The study population was selected using a stratified random sampling method. Three cities (Hangzhou, Nanjing, and Wenzhou) were randomly selected from the East China region, and then five colleges were randomly selected from these three cities. The survey was conducted from September 2022 to December 2022. According to the sample size calculation method (Preacher and Kelley, 2011), the estimated sample size ranges from 380 to 759. We expanded our analysis to include a larger sample of 1,420 undergraduate students from freshman to senior year.

The study included participants who met the following criteria: (1) aged between 17 and 25 years old, and (2) enrolled as undergraduate students at the university. The focus on undergraduate students was attributed to their unique developmental stage and specific academic pressures. Additionally, physical education classes, which are mandatory only for undergraduates, are not typically a part of graduate programs in China. This focus also allowed us to maintain sample homogeneity and data consistency, thereby improving the reliability and validity of our results. All participants provided informed consent and participated in the survey voluntarily. The study excluded: (1) graduate students and intern students at the school, and (2) individuals employed by the schools.

An initial survey was administered to eight college students across various academic levels to verify the questionnaire’s validity. The primary aim of the pilot survey was to validate the comprehensibility and effectiveness of our questionnaire. We aimed to ensure that the questionnaire’s questions could be widely understood and effectively gather the information required for our research. Based on the respondents’ feedback, suitable modifications were made to the questionnaire.

The investigators, all of whom are public physical education teachers at these universities, received unified training via Tencent meetings. After obtaining approval from the school leadership, they collaborated to complete the questionnaire collection. The questionnaires were collected through face-to-face interviews. Before data entry, errors were identified and corrected, and missing responses were completed, ensuring logical consistency. Questionnaires with clear logical inconsistencies and numerous missing responses were excluded. This approach aimed to enhance the precision of variable measurements within each subgroup. The selected colleges served as the targets for the survey.

The study’s data analysis was conducted using IBM SPSS 22.0 and the PROCESS V3.5 plug-in, chosen for their robust analytical capabilities, compatibility, and the plug-in’s mediation and moderation analysis features. The variables examined in the correlation and regression analyses were carefully selected based on their relevance to the study’s objectives. These variables were subjected to descriptive analysis to provide a summary of their basic features and distributions. ANOVA was conducted to compare the means of these variables across different groups, providing insights into their relationships. Regression analysis was used to understand the predictive relationships between these variables. To validate the mediating effects between variables, a bias-corrected nonparametric bootstrap 95% confidence interval estimation was chosen. This method, involving 5,000 repetitions of sampling, was selected for its advantages in bias correction, skewness adjustment, versatility, and ease of implementation.

Additionally, given that the data in our study is self-reported, there is a potential for common method bias. During the testing process with college students, we emphasized the anonymity and confidentiality of the questionnaire, and clarified that the data is solely for scientific research purposes, in order to control the sources of common method bias as much as possible. Furthermore, we employed Harman’s single-factor test (Baumgartner et al., 2021) to check for common method bias.

From September 2022 to December 2022, this survey-based, cross-sectional study collected data from 1,380 college students across five universities in three cities. Specifically, a total of 1,420 questionnaires were distributed to randomly selected students in physical education classes across these colleges. Of these, 1,397 were recovered, and after excluding invalid questionnaires, 1,380 valid responses remained. The recovery rate was 98.3%, and the validity rate was 97.1%. The high recovery rate can be attributed to the strong emphasis on academic participation in these institutions, and no additional incentives were provided. The respondents included 649 freshmen (47%), 469 sophomores (34%), 193 juniors (14%), and 69 seniors (5%). There were 773 male students (56%) and 607 female students (44%).

The results of an unrotated exploratory factor analysis using Harman’s one-factor test extracted a total of 11 factors with eigenvalues >1. The maximum variance explained by a single factor was 23.76%, which is less than the critical criterion of 40%. These results indicate that there is no serious common method bias in this study. Common method bias, if present, could potentially inflate the relationships between variables due to shared method variance, leading to erroneous conclusions. However, our test results alleviate these concerns. The fact that no single factor emerged or accounted for the majority of the covariance among the variables suggests that the relationships observed in our study are likely to be attributed to the constructs under investigation rather than to any potential method bias.

In this study, we explored the potential influences of both academic year (encompassing freshman, sophomore, junior, and senior stages) and major (including fields such as medicine, foreign languages, law, computer science, education, psychology, materials engineering, chemistry, pharmacy, electronic information, and others) on the prevalence of negative emotions among college students. We employed analysis of variance (ANOVA) with the aim of discerning any statistically significant disparities in the cumulative scores of negative emotions across diverse academic years and majors. Contrary to our expectations, the results revealed that these factors did not exert a significant impact on the reported intensity of negative emotions, as evidenced by p-values of 0.405 for academic year and 0.117 for major. This finding implies a uniformity in the experience of negative emotions across various academic progressions and fields of study. Consequently, our analysis lends support to the hypothesis that the intensity of negative emotions experienced by college students is not significantly influenced by their respective academic year or chosen major.

The means (M), standard deviations (SD) and correlation matrices for physical activity, psychological resilience, coping styles and negative emotions are given in Table 1. The descriptive results of each variable indicate that the average amount of physical activity among college students is at a low level (Mean = 16.911, SD = 17.108). The large standard deviation suggests a wide range of physical activity levels among the students. This could be due to a variety of factors, such as differences in personal habits, academic workload, or access to sports facilities. Psychological resilience, categorized into tenacity, self-improvement, and optimism, along with coping styles, divided into positive and negative coping, and negative emotions, subdivided into depression, anxiety, and stress, form the core variables of our study. When encountering stress, college students tend to use positive coping strategies (Mean = 36.796, SD = 5.640) than negative coping (Mean = 18.944, SD = 4.585).

Our correlation statistical analysis (Table 1) reveals significant relationships between physical activity, psychological resilience, coping styles, and negative emotions. Specifically, physical activity is positively correlated with both psychological resilience (r = 0.270, p < 0.001) and positive coping (r = 0.166, p < 0.001), but negatively correlated with negative emotions (r = −0.097, p < 0.001).

Moreover, psychological resilience and positive coping share a significant positive correlation (r = 0.602, p < 0.001). Within the sub-dimensions of psychological resilience, tenacity (r = 0.556, p < 0.001), self-improvement (r = 0.591, p < 0.001), and optimism (r = 0.515, p < 0.001) all exhibit positive correlations with positive coping.

Psychological resilience is significantly negatively correlated with negative emotions (r = −0.222, p < 0.001), including its sub-dimensions: depression (r = −0.253, p < 0.001), anxiety (r = −0.186, p < 0.001), and stress (r = −0.199, p < 0.001). Similarly, positive coping is negatively correlated with negative emotions (r = −0.187, p < 0.001) and its sub-dimensions: depression (r = −0.213, p < 0.001), anxiety (r = −0.155, p < 0.001), and stress (r = −0.167, p < 0.001). While, negative coping is positively correlated with negative emotions (r = 0. 399, p < 0.001) and its sub-dimensions: depression (r = 0. 389, p < 0.001), anxiety (r = 0. 353, p < 0.001), and stress (r = 0. 404, p < 0.001).

These results suggest that individuals can develop a higher level of psychological resilience through physical activity, enabling them to better handle various challenges. It appears that college students prefer to employ positive coping strategies when dealing with stress and negative emotions. This preference may indicate that during physical activity, individuals tend to adopt positive coping mechanisms, such as seeking support and problem-solving, to minimize the experience of negative emotions.

Previous studies have highlighted significant gender differences in physical activity levels (McCarthy and Warne, 2022). This study further explores gender differences among these variables, as shown in Table 2. We applied an independent samples t-test to compare the gender difference. As shown in Table 2, male students are significantly more physically active than female students (p < 0.001). The corresponding Cohen’s d value is 0.453, suggesting a near-medium effect size. This aligns with previous findings suggesting that male students are generally more inclined towards active involvement in physical activities (Sisay, 2021; Xu et al., 2021). These findings underscore the potential benefits of promoting physical activity as a strategy to mitigate negative emotions among college students, particularly in males.

Regarding coping styles, significant differences were observed between male and female students in both positive (p = 0.002, Cohen’s d = 0.173) and negative coping (p = 0.028, Cohen’s d = 0.118) (Table 2). Specifically, female students tended to employ more positive strategies to manage stress, aligning with the general characteristics of emotional expression and support-seeking in female students (Chaplin, 2015). These results indicate that although the effect sizes of positive and negative coping strategies are small to tiny, they are still significant in terms of gender differences, suggesting that gender differences in coping strategies may have important implications for stress management and emotional regulation.

The study also found that male students’ overall negative emotion scores were significantly higher than female students (p < 0.001) (Table 2). Further analysis revealed significant gender differences in the dimensions of negative emotions, specifically depression (p < 0.001, Cohen’s d = 0.211), anxiety (p < 0.001, Cohen’s d = 0.255) and stress (p < 0.001, Cohen’s d = 0.235). These Cohen’s d values reflect “small” to “near-medium” effect sizes for gender differences in depression, anxiety, and stress. This finding corroborates past studies, suggesting that male students may be less likely to express emotions, leading to an accumulation of negative emotions (MacArthur, 2019; Sagar-Ouriaghli et al., 2020). It further underscores the importance of taking into account the extent and nuances of gender differences in the influence on negative emotions when formulating mental health interventions for college students.

In developing our model, we considered cities and colleges as potential random effect factors, hypothesizing they might influence the dependent variable—total negative emotions score. However, our analysis revealed that the variance components for each random effect were zero. This suggests that the influence of cities and colleges as random factors is either extremely limited or virtually non-existent, indicating a consistent influence across different locations and institutions on the dependent measures.

To further understand the underlying mechanisms affecting negative emotions, we tested the mediating effects of psychological resilience and positive coping styles in the relationship between physical activity and negative emotions. We used Model 6 in SPSS PROCESS V3.3 (Hayes, 2013) to conduct the analyses and showed the results in Table 3.

Previous research has primarily employed scale measurements and experimental methods to gather data and quantitatively evaluate the impact of various physical activity programs and the extent of physical activity on the negative emotions of college students (Stults-Kolehmainen and Sinha, 2014; Churchill et al., 2021). These studies have proposed intervention strategies and methods, demonstrating that active physical activity can alleviate negative emotions, enhancing students’ ability to cope with stress and bolstering their psychological resilience. Our study, however, concentrates on the direct and indirect effects of physical activity on the negative emotions of college students. It elucidates the relationship, mechanisms, and effects of physical activity on negative emotions by incorporating elements of psychological resilience and positive coping styles. The findings indicate a significant negative and indirect correlation between physical activity and negative emotions in college students. The total effect of physical activity significantly predicts a decrease in negative emotions, as evidenced in Table 3, thereby confirming Hypothesis H1.

Stressful emotions can disrupt an individual’s internal balance, leading to a reconfiguration of perceptions and a variety of outcomes. Psychological resilience, a beneficial personality trait, serves as a crucial safeguard. It empowers individuals to withstand stress, swiftly adapt to their environment, and mitigate the harmful effects of adverse events (Troy et al., 2023). In the context of college students, psychological resilience plays a pivotal role in preserving their physical and mental equilibrium when they face stressful situations (Gong et al., 2023). Individuals with lower levels of psychological resilience may experience physical discomfort (e.g., shortness of breath, rapid heartbeat, dizziness), mental depression, and cognitive blankness when dealing with stress (Doan et al., 2023).

Psychological resilience can be bolstered through various means, such as self-improvement, setting appropriate life goals, moderating expectations, and maintaining harmonious family relationships. These strategies can assist college students in maintaining a positive psychological state (Wu et al., 2020; Kuang et al., 2024). However, the specific mechanisms through which physical activities strengthen psychological resilience and alleviate negative emotions have not been thoroughly studied.

Psychological resilience, a multidimensional construct with hierarchical characteristics, has been less explored in previous research, which has primarily focused on goals, emotions, and cognition (Chaplin, 2015; Zhang et al., 2022). This study aims to delve into three dimensions of resilience: individual self-improvement, optimism, and resilience, using a comprehensive scale of 25 items. Our findings reveal a robust positive correlation between physical activity and psychological resilience, a correlation that persists even when other control factors are considered. We extend the scope of previous research by demonstrating that physical activity primarily alleviates negative emotions through the comprehensive mediating role of psychological resilience. The divergence between our findings and those of previous studies (Wu et al., 2020; Xu et al., 2021; Gong et al., 2023) underscores the importance of considering the multidimensional nature of psychological resilience in future research, as the differences may be attributed to the varied scales used to measure psychological resilience.

This study affirms Hypothesis H2, indicating that individuals who engage in high levels of physical activity demonstrate increased resistance to negative emotions under stressful circumstances. This finding is consistent with the perspective that psychological resilience is a dynamic process influenced by various factors such as optimism, resilience, and self-improvement (Godara et al., 2022; Ijntema et al., 2023). These traits equip individuals with the capacity to adopt a more constructive problem-solving approach, thereby reducing the negative impact of risk factors and promoting individual adaptation and growth. This underscores the importance of physical activity in fostering psychological resilience and managing stress effectively.

Coping strategies are generally categorized into positive and negative types. Positive coping acts as a regulatory mechanism for stress, facilitating adaptive responses, whereas negative coping is often characterized by emotional distress and avoidance behaviors (Fteiha and Awwad, 2020). Our findings align with previous research styles (Moeller et al., 2020; Chen et al., 2022), indicating that regular physical activity is significantly associated with the adoption of positive coping. However, it appears to have no significant association with negative coping strategies, as indicated in Table 1. Engaging in physical activity is not merely an exercise but a proactive process that nurtures positive coping mechanisms. Despite this, our current study did not support the hypothesis (H3) that physical activity mitigates negative emotions in college students by significantly mediating through positive coping strategies, as shown in Table 3.

The sustained practice of physical activity has the potential to bolster psychological resilience and improve coping strategies among college students (Xu et al., 2021; Chen et al., 2022). This development promotes an active and engaged approach to challenges, as opposed to a passive or avoidance-based one. It empowers students to utilize their cognitive resources and physical vigor, to actively recalibrate their self-perception, regulate their emotions, and tackle obstacles directly. Thus, it is plausible that physical activity may enhance psychological resilience, thereby transforming their coping strategies.

Stressful events inevitably impact individuals, often leading to negative emotions and impaired cognitive functions. However, individuals with higher psychological resilience are more likely to employ positive coping strategies to address these issues. Prior research has indicated that factors such as interpersonal support, peer relationships, and sports friendships at school can foster a willingness to exercise among college students, thereby enhancing their physical activity behavior (Zhang et al., 2022; Zou et al., 2023). Additionally, family parental support, social support, and teacher support can also bolster college students’ motivation and engagement in physical activity. These factors can aid college students in strengthening their commitment to exercise, implementing exercise behaviors, and fostering psychological resilience. This, in turn, enables them to adopt positive coping styles when faced with stress, thereby mitigating negative emotions and other mental health issues.

Our study’s findings suggest that psychological resilience can inspire college students to maintain an optimistic attitude, perseverance, and self-improvement efforts in the face of adversity. They can positively adjust their cognition, confront the problem rather than evade it, and utilize their resources and skills to manage difficulties and alleviate negative emotions. This study further corroborates the fully mediating role of psychological resilience and positive coping styles between physical activity and negative emotions among college students. In other words, physical activity primarily mitigates negative emotions by enhancing psychological resilience and positive coping styles. Thus, hypothesis H4 is validated.

Our study also found significant gender differences in the emotional responses of college students, aligning with previous research (Chaplin, 2015; Graves et al., 2021; McCarthy and Warne, 2022). Female students demonstrated higher levels of positive coping, while male students exhibited more negative emotions, including anxiety, depression, and stress. Notably, male students’ overall negative emotion scores were significantly higher than those of female students. These gender differences have practical implications for designing interventions or support programs for male and female students. For male students, an increased focus can be placed on emotion management and stress coping. This could include counseling services and training programs that enhance their emotional regulation skills. For female students, the emphasis could be on fostering the development of positive coping mechanisms. This might involve promoting positive thinking, self-affirmation, and enhancing self-efficacy.

This study underscores the importance of focusing on the physical activity behavior of college students. It advocates for attention to the university physical education curriculum, calling for curriculum reform and innovation, and the provision of optimal physical education courses. The study further emphasizes the need for qualified physical education teachers who can guide students and help them realize the benefits of physical activity in terms of physical fitness, personality development, and willpower. In addition, the study suggests the implementation of specialized courses tailored to students of different genders. These courses could provide customized exercise programs and emotional management strategies to cater to their unique needs. By understanding and addressing these gender differences, colleges can more effectively support the mental health and well-being of all students. This approach fosters a healthier and more inclusive campus environment, thereby enhancing the overall student experience.

While this study offers valuable insights into the relationship between physical activity, psychological resilience, positive coping styles, and negative emotions, it’s important to interpret the results with caution due to certain limitations. The self-report method used to measure the intensity, duration, and frequency of participants’ daily physical activity may not accurately represent their actual levels. Future research could benefit from more objective data collection methods, such as using wearable electronic devices. Additionally, this cross-sectional study limits the understanding of potential causal relationships over time. A longitudinal design in future studies could provide more comprehensive insights. Furthermore, potential confounding variables, such as financial means, access to resources, parental support, and upbringing, could influence the observed relationships. Future research should consider these variables for a more holistic understanding of the interplay between physical activity, coping skills, and emotional well-being.