The relationship between occupational stress and burnout among primary and secondary school physical education teachers: evidence from multiverse-style analysis and diary method

Abstract

Background

To investigate the relationship between occupational stress and burnout among primary and secondary school physical education (PE) teachers and its underlying mechanisms.

Method

Study 1 utilized a large cross-sectional sample of 4,397 teachers and employed multiverse-style analysis to examine the robustness of the relationship between occupational stress and burnout. Study 2 applied a 21-day diary method to a sample of 324 teachers to explore the intrinsic mechanisms through which occupational stress affects burnout.

Results

Study 1 found a positive and robust relationship between occupational stress and burnout. Study 2 revealed that rumination and negative emotions act as a chain mediation mechanism in the influence of occupational stress on burnout.

Conclusion

These findings have significant practical and theoretical implications for improving the quality of physical education and students’ physical fitness.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40359-025-03351-2.

Introduction

Physical education (PE) teachers in primary and secondary schools play an indispensable role in the overall development of students. They not only promote students’ physical health but also influence the development of psychological and social skills through physical education. Recent studies have shown that PE teachers support students’ overall well-being by fostering “physical literacy” [1, 2]. This not only involves improving motor skills but also includes understanding the importance of physical activity for health, cultivating positive health behaviors, and enhancing psychological resilience in coping with challenges and stress [3, 4]. The positive impact of physical education on students’ mental health has also been widely recognized. Participation in physical activities helps reduce anxiety and stress, and through collaborative sports and team projects, students develop social skills such as teamwork and communication, which are crucial for their overall personality development [1]. However, PE teachers in China are often overlooked in the educational system, primarily because PE courses are seen as secondary subjects and do not receive the same level of importance as academic courses [5]. This bias results in PE teachers being at a disadvantage in terms of resource allocation, career development opportunities, and social recognition [6]. These factors significantly erode physical education teachers’ job satisfaction and aggravate their occupational burnout [7]. Burnout is a work-related syndrome caused by the prolonged mismanagement of occupational stress, and mainstream theory conceptualizes it along three dimensions: emotional exhaustion, depersonalization, and diminished personal accomplishment [8]. The World Health Organization likewise defines burnout as a condition resulting from chronic workplace stress, chiefly manifested as energy depletion, negative detachment from one’s work, and a reduced sense of professional efficacy [9]. From the perspective of the Job Demands–Resources (JD-R) model, when job demands surpass the resources available to an individual (e.g., energy, organizational support), resource depletion ensues, ultimately leading to emotional and cognitive exhaustion. Studies indicate that, as members of a highly interpersonal service occupation, teachers are especially prone to burnout; compared with teachers of other subjects, physical education teachers are more likely to experience symptoms such as emotional exhaustion and a reduced sense of accomplishment [10].

In addition, occupational stress among PE teachers is an area worthy of further exploration, as this profession has unique job demands and environmental pressures. PE teachers not only need to teach physical skills but also take on extracurricular activities, manage competitions, and perform additional duties. The physical labor involved and the responsibility for students’ safety further increase their stress levels. These factors significantly affect PE teachers’ job satisfaction and contribute to burnout [7]. Such stress and burnout not only harm PE’ physical and mental well-being, reduce the quality of instruction, and consequently affect students’ learning experiences, but may also lead to an increase in teacher turnover, thereby negatively impacting the stability and normal functioning of the educational system [11, 12]. Most existing research focuses on burnout and mental health issues among the general teaching population, with relatively little attention paid to the specific group of PE teachers [4, 13]. Therefore, further research on the occupational stress and burnout of PE teachers will not only help improve their job satisfaction and work efficiency but also contribute significantly to enhancing the overall quality of physical education.

The relationship between occupational stress and burnout

A considerable body of research indicates a significant positive relationship between occupational stress and burnout. Prolonged high-intensity occupational stress can lead to symptoms of burnout such as emotional exhaustion, depersonalization, and reduced personal accomplishment [8]. This relationship primarily arises because an individual’s psychological resources are overly depleted in high-pressure environments, rendering them unable to effectively meet work demands, leading to psychological and physical fatigue [14, 15]. Additionally, factors such as lack of job autonomy and social support in the work environment can exacerbate occupational stress, further increasing the incidence of burnout [16, 17]. However, some studies have pointed out that moderate occupational stress can stimulate individuals’ motivation, enhancing job performance and satisfaction, thereby reducing burnout [18, 19]. This phenomenon can be explained by the Yerkes-Dodson Law, which states that a moderate level of stress can enhance an individual’s attention and focus, contributing to optimized job performance [20]. Additionally, research by Cavanaugh’s team suggests that challenging job tasks can promote employees’ positive coping strategies and professional growth, thus reducing feelings of burnout [21]. Therefore, moderate stress can sometimes have a motivating effect and reduce the occurrence of burnout. Thus, while occupational stress can exacerbate burnout among primary and secondary school physical education teachers in some cases, its general effectiveness still needs further research and validation.

Mechanisms by which occupational stress influences burnout

According to the Stress-Strain-Outcome (SSO) model by Koeske et al. [22], stress induces strain (i.e., the destructive impact of stress on an individual’s psychological, physiological, and behavioral aspects), leading to various psychological, behavioral, and social problems as outcomes. In the SSO model, strain essentially acts as a mediating variable. Different studies have varied definitions of strain, with most considering it to be composed of cognitive and emotional elements [23, 24]. In this study, occupational stress is treated as the primary stressor and occupational burnout as the eventual outcome, with PE teachers’ psychological strain process situated in between. Following this theoretical framework, we posit that occupational stress triggers a cognitive-emotional chain reaction—first eliciting cognitive strain, then arousing emotional responses, and ultimately shaping burnout. Accordingly, we selected two representative strain variables, one cognitive and one emotional, for examination.

Rumination was chosen as the cognitive strain mediator because it is a common and prototypical negative cognitive response in stressful contexts. Rumination refers to the repetitive immersion in thoughts about negative events and emotions, whereby individuals find it difficult to disengage from continual thoughts about the stressor. A substantial body of research shows that rumination is a maladaptive coping style that prolongs and magnifies negative affect and occupational burnout [25]. Elevated daily stress is often accompanied by increased rumination, which intensifies subsequent negative emotion. Specifically within teaching, rumination has been shown to positively predict teachers’ burnout levels [26]. Rumination thus effectively captures the cognitive stress-response pathway, and positioning it as a mediator helps delineate how occupational stress influences negative emotion and burnout via cognition.Negative emotion represents the emotional-response mediator. According to stress-process theory, individuals under occupational stress commonly experience anxiety, depression, anger, and other negative affects. These emotions constitute the direct emotional expression of stress and form the affective component of strain. In work settings, persistent negative emotion drains teachers’ affective resources and gradually evolves into chronic fatigue and emotional exhaustion—core features of occupational burnout. Empirical evidence corroborates the tight link between negative emotions and burnout [27]. This connection indicates that negative emotion functions as a crucial bridge between stress and burnout. Consequently, treating negative emotion as a mediator effectively captures teachers’ typical emotional responses to stress and reveals how emotion channels the impact of stress on burnout.Together, rumination and negative emotion serve as representative cognitive and emotional mediators, aligning with the SSO model’s framework in which “stress leads to strain, leading to outcomes.” The resulting cognitive-emotional chain mediation path is logically clear: when PE teachers experience occupational stress, they first develop negative cognitions (ruminating on work difficulties), which in turn heighten negative emotions (e.g., frustration, anxiety), and the cumulative affective exhaustion eventually culminates in occupational burnout. Selecting these two variables helps operationalize abstract cognitive strain and emotional response—both rumination and negative emotion have well-validated measures and extensive empirical support—thereby enabling rigorous testing of the mediating mechanism through which stress exerts its effects.

In summary, the primary objective of this study is to address the lack of attention in previous research regarding the occupational stress and burnout of PE teachers, thereby filling this gap in the field. Due to the unique characteristics of their work—such as intense physical labor, student behavior management, and organizing extracurricular activities—PE teachers face distinctive sources of stress that significantly impact their occupational health and burnout. This paper consists of two sub-studies. Study 1 employs a multiverse analysis to investigate the robustness of the relationship between occupational stress and burnout in a large sample. Study 2 uses a diary method to examine whether rumination and negative emotions mediate the effects of occupational stress on burnout. Specifically, the following hypotheses are proposed: Study 1: There is a robust link between occupational stress and burnout (H1). Study 2: T1 occupational stress leads to increased T2 rumination and T2 negative emotions over time, which in turn positively affect T3 burnout among PE teachers. Moreover, rumination and negative emotions not only act as individual mediators (H2 and H3) but also form a sequential mediation (H4).

Study 1: relationship between occupational stress and burnout

Participants

The participants were from a large project focused on the mental health of primary and secondary school physical education teachers in mainland China. A cluster sampling method was employed, selecting 385 schools from 11 provinces and cities in the eastern and central regions of China in 2022. Surveys were distributed and collected at the school level. A total of 4800 questionnaires were distributed, and after excluding uncompleted or invalid responses, 4397 valid questionnaires were obtained. Among the participants, there were 2,863 males, accounting for 53.9% of the total. The average age was 30.81 years. The majority of respondents held a bachelor’s degree or higher, comprising 58.2%, followed by those with a junior college degree or below, who made up 41.8%. Participants with five years or less of teaching experience accounted for 52.2%, those with 6–10 years of teaching experience made up 26.1%, and those with more than 10 years of experience constituted 21.7%. The majority of participants came from public schools, representing 81.1% of the sample.Missing values were addressed using the EM imputation method, with a missing rate ranging from 0.30 to 8.20% and an average missing rate of 1.55%. Using the web calculator from Statistics Kingdom, it was found that a minimum sample size of 1230 was needed to achieve a statistical power of 0.80.

Measures

Occupational stress scale

The Occupational Stress Scale used in this study was developed by Lin Xiaoqun and colleagues, specifically designed for primary and secondary school physical education teachers [28]. The scale consists of five dimensions: student-related stress, parent-related stress, exam-related stress, work-related stress, and career-related stress, comprising a total of 27 items. Example items include: “The school leadership does not support physical education” and “Concern about injury accidents during physical education classes.” The scale adopts a 5-point Likert scoring system, with responses ranging from “very stressful” (5 points), “quite stressful” (4 points), “moderately stressful” (3 points), “somewhat stressful” (2 points), to “not stressful” (1 point). In this study, the Cronbach’s alpha coefficient for the scale was 0.82.

Teacher burnout scale

This study employed the Teacher Burnout Scale (TBS), developed by Wang Guoxiang and colleagues, which consists of 22 items covering three dimensions: emotional exhaustion, depersonalization, and reduced personal accomplishment [29]. Sample items include: “I feel exhausted after a day of work” and “I feel disheartened by working in early childhood education.” The scale uses a 5-point Likert scoring system, ranging from 1 (“never”) to 5 (“always”). Items in the reduced personal accomplishment dimension are reverse-scored, while items in the emotional exhaustion and depersonalization dimensions are positively scored. A higher total score indicates a higher level of burnout among teachers. In this study, the Cronbach’s alpha coefficient for the scale was 0.91.

Demographic questionnaire

A self-administered demographic questionnaire was used to collect participants’ gender, age, teaching experience, grade levels they teach, educational qualifications, and the type of school they work in .

Analysis procedures

Because all data in this study were collected via self-report questionnaires, common-method bias could threaten the reliability of the results. We therefore implemented multiple strategies at both the design and analytic stages.Design-stage procedural controls. Item order was randomized and several items were reverse-worded to minimize systematic error stemming from fixed sequencing or uniform wording.Analytic diagnostics. We conducted Harman’s single-factor test to determine whether common-method bias was serious. All measured items were entered into an unrotated exploratory factor analysis; multiple factors with eigenvalues greater than 1 emerged, and the first factor accounted for less than the conventional 40% threshold of total variance. Hence, the present data show no evidence of substantial common-method bias (see Supplementary Materials for details).

To reduce the subjectivity of single analytical decisions and to evaluate the robustness of our findings, we applied a multiverse-style analysis. This approach entails specifying all reasonable analytic choices—for example, different strategies for handling missing data, alternative variable codings, or varying sets of covariates—then running each analytic pipeline separately and synthesizing the resulting distribution of outcomes. Prior research demonstrates that multiverse-style analysis can reveal the stability or fragility of empirical findings and highlight sensitivity factors, thereby bolstering the credibility of conclusions (see Supplementary Materials for an extended rationale).

The data possess a “participant (Level 1) — class/school (Level 2)” hierarchical structure. Ignoring this dependency and fitting single-level regressions typically underestimates standard errors and inflates statistical significance. We therefore computed the intraclass correlation coefficient (ICC), which quantifies the proportion of total variance attributable to between-group rather than within-group differences [30]. The ICC calculation formula is as follows: ICC = τ₀₀ / (τ₀₀ + σ²), where τ₀₀ is the random-intercept variance and σ² is the residual variance. ICC values range from 0 to 1: values < 0.05 suggest negligible clustering, 0.05–0.15 indicate moderate clustering, and > 0.15 denote strong clustering [31]. For occupational stress and burnout, the null models yielded ICCs of 0.119 and 0.101, respectively, indicating that 11.9% and 10.1% of the total variance resided at the class/school level—above the common 0.05 benchmark and thus evidencing non-trivial between-group heterogeneity .All multiverse pipelines were estimated with multilevel models implemented in the specr package in R.

Results

Figure 1 displays the descriptive results of the multiverse-style analysis. Initially, there are 64 combinations of strategies for occupational stress and burnout (with 1 independent variable indicator, 6 covariate indicators, and 1 dependent variable indicator, as shown in Fig. 2). The predictive effects of occupational stress on burnout range from 0.018 to 0.038, and all 64 strategy combinations are statistically significant. Statistical inferences indicate that occupational stress positively predicts burnout, Median β = 0.030, p < 0.001; partial r² = 0.011;NSRPD(number of significant results in the predominant direction) = 64/64, p < 0.001。.

Fig. 1.

Mediation Diagram for Testing the Chain Mediation Effect

Fig. 2.

Combination of multiverse-style analysis strategies and their curves

Discussion

The relationship between teachers’ occupational stress and burnout has long been a central topic in educational psychology. Yet the empirical conclusions remain mixed, owing to methodological choices, control variables, and sample characteristics that differ across studies. Some scholars emphasize the moderating roles of contextual factors and individual resources [32], whereas others contend that the linkage between stress and burnout may be more complex [20, 21]. Consequently, the field still questions the robustness of the stress–burnout association, underscoring the need for more comprehensive and rigorous analytic strategies.Compared with earlier single-equation regression studies, Study 1 adopted a large cross-sectional sample and implemented statistical inference across multiple analytic strategy combinations through multiverse-style analysis, thereby minimizing subjective researcher decisions and revealing more generalizable patterns. The multiverse-style analysis results showed that occupational stress consistently and significantly predicted occupational burnout under every analytic specification. Such stability provides a solid empirical foundation for inferring the authenticity of the stress–burnout relationship. Logically, if a relationship appears only under a particular model or assumption but disappears or reverses once analytic conditions change, it is likely a statistical artifact or methodological bias. In contrast, the present findings indicate that, regardless of which covariates are controlled or which measurement indices are used, the effect of stress on burnout remains positive and significant. This consistent pattern suggests that the stress–burnout linkage is an inherent association reflected by the data rather than a by-product of analytic decisions.

The conclusion aligns with classic theoretical frameworks and extant empirical evidence. Both the JD-R model and the SSO model posit that high job demands deplete personal resources, leading to cumulative psychological strain that ultimately manifests as adverse outcomes such as burnout [22, 33]. Primary and secondary school physical education teachers face multiple occupational stresses in their work, mainly stemming from heavy teaching loads, challenges in student management, lack of career development opportunities, and high expectations from society and parents. These stress factors interact and lead to burnout among physical education teachers. Specifically, the heavy teaching tasks and intense physical labor place significant physical and psychological burdens on physical education teachers, leading to widespread emotional exhaustion due to prolonged high-pressure conditions [34, 35]. Furthermore, challenges in student management, including dealing with students’ misbehavior and maintaining classroom order, intensify teachers’ psychological stress and increase the risk of depersonalization [36]. Finally, physical education teachers have fewer opportunities for career advancement, lacking clear promotion pathways and professional recognition, which makes it difficult for them to feel a sense of accomplishment in their jobs, further exacerbating symptoms of burnout related to low personal accomplishment [37, 38].

Study 2: intrinsic mechanisms through which occupational stress affects burnout

Participants

In Anhui Province, China, 35 schools were randomly selected, and 347 primary and secondary school physical education teachers were invited to participate in a preliminary assessment of occupational stress (T1). They then took part in a 21-day follow-up survey using the diary method, where they reported daily rumination (T2) and negative emotions (T2). A post-test (T3) on burnout was conducted at the end of this period. Participants who failed to respond for more than three days at T2 or T3 were excluded from the study. Ultimately, 324 participants remained valid for analysis. Among the respondents, there were 158 males, accounting for 45.7% of the total. The average age was 29.56 years. The majority of respondents held a bachelor’s degree or higher, making up 55.9% of the sample. Respondents with five years or less of teaching experience represented the smallest proportion, at 43.7%. Most of the respondents came from public schools, accounting for 72.8% of the total.Missing values were handled using the EM imputation method, with a missing proportion ranging from 0.30 to 8.20%, and an average missing rate of 1.55%. Using the web calculator from Statistics Kingdom, it was determined that a minimum sample size of 307 was required to achieve a statistical power of 0.80.

Measures

Rumination questionnaire

The Daily Rumination Scale is adapted from the rumination subscale of the Cognitive Emotion Regulation Questionnaire [39]. Participants assess their own state based on their daily situations. The rumination scale consists of 4 items (e.g., “I am absorbed by thoughts and feelings about what I have experienced”), using a 5-point scoring system (1 = never, 5 = always). Higher scores indicate higher levels of rumination. In this study, the Cronbach’s alpha coefficient for this scale was 0.83.

Negative emotion questionnaire

Daily negative emotions are assessed using a single-item measure [40] (“How bad was your mood today? Score from 1 to 100, where 1 means not bad at all, and 100 means extremely bad”). The Cronbach’s alpha coefficient for this measure in this study was 0.88.

The tools used to measure occupational stress, occupational burnout, and demographic variables are the same as those used in Study 1.

Analysis procedures

Step 1, Common Method Bias(Same as Study 1), Descriptive and Correlational Statistics in SPSS 24.0 Initially, common method bias, as well as descriptive and correlational statistics, are assessed in SPSS 24.0. The results of these analyses are detailed in the supplementary materials.

Step 2, Testing the Mediating Role of Rumination and Negative Emotions Using PROCESS Model 6 To test the mediating effects of rumination and negative emotions, the SPSS macro PROCESS Model 6 is employed for sequential mediation effect analysis. This involves controlling for covariates and establishing a direct pathway from occupational stress to burnout among primary and secondary school physical education teachers. Rumination and negative emotions are then incorporated into the mediation model, constructing a path from rumination to negative emotions, thus forming a sequential mediation model. Statistical Testing Using Bootstrap The regression coefficients are tested for significance using Bootstrap resampling (5,000 samples), to obtain standard errors and 95% bias-corrected confidence intervals (CI) for parameter estimates. A p-value of less than 0.05 is considered statistically significant. The ICC values for occupational stress, rumination, negative emotions, and burnout are 0.011, 0.000, 0.014, and 0.027, respectively. According to McGraw et al.‘s standards [30], these values are less than 0.05, which suggests a low level of data nesting by groups. Therefore, a multilevel model is not necessary for this analysis.

Results

The model involved occupational stress as the independent variable, rumination and negative emotions as mediators, and burnout as the dependent variable, constructing a chained mediation model (Fig. 1). The results showed: T1 occupational stress significantly and positively predicted T3 burnout (β = 0.382, p < 0.001), confirming Hypothesis 1; T1 occupational stress significantly and positively predicted T2 rumination (β = 0.363, p < 0.001) and T2 negative emotions (β = 0.221, p < 0.05); T2 rumination positively predicted T2 negative emotions (β = 0.415, p < 0.001) and positively predicted T3 burnout (β = 0.379, p < 0.01); T2 negative emotions positively predicted T3 burnout (β = 0.458, p < 0.01).

Further examination of the mediation effects (see Table 1) indicated that the 95% confidence intervals of the total mediation effects of rumination and negative emotions on the impact of T1 occupational stress on T3 burnout did not include zero; the results demonstrated that T2 rumination and T2 negative emotions are mediators in the influence of T1 occupational stress on T3 burnout (effect size = 0.308, 95% CI[-0.219 ~ -0.138]). This mediation effect consists primarily of the following three pathways: (1) T1 occupational stress → T2 rumination → T3 burnout, with a confidence interval for the mediation effect that does not include zero, indicating a significant mediation effect (effect size = 0.138, 95% CI [0.040 ~ 0.236]), confirming Hypothesis 2; (2) T1 occupational stress → T2 negative emotions → T3 burnout, with a confidence interval for the mediation effect that does not include zero, indicating a significant mediation effect (effect size = 0.101, 95% CI[0.004 ~ 0.197]), confirming Hypothesis 3; (3) T1 occupational stress → T2 rumination → T2 negative emotions → T3 burnout, with a confidence interval for the mediation effect that does not include zero, indicating a significant mediation effect (effect size = 0.069, 95% CI [0.008 ~ 0.029]), confirming Hypothesis 4. These results suggest that T2 rumination and T2 negative emotions have an indirect impact on alleviating T3 burnout from T1 occupational stress among Chinese middle and primary school physical education teachers. Additionally, T2 rumination and T2 negative emotions further affect the impact of T1 occupational stress on T3 burnout through a chained mediation mechanism.

Table 1.

Bootstrap analysis for mediation effect test

Effect	Path Relationship	Effect	Boot SE	Boot LL CI	Boot UL CI
Direct Effect	T1 occupational stress→T3 burnout	0.382	0.016	0.284	0.481
Mediating Effect	T1 occupational stress→T2 rumination→T3 burnout	0.138	0.012	0.040	0.236
	T1 occupational stress→T2 negative emotions→T3 burnout	0.101	0.006	0.004	0.197
	T1occupational stress→T2 rumination→T2 negative emotions→T3 burnout	0.069	0.018	0.008	0.029
Total Mediating Effect		0.308	0.027	0.019	0.138
Total Effect		0.690	0.045	0.180	0.329

Discussion

Study 2 utilized the diary method to reveal the intrinsic impact mechanism between occupational stress and burnout from a time series perspective. Rumination plays a crucial mediating role between teachers’ occupational stress and burnout. When faced with occupational stress, teachers often repeatedly contemplate the setbacks and challenges in their work, which makes it difficult for them to detach from the sources of stress, leading to emotional exhaustion and burnout [41, 42]. Research indicates that rumination not only prolongs the duration of stress but also amplifies its negative effects, resulting in the individual’s inability to effectively recover and cope [43]. For example, when teachers continue to think about classroom management issues or student behavior problems after work, their psychological and physiological recovery processes are disrupted, preventing them from obtaining necessary rest and relaxation, which further intensifies burnout. Additionally, rumination is associated with high levels of anxiety and depression, which can further exacerbate the state of teachers’ burnout [44].

Study 2 also identified the mediating role of negative emotions in the relationship between occupational stress and burnout among middle and primary school physical education teachers. Specifically, occupational stress exacerbates burnout by elevating negative emotions in these teachers. Occupational stress leads to various negative emotions in teachers, such as anxiety, depression, and anger, which in turn intensify the symptoms of burnout [45]. Negative emotions not only directly impact teachers’ psychological health but also diminish their job satisfaction and professional identity, thereby increasing the risk of burnout [46]. Research shows that when faced with high workloads and complex student management issues, teachers often experience intense feelings of frustration and helplessness, which erode their psychological resilience and work motivation [47]. This emotional response not only affects teachers’ performance but also makes it difficult for them to recover and relax in their daily lives, creating a vicious cycle that further intensifies job burnout [48].

Study 2 established a chained mediation effect of rumination and negative emotions between occupational stress and burnout among middle and primary school physical education teachers, indicating that occupational stress first increases rumination in these teachers, which then enhances their negative emotions, ultimately aggravating burnout. The JD-R Model provides a robust framework for explaining this phenomenon. According to the JD-R model, high job demands and low resource availability in the work environment lead to the depletion of an individual’s psychological and physiological resources, which in turn triggers burnout [33]. In this model, the balance between job demands and resources determines the levels of work pressure and burnout experienced by employees. When teachers face high levels of job demands, their psychological resources are excessively consumed, leading to increased rumination and negative emotions [49]. Rumination involves repeatedly thinking about stress-related negative events, a pattern of thought that not only prolongs the impact of stress but also amplifies its negative effects, making it difficult for teachers to detach from sources of stress [50]. This persistent psychological burden further erodes teachers’ psychological resources, leading to emotional exhaustion and depersonalization, both key components of burnout. Negative emotions, as a form of psychological strain, are the direct result of high job demands and rumination. Such emotions, including anxiety, depression, and anger, not only affect teachers’ psychological health but also diminish their job satisfaction and professional identity, thereby intensifying burnout [51].

Notably, the direct effect of the independent variable (occupational stress) on the dependent variable (burnout) remains significant after the mediators are entered, indicating partial mediation. Theoretically, this finding means that the impact of occupational stress on burnout is not transmitted solely through rumination and negative emotion. Although these two mediators significantly convey the effect, they do not exhaust all operative mechanisms; occupational stress itself still retains a direct influence on burnout. On the one hand, high levels of occupational stress may directly drain teachers’ physical and psychological resources (e.g., physical overexertion, emotional exhaustion), thereby triggering burnout without necessarily passing through rumination or emotional change. This aligns with Conservation of Resources theory: when stress is excessive and resources scarce, individuals suffer resource loss and energy depletion that directly precipitate burnout [52]. On the other hand, the presence of partial mediation implies additional unmeasured mediators or pathways—for example, work-environment factors (such as low social support or role conflict) or personal factors (such as coping style or physical fatigue) that intervene in the stress–burnout link. When such factors are not controlled, a residual direct effect of occupational stress on burnout is to be expected. This residual effect is consistent with the complex, multifaceted etiology of occupational burnout: among primary and secondary school physical education teachers, burnout is not shaped by a single pathway but rather by the confluence of multiple processes [10, 53].

General discussion

This study explored the robust relationship between occupational stress and burnout among middle and primary school physical education teachers, as well as the mediating roles of rumination and negative emotions. Study 1 confirmed Hypothesis 1—that there is a robust and practically significant relationship between occupational stress and burnout, warranting attention from researchers and society. Study 2 confirmed Hypotheses H2, H3, and H4. Specifically, baseline occupational stress triggered an increase in rumination and negative emotions over time, which in turn exacerbated burnout at follow-up, with rumination and negative emotions also forming a chained mediation.

In this research, by integrating the SSO model, we further enhanced the understanding of the pathways influencing burnout. Through the diary method design, this study revealed how rumination and negative emotions act as critical mediating links in the strain process, transmitting occupational stress to burnout. Particularly in daily teaching and teacher work pressures, this continuous psychological burden is not limited to short-term psychological stress responses but forms a long-term psychological stress state, which has a direct and profound impact on teachers’ burnout. By conducting a detailed study of the occupational stress and burnout among middle and primary school physical education teachers, this research not only expands the existing theories in educational psychology and occupational health psychology but also provides empirical foundations and strategic recommendations for practically addressing stress issues in educational settings.

This study inevitably has some limitations. First, it relied on self-report methods, which may introduce subjective biases. Second, there may be complex interactions between occupational stress, rumination, and negative emotions, rather than merely a unidirectional chained mediation relationship. For instance, in this study, the interaction between rumination and negative emotions could impact burnout, a topic that might be further explored in future research. Lastly, although multiverse-style analysis and longitudinal mediation provide effective methods for exploring causal relationships between variables, caution should still be exercised when interpreting ultimate causal relationships. Even with advanced statistical models, it is challenging to entirely eliminate all potential confounding variables or definitively prove causal relationships. In future research, we plan to employ rigorous experimental designs to explore the causal relationships between variables.

Conclusion

This research, through two studies, explored the impact of occupational stress on burnout among middle and primary school physical education teachers and its underlying mechanisms. Occupational stress not only directly exacerbates the burnout issue in these teachers but can also achieve the same effect by elevating rumination and negative emotions.

By accurately identifying and targeting key mediating variables of burnout, schools and education administrators can implement various strategies to help PE teachers manage rumination and negative emotions, thereby improving their job satisfaction and work efficiency. These strategies include, but are not limited to, providing mental health training and counseling to help teachers recognize and cope with rumination; organizing mindfulness training and offering emotional management skills workshops to teach teachers how to identify and manage negative emotions; promoting work-life balance by reducing administrative burdens and encouraging leisure activities to alleviate stress. Systematic interventions like these can enhance PE teachers’ job satisfaction and teaching quality, ultimately promoting students’ physical health and psychological development more effectively.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

PE

Physical education

SSO

Stress-strain-outcome

OS

Occupational stress

TE

Teaching experience

EQ

Educational qualification

GL

Grade levels

TK

Type of kindergarten

NSRPD

Number of significant results in the predominant direction

JD-R

Job demands-resources

CI

Confidence intervals

Author contributions

XL: conceptualization, writing. HX and XC: analyzed and interpreted the data. JZ: funding acquisition. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the Fundamental Research Funds for the Central Universities (2024CDJSKXYTY03).

Data availability

Data will be made available on request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the ethical standards of Hefei College of Finance and Economics’s Review Board and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. All participants provided their written informed consent to participate in this study. Furthermore, explicit consent was obtained to publish the data gathered from the participants. The consent forms used specifically stated that the participants acknowledged their anonymized data could be used for scientific publication. This procedure was approved by Sichuan Institute of Industrial Technology’s Ethics Committee(number 2022J51).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.