institutions in the central district of Çorum province in Türkiye during the 2023–2024 academic year. Data were collected using a personal information form, the "Academic Expectations Stress Inventory," the "Self-Regulation Scale," and the "School Belonging Scale." The obtained data were analyzed using the SPSS V25 package program, and descriptive statistics as well as correlation analyses were conducted. In addition, mediation (Model 4) and moderation (Model 59) analyses were performed, using the SPSS PROCESS Macro V4.1 developed by Hayes, with a 95% confidence interval and 5000 bootstrap samples to test the significance of indirect (mediation) and moderating (moderation) effects. According to the descriptive findings, students' mean scores of academic stress were at a moderate level (X̄ = 3.04 ± 0.86); their mean scores of self-regulation were also at a moderate level (X̄ = 3.26 ± 0.38). However, the mean scores of school belonging were found to be low (X̄ = 2.17 ± 0.44). In terms of comparative findings, statistically significant differences were found between gender groups in self-reinforcement (t1005 = -3.853, p < .05), self-monitoring (t1005 = 2.994, p < .05), and school belonging (t1005 = 3.052, p < .05). These differences favored male students in school belonging, and female students in self-monitoring and self-reinforcement. Regarding the variable of regular sports participation, significant differences were found in academic stress (t1005 = 2.338, p < .05) and self-monitoring (t1005 = 2.378, p < .05), in favor of students who regularly engaged in sports. Regarding the variable of sports ranking, significant differences were found in academic stress (t1005 = -2.440, p < .05), self-reinforcement (t1005 = -3.105, p < .05), and school belonging (t1005 = 2.302, p < .05). These differences favored students with a sports ranking in terms of academic stress and self-reinforcement, while they favored students without a sports ranking in terms of school belonging. No significant differences were found among groups with respect to school type. With regard to relational analyses, Pearson product-moment correlation coefficients (r) ranged from -.29 to .21. Eight out of ten correlations were significant, three of which were positive and five were negative. According to the predictive findings, the hypothesized regression model was found to be significant (F(4,1002) = 27.908, p < .001), indicating that academic stress, self-reinforcement, self-monitoring, and self-regulation together explained 10% of the total variance in school belonging (R² = .100). Mediation analysis results showed that the direct effect of academic stress on school belonging was significant (b = -0.043, p = .008). When the significance of indirect effects was tested, it was confirmed that both self-reinforcement (b = -0.013, 95% CI [-0.0234, -0.0059]) and self-monitoring (b = 0.032, 95% CI [0.0197, 0.0455]) had partial mediation effects. The model explained 10.02% of the total variance (R² = .1002, p < .001). In conclusion, academic stress negatively affected school belonging, while self-regulation had a partial mediating role in this relationship, and the sports age variable was found to have a moderating effect on the model.