Login
Register
Submit a Paper
Propose a Special lssue
Open Access
ARTICLE
The Relationships among Exercise Participation, Self-Compassion and Academic Stress in Classroom Contexts: Based on Latent Profiles and Mediation Analyses
1 College of Physical Education and Sport, Beijing Normal University, Beijing, China
2 College of Education for the Future, Beijing Normal University, Zhuhai, China
3 School of Physical Education and Sports Science, Qufu Normal University, Qufu, China
4 Xinjiang Urumqi No. 9 Middle School, Urumqi, China
5 Xinjiang Urumqi No. 12 Middle School, Urumqi, China
* Corresponding Authors: Yuyang Nie. Email: ; Cong Liu. Email:
(This article belongs to the Special Issue: Latent Profile Analysis in Mental Health Research: Exploring Heterogeneity through Person Centric Approach)
International Journal of Mental Health Promotion 2026, 28(5), 7 https://doi.org/10.32604/ijmhp.2026.076017
Received 12 November 2025; Accepted 25 February 2026; Issue published 28 May 2026
View Full Text
Download PDFAbstract
Background: Physical exercise is recognized as an effective means of alleviating academic stress, and physical education (PE) classes constitute a primary source of such activity for middle school students. This study aimed to delve into the diversity of PE class participation patterns among these students, examine their relationship with academic stress, and specifically investigate the mediating role of self-compassion in this process. Methods: A cross-sectional survey was conducted among 849 Chinese middle school students. Data were collected via online questionnaires using validated measurement instruments, which included the degree of participation in PE classes, academic stress, and the self-compassion scale. SPSS 27.0 was used to perform correlation and mediation analyses, and Mplus 8.3 was utilized for latent profile analysis (LPA). Results: The study identified four distinct patterns of participation: the avoidant group (13.31%), the moderate participation group (10.75%), the interest-driven group (29.71%), and the active participation group (46.23%). Additionally, compared to the “Avoidant group”, students in the other three groups showed significantly higher levels of self-compassion and reported significantly lower levels of academic stress. Specifically, the relative indirect effects for the Moderately Engaged group, Interest-Driven Engagers group, and Actively Engaged group were −0.060 (95% CI: [−0.180, 0.055]), −0.123 (95% CI: [−0.220, −0.021]), and −0.234 (95% CI: [−0.333, −0.137]), respectively. Conclusion: These results underscore the importance of PE participation patterns and highlight that optimizing PE class design to stimulate students’ intrinsic interest, thereby enhancing their engagement, represents an effective strategy for promoting the overall psychosocial well-being of middle school students.Keywords
Supplementary Material
Supplementary Material FileAcademic stress refers to a variety of pressures felt by students during the learning process [1]. Heavy assignments, large amounts of reading materials, and the need to master knowledge points and exams may cause great stress to students [2,3]. A report published by the World Economy and the Organization for Cooperation and Development (OECD, 2017) states that with rising academic expectations, students around the world are facing pressure from both academic and social experiences [4]. In China, the phenomenon of high school students facing academic pressure is more prevalent due to the pressure of promotion in the middle and high school entrance exams; according to a survey, as many as 58.9% of high school students in China suffer from academic pressure for a long period of time [5].
This widespread academic stress has a particularly profound impact on secondary school students who are currently in adolescence. Secondary school students are in the midst of adolescence and young adulthood, which is a critical transitional stage from childhood to adulthood. Along with significant physical, cognitive, and emotional changes [6], they are psychologically relatively vulnerable and more likely to experience negative emotions such as anxiety, depression, and stress [7]. Research has found that moderate stress can actually have positive effects, such as improving students’ attention and memory, which can facilitate learning [8]. This is because stress activates areas of the brain, such as the prefrontal cortex, which are closely related to cognitive control and decision-making. However, too much stress can not only impair these important cognitive functions [9] but also exacerbate adolescents’ already existing psychological vulnerabilities, leading to a variety of mental health problems such as depression, anxiety, and academic burnout [10,11]. An empirical study from China also confirms that higher levels of academic stress are associated with poorer psychological status afterward [12]. On the contrary, lower levels of negative stressful experiences can help to improve teacher-student relationships, increase school engagement, and promote academic achievement [13]. Therefore, paying attention to and managing students’ academic stress not only helps to improve their academic performance but also protects their psychological well-being.
It is crucial to find effective coping strategies to address the negative impacts of the aforementioned academic stress. Physical activity is often regarded as a protective factor for adolescent mental health [14]. Research confirms that moderate physical activity is effective in reducing academic stress and positively affecting the mental health of adolescents [15,16]. Emotion regulation theories support the rationale for this behavior, as exercise promotes the brain’s release of endorphins, the “happy hormones” that elevate mood and alleviate negative emotions such as anxiety and depression [17]. In addition, exercise can help students temporarily detach themselves from academic stress by diverting their attention, thus reducing stress [18]. However, in China, parents and schools generally worry that extracurricular sports activities will impact academics, and this concern limits students’ opportunities to participate in extracurricular sports to some extent [19]. Therefore, the nationally regulated physical education (PE) classes have become an important way for secondary school students to obtain stable exercise time, directly influencing their overall physical activity levels. However, physical activity exhibits a “dose effect”, meaning that specific intensity and frequency are required to achieve benefits [20]. Consequently, the degree of student engagement in PE classes is directly related to the effectiveness of physical activity in alleviating academic stress.
In exploring the mechanisms by which physical activity alleviates stress, self-compassion has attracted researchers’ attention as a significant psychological variable. Self-compassion refers to the ability to treat oneself with kindness and understanding during times of pain or failure, recognizing that these difficulties are shared human experiences, and maintaining a positive mindset and awareness of one’s own thoughts and emotions [21]. It plays a key psychological protective role for adolescents, helping them face their deficiencies and dilemmas with an objective and friendly attitude, thereby managing academic stress more effectively [22]. It is important to note that self-empathy is not inherent by nature but is a trait that can be cultivated [23]. Research suggests that elevated feelings of self-compassion can mitigate psychological factors that impede willingness to be physically active, which in turn drives individuals to engage in physical activity [24]. In turn, physical activity is also beneficial in fostering self-compassion. A meta-analysis by Wong et al. confirms that various forms of physical activity, such as aerobic exercise, yoga, and strength training, can enhance individuals’ levels of self-compassion [25]. In addition, as an important component of self-compassion, positive thought-based interventions have also been recognized as effective stress reducers [26,27]. Therefore, self-compassion may mediate the association between secondary school students’ sports participation in PE classes and academic stress.
However, secondary school students’ participation in PE classes varies significantly, directly affecting class effectiveness. While previous studies have linked sports participation, self-compassion, and academic stress, most have focused on general physical activity rather than the specific context of PE classes. Therefore, this study uses latent profile analysis (LPA) to classify students based on their specific participation behaviors, avoiding subjective categorization [28]. On this basis, we will further explore the relationship between different types of engagement and academic stress and examine the possible role of self-compassion in the relationship.
In summary, the present study attempted to identify different profiles of secondary school students’ sports participation in PE classes by means of potential profile analysis, to explore the relationship between different profiles and academic stress, and to examine whether self-compassion plays a mediating role between potential profiles of sports participation in PE classes and academic stress. In order to provide empirical evidence to alleviate students’ academic stress and promote healthy development. Accordingly, the following hypotheses were formulated:
Hypothesis 1: Secondary school students present different potential profiles on sports participation in physical education classes.
Hypothesis 2: Different potential profiles differ in self-compassion and academic stress.
Hypothesis 3: Self-compassion mediates the relationship between physical education participation and academic stress.
2.1 Participants and Procedures
This study was conducted in June 2025 as a cross-sectional survey in a middle school in Xinjiang, using a stratified random sampling method to recruit participants across different grades. The study strictly adhered to the Declaration of Helsinki and related laws, regulations, and ethical codes to ensure the morality and legitimacy of the study. The study was approved by the Ethics Committee of the Faculty of Psychology at Beijing Normal University (IRB No.: BNU202506160166).
After identifying the research topic, we integrated the PE Class Exercise Participation, Self-Compassion, and Academic Stress Scales into the Questionnaire Star platform. Before the questionnaire was distributed, we trained the participants in the study. After obtaining informed consent from the participants and their guardians, they filled out the questionnaire online through an exclusive link. The entire study process was executed based on the STROBE checklist for cross-sectional studies to ensure transparency of the study and completeness of reporting (Supplementary Material S1) [29]. Through these rigorous measures, we sought to enhance the reliability and validity of the study. A total of 872 data sets were collected in this study. After screening and excluding 23 data sets with homogeneous responses or excessively short completion times, 849 valid data sets were obtained, resulting in a validity rate of 97.1%. In the sample, females accounted for 48.6% and males for 51.4%, with a mean age of 14.9 ± 1.63 years. Regarding grade distribution, there were 237 students in Grade 7 (28%), 218 in Grade 8 (26%), 167 in Grade 10 (20%), and 227 in Grade 11 (27%), with a total sample size of 849.
2.2.1 Sports Participation in Physical Education Classes
This study used the Measurement of Students’ Sports Participation in PE Scale developed by Agbuga et al. to assess students’ participation in PE classes [30]. The scale consists of three dimensions: behavioral participation, effective participation, and cognitive participation, with a total of 13 questions, each of which is scored on a 5-point Likert scale, ranging from 1 to 5 on a scale from “very inconsistent” to “very consistent”. The total score of the scale ranges from 13 to 65, with higher scores indicating more active participation in PE classes. The Chinese version was revised by Zhang and has been well-validated with secondary school students [31]. The Cronbach’s alpha of the scale in this study was 0.952.
This study used the short version of the Self-Compassion Scale (SCS-SF) to assess the level of self-compassion of the subjects [32]. Its Chinese version was revised by Gong et al. [33]. The scale contains 12 items based on a three-dimensional six-factor model, measuring self-kindness, common humanity, and positive thinking. The scale is scored on a 5-point scale, with higher scores indicating higher levels of self-compassion. The Cronbach’s alpha coefficient for this scale in this study was 0.755.
In this study, the Academic Stress Scale for Secondary School Students, developed by Xu et al., was used to assess students’ academic stress [34]. The scale is based on Likert’s 5-point self-rating scale, with scores ranging from 1 to 5, from “very inconsistent” to “fully consistent”. The scale consists of 21 items, and the total score ranges from 21 to 105, with higher scores indicating higher levels of academic stress among secondary school students. The Cronbach’s alpha coefficient of the total scale in the original study was 0.81, and the Cronbach’s alpha of the scale in this study was 0.908.
After completing the initial cleaning and organizing of the data, we first used Mplus 8.3 software (Muthén & Muthén, Los Angeles, CA, USA) to perform an LPA, which aims to identify potential categories formed in the student population based on the characteristics of sports participation in PE classes. In accordance with the recommendations of Weller et al., we reported the corresponding fit indices, entropy, posterior probabilities, and the smallest class proportion [35]. Next, correlation analysis was performed using SPSS 27.0 software (IBM Corp., Armonk, NY, USA) in order to examine the degree of association among the variables. Also, based on the different participation types identified by the LPA, we further executed a one-way analysis of variance to compare the differences in the distribution of each type of student on the other variables. Finally, still using SPSS 27.0 and its PROCESS 4.2 plug-in, a mediated-effects model was constructed with potential profile categories of sports participation as the independent variable, academic stress as the dependent variable, and self-control as the mediator, and controlling for the effects of gender and age. The analysis was conducted using Bootstrap methods with 5000 resampling iterations to construct 95% confidence intervals (CIs). We assessed the significance of the mediating effect by determining whether the CI contained zero: if the CI did not contain zero, the corresponding mediating effect was considered significant.
3.1 Potential Profiles of Secondary School Students’ Sport Participation in PE Classes
The results show (Table 1) that the values of the information criterion indicators Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Adjusted Bayesian Information Criterion (ABIC) exhibit a decreasing trend with the increase in the number of potential categories, indicating that the model’s goodness-of-fit to the data is gradually improving.
Further combining the results of Lo-Mendell-Rubin adjusted Likelihood Ratio Test (LMR-LRT) and Bootstrap Likelihood Ratio Test (BLRT), the p-values of each test reached the significance level (p < 0.05) from the dichotomous model to the quaternary model, implying that the subsequent models (i.e., those containing more categories) significantly outperform their predecessor models. In addition, the Entropy index of each categorical model remained above 0.9, indicating a high categorization accuracy of over 90%.
However, when the models were extended to five classifications, the p-value of the LMR-LRT test was greater than 0.05, failing to reach the significance level of 0.05. This suggests that the five-classification model did not provide a significant improvement in fit compared to the four-classification model. Therefore, the four-category model is the best choice. It not only has an entropy greater than 0.9 but also has a lower information criterion value and a reasonable distribution of categories. The average posterior probabilities for each class were: The Avoiders (0.994), The Moderately Engaged (0.908), Interest-Driven Engagers (0.924), and The Actively Engaged (0.959). With the exception of minimal overlap between the Moderately Engaged and Interest-Driven Engagers, the samples in each category demonstrated extremely high classification accuracy.
Fig. 1 visually shows the mean scores of these four identified potential categories on the 13 entries of sports participation in PE class. The horizontal coordinates in the figure represent the 13 entries of the sport participation assessment, and the vertical coordinates represent the mean scores of the different categories on these entries. Based on the differences in students’ sports participation in PE classes, we categorized them into four types: active participants, interest-driven participants, moderate participants, and avoidant participants.
Table 1: Latent profile model of physical activity participation in middle school physical education classes and its indicators.
| Class | LL | AIC | BIC | ABIC | Entropy | LMR | BLRT | Type (%) |
|---|---|---|---|---|---|---|---|---|
| 1 | −17,809.27 | 35,670.54 | 35,793.88 | 35,711.31 | - | - | - | 100 |
| 2 | −14,689.14 | 29,458.28 | 29,648.04 | 29,521.01 | 0.98 | <0.01 | <0.01 | 20.98/79.02 |
| 3 | −13,930.11 | 27,968.21 | 28,224.39 | 28,052.90 | 0.92 | <0.01 | <0.01 | 16.15/52.0/31.84 |
| 4 | −13,709.76 | 27,555.53 | 27,878.12 | 27,662.17 | 0.91 | 0.04 | <0.01 | 13.31/10.75/29.71/46.23 |
| 5 | −13,618.54 | 27,401.07 | 27,790.08 | 27,529.68 | 0.90 | 0.71 | <0.01 | 13.25/7.59/25.50/8.37/45.29 |
Figure 1: Mean scores of middle school students on each item based on different profiles of physical education class participation.
3.2 Results of Correlation Analysis
According to the results of the correlation analysis presented in Table 2, academic stress is significantly and negatively correlated with participation in PE class sports (r = −0.315, p < 0.01) and self-compassion (r = −0.425, p < 0.01). Meanwhile, a notable positive association exists between self-compassion and attendance in PE sessions (r = 0.239, p < 0.01). The research indicated a negative correlation between age and academic stress, although gender exhibited no significant link with any other variable.
Table 2: Analysis of bivariate correlations among various variables.
| Variable | Grade | Gender | Age | AS | SP | SCS |
|---|---|---|---|---|---|---|
| Grade | - | |||||
| Gender | −0.033 | - | ||||
| Age | 0.948** | −0.005 | - | |||
| AS | −0.126** | 0.033 | −0.133** | - | ||
| SP | 0.072* | −0.001 | 0.062 | −0.315** | - | |
| SC | 0.033 | 0.005 | 0.031 | −0.425** | 0.239** | - |
3.3 Results of Univariate Analysis
Univariate analysis was conducted based on the four categories of secondary school students’ sports participation (LPA) in PE classes. From Table 3, it can be seen that there are significant differences in gender, age, grade, self-compassion score, and academic stress score across categories. Specifically, the higher the level of sports participation in PE classes, the higher the participants’ self-compassion scores and the lower their perceived academic stress. Female students had a higher percentage in the active participation group, but also had a much higher percentage in the avoidance group than male students.
Table 3: Statistical characteristics of different categories.
| Characteristic | Overall (N = 849) | The Avoiders (N = 1) | The Moderately Engaged (N = 92) | Interest-Driven Engagers (N = 249) | The Actively Engaged (N = 397) | p-Value |
|---|---|---|---|---|---|---|
| Gender, n (%) | <0.001 | |||||
| Male | 413 (48.6%) | 39 (35.1%) | 61 (66.3%) | 137 (55.0%) | 176 (44.3%) | |
| Women | 436 (51.4%) | 72 (64.9%) | 31 (33.7%) | 112 (45.0%) | 221 (55.7%) | |
| Grade, n (%) | <0.001 | |||||
| Grade 7 | 237 (27.9%) | 35 (31.5%) | 30 (32.6%) | 94 (37.8%) | 78 (19.6%) | |
| Grade 8 | 218 (25.7%) | 26 (23.4%) | 26 (28.3%) | 53 (21.3%) | 113 (28.5%) | |
| Grade 10 | 167 (19.7%) | 13 (11.7%) | 9 (9.8%) | 36 (14.5%) | 109 (27.5%) | |
| Grade 11 | 227 (26.7%) | 37 (33.3%) | 27 (29.3%) | 66 (26.5%) | 97 (24.4%) | |
| Age, Mean (SD) | 14.90 (1.63) | 14.97 (1.68) | 14.64 (1.69) | 14.69 (1.67) | 15.06 (1.56) | 0.023 |
| SC, Mean (SD) | 42 (7) | 39 (9) | 40 (7) | 41 (6) | 43 (6) | <0.001 |
| AS, Mean (SD) | 52 (16) | 62 (20) | 56 (15) | 53 (13) | 48 (14) | <0.001 |
3.4 Mediating Effects of Self-Compassion
This study draws on the suggestion of Fang et al. to use relative mediation analysis for independent variables containing three or more categories (k ≥ 3) [36]. First, considering the possible interference of demographic factors, we included gender and age as control variables in the model. Second, for the categorization process, we set “avoidance” as the reference group. Then, we coded dummies for the potential profile categories of the independent variable “self-compassion” and generated three dummy variables: D1 (vs. passive involvement vs. avoidance), D2 (vs. interest-driven vs. avoidance), and D3 (vs. active involvement vs. avoidance). See Table 4 for the mediating effect pathways of self-compassion and their specific effect sizes.
Fig. 2 illustrates the modeling of the mediating role of self-compassion between potential profiles of sports participation in PE classes and academic performance. The results of the analysis showed that the relative direct effect reached significant levels in the passive participation, interest-driven, and active participation groups compared to the avoidance group. In addition, the relative indirect effect showed a significant effect only in the interest-driven and active participation groups, while no significant difference was observed in the passive participation group.
Table 4: Self-compassion mediating effect test.
| Types | β | SE | 95% CI |
|---|---|---|---|
| Relative Direct Effect | |||
| The Moderately Engaged | −0.330 | 0.125 | −0.575, −0.085 |
| Interest-Driven Engagers | −0.423 | 0.101 | −0.622, −0.225 |
| The Actively Engaged | −0.669 | 0.096 | −0.857, −0.481 |
| Relative Indirect Effect | |||
| The Moderately Engaged | −0.060 | −0.060 | −0.180, 0.055 |
| Interest-Driven Engagers | −0.123 | 0.051 | −0.220, −0.021 |
| The Actively Engaged | −0.234 | 0.050 | −0.333, −0.137 |
Figure 2: The mediating pathway diagram of self-compassion between physical activity participation in physical education classes and academic stress. Note: SC: Self-compassion; AS: Academic stress. Solid lines represent significant paths, while dashed lines indicate non-significant paths.
This study examined the physical activity participation patterns of secondary school students in PE classes, utilizing LPA grounded in emotion regulation theory. The results of the analysis showed that the student population could be categorized into four different participation profiles: the avoidance group (13.31%), moderate participation group (10.75%), the interest-driven group (29.71%), and the active participation group (46.23%). This finding confirms research Hypothesis 1. Additionally, the study examined the differences in self-compassion and academic stress levels among secondary school students across various participation profiles. Students in the passive, interest-driven, and active participation groups exhibited higher levels of self-compassion and lower perceptions of academic stress compared to those in the “avoidance group”, confirming Hypothesis 2. This finding further revealed that sports participation was positively associated with self-compassion and negatively associated with academic stress, which is consistent with the findings of previous studies by August et al. and Wunsch et al., and collectively emphasizes the positive effects of physical activity participation on adolescents’ physical and mental health [37,38].
The results indicated that both the interest-driven and active involvement groups were effective in forecasting people’s levels of self-compassion, in contrast to the exercise-avoidance group. The prognostic effect of the passive participation group was insignificant. This disparity may arise from variations in motivational characteristics. The exercise avoidance and passive participation groups had a comparable characteristic: an absence of drive for physical activity. Despite students in the passive participation group engaging to a degree in their PE lessons, this involvement failed to provide substantial beneficial consequences owing to insufficient intrinsic desire. According to self-determination theory, individuals are more likely to experience positive emotions, such as joy, vigor, and pride, when they participate in sports based on intrinsic motivation (e.g., enjoyment, sense of accomplishment, or excitement) [39]. On the contrary, in the passive participation group, due to insufficient motivation, they struggled to achieve full psychological benefits from exercise, which may explain the non-significant predictive effect of their self-compassion.
However, in the analysis of variance, the self-compassion score of the passive participation group was higher than that of the exercise avoidance group. This may indicate that even limited participation in exercise may still result in a marginally positive impact compared to complete exercise avoidance. Further comparing the interest-driven and active participation groups with the exercise avoidance group, we found that the first two groups were significant predictors of self-compassion, with relative direct effects of β = 0.331 (p < 0.01) and β = 0.631 (p < 0.01), respectively. de Bruin et al. conducted a five-week intervention demonstrating that physical activity could effectively improve attentional control, executive functioning, awareness of positive thoughts, self-compassion, and worry [40]. A randomized controlled trial (RCT) by Mothes et al. showed that regular physical activity also improves positive self-perceptions, such as dispositional positive thoughts, in healthy men [41]. Similarly, tai chi has been shown to significantly increase individuals’ levels of self-compassion in an RCT [42], which is an important component of Chinese PE classes.
Research indicates that individuals who are passively involved, motivated by interest, or actively participating in physical activity exhibit markedly lower levels of academic stress compared to those who opt to forgo exercise within the college demographic, underscoring the significance of physical activity as a viable strategy for stress management [43]. In terms of physiological mechanisms, physical activity can reduce the stress response by reducing stress-induced cortisol release or relieving tension [44], while systematic evaluations by De Nys et al. and Beserra et al. have also confirmed the positive effects of exercise on reducing cortisol levels [45,46]. However, it is worth noting that exercise modulates the hypothalamic-pituitary-adrenal axis (HPA axis) and influences the cortisol arousal response (CAR), which may require reaching a specific exercise threshold [47]. This is particularly important for middle school students: they tend to be more likely to reach this threshold if they are more actively engaged in PE classes, which can lead to more effective academic stress relief. In addition, active participation in PE, whether performing aerobic exercises or participating in team sports such as basketball and soccer, often creates a more positive learning atmosphere and a stronger sense of social support. This positive atmosphere facilitates the acquisition of motor skills and effectively relieves stress [48,49], and social support itself has been suggested to buffer the HPA axis and the cardiovascular system from stress responses [50]. As a result, even students who are less willing to participate are susceptible to the enthusiastic atmosphere and thus get enough exercise to reach the exercise threshold needed to lower cortisol levels. In the process, they can harvest the social support and psychological resources necessary to cope with stress. This suggests that measures to motivate students to actually engage in exercise in the PE classroom are critical to their healthy development.
The results of this study also support the mediating role of self-compassion between the potential profiles of exercise participation in PE class and academic stress, thus confirming Hypothesis 3. The mediating effect of self-compassion was more significant in the remaining groups compared to the exercise avoidance group. Specifically, sports participation in PE classes contributed to adolescents’ development of self-compassion, which in turn significantly influenced individuals’ stress responses [51,52,53]. Compared with adolescents with lower levels of self-compassion, adolescents with higher levels of self-compassion exhibit more moderate physiological stress responses when facing stress (e.g., lower cortisol secretion, greater heart rate variability, complete recovery from stress, and smaller elevations in heart rate and blood pressure) [54]. In addition, interventions targeting self-compassion and positive thinking have been shown to be effective in reducing adolescents’ perceived stress levels [23]. In conjunction with our previous discussion, it is clear that incorporating aerobic exercise, strength training, and tai chi programs into PE classes is an effective way to develop self-compassion. Therefore, increased exercise participation in PE classes by secondary school students can be considered as an intervention to promote self-compassion, which in turn can help to regulate their academic stress response and maintain their physical and mental health.
Research Limitations and Future Perspectives
This study has some methodological limitations. First, the cross-sectional research design adopted is essentially correlational, making it difficult to draw strong inferences regarding the causal relationships between physical activity participation, self-compassion, and academic stress, allowing only for a preliminary exploration of their directional relationships. A design such as a longitudinal study or a RCT would be a better choice in the future. Second, over-reliance on subjective reports may introduce measurement errors when measuring exercise participation in PE classes. In the future, physiological indicators (e.g., heart rate monitoring) or wearable devices (e.g., accelerometers) can be introduced for objective measurements, which will significantly improve the reliability and validity of the data. Finally, the present study only included self-compassion and failed to adequately consider other moderators or mediators that may influence the relationship between variables, such as self-efficacy and mental toughness. Future research could be conducted on these variables.
This study revealed that the participation of secondary school students in sports during PE class does not follow a single pattern but rather shows significant heterogeneity, which can be categorized into four types: avoidance group, passive participation group, interest-driven group, and active participation group. Further analysis revealed statistically significant differences among these four groups in terms of gender, age, grade level, level of self-compassion, and perceived academic stress. Of particular importance, this study confirms that self-compassion plays a significant mediating role between the type of sports participation in PE class and academic stress: students’ sports participation patterns influence their academic stress by affecting their level of self-compassion. Consequently, we advise PE instructors to refine the structure of their curriculum to augment student involvement in PE classes and foster their physical and mental health growth.
Acknowledgement:
Funding Statement: National Social Science Fund of China “14th Five-Year Plan” 2021 General Research Topic in Education: “Study on Higher Education Quality Assurance System in the Guangdong-Hong Kong-Macao Greater Bay Area from the Perspective of Regional Development” (Project Number: BIA210191).
Author Contributions: Guofeng Qu, Cong Liu and Yuyang Nie: Conceptualization, Data Analysis, Interpretation, Visualization and Writing—Original Draft. Jing Liu and Fengwei Jia: Data Validation, Interpretation and Manuscript Review. Xishuai Wang: Technical Support for Data Analysis. Guoyue Tang and Zhonghu Gu: Data Collection. Yuyang Nie and Cong Liu: Supervision, Project Administration and Writing—Review & Editing. All authors reviewed and approved the final version of the manuscript.
Availability of Data and Materials: Data supporting this study can be obtained from the corresponding authors.
Ethics Approval: The study was approved by the Ethics Committee of the Faculty of Psychology at Beijing Normal University (IRB No.: BNU202506160166). Informed consent was obtained from all participants and their guardians.
Conflicts of Interest: The authors declare no conflicts of interest.
Supplementary Materials: The supplementary material is available online at https://www.techscience.com/doi/10.32604/ijmhp.2026.076017/s1.
References
1. Han L , Gao X , Wang X , Ren W . The relationship between academic stress and educational anxiety: learning anxiety and learning weariness as mediators. Curr Psychol. 2024; 43( 40): 31592– 603. doi:10.1007/s12144-024-06738-3. [Google Scholar] [CrossRef]
2. Davis JD , McDuffie AR , Drake C , Seiwell AL . Teachers’ perceptions of the official curriculum: problem solving and rigor. Int J Educ Res. 2019; 93: 91– 100. doi:10.1016/j.ijer.2018.10.002. [Google Scholar] [CrossRef]
3. Ahmad I , Gul R , Zeb M . A qualitative inquiry of university student’s experiences of exam stress and its effect on their academic performance. Hum Arenas. 2024; 7( 4): 778– 88. doi:10.1007/s42087-022-00285-8. [Google Scholar] [CrossRef]
4. OECD . Education at a glance 2017: OECD indicators. Paris, France: OECD; 2017. [Google Scholar]
5. Luo Y , Cui Z , Zou P , Wang K , Lin Z , He J , et al. Mental health problems and associated factors in Chinese high school students in Henan Province: a cross-sectional study. Int J Environ Res Public Health. 2020; 17( 16): 5944. doi:10.3390/ijerph17165944. [Google Scholar] [CrossRef]
6. Nahman-Averbuch H , Li R , Boerner KE , Lewis C , Garwood S , Palermo TM , et al. Alterations in pain during adolescence and puberty. Trends Neurosci. 2023; 46( 4): 307– 17. doi:10.1016/j.tins.2023.01.006. [Google Scholar] [CrossRef]
7. Helland Lindborg T , Kristensson Hallström I , Synnøve Litland A , Åvik Persson H . Adolescent boys’ experiences of mental health and school health services—an interview study from Norway. BMC Public Health. 2024; 24( 1): 1464. doi:10.1186/s12889-024-18952-6. [Google Scholar] [CrossRef]
8. Malik S . Examining cognitive functioning and academic stress in high school students implications for education and well-being. Int J Res Appl Sci Eng Technol. 2024; 12( 3): 2747– 57. doi:10.22214/ijraset.2024.59339. [Google Scholar] [CrossRef]
9. Scrimin S , Mason L , Moscardino U . School-related stress and cognitive performance: a mood-induction study. Contemp Educ Psychol. 2014; 39( 4): 359– 68. doi:10.1016/j.cedpsych.2014.09.002. [Google Scholar] [CrossRef]
10. Liu C , She X , Lan L , Wang H , Wang M , Abbey C , et al. Parenting stress and adolescent academic burnout: the chain mediating role of mental health symptoms and positive psychological traits. Curr Psychol. 2024; 43( 8): 7643– 54. doi:10.1007/s12144-023-04961-y. [Google Scholar] [CrossRef]
11. Bai Q , Bai S , Dan Q , Lei L , Wang P . Mother phubbing and adolescent academic burnout: the mediating role of mental health and the moderating role of agreeableness and neuroticism. Pers Individ Differ. 2020; 155: 109622. doi:10.1016/j.paid.2019.109622. [Google Scholar] [CrossRef]
12. Wang J , Wang Z , Yang Y , Wang T , Lin H , Zhang W , et al. Academic burden and emotional problems among adolescents: a longitudinal mediation analysis. J Adolesc. 2025; 97( 4): 989– 1001. doi:10.1002/jad.12471. [Google Scholar] [CrossRef]
13. Forrest CB , Bevans KB , Riley AW , Crespo R , Louis TA . Health and school outcomes during children’s transition into adolescence. J Adolesc Health. 2013; 52( 2): 186– 94. doi:10.1016/j.jadohealth.2012.06.019. [Google Scholar] [CrossRef]
14. Kaiser-Jovy S , Scheu A , Greier K . Media use, sports activities, and motor fitness in childhood and adolescence. Wien Klin Wochenschr. 2017; 129( 13–14): 464– 71. doi:10.1007/s00508-017-1216-9. [Google Scholar] [CrossRef]
15. Theodoratou M , Dritsas I , Saltou M , Dimas V , Spyropoulos A , Nikolopoulou E , et al. Physical exercise and students’ mental health. Eur Psychiatr. 2016; 33( S1): s219. doi:10.1016/j.eurpsy.2016.01.533. [Google Scholar] [CrossRef]
16. Cheng R , Yang L , Kang SJ . A study on the relationship between high school students’ sleep quality, physical exercise, academic stress, and subjective well-being. BMC Psychol. 2025; 13( 1): 180. doi:10.1186/s40359-025-02497-3. [Google Scholar] [CrossRef]
17. Matten SJ , Pausch MJ . The psyche—thoughts, feelings, body, behavior. Wiesbaden, Germany: Springer Fachmedien Wiesbaden; 2024. p. 17– 48. doi:10.1007/978-3-658-45432-6_4. [Google Scholar] [CrossRef]
18. Mikkelsen K , Stojanovska L , Polenakovic M , Bosevski M , Apostolopoulos V . Exercise and mental health. Maturitas. 2017; 106: 48– 56. doi:10.1016/j.maturitas.2017.09.003. [Google Scholar] [CrossRef]
19. Wang D , Xiong R , Zhang J , Han X , Jin L , Liu W , et al. Effect of extracurricular after-school physical activities on academic performance of schoolchildren: a cluster randomized clinical trial. JAMA Pediatr. 2023; 177( 11): 1141– 8. doi:10.1001/jamapediatrics.2023.3615. [Google Scholar] [CrossRef]
20. Fornaguera M , Dowda M , Pate R . Dose-response associations between physical activity and mental health among USA adolescents. Eur J Public Health. 2024; 34( Supplement_3): ckae144.2209. doi:10.1093/eurpub/ckae144.2209. [Google Scholar] [CrossRef]
21. Barnard LK , Curry JF . Self-compassion: conceptualizations, correlates, & interventions. Rev Gen Psychol. 2011; 15( 4): 289– 303. doi:10.1037/a0025754. [Google Scholar] [CrossRef]
22. Neff KD , McGehee P . Self-compassion and psychological resilience among adolescents and young adults. Self Identity. 2010; 9( 3): 225– 40. doi:10.1080/15298860902979307. [Google Scholar] [CrossRef]
23. Bluth K , Gaylord SA , Campo RA , Mullarkey MC , Hobbs L . Making friends with yourself: a mixed methods pilot study of a mindful self-compassion program for adolescents. Mindfulness. 2016; 7( 2): 479– 92. doi:10.1007/s12671-015-0476-6. [Google Scholar] [CrossRef]
24. Rapoport O , Bengel S , Möcklinghoff S , Neidhardt E . Self-compassion moderates the influence of procrastination on postponing sporting activity. Pers Individ Differ. 2022; 185: 111242. doi:10.1016/j.paid.2021.111242. [Google Scholar] [CrossRef]
25. Wong MYC , Chung PK , Leung KM . The relationship between physical activity and self-compassion: a systematic review and meta-analysis. Mindfulness. 2021; 12( 3): 547– 63. doi:10.1007/s12671-020-01513-4. [Google Scholar] [CrossRef]
26. Davis KM , Wojcik CM , Baillie AJ , Foley E , Goddard T , Lau MA , et al. Mechanisms of mindfulness: a longitudinal study of a mindfulness-based stress reduction program. Mindfulness. 2024; 15( 5): 1188– 207. doi:10.1007/s12671-024-02359-w. [Google Scholar] [CrossRef]
27. Sawyer M , Stanton MV , Southerland AC , Kelly CN , Gutierrez L . A systematic review evaluating the efficacy of mindfulness-based interventions on reducing stress in Hispanic/Latino/a/x college students. Mindfulness. 2025; 16( 5): 1130– 58. doi:10.1007/s12671-025-02562-3. [Google Scholar] [CrossRef]
28. Wang M . Latent variable modeling and mplus applications—advanced chapter. Chongqing, China: Chongqing University Press; 2018. (In Chinese). [Google Scholar]
29. von Elm E , Altman DG , Egger M , Pocock SJ , Gøtzsche PC , Vandenbroucke JP . The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007; 147( 8): 573– 7. doi:10.7326/0003-4819-147-8-200710160-00010. [Google Scholar] [CrossRef]
30. Agbuga B , Xiang P , McBride RE , Su X . Student perceptions of instructional choices in middle school physical education. J Teach Phys Educ. 2016; 35( 2): 138– 48. doi:10.1123/jtpe.2015-0010. [Google Scholar] [CrossRef]
31. Zhang BG . The influence of interpersonal support on senior high school students’ physical activity participation in specialized physical education [ dissertation]. Shanghai, China: Shanghai University of Sport; 2023. (In Chinese). [Google Scholar]
32. Neff KD . The development and validation of a scale to measure self-compassion. Self Identity. 2003; 2( 3): 223– 50. doi:10.1080/15298860309027. [Google Scholar] [CrossRef]
33. Gong HL , Jia HL , Guo TM , Zou LL . The Revision of self-compassion scale and its reliability and validity in adolescents. Psychol Res. 2014; 7( 1): 36– 79. (In Chinese). [Google Scholar]
34. Xu JJ , Cao JF , Cui LZ , Zhu P . Preliminary development of a questionnaire for measuring learning stress among middle school students. Chin J Sch Health. 2010; 31( 1): 68– 9. doi:10.16835/j.cnki.1000-9817.2010.01.032. [Google Scholar] [CrossRef]
35. Weller BE , Bowen NK , Faubert SJ . Latent class analysis: a guide to best practice. J Black Psychol. 2020; 46( 4): 287– 311. doi:10.1177/0095798420930932. [Google Scholar] [CrossRef]
36. Fang J , Wen ZL , Zhang MQ . Mediation analysis of categorical variables. Psychol Sci. 2017; 40( 2): 471– 7. (In Chinese). doi:10.16719/j.cnki.1671-6981.20170233. [Google Scholar] [CrossRef]
37. August KJ , Malik D , Markey CH , Woods K , Gerwitz GC . Additive and interactive associations among body appreciation, self-compassion, and gender in understanding college students’ health behaviors. Body Image. 2023; 47: 101634. doi:10.1016/j.bodyim.2023.101634. [Google Scholar] [CrossRef]
38. Wunsch K , Kasten N , Fuchs R . The effect of physical activity on sleep quality, well-being, and affect in academic stress periods. Nat Sci Sleep. 2017; 9: 117– 26. doi:10.2147/NSS.S132078. [Google Scholar] [CrossRef]
39. Nam IS . The effect of participation motivation on exercise emotion and psychological well-being of marathon participants. J Korea Acad Ind Coop Soc. 2013; 14( 9): 4285– 95. doi:10.5762/kais.2013.14.9.4285. [Google Scholar] [CrossRef]
40. de Bruin EI , van der Zwan JE , Bögels SM . A RCT comparing daily mindfulness meditations, biofeedback exercises, and daily physical exercise on attention control, executive functioning, mindful awareness, self-compassion, and worrying in stressed young adults. Mindfulness. 2016; 7( 5): 1182– 92. doi:10.1007/s12671-016-0561-5. [Google Scholar] [CrossRef]
41. Mothes H , Klaperski S , Seelig H , Schmidt S , Fuchs R . Regular aerobic exercise increases dispositional mindfulness in men: a randomized controlled trial. Ment Health Phys Act. 2014; 7( 2): 111– 9. doi:10.1016/j.mhpa.2014.02.003. [Google Scholar] [CrossRef]
42. Nedeljkovic M , Wirtz PH , Ausfeld-Hafter B . Effects of taiji practice on mindfulness and self-compassion in healthy participants—a randomized controlled trial. Mindfulness. 2012; 3( 3): 200– 8. doi:10.1007/s12671-012-0092-7. [Google Scholar] [CrossRef]
43. Nguyen-Michel ST , Unger JB , Hamilton J , Spruijt-Metz D . Associations between physical activity and perceived stress/hassles in college students. Stress Health. 2006; 22( 3): 179– 88. doi:10.1002/smi.1094. [Google Scholar] [CrossRef]
44. Fuchs R , Klaperski S . Stressregulation durch sport und bewegung. In: Handbuch stressregulation und sport. Berlin/Heidelberg, Germany: Springer; 2017. p. 205– 26. doi:10.1007/978-3-662-49322-9_9. [Google Scholar] [CrossRef]
45. De Nys L , Anderson K , Ofosu EF , Ryde GC , Connelly J , Whittaker AC . The effects of physical activity on cortisol and sleep: a systematic review and meta-analysis. Psychoneuroendocrinology. 2022; 143: 105843. doi:10.1016/j.psyneuen.2022.105843. [Google Scholar] [CrossRef]
46. Beserra AHN , Kameda P , Deslandes AC , Schuch FB , Laks J , de Moraes HS . Can physical exercise modulate cortisol level in subjects with depression? A systematic review and meta-analysis. Trends Psychiatry Psychother. 2018; 40( 4): 360– 8. doi:10.1590/2237-6089-2017-0155. [Google Scholar] [CrossRef]
47. Anderson T , Wideman L . Exercise and the cortisol awakening response: a systematic review. Sports Med Open. 2017; 3( 1): 37. doi:10.1186/s40798-017-0102-3. [Google Scholar] [CrossRef]
48. Barney DC , Pleban FT , Lewis T . Relationship between physical activity and stress among junior high school students in the physical education environment. Phys Educ. 2019; 76( 3): 777– 99. doi:10.18666/tpe-2019-v76-i3-8966. [Google Scholar] [CrossRef]
49. Virtanen A , Lauritsalo K , Mäkinen T , Hurskainen H , Tynjälä P . The role of positive atmosphere on learning generic skills in higher education—experiences of physical education students. Front Educ. 2022; 7: 886139. doi:10.3389/feduc.2022.886139. [Google Scholar] [CrossRef]
50. Arch JJ , Brown KW , Dean DJ , Landy LN , Brown KD , Laudenslager ML . Self-compassion training modulates alpha-amylase, heart rate variability, and subjective responses to social evaluative threat in women. Psychoneuroendocrinology. 2014; 42: 49– 58. doi:10.1016/j.psyneuen.2013.12.018. [Google Scholar] [CrossRef]
51. MacBeth A , Gumley A . Exploring compassion: a meta-analysis of the association between self-compassion and psychopathology. Clin Psychol Rev. 2012; 32( 6): 545– 52. doi:10.1016/j.cpr.2012.06.003. [Google Scholar] [CrossRef]
52. Chishima Y , Mizuno M , Sugawara D , Miyagawa Y . The influence of self-compassion on cognitive appraisals and coping with stressful events. Mindfulness. 2018; 9( 6): 1907– 15. doi:10.1007/s12671-018-0933-0. [Google Scholar] [CrossRef]
53. Hu Y , Wang Y , Sun Y , Arteta-Garcia J , Purol S . Diary study: the protective role of self-compassion on stress-related poor sleep quality. Mindfulness. 2018; 9( 6): 1931– 40. doi:10.1007/s12671-018-0939-7. [Google Scholar] [CrossRef]
54. Bluth K , Roberson PNE , Gaylord SA , Faurot KR , Grewen KM , Arzon S , et al. Does self-compassion protect adolescents from stress? J Child Fam Stud. 2016; 25( 4): 1098– 109. doi:10.1007/s10826-015-0307-3. [Google Scholar] [CrossRef]
Cite This Article
Copyright © 2026 The Author(s). Published by Tech Science Press.This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Downloads
Citation Tools
