The relationship between coaching leadership practices and athlete burnout among college athletes in China: the mediating role of psychological resilience and emotion regulation Authors: Xiaoying Chen, Yuan Yue, Jun Xiang, Fang Zhang Published: 2026 ABSTRACT AbstractAs a key indicator in sports psychology reflecting athletes’ complex physiological and psychological responses to challenges or threats, athlete burnout has become a focal issue in research on athlete-coach interactions and athletic performance.Objective To investigate the impact of coaches’ leadership behaviors on athlete burnout among Chinese collegiate student-athletes and to validate the mediating roles of psychological resilience and emotion regulation in this relationship.Method Using convenience sampling combined with cluster sampling, 519 student athletes from multiple Chinese universities were selected as participants. The study employed the Coach Leadership Behavior Scale, the Burnout Scale, the Psychological Resilience Scale, and the emotion regulation Difficulty Scale for measurement. Data analysis utilized SPSS 21.0 and the Hayes Process macro for correlation analysis and mediation effect testing.Results (1) Significant gender differences emerged: Male athletes exhibited lower levels of psychological resilience and emotion regulation difficulties (i.e., stronger regulation abilities), with significantly lower sports burnout scores than female athletes (P < 0.05). No significant differences were found based on age or athletic level (P > 0.05). (2) Coach leadership behaviors showed a significant negative correlation with athletic burnout (R = -0.543, P < 0.01) and exerted a direct negative predictive effect on burnout (β = -0.276, t = -7.726, P < 0.001). Positive leadership behavior significantly enhanced psychological resilience (β = 0.615, t = 12.956, P < 0.001) and reduced emotion regulation difficulties (β =-0.472, t=-7.929, P < 0.001). Psychological resilience positively predicted emotion regulation levels (β = -0.513, t = 10.696, P < 0.001) and indirectly reduced burnout (β = -0.116, t = -3.857, P < 0.001); Reduced emotion regulation difficulties also significantly decreased burnout (β = -0.160, t = -6.394, P < 0.001); (3) Mediational analysis revealed that psychological resilience and emotion regulation difficulties jointly mediated the relationship between leadership behaviors and athlete burnout. Direct effects accounted for 58.35% of the total effect, while indirect effects accounted for 41.65%. Examining individual pathways: psychological resilience alone mediated 15.01%, emotion regulation difficulties alone mediated 15.86%, and their chained mediation effect accounted for 10.78%. This indicates that positive leadership behaviors alleviate sports burnout through dual pathways: enhancing psychological resilience and reducing emotion regulation difficulties.Conclusion Coaches’ positive leadership behaviors significantly reduce athletes’ burnout levels by enhancing psychological resilience and reducing emotion regulation difficulties. Psychological resilience and emotion regulation difficulties play crucial chain-mediated roles in this mechanism. This finding expands the theoretical implications of leadership behaviors and mental health in sports psychology, providing an empirical foundation for psychological promotion, fatigue management, and performance enhancement among college athletes. Future research should integrate longitudinal tracking, multi-source data, and AI-based behavioral analysis methods to enhance model explanatory power and generalizability, thereby revealing the dynamic mechanisms linking sports leadership to athletes’ psychological adaptation.Keywords: Coaching leadership behavior, Athlete burnout, Psychological resilience, Emotion regulation, Chain mediation, College athletes Introduction University athletes in China increasingly face substantial pressures as they simultaneously fulfill the roles of students and competitive performers. Intensive training schedules, academic responsibilities, and competitive expectations often place athletes under sustained psychological strain [1]. Prolonged exposure to such demands may lead to athlete burnout, typically characterized by emotional and physical exhaustion, a diminished sense of sport value, and reduced feelings of accomplishment [2, 3]. Athlete burnout not only impairs sport performance but may also negatively affect athletes’ psychological well‑being and academic persistence. Consequently, understanding the psychological mechanisms that contribute to athlete burnout has become an important issue in sport psychology and athlete development research. Among the contextual factors shaping athletes’ experiences, coaching leadership behaviors are particularly influential. Within university sport systems, coaches not only organize training and competition but also shape the motivational climate and emotional environment experienced by athletes. Leadership styles emphasizing strict control may enhance discipline but can simultaneously suppress athletes’ autonomy and interpersonal connectedness, thereby increasing burnout risk [4]. In contrast, autonomy‑supportive leadership behaviors-such as encouraging participation in decision‑making and providing emotional understanding-have been shown to enhance motivation quality and psychological well‑being [5]. However, despite growing recognition of the importance of coaching leadership, the psychological mechanisms through which leadership behaviors influence athlete burnout remain insufficiently clarified. Self‑Determination Theory (SDT) provides an important theoretical framework for understanding how social environments influence individuals’ motivation and psychological functioning [6]. According to SDT, optimal psychological development depends on the satisfaction of three basic psychological needs: autonomy, competence, and relatedness. When these needs are fulfilled, individuals are more likely to experience intrinsic motivation, psychological resilience, and emotional well‑being; when these needs are frustrated, individuals are more vulnerable to stress, disengagement, and burnout [7, 8]. Within sport contexts, supportive coaching leadership can facilitate need satisfaction by promoting autonomy support, competence development, and positive interpersonal relationships. These leadership behaviors may therefore foster internal psychological resources that help athletes adapt to stress and maintain psychological stability. In particular, psychological resilience reflects an individual’s capacity to adapt to adversity and sustain psychological functioning under pressure, whereas emotion regulation refers to the strategies individuals use to manage emotional responses in challenging situations. From a psychological resource perspective, resilience may serve as a foundational capacity that enables athletes to employ more adaptive emotion regulation strategies when facing training stress and competitive pressure. Despite the relevance of these processes, previous research has often examined psychological resilience and emotion regulation independently when explaining athletes’ psychological adjustment. As a result, limited attention has been given to how these mechanisms may operate together in linking coaching leadership to athlete burnout. Building on SDT and resource‑based perspectives, the present study proposes a sequential psychological mechanism in which coaching leadership behaviors influence athlete burnout through the progressive pathway of psychological resilience and emotion regulation. Specifically, supportive leadership behaviors may strengthen athletes’ psychological resilience by satisfying their basic psychological needs and promoting adaptive cognitive appraisals of stress. Enhanced resilience may subsequently facilitate more effective emotion regulation strategies, which ultimately reduce emotional exhaustion and burnout. By integrating leadership behavior, psychological resources, and emotional coping processes into a unified analytical framework, this study aims to clarify the psychological mechanisms underlying athlete burnout among Chinese university athletes and provide theoretical and practical implications for coaching leadership and athlete psychological support. Literature review and research hypotheses Coaching leadership behavior and athlete burnout Scholars have noted a relationship between coaching leadership behavior and athlete burnout. First, coaching leadership behaviors typically encompass multiple styles such as supportive, democratic, and authoritarian. Different leadership approaches profoundly influence athletes’ psychological states and behavioral performance [6]. From a psychological mechanism perspective, based on Self-Determination Theory (SDT), coaches’ supportive and democratic leadership behaviors can promote athletes’ psychological needs fulfillment (autonomy, competence, and relatedness). This enhances their intrinsic motivation and well-being while reducing the risk of burnout. Conversely, authoritarian and high-pressure leadership tends to cause psychological need frustration, leading to emotional exhaustion and diminished sense of accomplishment under prolonged stress and lack of autonomy [5, 6]. Secondly, according to the stress-coping theory, athletes’ cognitive assessments of situations and coping strategies when facing high-intensity training and competitive pressures are significantly influenced by coaching leadership styles. Supportive leadership helps athletes perceive stress as a challenge, stimulating positive coping strategies; authoritarian leadership, however, may amplify threat assessments, inducing chronic psychological stress and burnout responses [9]. Concurrently, extensive empirical research confirms that under conditions of psychological need satisfaction and positive stress appraisal, athletes exhibit significantly reduced levels of emotional exhaustion and depersonalization [10]. Moreover, coaches’ leadership behaviors not only directly impact athletes’ mental health but also indirectly influence training outcomes and competitive performance by regulating motivation quality, training commitment, and team atmosphere [4]. Thus, the need-satisfaction–need-frustration mechanism and the stress-evaluation pathway jointly explain the psychological link between coaching leadership and athletic burnout, providing a more systematic theoretical framework for understanding this relationship among collegiate athletes. In summary, coaches adopting supportive and democratic leadership styles can promote athletes’ psychological needs fulfillment and positive stress perception, thereby reducing risks of emotional exhaustion and diminished sense of accomplishment, ultimately decreasing sports burnout incidence. Thus, Hypothesis 1 is proposed: coaching leadership behaviors exert a significant negative predictive effect on sports burnout among college athletes. Mediating role of psychological resilience One of the mediating mechanisms in this study is the mediating effect of psychological resilience. Psychological resilience, as a key psychological resource enabling individuals to maintain positive adaptation when facing stress, setbacks, and challenges, has garnered significant attention in the field of sports psychology in recent years. Existing research indicates that coaches’ leadership behaviors can significantly influence the formation and development of athletes’ psychological resilience. Specifically, supportive and democratic leadership styles foster a positive training environment, satisfying athletes’ three psychological needs: autonomy, competence, and relatedness [11]. According to self-determination theory, when fundamental psychological needs are met, individuals exhibit heightened motivation and psychological resilience. By encouraging autonomous decision-making, providing emotional support, and offering positive feedback, coaches can strengthen athletes’ intrinsic commitment to training goals, thereby enhancing their self-efficacy and stress resilience. Conversely, neglectful or authoritarian leadership leads to unmet needs, causing athletes to perceive stress as threatening and consequently diminishing their psychological resilience levels [12]. Regarding the relationship between psychological resilience and burnout, research indicates that psychological resilience serves as a crucial psychological buffer against burnout. According to the stress-resource model, individuals possessing sufficient psychological resources (such as resilience) under high training loads can reduce burnout incidence through positive stress reappraisal and effective emotion regulation mechanisms. Athletes with high psychological resilience tend to perceive training stress as a growth challenge rather than a threat, thereby maintaining emotional stability and a sense of accomplishment [13]. Conversely, those with lower psychological resilience are more likely to fall into a vicious cycle of emotional exhaustion and diminished achievement. In summary, psychological resilience serves not merely as a simple mediating variable in the “coach leadership behavior–athletic burnout” causal chain, but as a key moderating pathway linking “external social support–internal coping mechanisms–psychological outcomes.” Supportive leadership fosters psychological resilience by fulfilling athletes’ psychological needs and facilitating positive stress reappraisal mechanisms. High psychological resilience, in turn, effectively mitigates the negative impact of training loads and external pressures on mental health, thereby reducing the risk of burnout. Therefore, this study proposes the following hypothesis 2:Psychological resilience significantly mediates the relationship between coaching leadership behaviors and athletic burnout among college athletes. Specifically, coaches’ positive leadership behaviors enhance athletes’ psychological resilience by fulfilling their psychological needs and promoting positive stress reappraisal, thereby indirectly reducing the occurrence of athletic burnout. Mediating role of emotion regulation The second mediating mechanism in this study is the mediating role of emotion regulation. emotion regulation refers to an individual’s ability to achieve psychological balance and adaptability by controlling their emotional responses when facing stress and challenges. According to stress-coping theory and self-determination theory, emotion regulation is not merely passive response control but a crucial process through which individuals satisfy fundamental psychological needs and actively evaluate stressors. When athletes engage in positive emotional appraisal and cognitive restructuring of training stress or setbacks, they are better able to maintain intrinsic motivation while reducing psychological exhaustion and devaluation of sport [14]. Existing research indicates that emotion regulation capacity is closely linked to athlete burnout. Athletes with strong emotion regulation abilities maintain psychological adaptation during high-intensity training. They effectively manage negative emotions through positive strategies like cognitive reappraisal and self-soothing, significantly reducing emotional exhaustion and burnout levels [8]. In contrast, athletes relying on negative strategies such as emotional suppression often experience heightened demand frustration and perceived lack of control, substantially increasing their risk of burnout [15]. Thus, emotion regulation can be regarded as a critical psychological resource for sustaining psychological needs fulfillment and motivational stability, serving as an important protective mechanism against exercise fatigue. Concurrently, coaches’ leadership styles play a central role in emotion regulation mechanisms. Within the framework of self-determination theory, supportive and democratic leadership behaviors promote athletes’ autonomy and relatedness fulfillment, providing a safe space for emotional expression and problem-solving [16]. For instance, research indicates that coaches who foster an open, trusting training environment and implement positive feedback mechanisms can enhance athletes’ emotional recognition and reappraisal, thereby improving their emotional adaptability [17]. Conversely, authoritarian or high-pressure leadership behaviors tend to frustrate psychological needs, inhibit emotional expression and problem-solving, and increase athletes’ reliance on negative emotion strategies, exacerbating mental exhaustion and exercise fatigue [18]. Collectively, coaches’ supportive leadership behaviors can enhance athletes’ psychological needs fulfillment and positive emotion appraisal capabilities, thereby improving their emotion regulation levels and reducing sports burnout [19]. Therefore, this study proposes Hypothesis 3: emotion regulation plays a significant mediating role between coaches’ leadership behaviors and sports burnout among college athletes. Chain mediation effect of psychological resilience and emotion regulation Although psychological resilience and emotion regulation have both been demonstrated to exert independent protective effects against athletic stress and burnout, the present study conceptualizes them as sequentially related psychological processes rather than merely parallel mediators. Psychological resilience reflects an individual’s capacity to adapt to adversity and maintain psychological stability under stress, representing a foundational psychological resource for coping with demanding situations. In contrast, emotion regulation refers to the strategies individuals employ to manage and modify their emotional responses during stressful experiences. Conceptually, resilience may provide the psychological foundation that facilitates the effective use of emotion regulation strategies.Existing research suggests that the effectiveness of emotion regulation often depends on an individual’s psychological flexibility and positive belief systems; individuals with higher levels of psychological resilience are more likely to adopt adaptive emotional strategies, such as cognitive reappraisal and constructive coping [8, 20]. Thus, resilience may function as an antecedent psychological resource that supports subsequent emotion regulation processes, allowing internal psychological resources to be translated into effective emotional management. Coaches’ leadership behaviors may exert indirect effects on athletic burnout by influencing athletes’ psychological and emotional responses to stress. Rather than representing two independent pathways, the present study proposes a sequential psychological mechanism linking psychological resources to emotional regulation outcomes. According to self-determination theory and cognitive appraisal models, supportive leadership behaviors satisfy athletes’ three fundamental psychological needs—autonomy, competence, and relatedness—thereby promoting the accumulation of psychological resources and strengthening psychological resilience [12, 16]. Athletes with higher psychological resilience are more likely to perceive training pressures and failures as challenges rather than threats, which facilitates the activation of adaptive emotion regulation strategies such as cognitive reappraisal and problem‑focused coping [20].Empirical studies further indicate that psychological resilience not only influences emotional states directly but also enhances psychological adjustment by promoting positive emotion regulation mechanisms [21]. For example, athletes with higher levels of resilience are more likely to engage in constructive emotion regulation strategies—such as reframing stressful events and reinforcing positive self‑affirmations—thereby significantly reducing their levels of burnout [21, 22]. These findings suggest that resilience may operate as an upstream psychological resource that enables more adaptive emotional regulation in stressful sport contexts. From the perspective of conservation of resources theory, psychological resilience functions as a core psychological resource that helps individuals withstand stress and prevent emotional exhaustion, while adaptive emotion regulation serves as a mechanism that maintains and replenishes these psychological resources. Accordingly, supportive coaching leadership may first enhance athletes’ psychological resilience by satisfying their psychological needs and encouraging positive cognitive appraisals of stressful situations. Enhanced resilience may subsequently facilitate more effective emotion regulation, ultimately reducing the likelihood of athlete burnout.Unlike a parallel mediation perspective, the present study proposes that psychological resilience precedes emotion regulation because resilience represents a broader psychological resource that supports athletes’ capacity to manage emotional responses in demanding sport environments. Based on the above theoretical and empirical reasoning, this study proposes Hypothesis 4: Psychological resilience and emotion regulation exert a chain-mediated effect between coaches’ leadership behaviors and athletic burnout among college athletes. In summary, scholars have identified the association between coaches’ leadership behaviors and athlete burnout, emphasizing the mediating and chain-mediating roles of psychological resilience and emotion regulation in this relationship. These findings provide theoretical support for the hypotheses of this study. Thus, the conceptual framework of this research is established (Fig. 1): (1) To examine the predictive role of coaches’ leadership behaviors on athlete burnout among Chinese college athletes; (2) To examine the mediating role of psychological resilience between coaching leadership behaviors and athlete burnout among Chinese university athletes; (3) To examine the mediating role of emotion regulation between coaching leadership behaviors and athlete burnout among Chinese university athletes; (4) To test the chain mediating role of psychological resilience and emotion regulation between coaching leadership behaviors and athlete burnout among Chinese university athletes. Research framework diagram Research participants and methods Research participants This study employed a combination of convenience sampling and cluster sampling to conduct centralized questionnaire testing among college athletes in physical education classes across multiple universities in China. Questionnaire distribution and administration were uniformly implemented by professionally trained physical education instructors and psychology majors within the classroom teaching context. As sample acquisition depended on course schedules and voluntary participant enrollment, strict random sampling conditions were not met.A total of 522 questionnaires were distributed, with 519 valid responses collected, yielding a response rate of 99.63%. Assessments were conducted collectively under standardized guidance to ensure procedural consistency. Inclusion criteria were: (1) current university athlete status; (2) participation in at least three provincial-level or higher athletic competitions within the past five years. Exclusion criteria included: missing data, non-university athlete status, or fewer than three competitions within the past five years. The subjects ranged in age from 17 to 24 years (mean age M = 20.00, standard deviation SD = 1.47), comprising 348 males (67.1%) and 171 females (32.9%). Athletic grade distribution was as follows: International Master 10 (1.9%), National Master 33 (6.4%), First-Class Athlete 153 (29.5%), Second-Class Athlete 106 (20.4%), Third-Class Athlete 217 (41.8%). No significant differences existed between gender, age, and athletic grade. Given that participants took the test within their classes and schools, the sample data may exhibit clustering effects. Since the test was anonymous, class or team codes could not be identified in the data, preventing direct control for cluster-wise standard errors in the model. To assess this potential influence, this study calculated intraclass correlation coefficients (ICC) and design effects. Results showed low ICC values (< 0.02), indicating minimal clustering effects. However, the authors acknowledged in the discussion that standard errors may be slightly underestimated, potentially leading to a marginally increased Type I error rate.Sample size was estimated using the formula N = Z² × [P × (1 − P)/E²] [23]. This study was approved by the Ethics Committee of Zhaoqing University (Approval No.: 2025030). All participants signed written informed consent forms detailing the study objectives, confidentiality principles, voluntary participation procedures, and researcher contact information. Data collection was completed between May 31, 2025, and June 20, 2025. Research methods Psychometric methods Coach Leadership Behavior Scale The Sports Leadership Behavior Scale developed by Chelladurai et al. [24], subsequently translated and adapted by Chinese scholar Yijun-Qiu [25], was employed to assess coaches’ leadership behavior levels within the Chinese sports cultural context. This 25-item scale retains the five dimensions established by Chelladurai: Training Guidance Behavior (e.g., “My coach pays attention to whether athletes are performing to their potential”), Democratic Behavior (e.g., “My coach obtains team approval before making important decisions”), Authoritarian Behavior (e.g., “My coach does not need to explain their actions”), Positive Feedback Behavior (e.g., “My coach informs athletes when they perform exceptionally well”), and Social Support Behavior (e.g., “My coach personally supports athletes”). The scale employs a 5-point Likert scoring method (1 = never, 2 = rarely, 3 = sometimes, 4 = often, 5 = always). The total score reflects the level of coaching leadership behavior, with higher scores indicating greater frequency of specific behaviors. Previous studies have demonstrated the Coach Leadership Behavior Scale’s reliability and validity when measuring coaching leadership among Chinese student-athletes [26]. In the sample of this study, the internal consistency coefficients for each sub-dimension were as follows: Training and Guidance (α = 0.78), Democracy (α = 0.73), Positive Feedback (α = 0.76), Social Support (α = 0.74), and Authoritarian Behavior (α = 0.68). The overall reliability coefficient was α = 0.681, falling within the acceptable range and reflecting the structural heterogeneity among different leadership style dimensions in this scale. Exploratory factor analysis yielded KMO = 0.736 and Bartlett’s sphericity test p < 0.001. A five-factor structure consistent with theory emerged, with item loadings ranging from 0.52 to 0.83 and no significant cross-loadings, validating the scale’s construct validity. Furthermore, based on theoretical consistency and analytical objectives, this study combined the dimensions of coaching, democracy, positive feedback, and social support into the “Autonomous Supportive Leadership Index” (α = 0.82) to measure coaches’ supportive and motivational leadership behaviors. Authoritarian behaviors were analyzed separately as the “Controlling Leadership Style” variable. All relevant regression and mediation analyses were conducted using SPSS 21.0 and the Hayes PROCESS plugin. While structural equation modeling was not employed, this methodology ensures robust results and reliable interpretation of validity. This approach balances theoretical foundations with statistical feasibility, providing a credible basis for examining the relationship between coaching leadership behaviors and athlete motivation. Athletes Burnout Scale The Athletes’ Burnout Questionnaire (ABQ) developed by Raedeke and Smith (2001) was used to assess athletes’ burnout symptoms [27]. Subsequently, the Chinese version of the ABQ, translated and revised by Zuo-song Chenand Ai-guang Zhou(2014) and validated among Chinese athletes, demonstrated sound psychometric properties [28]. The scale comprises 15 items scored on a 5-point Likert scale (1 = Strongly Disagree, 2 = Somewhat Disagree, 3 = Undecided, 4 = Somewhat Agree, 5 = Strongly Agree) and measures three dimensions: Reduced Sense of Achievement (e.g., “I feel extremely fatigued during training and competitions”), Emotional and Physical Exhaustion (e.g., “Training and competitions make me feel averse”), and Depreciation of Sport (e.g., “I don’t care whether I perform at my expected level”). The total score is calculated by summing the item scores, with higher scores indicating greater levels of burnout. Previous studies have employed the Sport Burnout Scale to measure burnout among Chinese athletes, demonstrating good reliability and validity [29]. In this study, all subscales and the total scale exhibited good internal consistency: Emotional/Physical Exhaustion (α = 0.781); Reduced Sense of Accomplishment (α = 0.743); Depreciation of Sport (α = 0.769); Total Scale (α = 0.752). Additionally, exploratory factor analysis (EFA) was conducted using SPSS 21.0. The EFA results indicated a reasonable scale structure with good construct validity (KMO = 0.851). KMO values ranged from 0.736 to 0.873, and Bartlett’s sphericity test was significant (p < 0.001). The extracted factors aligned with the theoretical structure, with item loadings ranging from 0.52 to 0.83 and no significant cross-loadings, indicating the scale possesses good construct validity. Psychological Resilience Scale The Psychological Resilience Scale developed by Connor et al. [30] was adopted and subsequently translated and adapted by Chinese scholars Xiao-nan Yu et al. [31] to assess psychological resilience levels across various groups within the Chinese sports culture context. The scale comprises 25 items across three dimensions: Resilience (e.g., “No matter what happens, I can handle it”), Strength (e.g., “I consider myself a strong person”), and Optimism (e.g., “I take pride in my achievements”). The scale employs a 5-point Likert scale (1 = strongly disagree, 2 = somewhat disagree, 3 = unsure, 4 = somewhat agree, 5 = strongly agree). Total scores are calculated by summing individual item scores, with higher scores indicating greater psychological resilience. In previous studies, the Psychological Resilience Scale has been used to measure psychological resilience among Chinese university students, demonstrating good reliability and validity [32]. In this study, the Cronbach’s α coefficient for the scale was 0.769. The reliability coefficients for each subscale were: Resilience α ≈ 0.812, Strength α ≈ 0.753, Optimism α ≈ 0.702, indicating high internal consistency and good reliability of the questionnaire. Exploratory factor analysis results confirmed a sound scale structure (KMO value: 0.873). The KMO value ranged from 0.736 to 0.873, and Bartlett’s sphericity test was significant (p < 0.001). The extracted factors aligned with the theoretical structure, with item loadings ranging from 0.52 to 0.83 and no significant cross-loadings. This indicates the sample data was highly suitable for factor analysis to validate the scale structure, suggesting good construct validity. Emotion Regulation Scale The Emotion Regulation Difficulty Questionnaire developed by Gratz et al. [33] was adopted to assess emotion regulation difficulties. Subsequently, the Emotion Regulation Scale translated by Chinese scholars was used to test individual emotion regulation levels in the Chinese context [34]. This 36-item scale comprises six dimensions: Difficulty Understanding Emotions (My feelings confuse me), Difficulty Controlling Impulses (When upset, I lose control), Difficulty Staying Goal-Oriented (When upset, I still manage to complete tasks), Difficulty Accepting Emotional Responses (When upset, I feel unable to improve my mood), Difficulty Effectively Using Regulation Strategies (When upset, it takes me a long time to feel better). This scale employs a 5-point Likert scoring method (1 = almost never, 2 = sometimes, 3 = about half the time, 4 = most of the time, 5 = almost always). The total score is calculated by summing the 36 items, with higher scores indicating greater difficulty in emotion regulation and lower emotion regulation ability. To maintain conceptual consistency in statistical analysis and result interpretation, the emotion regulation variable is uniformly referred to as “emotion regulation difficulty” throughout the text, where lower scores represent stronger emotion regulation abilities. In previous studies, the emotion regulation Difficulty Scale has been used to measure emotion regulation difficulties among Chinese university students, demonstrating good reliability and validity [35]. In this study, the Cronbach’s Alpha coefficient for the scale was 0.809, with all subscales demonstrating good reliability (emotional awareness difficulty α = 0.79, impulse control difficulty α = 0.81, goal-directed difficulty α = 0.83, etc.), indicating strong internal consistency within this sample. Exploratory factor analysis results confirmed a sound scale structure (KMO value = 0.872). KMO values ranged from 0.736 to 0.873, with Bartlett’s sphericity test significant (p < 0.001). The extracted factors aligned with the theoretical structure, and item loadings ranged from 0.52 to 0.83 with no significant cross-loadings, indicating the scale possesses good construct validity. Ethical considerations This study strictly adhered to the ethical guidelines of the Declaration of Helsinki (1964) and its subsequent revisions throughout its design and implementation. It was approved by the Ethics Review Committee of Zhaoqing University (Approval No.: 2025030). Prior to data collection, the research team fully explained the study’s background, objectives, data confidentiality principles, and participants’ right to withdraw to all potential participants, ensuring their informed and voluntary participation. No personally identifiable information was collected during the research process. Anonymous recording and secure storage measures were employed, with data used solely for academic research analysis. Deceptive methods were never utilized, nor were any situations involving potential physical or psychological risks to participants encountered. This approach effectively safeguarded participants’ rights to informed consent, privacy, and safety.It is worth noting that the sample size of this study (N = 519) meets the basic requirement of 5 to 10 times the number of items per factor in factor analysis, ensuring sufficient testing power for the primary regression and mediation effect analyses. Furthermore, as all statistical analyses were conducted using SPSS and the Hayes PROCESS macro—which does not support structural equation modeling—confirmatory factor analysis (CFA) was not performed, nor were model fit indices (e.g., CFI/TLI, RMSEA, SRMR) reported. Despite the absence of CFA validation, the factor structures and reliability results of all scales align with existing Chinese-language validation studies, supporting their applicability within the Chinese athlete sample. Mathematical statistics methods Statistical analysis of the study data was conducted using SPSS 26.0 software, AMOS structural equation modeling software, and the PROCESS macro developed by Hayes. First, descriptive statistics were performed in SPSS 26.0 on demographic information and test data for coaching leadership behavior, psychological resilience, emotion regulation, and athlete burnout, clearly presenting the basic characteristics and distributions of each variable. Differences between groups were assessed through comparison tests. Second, to further examine the construct validity and structural stability of the measurement instruments, confirmatory factor analysis CFA was conducted using AMOS to validate the measurement model including coaching leadership behavior, psychological resilience, emotion regulation difficulties, and athlete burnout. The CFA was performed based on the maximum likelihood estimation method to test whether the observed variables adequately represented the latent constructs. Model fit was evaluated using several commonly recommended indices, including the chi-square to degrees of freedom ratio χ² df, the comparative fit index CFI, the Tucker-Lewis index TLI, the root mean square error of approximation RMSEA, and the standardized root mean square residual SRMR. Generally, χ² df values below 3.0, CFI and TLI values above 0.90, and RMSEA and SRMR values below 0.08 indicate an acceptable model fit, suggesting that the measurement model demonstrates satisfactory construct validity. Third, to address potential common method variance, the Harman single-factor test was applied using SPSS 26.0. Fourth, Pearson bivariate correlations were examined using SPSS 26.0 to clarify pairwise relationships among coaching leadership behavior, psychological resilience, emotion regulation, and athlete burnout among college athletes. Finally, the PROCESS plugin Model 6 combined with Bootstrap sampling 5000 iterations was employed to examine the chained mediating effects of coaching leadership behavior, psychological resilience, emotion regulation, and athlete burnout. A significance level of P < 0.05 was set to ensure the scientific reliability of the research conclusions. Results Descriptive statistics of coaching leadership behavior, burnout, psychological resilience, and emotion regulation To comprehensively present the psychological and behavioral profiles of the research sample, descriptive statistics are provided for the distribution of four core variables. Differences across gender, age, and athletic level are shown in Tables 1, 2 and 3. The following data primarily serve to describe sample characteristics and illustrate trends; related differential analyses are exploratory findings and are not incorporated into primary hypothesis testing or theoretical inferences. Table 1 indicates that gender exhibits significant differences in athlete burnout, psychological resilience, and emotion regulation (P < 0.05), while no significant differences are observed in coaching leadership behaviors (P > 0.05). Overall, male athletes demonstrate slightly higher mean scores in psychological resilience and emotion regulation compared to female athletes, whereas female athletes exhibit slightly higher burnout scores. Results in Tables 2 and 3 indicate that differences in age and athletic level did not reach statistical significance (P > 0.05). These statistical trends provide contextual information for subsequent variable relationship analyses. It is important to note that group differences were used solely for sample description and trend exploration and did not constitute the primary hypotheses tested in this study. Table 1Gender differences in coaches’ leadership behavior, athlete burnout, psychological resilience, and emotion regulationVariableGenderNumber (%)MSDt p Coaching Leadership Behaviormale348(67.05%)89.8511.57-0.1590.874female171(32.95%)90.0211.11total51989.9011.41Athlete Burnoutmale348(67.05%)42.209.654.8780.000female171(32.95%)44.8210.13total51943.069.88Psychological Resiliencemale348(67.05%)99.4514.182.2920.022female171(32.95%)92.9614.36total51997.3114.55Emotion Regulationmale348(67.05%)111.4017.25-2.8690.004female171(32.95%)107.7017.41total519110.1817.37 Gender differences in coaches’ leadership behavior, athlete burnout, psychological resilience, and emotion regulation Table 2Differences in athletic level manifested in coaching leadership behavior, athlete burnout, psychological resilience, and emotion regulationVariableAthletic LevelNumber (%)MSDt p Coaching Leadership BehaviorThird Class217(41.81%)90.2910.930.5770.679Second Class106(20.42%)89.312.8First Class153(29.48%)90.1411.4Master Athlete33(6.36%)87.5811.07International Master Athlete10(1.93%)927.33total51989.911.41Psychological ResilienceThird Class217(41.81%)96.3913.691.0160.398Second Class106(20.42%)96.9214.72First Class153(29.48%)98.2315.98Master Athlete33(6.36%)98.2113.55International Master Athlete10(1.93%)104.49.54total51997.3114.55Emotion RegulationThird Class217(41.81%)110.4516.140.8920.469Second Class106(20.42%)111.4518.96First Class153(29.48%)108.3917.82Master Athlete33(6.36%)110.7318.08International Master Athlete10(1.93%)116.416.6total519110.1817.37Athlete BurnoutThird Class217(41.81%)43.599.520.9390.441Second Class106(20.42%)42.3110.05First Class153(29.48%)42.8810.45Master Athlete33(6.36%)44.219.77International Master Athlete10(1.93%)38.66.33total51943.069.88 Differences in athletic level manifested in coaching leadership behavior, athlete burnout, psychological resilience, and emotion regulation Table 3Age differences in coaches’ leadership behavior, athlete burnout, psychological resilience, and emotion regulationVariableAgeNumber (%)MSDF p Coaching Leadership Behavior17.0022(4.24%)91.007.460.9300.48318.0055(10.60%)90.1111.4719.00117(22.54%)90.8510.6820.00141(27.17%)90.9911.3221.00108(20.81%)88.1211.8922.0049(9.44%)88.8810.8323.0022(4.24%)88.3616.1824.005(0.96%)85.4015.22total51989.9011.41Psychological Resilience17.0022(4.24%)103.1811.131.0760.37818.0055(10.60%)97.6914.5819.00117(22.54%)97.9513.3220.00141(27.17%)97.9415.5821.00108(20.81%)95.6213.4822.0049(9.44%)96.6316.7923.0022(4.24%)93.1416.8824.005(0.96%)96.0010.68total51997.3114.55Emotion Regulation17.0022(4.24%)113.5014.810.5580.79018.0055(10.60%)110.5617.8419.00117(22.54%)111.7916.3220.00141(27.17%)110.2516.5021.00108(20.81%)108.8918.9922.0049(9.44%)109.1619.0923.0022(4.24%)106.3218.4624.005(0.96%)106.4016.77total519110.1817.37Athlete Burnout17.0022(4.24%)42.9110.190.5540.79318.0055(10.60%)43.317.8019.00117(22.54%)42.749.7720.00141(27.17%)42.1110.4221.00108(20.81%)43.939.6922.0049(9.44%)44.5710.8623.0022(4.24%)43.4510.6024.005(0.96%)40.209.83total51943.069.88 Age differences in coaches’ leadership behavior, athlete burnout, psychological resilience, and emotion regulation Measurement model To further examine the construct validity of the measurement instruments, a confirmatory factor analysis CFA was conducted using AMOS with the maximum likelihood estimation method. A four-factor measurement model was specified, including coaching leadership behaviors, psychological resilience, emotion regulation difficulties, and athlete burnout. The observed indicators were dimension-level composite scores calculated as the mean of items within each subscale. The results indicated that the measurement model demonstrated an acceptable fit to the data χ²/df = 3.094, CFI = 0.925, TLI = 0.909, RMSEA = 0.064, GFI = 0.921, RMR = 0.022, suggesting that the proposed factor structure adequately represented the relationships among the observed indicators and latent constructs.All factor loadings were statistically significant and exceeded the recommended threshold of 0.50, indicating satisfactory construct validity. Furthermore, the standardized factor loadings were within acceptable ranges, supporting the adequacy of the measurement model. In addition, the reliability coefficients of all constructs remained within acceptable ranges, further supporting the stability of the measurement structure. Common method bias test Since the data for this study were primarily collected via questionnaires, the potential risk of common method variance (CMV) cannot be overlooked. This bias may compromise the reliability of results and the validity of causal inferences. To mitigate such risks, the Harman single-factor test was first employed for preliminary detection. The specific procedure involved including all measurement items simultaneously in an unrotated principal component analysis. Results revealed 39 factors with eigenvalues exceeding 1, with the first factor explaining 9.86% of the total variance—significantly below the critical threshold of 40% [36, 37]. This indicates no single dominant factor explaining substantial variance in the data, suggesting a low preliminary risk of common method variance. However, the Harman one-factor test serves only as a preliminary screening method. Its sensitivity to potential method factors and systematic measurement errors is limited, and it cannot completely rule out residual CMV risks. Therefore, this study maintains a cautious stance and explicitly acknowledges this risk when interpreting and discussing the results. Future research may consider employing marker variable methods or latent method factor models to more thoroughly identify and control method variance, thereby enhancing the robustness and credibility of conclusions. To further validate model robustness and the validity of statistical assumptions, this study concurrently conducted fundamental modeling diagnostics and hypothesis tests: (1) Linearity and residual distribution tests: Model fit was examined via scatterplots and residual plots, confirming that linearity and residual normality assumptions were generally satisfied; (2) Outlier and influential case diagnostics: Calculations of Cook’s D, leverage values, and studentized residuals revealed no single observation exerting significant influence on the overall model; (3) Multicollinearity testing: The variance inflation factors (VIF) for all primary independent variables were below 5, confirming no severe multicollinearity issues. Additionally, given that questionnaire data were collected in classroom or team settings, clustering effects among sample units may exist, potentially leading to correlated error terms. This study employed cluster-robust standard errors to adjust estimates, reducing bias from intra-cluster correlations and enhancing the robustness of significance estimates. If research constraints prevent this method, the limitation must be explicitly stated in result interpretations. In summary, this study found no significant common method bias through the Harman test. However, it cautiously acknowledges the possibility of residual risk. By combining multiple model assumption diagnostics and robustness control methods, the data quality and result credibility underwent systematic review and validation, ensuring the scientific rigor and reliability of the research conclusions. Correlation analysis of coaching leadership, burnout, psychological resilience, and emotion regulation Table 4 indicates that coaching leadership behaviors are significantly negatively correlated with burnout (R = -0.543, P < 0.01) and significantly positively correlated with psychological resilience; burnout is significantly negatively correlated with psychological resilience. Additionally, coaching leadership behavior showed a significant negative correlation with emotion regulation difficulties (R = -0.520, P < 0.01), while psychological resilience also exhibited a significant negative correlation with emotion regulation difficulties (R = -0.580, P < 0.01). Since higher DERS scores indicate greater emotion regulation difficulties (i.e., weaker regulation capacity), these findings suggest that positive coaching leadership behaviors correlate with lower emotion regulation difficulties (higher regulation capacity). Conversely, higher levels of athlete burnout are associated with more pronounced emotion regulation difficulties. This finding aligns with the theoretical expectations of the emotion regulation difficulties model, suggesting that strong psychological resilience and supportive leadership may help reduce individuals’ emotion regulation barriers, thereby decreasing athletes’ burnout levels. Table 4Correlation analysis statistics for coaches’ leadership behaviors, athlete burnout, psychological resilience, and emotion regulationVariableM±༳༤Coaching Leadership BehaviorAthlete BurnoutPsychological ResilienceEmotion RegulationCoaching Leadership Behavior89.90 ± 11.411Athlete Burnout43.06 ± 9.88-0.543**1Psychological Resilience97.31 ± 14.550.484**-0.501**1Emotion Regulation110.18 ± 17.370.520**-0.549**0.580**1N = 519. ** p < 0.01. The same applies below Correlation analysis statistics for coaches’ leadership behaviors, athlete burnout, psychological resilience, and emotion regulation N = 519. ** p < 0.01. The same applies below Testing the mediating effects of coaching leadership behavior and emotion regulation Using coaching leadership behavior as the independent variable, psychological resilience and emotion regulation difficulties as mediating variables, and sports burnout as the dependent variable, the SPSS macro plugin PROCESS provided by Hayes [38] was employed. Model 6 was selected from the Templates. A bias-corrected nonparametric percentile bootstrap test (5000 repeated samples) was employed to calculate the 95% confidence interval, enabling chained mediation effect analysis. Gender was controlled within the model to eliminate its potential confounding effects on psychological resilience and emotion regulation difficulties. Age and competitive level were theoretically weakly associated with this model and showed no significant correlations in pre-tests; thus, they were excluded from the model and noted in the text. Regarding specific pathways (Table 5; Fig. 2): The direct effect of “coach leadership behavior” on “sport burnout” remained significant (β = 0.615, p < 0.001) even after incorporating psychological resilience and emotion regulation difficulties as mediators. This result validates Hypothesis H1: Positive coaching leadership behavior directly reduces athletes’ burnout levels, indicating that effective leadership styles themselves serve as crucial psychological protective factors. They directly buffer training stress through emotional support, fair communication, and motivational mechanisms. Coach Leadership Behavior → Psychological Resilience → Athletic Burnout: Indirect Effect = -0.071, Boot SE = 0.019, 95% CI [-0.113, -0.038], accounting for 15.01% of the total effect. This indicates that positive leadership behavior significantly enhances athletes’ psychological resilience, thereby buffering athletic burnout. Although the effect size is small to moderate, it holds significant implications for sports psychology intervention practices, suggesting that strengthening psychological resilience may be an effective pathway to reduce burnout; Coach Leadership Behavior → Difficulty in emotion regulation → Athletic Burnout: Indirect Effect = -0.075, Boot SE = 0.015, 95% CI [-0.110, -0.049], accounting for 15.86% of the total effect. This pathway exhibits a slightly higher effect size, indicating that leadership behavior indirectly reduces burnout levels by improving emotion regulation capabilities. Essentially, this pathway reflects how coaches’ supportive communication and emotional care promote athletes’ emotional self-regulation, thereby enhancing psychological adaptability for sustained training; Coach Leadership Behavior → Psychological Resilience → Difficulties in emotion regulation → Athletic Burnout: Chain Mediation Effect = -0.051, Boot SE = 0.010, 95% CI [-0.072, -0.034], accounting for 10.78% of the total effect. This pathway indicates that psychological resilience serves as the psychological foundation for emotion regulation. Coaches’ positive leadership styles enhance psychological resilience, which indirectly reduces emotion regulation difficulties and further lowers the risk of burnout. Standardized path coefficients reveal that coaching leadership behavior exerts the most significant effect on psychological resilience (β = 0.615, P < 0.001), while psychological resilience significantly influences emotion regulation difficulties (β = 0.513, P < 0.001), indicating that self-psychological resources play a crucial mediating role. All three indirect paths were significant, with confidence intervals not containing zero, validating hypotheses H2-H4. Table 5Regression analysis of the chain mediation model between coaching leadership behaviors and burnout in sports(gender control)VariablePsychological ResilienceEmotion RegulationExerciseBurnoutOverallEffectβtβtβtβtCoaching Leadership Behavior0.61512.956**0.4727.929**-0.276-7.726**-0.473-14.892**Psychological Resilience0.51310.696**-0.116-3.857**Emotion Regulation-0.160-6.394**R20.2920.4130.4210.314F52.89372.21661.93458.883**P < 0.001 Regression analysis of the chain mediation model between coaching leadership behaviors and burnout in sports(gender control) **P < 0.001 Chain mediation model of psychological resilience and emotion regulation between coaching leadership behavior and athlete burnout After examining the standardized effect sizes and significance tests for the path from coaching leadership behavior to athlete burnout (Table 6), it was found that the Bootstrap 95% confidence interval for the total indirect effect of coaching leadership behavior on athlete burnout did not include zero. This indicates that “psychological resilience” and “emotion regulation difficulties” exerted significant mediating effects between coaching leadership behavior and athlete burnout. Tables 5 and 6 presents the standardized path coefficients and mediating paths. The overall model fits well (R² for burnout = 0.421, F = 61.934, P < 0.001). The total effect of coaching leadership behavior on sports burnout was β = -0.473, 95% CI [-0.535, -0.410], comprising a direct effect of β = -0.276, 95% CI [-0.346, -0.205]; The total indirect effect was β = -0.197, 95% CI [-0.247, -0.151], accounting for 41.65% of the total effect, indicating that psychological resilience and emotion regulation difficulties jointly served as key psychological mechanism pathways. Table 6Results of the chain mediation effect test of coaches’ leadership behavior on athlete burnoutEffect TypeEffect ValueBoot SEBootstrap95% CIEffect Proportion Lower LimitUpper LimitTotal Effect-0.4730.032-0.535-0.410100%Direct Effect-0.276 0.036-0.346-0.20558.35% Coach Leadership Behavior → Psychological Resilience → Athlete Burnout-0.0710.019-0.113 -0.03815.01%Coach Leadership Behavior → Emotion Regulation → Athlete Burnout-0.075 0.015-0.110-0.04915.86%Coach Leadership Behavior → Psychological Resilience → Emotion Regulation → Athlete Burnout-0.051 0.010 -0.072-0.03410.78%Total Indirect Effect-0.197 0.024 -0.247-0.15141.65% 5000-sample bootstrap percentiles for bias correction; confidence intervals do not include zero; gender is a control variable Results of the chain mediation effect test of coaches’ leadership behavior on athlete burnout 5000-sample bootstrap percentiles for bias correction; confidence intervals do not include zero; gender is a control variable The standardized effect coefficients reveal that the chained mediation pathway (psychological resilience–emotion regulation difficulties) accounts for approximately 10% of the total effect, while each single psychological mechanism contributes about 15% through indirect pathways. Thus, roughly 40% of the total effect is mediated by psychological mechanisms—a moderate effect size in sports psychology—indicating that enhancing psychological resilience and emotion regulation skills can produce significant psychological stress-reducing effects within training systems. Beyond merely reporting significance, this study emphasizes the practical significance of effect sizes: even if the absolute values are modest, their cumulative impact is particularly crucial in the long-term fight against burnout, holding clear application value (e.g., designing psychological training programs, coaching communication skills training). In summary, the model remains robust after controlling for gender, with all paths statistically significant and directionally consistent with theoretical expectations. This supports a chain mechanism whereby coaches’ positive leadership behaviors indirectly reduce exercise fatigue by enhancing psychological resilience and improving emotion regulation difficulties. Discussion The findings of this study indicate: (1) Coaching leadership behaviors are significantly negatively associated with athletic burnout, suggesting that more supportive leadership environments tend to be related to lower levels of athlete burnout; (2) Psychological resilience and emotion regulation function as significant mediating variables in the relationship between coaching leadership behaviors and athlete burnout, indicating that leadership behaviors may be associated with athletes’ internal psychological resources and emotional coping processes; (3) emotion regulation difficulties are significantly positively associated with athlete burnout, indicating that greater difficulties in regulating emotions tend to coincide with higher levels of athlete burnout, whereas athletes with stronger emotion regulation capacities may be better able to cope with competitive pressure and psychological fatigue. It is important to note that the Difficulties in Emotion Regulation Scale (DERS) used in this study is a reverse-scored construct, where higher scores reflect greater difficulties in emotion regulation rather than stronger regulation abilities. To avoid conceptual misunderstandings and directional biases, this paper employs “emotion regulation difficulties” as the variable name in its interpretation, so that higher scores indicate greater emotional dysregulation and weaker regulatory capacity, while lower scores reflect relatively stronger emotion regulation ability. Based on this logical clarification, the findings show a significant negative association between coaching leadership behaviors and emotion regulation difficulties, suggesting that more supportive leadership contexts tend to be related to lower levels of emotional dysregulation among athletes. This pattern may reflect that supportive coaching behaviors are associated with more stable emotional responses and stronger psychological recovery capacity among athletes, which in turn may be linked to lower levels of athlete burnout. Statistically, this study did not reverse-score the raw data. Instead, it revised terminology and calibrated theoretical logic to align result interpretations with the scale’s measurement intent. This clarification helps ensure conceptual consistency and improves the interpretive accuracy of the relationships among leadership behavior, psychological resilience, and emotion regulation within the proposed theoretical framework. Coaching leadership behaviors and athlete burnout The findings of this study indicate that coaches’ leadership behaviors are significantly negatively associated with athlete burnout among university athletes in China, supporting Hypothesis 1. This result suggests that athletes who perceive higher levels of supportive coaching leadership tend to report lower levels of athlete burnout. According to Self-Determination Theory (SDT), the satisfaction of basic psychological needs—autonomy, competence, and relatedness—is essential for sustaining intrinsic motivation and psychological well-being. When coaches adopt supportive and democratic leadership behaviors, athletes are more likely to feel respected, trusted, and understood, which may be associated with greater motivation and lower emotional exhaustion in sport contexts [39]. Previous research similarly indicates that supportive coaching behaviors are negatively correlated with athletes’ emotional exhaustion and sport devaluation, whereas authoritarian leadership behaviors are positively associated with athlete burnout [6, 40]. Furthermore, supportive leadership behaviors may help create a training environment characterized by trust, constructive feedback, and emotional support, which can be associated with stronger self-efficacy, team belonging, and psychological stability among athletes [41]. From the perspective of SDT, when athletes perceive autonomy support and empathetic communication from coaches, their psychological needs are more likely to be satisfied, which may facilitate adaptive coping and resilience when facing training stress or competitive setbacks [42]. Consistent with previous studies, democratic and supportive coaching behaviors are often associated with lower levels of athlete burnout because they promote psychological safety and perceived competence within the training environment [43, 44]. These findings collectively highlight the potential role of leadership climate in shaping athletes’ psychological experiences in sport settings. Exploratory analyses further showed that male athletes reported slightly higher levels of psychological resilience and emotion regulation, whereas female athletes displayed somewhat higher levels of athlete burnout. These differences should be interpreted cautiously and may reflect contextual and sociocultural influences rather than inherent gender distinctions. Gender role socialization theory suggests that males may be encouraged to demonstrate confidence and psychological toughness in sport contexts, which may be associated with greater perceived resilience and emotional control [45]. At the same time, variations in sport disciplines, training environments, resource allocation, and cultural expectations may also shape athletes’ psychological experiences [46–48]. Therefore, these group differences should be regarded as preliminary descriptive patterns rather than definitive conclusions, and future research should further examine potential moderating factors using more advanced analytical approaches. Mediating role of psychological resilience The findings of this study indicate that psychological resilience significantly mediates the association between coaching leadership behaviors and athlete burnout among university athletes in China, supporting Hypothesis 2. This result suggests that coaching leadership behaviors may be related to athlete burnout partly through athletes’ internal psychological resources. Previous research indicates that athletes with higher psychological resilience tend to maintain greater emotional stability and psychological balance under training and competition pressures, which is often associated with lower levels of emotional exhaustion and sport devaluation [49]. From the perspective of resilience theory, psychological resilience represents a core psychological resource that enables athletes to adapt to stress and setbacks, facilitating more adaptive coping and emotional adjustment in demanding sport contexts [13]. Accordingly, athletes with stronger resilience may be better able to manage training demands and competitive stress, which may in turn be linked to lower levels of athlete burnout. In addition, coaching leadership behaviors are closely associated with athletes’ psychological resilience within the sport environment. Drawing on Self-Determination Theory (SDT), supportive, democratic, and participatory leadership styles may help satisfy athletes’ basic psychological needs for autonomy, competence, and relatedness, thereby strengthening their psychological resources and adaptive capacity [50]. Prior studies suggest that coaching environments characterized by trust, emotional support, and constructive feedback are positively related to athletes’ confidence and stress tolerance, creating favorable conditions for the development of psychological resilience [51]. Taken together, these findings suggest that psychological resilience may represent an important psychological pathway linking coaching leadership behaviors with athlete burnout. However, because the present study employs a cross-sectional design, the results should be interpreted as associations rather than causal effects. Future research using longitudinal or experimental designs would be valuable for further examining the temporal dynamics of leadership, psychological resilience, and athlete burnout in sport contexts [10, 52]. Mediating role of emotion regulation This study further examined the mediating role of emotion regulation in the relationship between coaching leadership behaviors and athlete burnout. The results indicate that emotion regulation difficulties are significantly positively associated with athlete burnout, while coaching leadership behaviors are negatively associated with emotion regulation difficulties, which is consistent with the scoring direction of the Difficulties in Emotion Regulation Scale (DERS). Lower levels of emotion regulation difficulty reflect stronger emotion regulation capacity, enabling athletes to cope more effectively with training demands and competitive pressure. The findings further suggest that emotion regulation difficulties partially mediate the association between coaching leadership behaviors and athlete burnout, indicating that coaching leadership may be related to athletes’ emotional adaptation processes within sport contexts. These results are broadly consistent with previous studies emphasizing the importance of emotion regulation in explaining psychological fatigue and burnout risk among athletes [53, 54, 56]. According to the process model of emotion regulation, individuals who frequently employ adaptive strategies-such as cognitive reappraisal and constructive emotional expression—tend to maintain better psychological balance under prolonged stress and may report lower levels of emotional exhaustion and sport devaluation [53, 54]. At the same time, previous research suggests that coaching leadership behaviors may be associated with athletes’ emotion regulation patterns through the emotional climate created within the training environment. Within the framework of Self‑Determination Theory, supportive and autonomy-supportive leadership behaviors can contribute to athletes’ perceptions of psychological safety and emotional support, which may encourage the use of more adaptive emotional coping strategies [55]. Studies have also indicated that coaches’ emotional support, constructive feedback, and open communication are related to athletes’ ability to recognize and express emotions, thereby facilitating healthier emotional adjustment in sport settings [56]. Taken together, these findings suggest that emotion regulation may represent an important psychological process through which coaching leadership behaviors are associated with athlete burnout. However, given the cross-sectional and self-report nature of the present study, these relationships should be interpreted as statistical associations rather than causal effects. Future research could employ longitudinal or experimental designs and incorporate multi‑source data (e.g., coach evaluations or behavioral observations) to further examine the dynamic mechanisms linking leadership behaviors, emotion regulation, and athlete burnout [57, 58]. Chain mediation effect of psychological resilience and emotion regulation The present study further examined the sequential mediating roles of psychological resilience and emotion regulation difficulties in the relationship between coaching leadership behaviors and athlete burnout. The results indicate that coaching leadership behaviors are negatively associated with athlete burnout, and this relationship is statistically linked to a sequential pathway involving psychological resilience and emotion regulation difficulties. Specifically, athletes who perceive more supportive coaching leadership tend to report higher levels of psychological resilience, which is in turn associated with lower levels of emotion regulation difficulties and lower levels of athlete burnout. These findings suggest a potential psychological process through which leadership environments may be related to athletes’ psychological well-being in collegiate sport settings. Psychological resilience and emotion regulation appear to be closely connected psychological processes. Individuals with higher levels of resilience are often better able to adapt to competitive stress and training demands, and may be more likely to employ adaptive emotion regulation strategies such as cognitive reappraisal or constructive coping when facing pressure [59]. Previous research has similarly reported that athletes with stronger psychological resilience tend to adopt more positive emotion regulation strategies and report relatively lower levels of burnout [12, 22]. From the perspective of Self-Determination Theory (SDT), supportive coaching leadership may contribute to the development of such psychological resources by satisfying athletes’ needs for autonomy, competence, and relatedness [50]. Coaching environments characterized by emotional support, constructive feedback, and open communication have been found to be positively associated with athletes’ psychological resilience and stress tolerance [42]. Within this context, emotion regulation may function as an important psychological link connecting resilience with athlete burnout, as athletes with stronger resilience tend to experience fewer difficulties in regulating emotional responses to stress. Overall, the sequential mediation pattern observed in this study highlights a potential psychological pathway linking coaching leadership behaviors, psychological resources, and emotional coping processes with athlete burnout. Consistent with previous findings, external resources such as coach support may be associated with improved mental health outcomes partly through their relationships with resilience and emotion regulation [12, 60]. However, given the cross-sectional and self-report design of the present study, these findings should be interpreted as associative patterns rather than causal effects. From a practical perspective, the results suggest that coaching practices emphasizing autonomy support, emotional understanding, and constructive feedback may be associated with stronger psychological resilience and healthier emotional adjustment among athletes. Future research may further explore these mechanisms using longitudinal designs, multi-source data, or intervention studies to better understand how leadership environments influence athlete burnout over time. Practical implications This empirical study reveals that coaches’ supportive leadership significantly and negatively predicts athletes’ burnout, with psychological resilience and emotion regulation serving as chain mediators between the two. Specifically, supportive leadership enhances athletes’ psychological resilience, which in turn strengthens their emotion regulation capabilities, thereby effectively reducing burnout risk. Psychological resilience exhibits a stronger mediating effect in the model, indicating it serves as the key psychological mechanism through which coaching leadership influences burnout. emotion regulation, meanwhile, provides a stable conduit between psychological resilience and burnout. Based on these findings, the study’s practical implications manifest in two primary areas: First, from the perspective of university sports team management and coach training, enhancing athletes’ psychological resilience and emotion regulation capabilities should be integrated into leadership style optimization programs. Emphasis should be placed on developing supportive and empowering leadership styles while improving coaches’ awareness and skills in psychological counseling and emotional management. Second, from the perspective of athlete psychological intervention practice, training to enhance psychological resilience (such as goal setting, cognitive restructuring, and stress exposure training) should be prioritized based on the model’s mediating effect strength, supplemented by systematic improvement in emotion recognition and regulation skills. Thus, the practical significance of this study lies in: it uses empirical evidence to reveal the intrinsic pathway through which coaching leadership behaviors improve athletes’ psychological resilience and emotion regulation, thereby mitigating burnout. This provides intervention rationale at the psychological mechanism level for university sports management, moving beyond mere experiential recommendations. Simultaneously, it offers quantifiable theoretical reference for constructing comprehensive intervention models targeting university athletes’ mental health in the future. Research limitations and future directions Although this study provides empirical evidence for understanding the associations between coaching leadership behaviors and athlete burnout among college athletes, several methodological and design limitations should be acknowledged, and the findings should be interpreted with appropriate caution. First, the cross‑sectional research design limits the ability to draw causal inferences among the variables examined in this study. The analyses capture relationships at a single point in time and therefore cannot determine temporal ordering or causal directionality. Although the proposed mediation model is theoretically grounded, the relationships identified should be interpreted as statistical associations rather than causal pathways. Future research could adopt longitudinal designs, experimental interventions, or dynamic structural equation modeling to better examine the temporal processes through which coaching leadership, psychological resilience, emotion regulation, and athlete burnout may be related over time. Second, the data were collected entirely through athlete self‑report questionnaires, which may introduce potential common method variance and response biases. Although procedural remedies such as anonymity and controlled questionnaire order were implemented, self‑report measures may still be influenced by social desirability, recall bias, or individual response tendencies. While Harman’s single‑factor test did not indicate a dominant common factor, this method provides only a preliminary diagnostic and cannot fully eliminate the possibility of common method variance. Future studies could improve measurement robustness by incorporating multi‑source data, such as coach assessments, peer reports, behavioral observations, or physiological indicators, and by employing statistical controls such as marker variables or latent method factor approaches. Third, the research sample was drawn from multiple university athletic teams, which may involve potential clustering effects at the team or institutional level. Athletes within the same team environment may share similar training climates or leadership experiences, which could influence the independence of observations and potentially affect standard error estimation. Future research may employ multilevel modeling or hierarchical data structures to better account for group‑level influences and to more accurately capture the nested nature of sport team environments. Fourth, although the overall reliability of the measurement scales reached acceptable levels, some sub‑dimensions approached lower reliability thresholds, which may affect measurement precision and construct differentiation. Future research could further strengthen measurement quality by conducting confirmatory factor analyses, testing measurement invariance across groups, and refining scale structures to enhance the stability and cross‑context applicability of the constructs.Although confirmatory factor analysis in the present study supported the overall measurement structure, future studies may further validate these scales across different athlete populations and cultural contexts to strengthen the generalizability of the measurement model. Finally, the present study did not comprehensively examine potential contextual moderators such as gender differences, sport types, or training environments. Although exploratory analyses suggested some group differences, these findings should be interpreted cautiously due to the cross‑sectional design and sample structure. Future studies could employ multi‑group structural equation modeling or moderated mediation analyses to explore how contextual and demographic factors may shape the relationships among coaching leadership behaviors, psychological resilience, emotion regulation, and athlete burnout. In summary, future research should adopt longitudinal and multi‑source research designs and incorporate multilevel analytical approaches to provide a more comprehensive understanding of the relationships identified in this study. Such approaches would help clarify the temporal dynamics and contextual influences underlying athlete burnout while improving the methodological rigor of research in this field. Conclusion This study systematically reveals the significant predictive role of coaching leadership behaviors on college student-athletes’ psychological resilience, emotion regulation, and athlete burnout, while clarifying the mediating and chain mediating effects of psychological resilience and emotion regulation. The findings not only enrich the theoretical understanding of the relationship between coaching leadership behaviors and athletes’ mental health within sports psychology but also provide scientific evidence for promoting mental health and enhancing athletic performance among college athletes. Notably, psychological resilience and emotion regulation serve as crucial bridges in the process by which coaching leadership behaviors influence athlete burnout, suggesting that practical interventions should prioritize cultivating athletes’ psychological resilience and emotion regulation capabilities. Although this study has limitations in sample representativeness and data collection methods, future research can further validate and expand these findings by expanding sample sizes and employing diversified assessment tools. This would enable deeper exploration of the underlying mechanisms through which coaching leadership influences athletes’ psychological and behavioral development. Recommendations Based on the findings and analysis, the following recommendations are proposed to enhance coaching leadership behaviors and mitigate athlete burnout: Optimize coaching leadership behaviors: University athletic departments should establish systematic coaching training programs focused on developing supportive and democratic leadership competencies. For supportive leadership, coaches should participate in foundational psychology courses to master psychological counseling techniques, enabling them to provide timely emotional support and encouragement when athletes experience low morale due to competition setbacks. For democratic leadership, conduct communication workshops to teach coaches how to listen to athletes’ perspectives and respect individual differences. During daily training, encourage athlete participation in planning sessions to foster an open, equal dialogue environment. This builds mutual trust and support between coaches and athletes, increasing psychological support and reducing athlete burnout risk. Optimize coaching leadership behaviors: University athletic departments should establish systematic coaching training programs focused on developing supportive and democratic leadership competencies. For supportive leadership, coaches should participate in foundational psychology courses to master psychological counseling techniques, enabling them to provide timely emotional support and encouragement when athletes experience low morale due to competition setbacks. For democratic leadership, conduct communication workshops to teach coaches how to listen to athletes’ perspectives and respect individual differences. During daily training, encourage athlete participation in planning sessions to foster an open, equal dialogue environment. This builds mutual trust and support between coaches and athletes, increasing psychological support and reducing athlete burnout risk. Prioritize Cultivating Athletes’ Psychological Resilience: University athletic departments should integrate psychological resilience training into routine management with scientifically designed programs. On one hand, specialized courses should be offered, featuring lectures by sports psychology experts covering topics like stress management and coping with setbacks. Through theoretical explanations and case studies, athletes learn stress relief techniques. Sharing sessions on adversity experiences and discussing coping strategies are also organized. On the other hand, training should be linked to real-world scenarios. Coaches set challenging tasks and situations—such as increasing training intensity or simulating unfavorable conditions—to build athletes’ psychological endurance. Timely feedback and guidance help them reflect on experiences, gradually enhancing psychological resilience and mental health. Prioritize Cultivating Athletes’ Psychological Resilience: University athletic departments should integrate psychological resilience training into routine management with scientifically designed programs. On one hand, specialized courses should be offered, featuring lectures by sports psychology experts covering topics like stress management and coping with setbacks. Through theoretical explanations and case studies, athletes learn stress relief techniques. Sharing sessions on adversity experiences and discussing coping strategies are also organized. On the other hand, training should be linked to real-world scenarios. Coaches set challenging tasks and situations—such as increasing training intensity or simulating unfavorable conditions—to build athletes’ psychological endurance. Timely feedback and guidance help them reflect on experiences, gradually enhancing psychological resilience and mental health. Strengthening emotion regulation Training: University athletic departments should provide comprehensive emotion regulation support. This includes offering psychological counseling for emotion regulation, organizing regular individual or group sessions, and utilizing psychological assessment tools to help athletes recognize their emotional patterns, evaluate their state, and identify potential issues. Group sessions encourage sharing emotional experiences and learning others’ coping methods. Concurrent skill workshops teach techniques like cognitive restructuring, positive self-affirmation, and emotional expression. Athletes are guided to reframe negative thoughts, adopt proactive perspectives, build confidence through positive language, express emotions healthily, and prevent or alleviate burnout. Strengthening emotion regulation Training: University athletic departments should provide comprehensive emotion regulation support. This includes offering psychological counseling for emotion regulation, organizing regular individual or group sessions, and utilizing psychological assessment tools to help athletes recognize their emotional patterns, evaluate their state, and identify potential issues. Group sessions encourage sharing emotional experiences and learning others’ coping methods. Concurrent skill workshops teach techniques like cognitive restructuring, positive self-affirmation, and emotional expression. Athletes are guided to reframe negative thoughts, adopt proactive perspectives, build confidence through positive language, express emotions healthily, and prevent or alleviate burnout. Acknowledgements We would like to thank all participants who helped us complete this study. Authors’ contributions Xiaoying Chen participated in the study design, data collection and analysis, and drafted the main manuscript text. Yuan Yue helps obtain data and prepare paper materials, charts, or tables.Jun Xiang helps to obtain data and prepare paper materials, charts or tables. Fang zhang revised the draft. All the authors reviewed the manuscript. Funding This research was supported by the following projects: The 2023 Guangdong Education Science Planning Project (2023GXJK528);Key Projects of the University Research Fund of Zhaoqing University 2024 (ZD202401).Zhaoqing University 2025 University Research Program (qn202514). Data availability The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request. Declarations Ethics approval and consent to participate The Ethics Review Committee of Zhaoqing University approved the present study (Approval No. 2025030). Prior to participation, all individuals provided informed consent. Research procedures complied with the Declaration of Helsinki and its subsequent amendments. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request.