Introduction

With the advent of the AI era and the widespread adoption of multimedia teaching, higher education is facing increasingly fierce competition and a rapidly changing instructional environment. Against this backdrop, how to attract and retain high-quality faculty has become a focal issue for both academia and educational administrators1. Job satisfaction (JSF), as a critical factor influencing the occupational stability of teachers, plays an essential role2. JSF refers to the positive emotional state that individuals experience when their work fulfills their personal values2. For college teachers, JSF specifically denotes a positive emotional orientation formed after a comprehensive evaluation of various dimensions of their teaching practices3. Social exchange theory4 further explains that when teachers experience intense levels of JSF, they often reciprocate by enhancing their organizational commitment, which is crucial for maintaining faculty stability. Within Chinese higher education, the increasing demands on teaching and research competitiveness have led college teachers to face high levels of job stress (JS), depressive symptoms, and reduced JSF1,5. Therefore, identifying effective strategies to enhance JSF among college teachers has become imperative.

JSF among teachers is shaped by both intrinsic and extrinsic factors. Intrinsic factors include emotional intelligence (EI), resilience (RES), self-efficacy, and positive emotions6,7,8,9. Teachers with elevated EI often show higher levels of work engagement and JSF10. Moreover, RES equips educators to handle stress and obstacles, further enhancing their JSF8. Extrinsic factors include social support, JS, and institutional working conditions11,12,13. Notably, among Chinese college teachers, JS is significantly associated with burnout and turnover intentions1.

RES, defined as the capacity to effectively cope with and adapt to adversity, trauma, or stress, has recently been recognized as a vital psychological resource for enhancing JSF14,15,16. Existing studies have consistently shown that RES not only helps reduce JS but also enhances EI17,18. Prior research has predominantly focused on the direct impact of RES, often overlooking its mediating function between JS, EI, and JSF. Therefore, the present study seeks to investigate the mechanisms by which JS, EI, and RES influence JSF in college teachers. Moreover, gender has been shown to be an important variable affecting how individuals cope with stress and regulate emotions in the workplace19,20. Female teachers may be more prone to emotional exhaustion due to higher expectations for emotional labor21, while male teachers are more inclined to adopt problem-focused strategies to cope with stress22. These gender differences may influence the pathways through which JS, EI, and RES affect JSF. Therefore, this research applies a multi-group analysis to uncover gender’s influence on the relationships among JS, EI, RES, and JSF, seeking to offer a more nuanced view of how gender shapes the psychological dynamics of job satisfaction.

Job stress (JS) and job satisfaction (JSF)

JS typically refers to the psychological or physiological responses that emerges when an individual’s characteristics collide with job demands, which can harm their performance and overall state23. Among college teachers, JS primarily originates from a range of situational stressors, such as excessive workload, time pressure, student discipline management, lack of teaching resources, weak professional identity, insufficient administrative support, and diverse task demands1,24. According to the JSF model proposed by Lent and Brown25, satisfaction in the workplace can be predicted by five categories of constructs: individual characteristics, self-belief, goal-oriented behaviors, external resources and obstacles, and workplace conditions. This indicates that JS serves as a critical external factor influencing teachers’ JSF. Empirical research consistently shows a significant negative relationship between JS and teachers’ JSF11,26,27. For instance, Wang, et al.27 indicated that higher JS levels were linked to lower JSF in their study on organizational commitment among teachers in three Chinese colleges. Additionally, Li, et al.11, through a cross-national survey of college teachers in four countries, further confirmed the negative predictive effect of JS on teacher satisfaction across different cultural and educational systems. Their research also underscored the advantages of distributed leadership in easing stress and boosting satisfaction. In summary, JS is a decisive factor affecting JSF among college teachers and should be given due attention in efforts to improve college teachers’ well-being.

Emotional intelligence (EI) and job satisfaction (JSF)

Mayer28 defines EI as the ability to perceive, understand, and regulate emotions, which supports personal growth and adaptation. Higher EI helps individuals manage both their own and others’ emotions6. Based on the Conservation of Resources (COR) Theory, individuals aim to acquire, maintain, and safeguard critical resources in their workplace. EI, as a key psychological resource, helps college teachers cope effectively with stressful situations, thereby reducing resource depletion29. Teachers with stronger EI are better able to maintain emotional stability during teaching, resist the loss of resources caused by negative events, and enhance their positive perceptions of work, leading to higher JSF30. Studies have consistently found a strong positive link between EI and JSF6,31,32,33. Rogowska and Meres6 emphasized that EI not only improves teachers’ JSF but also positively affects their life satisfaction and overall well-being. Therefore, from the perspective of COR theory, EI serves as an essential internal psychological resource that plays a vital role in enhancing teachers’ JSF and provides significant protective value in high-pressure educational environments.

The mediating role of resilience (RES)

According to resilience theory34, RES is a critical adaptive capacity that reflects an individual’s ability to cope with, adjust to, and recover from adversity, stress, or major challenges35. Previous research has demonstrated that RES can effectively enhance individuals’ JSF36,37,38,39. Ibrahim and Hussein36 emphasized that employees with high levels of RES possess greater self-awareness and adaptive flexibility, enabling them to respond to changes, solve problems on the spot, and quickly adapt to varying work contexts, thereby significantly improving their JSF. The COR theory29 offers a solid foundation for understanding the mediating role of RES. It suggests that individuals aim to acquire and protect resources to combat the depletion caused by stressors. In the context of college teaching, persistent JS can exhaust time, emotional energy, and cognitive resources. As a core psychological resource, RES not only mitigates the loss of these resources (resource preservation) but also facilitates the acquisition of new adaptive resources (resource gain), thereby buffering the negative impact of JS on JSF. Moreover, RES enhances individuals’ efficiency in utilizing EI, allowing teachers to respond more constructively to workplace emotional events and thus improving their overall emotional regulation and professional attitudes17,18. Although prior studies have acknowledged the positive role of RES in mitigating stress, improving job performance, and enhancing EI18,40,41, there remains a lack of systematic empirical investigation into how RES mediates the combined effects of JS and EI on college teachers’ JSF. Therefore, this study incorporates RES into the model to examine its mediating role in this pathway, aiming to enrich both theoretical understanding and practical applications.

The moderating role of gender

Research shows that teachers’ perceptions of their work environments vary by gender. Existing evidence indicates that female college teachers tend to experience more JS than males, displaying higher levels of burnout and lower JSF19,20,42. Zhou19 notes that female teachers often bear a heavier role load in the workplace, as they are expected to complete teaching duties while also managing family responsibilities and adhering to traditional gender roles in society. This “dual role” often becomes a primary source of stress for them20. Additionally, there are gender differences in EI. Studies have shown that women tend to develop stronger emotional awareness and regulation skills during socialization, which aligns with traditional gender roles that view women as more emotional and empathetic21. Consequently, female teachers often score higher in the EI construct, and greater EI is correlated with more effective teaching performance and classroom management, thereby enhancing their JSF43. On the other hand, research has found that male teachers tend to demonstrate greater RES and are more likely to cope with stress through proactive strategies, thus improving JSF44,45. These gender differences not only reflect variations in individual stress coping and emotional management but also point to gender as a moderating factor in promoting teachers’ job well-being. Based on the above research, this study treats gender as a potential moderating construct and investigates its role in the pathways through which JS, EI, and RES influence JSF among college teachers. The aim is to reveal the heterogeneity in career adaptation mechanisms between different gender groups, providing empirical evidence for gender-sensitive teacher support policies.

Current study

In summary, although previous studies have explored the effects of JS, EI, and RES on teachers’ JSF from various perspectives, research that systematically integrates these three relationships with RES as a mediating variable remains scarce, especially within the high-pressure profession of college teaching. By combining resilience theory34 and the COR Theory29, a deeper understanding can be gained of how individuals, when facing high workloads and emotional challenges, mobilize psychological resources to achieve emotional regulation and enhance JSF. Therefore, the goal of this study is to develop a theoretical model that thoroughly examines the impact pathways of JS and EI on JSF among college teachers, exploring the mediating role of RES and the moderating influence of gender. The following hypotheses are proposed:

H1

JS has a significant negative effect on JSF among college teachers.

H2

EI has a significant positive effect on JSF among college teachers.

H3

RES mediates the relationship between JS and JSF.

H4

RES mediates the relationship between EI and JSF.

H5

Gender significantly moderates the relationship between JS and JSF among college teachers.

H6

Gender significantly moderates the relationship between EI and JSF among college teachers.

H7

Gender significantly moderates the relationship between RES and JSF among college teachers.

Fig. 1
figure 1

Research model.

Full size image

Methods

Participants

This study conducted questionnaire data collection from February to April 2025, using a stratified random sampling method. Several universities in China were randomly selected, and then, within each university, random sampling was conducted by academic year to ensure the representativeness of the sample. The design and distribution of the questionnaires were carried out through the Wenjuanxing platform (https://www.wjx.cn/), and the survey link was distributed to the target population via social media, email, and university internal networks. To reduce self-selection bias, two measures were taken in this study: First, prior to the release of the questionnaire, with the assistance of various university departments, a unified invitation to participate in the survey was sent to all sampled college teachers to reduce the likelihood of only those with high JSF participating. Second, basic demographic information was collected to examine potential systematic differences during the data analysis phase. Additionally, all participants signed informed consent, agreeing to the research purpose and data use.

According to by Kline46, each item on the questionnaire should be answered by at least 10 respondents. The present questionnaire consisted of 82 items. Considering an anticipated attrition rate of approximately 20%, the final required sample size was 984 (82 items × 10 respondents + 20% buffer). 1,050 questionnaires were sent out, and 1,034 were returned. Fourteen were excluded during data cleaning for the following reasons: (1) more than 10 items were left unanswered (questionnaires missing more than 20% of items were deemed invalid); and (2) respondents selected the same extreme option (e.g., “strongly agree” or “strongly disagree”) for ≥ 80% of the items, which could distort the scale results and compromise the accuracy of the analysis47. Ultimately, 1,020 valid questionnaires were analyzed, with 567 male and 453 female participants (see Table 1 for demographics).

Table 1 Demographic characteristics of the sample.
Full size table

Measures

Job stress

The Job Stress Scale is a self-report measure that evaluates the level of JS in participants1,48. The scale consists of 24 items and includes five dimensions: Job Security, Teaching Security, Interpersonal Relationships, Workload, and Work Enjoyment. Job Security contains 8 items (e.g., “To what extent are you worried about unemployment?”), with a Cronbach’s α of 0.937. Teaching Security includes 5 items (e.g., “How stressed do you feel about managing difficult students?”), with a Cronbach’s α of 0.893. Interpersonal Relationships comprises 4 items (e.g., “How stressed do you feel when your supervisor does not trust you?”), with a Cronbach’s α of 0.883. Workload consists of 3 items (e.g., “How stressed do you feel about long working hours?”), with a Cronbach’s α of 0.848. Work Enjoyment includes 4 items (e.g., “How stressed do you feel about not enjoying your work?”), with a Cronbach’s α of 0.877. The scale uses a 5-point Likert scale from 1 (no stress) to 5 (extreme stress), with higher scores indicating higher JS. The Cronbach’s α was 0.973, indicating strong reliability.

Emotional intelligence

The Emotional Intelligence Scale is a self-report instrument used to assess participants’ levels of EI49. The scale contains 16 items across four dimensions: Self-Emotions Appraisal, Others-Emotions Appraisal, Use of Emotion, and Emotion Regulation. Self-Emotions Appraisal includes 4 items (e.g., “I often know clearly why I have certain feelings”), with a Cronbach’s α of 0.824. Others-Emotions Appraisal includes 4 items (e.g., “I can accurately sense my friends’ emotions through their behavior”), with a Cronbach’s α of 0.827. Use of Emotion includes 4 items (e.g., “I set goals for myself and strive to achieve them”), with a Cronbach’s α of 0.821. The Emotion Regulation scale contains 4 items (e.g., “I can manage my emotions and approach difficulties rationally”), with a Cronbach’s α of 0.824. Using a 5-point Likert scale, higher scores indicate higher EI. The overall Cronbach’s α was 0.950, showing high reliability.

Resilience

The Teacher Resilience Inventory (TRI) is a self-report tool for assessing RES levels, consisting of 20 items across five dimensions: physical, emotional, psychological, social, and spiritual resilience8. Physical resilience includes 4 items (e.g., “I rarely experience physical discomfort due to complex teaching procedures”), with a Cronbach’s α of 0.846. Emotional resilience includes 4 items (e.g., “I am able to understand my emotions and recognize how they influence my performance as a teacher”), with a Cronbach’s α of 0.803. Psychological resilience includes 4 items (e.g., “When I have clear professional goals and strive for them, I perform my teaching duties at my best”), with a Cronbach’s α of 0.803. Social resilience includes 4 items (e.g., “I am good at building positive relationships with students/colleagues in new environments”), with a Cronbach’s α of 0.808. Spiritual resilience includes 4 items (e.g., “Even in the face of resistance, I insist on teaching in accordance with professional ethics”), with a Cronbach’s α of 0.801. Using a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree), with higher scores indicating higher levels of RES. The Cronbach’s α coefficient was 0.956, demonstrating strong reliability.

Job satisfaction

The Job Satisfaction Scale is a self-report instrument used to evaluate JSF levels1. It includes 20 items (e.g., “I stay busy at work and manage my time well”) and uses a 5-point Likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree), with higher scores reflecting greater JSF. The Cronbach’s α was 0.957, indicating excellent reliability.

Two experts, one bilingual in Chinese and English and the other a linguist specializing in Chinese, translated the scales from English to Chinese, focusing on content and linguistic accuracy. Discrepancies were resolved through a collaborative review, and the final version was pilot-tested with 20 Chinese college teachers to address translation issues, resulting in the finalized Chinese adaptation.

Data analysis

This study used Partial Least Squares Structural Equation Modeling (PLS-SEM) to assess both the measurement and structural models. PLS-SEM has several strengths, including: first, it is a data analysis tool focused on prediction50. Second, it can handle small sample sizes and still exhibit good robustness even when the data distribution is non-normal50. Third, it is suitable for analyzing relatively complex models51. Given these characteristics, and considering that the sample size in this study was 1020 and the model structure involved 4 variables and 85 observed indicators, which introduced a certain degree of complexity, PLS-SEM was deemed highly appropriate for this research.

Results

Measurement validation

Reliability was examined using Cronbach’s α and Composite Reliability (CR), while convergent validity was measured through Average Variance Extracted (AVE)50. As shown in Table 2, all constructs had Cronbach’s α values of 0.7 or higher, CR values above 0.8, and AVE values exceeding 0.5, reflecting high reliability and convergent validity50.

Table 2 Reliability and validity assessment.
Full size table

Confirmatory Factor Analysis (CFA) was performed on the final model to check its structural, convergent, and discriminant validity. As shown in Table 3, the model’s fit indices met the required thresholds52,53.

Table 3 CFA results for the final model.
Full size table

The Fornell-Larcker criterion evaluates discriminant validity54. As shown in Table 4, the square root of each construct’s AVE exceeds its highest correlation with any other variable55, confirming discriminant validity.

Table 4 The Fornell-Larcker criterion.
Full size table

Another method for assessing discriminant validity is the Heterotrait-Monotrait Ratio (HTMT). Table 5 shows that all values fall below the 0.90 threshold55, thus confirming discriminant validity.

Table 5 Heterotrait-Monotrait Ratio.
Full size table

Structural model

The structural model was evaluated using multicollinearity tests, path coefficient significance tests, and the coefficient of determination (R²) to assess its reliability and explanatory power.

Collinearity test

The study assessed multicollinearity using the Variance Inflation Factor (VIF). Table 6 shows that the VIF values for all constructs ranged from 1.157 to 1.317, indicating no multicollinearity issue in the model, as all values are below the 3.3 threshold50.

Table 6 Collinearity test of the structural model.
Full size table

Path hypotheses

The significance tests in the structural model aimed to determine whether the exogenous constructs significantly influence the endogenous constructs. After controlling for covariates such as age, Table 7; Fig. 2 show that, among the direct significant predictors of JSF, EI (β = 0.372; t = 8.604; p = 0.000) and RES (β = 0.241; t = 6.012; p = 0.000) both significantly and positively influence JSF among college teachers, while JS (β = − 0.061; t = 2.108; p = 0.035) significantly negatively affects JSF. Furthermore, the study also found that EI (β = 0.365; t = 10.347; p = 0.000) significantly and positively influences RES among college teachers, while JS (β = − 0.094; t = 2.937; p = 0.003) significantly and negatively impacts RES. Therefore, H1 and H2 were supported.

Table 7 Path hypothesis testing Results.
Full size table
Fig. 2
figure 2

Mediating effect model. Note: JS, job stress; RES, resilience; EI, emotional intelligence; JSF, job satisfaction.

Full size image

R2 and Q2 of the structural model

The R2 indicates the extent to which the independent variables explain the dependent variables. Chin56 classifies R2 values as strong (0.67), moderate (0.33), and weak (0.19). As shown in Table 8, the R2 values for RES and JS are 0.168 and 0.295, respectively. Based on Hair Jr, et al.57, relying solely on R2 to assess model prediction accuracy is insufficient, so this study also uses Q2 to test predictive relevance58,59. A Q2 value greater than 0 signifies that the model has predictive relevance56. Table 8 shows Q2 values of 0.497 and 0.504 for RES and JS, respectively, confirming strong predictive relevance for all constructs.

Table 8 Predictive relevance of the structural model.
Full size table

Mediating effect analysis

This study used the bootstrap method in PLS-SEM60 to analyze the mediating role of RES in the relationship between JS, EI, and JSF. As shown in Table 9, RES partially mediates the effect of EI (β = 0.088, t = 5.228, p = 0.000) on JSF, and it also partially mediates the effect of JS (β = − 0.023, t = 2.565, p = 0.010) on JSF. Therefore, H3 and H4 were supported.

Table 9 Mediating effect analysis.
Full size table

Multi-group analysis

Measurement invariance analysis

Multigroup PLS analysis was employed in this study to examine differences in path coefficients across genders. This approach is widely utilized in current research61,62. Prior to testing the moderating effects, we conducted a measurement invariance test using the Measurement Invariance of Composite Models (MICOM) method to address any possible measurement bias concerns63. According to the framework outlined by Hair Jr, et al.57, effective multigroup analysis requires two key conditions: (1) structural invariance (i.e., the same model parameters and estimation methods are used across groups); and (2) composite invariance (i.e., the groups share the same indicator weights).

In this study, Smart-PLS 4.0 automatically ensured the establishment of structural invariance. For assessing composite invariance, we used a permutation algorithm for verification63. As shown in Table 10, composite invariance was verified for all constructs except RES (i.e., the correlation values exceeded the 5th percentile of the empirical distribution)53. Structural invariance was automatically met, as the same estimation algorithm was used for both groups63.

Table 10 MICOM step 2 compositional invariance: across males vs. females.
Full size table

Multigroup analysis

The results of the multi-group analysis indicate that, under the condition of ensuring measurement invariance, gender significantly moderates the relationships between JS, EI, RES, and JSF (see Table 11). The specific results are as follows: First, a significant difference was found between the male and female groups in the path “JS → JSF” (difference = − 0.262, t = 3.117, p = 0.002). The path coefficient for males was lower than for females, suggesting that female teachers’ JSF is more likely to decrease under JS, while the change in JSF among male teachers is relatively smaller. This result supports H5, indicating that females may face a greater psychological burden when coping with JS. Second, a significant difference was also found between the male and female groups in the path “EI → JSF” (difference = − 0.363, t = 4.071, p = 0.000). The male group’s path coefficient was lower than the female group’s, suggesting that EI has a more significant impact on JSF for female teachers. That is, female teachers’ EI can enhance their JSF to a greater extent, thus confirming H6. Finally, a significant difference was observed between the male and female groups in the path “RES → JSF” (difference = 0.229, t = 2.950, p = 0.003). The male group’s path coefficient exceeded that of the female group, indicating that male teachers’ RES has a more significant impact on JSF. In other words, male teachers are better able to cope with challenges at work through RES, thus enhancing their JSF, confirming H7. These differences may be related to gender-specific stress coping patterns and emotional regulation mechanisms64,65.

Table 11 Multigroup path analysis (male and female).
Full size table

Discussion

This study explored the effects of JS and EI on JSF among college teachers and further analyzed the mediating role of RES. Using PLS-SEM, the study confirmed all hypotheses. The results indicate that RES plays a significant mediating role in alleviating the negative impact of JS on JSF and in enhancing the positive effect of EI on JSF. Multi-group analysis showed that teachers of different genders experience varying degrees of influence from JS, EI, and RES on JSF.

Job stress (JS) and job satisfaction (JSF)

JS significantly negatively affected JSF among college teachers, supporting H1. This finding aligns with the results of Li, et al.11, who investigated JSF among teachers in four countries and found that high levels of JS significantly reduce teachers’ JSF. In our study, the negative impact of JS on JSF was also confirmed. When faced with excessive workloads, college teachers often feel emotionally exhausted and psychologically distressed, which not only affects their daily teaching performance but also lowers their overall satisfaction with work26. Furthermore, persistent JS may exacerbate teachers’ feelings of burnout, further affecting their professional commitment and work efficiency27. These findings stress the need to reduce JS to improve JSF and professional well-being among teachers.

Emotional intelligence (EI) and job satisfaction (JSF)

EI significantly positively influenced JSF among college teachers, supporting H2. This result is consistent with the study by Rogowska and Meres6, which conducted a cross-sectional study of 322 teachers and confirmed that EI effectively enhances JSF, thereby improving teachers’ life happiness and overall satisfaction. In our study, on one hand, EI helped teachers better understand and manage their own emotions, improving their emotional regulation ability and enabling them to cope more effectively with various emotional challenges and stress at work. Teachers with higher EI are better able to remain calm in complex work environments, which enhances their enthusiasm and satisfaction at work30. On the other hand, high EI also facilitated good interactions between teachers, colleagues, and students, improving the quality of interpersonal relationships at work66. This positive interaction and communication ability helped teachers gain more support and recognition at work, thereby increasing their overall JSF.

The mediating role of resilience (RES)

RES mediates the effect of JS on JSF among college teachers, supporting H3. While previous studies have not directly verified this mediation mechanism, independent evidence has provided theoretical support for the relationship among these three constructs. First, this study confirms the classical finding that JS negatively predicts JSF11,26, which is consistent with the COR theory of Hobfoll29. When college teachers face persistent JS, the depletion of their psychological resources leads to negative evaluations of their work. Second, RES, as a positive trait for adapting to stress, has been shown to enhance JSF by promoting positive emotional regulation67 and increasing coping efficacy68. Therefore, strengthening RES can help teachers effectively cope with external stressors, such as heavy teaching workloads, time constraints, and management challenges, thereby mitigating the impact of these stressors on JSF.

RES also mediates the effect of EI on JSF among college teachers, supporting H4. Although the direct effect of EI on JSF has been widely validated31,33, its underlying mechanisms still require further exploration. The results of this study show that EI not only directly enhances JSF but also indirectly promotes JSF by increasing RES. This mediating effect may arise from the dual role of EI in stress buffering and positive emotional regulation. Teachers with elevated EI are better equipped to handle occupational stress69, which reduces the buildup of negative emotions like anxiety and burnout70, thus enhances RES. RES plays a key role in this process; it is not only a mediator of emotional intelligence’s positive impact on JSF but also an important safeguard for teachers to maintain long-term professional adaptation. This study is the first to verify this mechanism within the college teacher population, addressing the gap in existing research, which has largely focused on the direct effects of EI while neglecting the mediating pathways.

The moderating role of gender

There is a significant gender difference in the effect of JS on JSF, supporting H5. Female teachers are more likely to experience a decline in JSF when facing JS, while male teachers show relatively smaller changes in JSF. This finding aligns with the research of Matud71 and Zhou19, showing that women use emotion-focused coping for JS, while men prefer problem-focused strategies72. This difference may stem from societal gender role expectations73, where women are often assigned higher emotional labor demands, making them more sensitive to the psychological impact of JS. Additionally, the “Tend-and-Befriend” theory from an evolutionary psychology perspective74 further supports this finding, indicating that women are more likely to seek social support and regulate their emotions under stress, while men are more likely to adopt a “Fight-or-Flight” strategy. Therefore, female college teachers may experience a decline in JSF due to long-term emotional exhaustion and excessive emotional labor75, which can affect their RES and career stability76. In contrast, male teachers often adopt problem-solving approaches to cope with stress, directly addressing and resolving specific work issues77, which may help them maintain higher JSF and mitigate the negative effects of stress.

The effect of EI on JSF also shows a significant gender difference, supporting H6. Female teachers experience a more significant improvement in JSF through EI. This finding aligns with the emotional intelligence theory of Sutherland78, which suggests that women are typically more sensitive and detailed in emotional recognition and management, allowing them to better regulate their emotional responses and improve their emotional state at work79. Female teachers may enhance their positive engagement with job tasks and increase JSF through higher emotional self-awareness and emotional regulation abilities when facing workplace challenges80. In contrast, while male teachers may not show significant differences in EI compared to females, they tend to cope with stress through external task resolution rather than emotional regulation72, resulting in a less pronounced effect of EI on JSF. This difference may be related to gender role expectations in socialization, where women are more often expected to demonstrate higher emotional expression and interpersonal skills21. As a result, they may receive more social support in emotional management, which in turn enhances their JSF.

There is also a significant gender difference in the effect of RES on JSF, supporting H7. Male teachers cope better with challenges at work through RES, leading to higher JSF. This result is consistent with Jia45, who found that males generally show better adaptability and stress resilience, helping them recover quickly and maintain a positive work attitude44. Men tend to view stress as an opportunity rather than a danger, adopting more constructive coping strategies such as problem-solving and goal-oriented behaviors81, which in turn enhances JSF. Compared to women, men may be more rational and intuitive in emotional control and stress management, enabling them to maintain higher RES when facing work difficulties22, thereby improving their overall JSF. The socialization process, which emphasizes “independence” and “decisiveness” for men, may lead them to focus more on self-regulation and control in stressful situations74, thereby increasing their JSF and adaptability through RES. This gender difference reflects the profound impact of gender roles on how RES functions.

Research implications

Theoretical implications

The theoretical innovations of this study lie in two main aspects. On one hand, this study developed a composite path model integrating multiple variables such as JS, EI, RES, and JSF, exploring how these variables jointly influence JSF among college teachers through mediation mechanisms. This model fills a theoretical gap regarding the mechanism by which JS and EI influence JSF and innovatively incorporates RES as a mediating variable, highlighting its critical role in the formation of teachers’ JSF. On the other hand, this research broadens the extension of resilience theory and the COR theory through the lens of gender differences and multi-group analysis. The study showed that gender significantly moderates the relationships between JS, EI, RES, and JSF, with female teachers more affected by JS and EI having a stronger impact on their JSF. In contrast, male teachers’ RES has a more significant impact on JSF. This finding provides an innovative theoretical contribution to current literature, accentuating the importance of gender in the influence of factors like JS and EI on teachers’ JSF.

Practical implications

At the teacher level, systematic interventions should be implemented to encourage teachers to actively develop emotional regulation and RES skills to enhance their adaptability to teaching stress and professional challenges. Specifically, based on the research findings that female teachers are more susceptible to emotional depletion, it is recommended that universities provide interventions focused on emotional regulation for female teachers, such as introducing mindfulness-based stress reduction (MBSR) training, conducting emotional expression and management workshops, and offering role conflict management courses. These measures can help enhance their awareness of stress, self-regulation, and ability to access social support, thus alleviating the decline in JSF caused by emotional labor. Meanwhile, since male teachers tend to cope with stress through problem-focused strategies, practical courses such as decision-making training and stress scenario simulations should be offered to enhance their goal-setting and problem-solving abilities when facing complex work situations, thereby increasing their sense of control over work outcomes and JSF. All teacher groups should be encouraged to participate regularly in mental health training, emotional awareness assessments, or reflection activities, such as using emotional recording tools, teaching scenario reflection writing, or accessing on-campus psychological counseling resources, to strengthen their long-term self-adjustment and recovery abilities. Additionally, based on the indirect suggestion from this study regarding career motivation variables, career goal-setting guidance and self-efficacy enhancement programs can further improve teachers’ sense of professional belonging and task engagement.

At the university management level, it is essential to establish gender-sensitive psychological support and teacher development systems. For female teachers, priority should be given to strengthening social support networks, such as setting up female teacher support groups, offering green channels for psychological counseling, and providing emotional assistance hotlines to reduce stress caused by family role conflicts or emotional labor. For male teachers, administrative processes should be optimized, and technical teaching support and resource integration services should be provided to reduce structural barriers encountered in problem-solving, thereby enhancing their work efficacy. Furthermore, universities can implement systematic EI and RES development programs, designing gender-specific modular courses, such as focusing on emotional expression, empathy, and emotional regulation for female teacher training, and focusing on task orientation, situational response, and RES training for male teachers’ capacity building. Managers should also assist teachers in achieving work-life balance by reasonably allocating teaching and management tasks, exploring flexible work systems, and implementing job rotation and paid emotional recovery systems, with particular attention to the continuous emotional load on female teachers and the support needs of male teachers in high-pressure tasks.

At the educational policy level, it is recommended to formulate teacher mental health development policies that recognize gender differences. For example, gender-oriented content should be incorporated into teacher qualification training and continuing education systems, with specialized training focused on emotional regulation and emotional labor management for female teachers, and practical courses on goal management and stress-constructive coping for male teachers. Policymakers should support the establishment of diversified psychological service platforms, with group counseling mechanisms and problem-solving-based personal growth paths in universities, ensuring that the service formats match gender preferences. Additionally, a precise matching mechanism between “stress type—gender traits—intervention program” should be promoted to ensure the effectiveness of the “RES→JSF” pathway while providing more differentiated and targeted psychological support resources. Ultimately, a comprehensive development support system that focuses on both teachers’ emotional well-being and enhances their professional efficacy and psychological adaptability is expected to be established.

Limitations and future research directions

A This study examined the relationships between JS, EI, RES, and JSF, verifying the mediating role of RES and the moderating role of gender, but it has some limitations. First, the reliance on questionnaire surveys may limit participants’ understanding of their true JSF. Future research could combine interviews and case studies for more authentic data. Second, the study focused on specific psychological factors, but future research should include variables like job engagement and personality traits for a more holistic model. Additionally, being cross-sectional, it doesn’t reveal long-term changes in JS, EI, RES, and JSF. Future studies should adopt longitudinal designs to explore their evolution and long-term effects. Lastly, the sample in this study was confined to a particular region, limiting its applicability to the wider population. To strengthen the generalizability of the results, future studies should involve a broader sample from multiple regions.

Conclusion

This study, based on a sample of 1020 college teachers, constructs and validates a structural equation model that includes both mediating and moderating pathways. The aim is to explore the mechanisms by which JS, EI, and RES affect JSF, and to further examine the moderating role of gender through multi-group analysis. The results indicate a significant negative correlation between JS and JSF, while EI is positively correlated with JSF. RES mediates the relationship between JS and JSF and also partially mediates the relationship between EI and JSF. Additionally, gender shows a significant moderating effect across several pathways: for female teachers, the negative impact of JS on JSF is more pronounced, and the positive influence of EI on JSF is more prominent; for male teachers, the positive effect of RES on JSF is more evident. Based on the validation of the proposed hypotheses, this study attempts to integrate psychological resource variables into the job satisfaction model, enriching the application perspectives of resilience theory and COR theory in high-pressure occupational groups. The findings provide theoretical support for understanding the psychological mechanisms of JSF among college teachers and offer practical insights for university administrators in developing intervention strategies to enhance teachers’ emotional regulation abilities and adaptive resources, particularly in designing gender-sensitive support programs.