Study on the Effect of Supervisory Style on Students’ Innovativeness From the Perspective of Dual Stress

Abstract

Supervisors serve as the primary responsible parties in the cultivation of graduate students, and their supervisory style exerts a crucial influence on students’ innovativeness. Grounded in the cognitive appraisal theory of stress, this study investigates the impact of different types of supervisor supervisory styles on graduate students’ innovativeness through paired questionnaire surveys of both supervisors and students. Furthermore, it examines the mediating role of stress and the moderating effect of psychological resilience. The findings indicate that supportive supervisory style positively influences graduate students’ innovativeness through challenge stress, whereas controlling supervisory style negatively affects innovativeness via hindrance stress. Additionally, psychological resilience strengthens the positive relationship between challenge stress and innovativeness, while mitigating the negative impact of hindrance stress on innovativeness. These results elucidate the underlying pathways through which different supervisory styles influence graduate students’ innovativeness via stress mechanisms, providing theoretical foundations and practical insights for optimizing supervisory practices and enhancing the innovative competencies of graduate students.

Keywords

supportive supervisory style controlling supervisory style challenge stress hindrance stress psychological resilience innovativeness

Introduction

As a vital force in scientific and technological innovation, graduate students’ innovativeness directly determines the potential for research breakthroughs and the efficacy of industrial transformation, serving as a key indicator of educational quality and a nation’s future competitiveness. Currently, graduate students’ performance in technological innovation has not fully met expectations, indicating significant shortcomings in the systematicness and effectiveness of innovativeness cultivation (H. Li & Li, 2023). Particularly in China, graduate education is grappling with issues such as the “laissez-faire” approach, “squeezed supervision,” and inadequate guidance, which severely constrain the quality of high-level innovative talent development (S. Bao & Yang, 2021). Against this backdrop, supervisors, as the primary responsible parties in graduate student cultivation, have seen their supervisory style become a core factor influencing students’ innovativeness (Han et al., 2022). Delving into the driving mechanisms of graduate students’ innovativeness from the perspective of supervisor supervisory styles not only helps clarify the pathways behind “dysfunctional supervision” but also provides a critical entry point for addressing the practical challenges in Chinese graduate education and constructing an efficient cultivation system, thereby systematically promoting the comprehensive enhancement of graduate students’ innovative competencies and research efficacy.

Existing research has explored the mechanisms through which supervisor supervisory styles influence graduate students’ innovativeness from various perspectives, including creative self-efficacy (Han et al., 2022), intrinsic motivation (Gu et al., 2017; S. Li et al., 2025), psychological capital (B. Yang et al., 2022), and network ties (Y. Zhang et al., 2024). However, on one hand, numerous studies have shown that both intrinsic and extrinsic motivations enable leaders to exert significant—whether positive or negative—influences on individual creativity (J. Zhou and Shalley, 2003). Within the extended research on leadership styles derived from this premise, the supportive and controlling leadership styles proposed by Oldham and Cummings are the most representative (Oldham and Cummings, 1996; X. Zhang & Bartol, 2010; Y. Zhou, 2023) and have been widely applied in educational contexts. Yet, existing studies on the impact of supervisor supervisory styles on graduate students’ innovativeness have predominantly focused on supportive supervisory styles, with limited attention paid to the effects of controlling supervisory styles. On the other hand, while prior research has emphasized the importance of individual cognitive complexity for creative thinking (Runco & Chand, 1995), few studies have examined the influence of supervisor supervisory styles on graduate students’ innovativeness from a cognitive perspective, even though supervisory styles are recognized as closely related to students’ cognitive changes (Mavrogalou-Foti et al., 2024). In this vein, this study aims to unveil the mediating mechanisms through which supportive and controlling supervisory styles affect graduate students’ innovativeness from a cognitive perspective.

Stress refers to an individual’s subjective perception and appraisal of external environmental demands (stressors). The cognitive appraisal theory of stress emphasizes that stressors trigger cognitive evaluations in individuals, leading to stress responses that subsequently guide their actions (Lazarus & Folkman, 1984). This provides a theoretical foundation for understanding the relationship between supervisor supervisory styles and graduate students’ innovativeness from a cognitive perspective, as supervisory styles can act as potential stressors for students (Tariq et al., 2023; Wu et al., 2013), and different types of stress significantly influence individual creativity (Antwi et al., 2019). Cavanaugh et al. (2000) further extended this theory by distinguishing between challenge stress, which arises from factors such as job scope, sense of responsibility, and working hours, and is conducive to skill development and achievement, and hindrance stress, which stems from role ambiguity and insecurity, and tends to constrain individual growth. We propose that supportive supervisory style is likely to induce challenge stress, whereas controlling supervisory style may trigger hindrance stress, thereby exerting differential effects on graduate students’ innovativeness.

Furthermore, how individuals perceive and respond to stress plays a crucial role in understanding its impact (Hartmann et al., 2020). According to the Job Demands-Resources (JD-R) model, job resources are pivotal in helping individuals cope with job demands and facilitate positive work outcomes (Bakker & Demerouti, 2007, 2024). Although this theory originated in the occupational context, its applicability to the educational domain has been widely validated (Awais et al., 2024). Psychological resilience, defined as the ability to dynamically integrate psychological resources and behavioral strategies to effectively adapt, recover, and even grow in the face of setbacks, adversity, or uncertainty, is considered a critical resource for coping with stress (Shin et al., 2012). Therefore, we theorize graduate students’ psychological resilience as a key boundary condition in the aforementioned mechanisms.

Building on these insights, this study aims to explain, based on the cognitive appraisal theory of stress, how supervisor supervisory styles influence graduate students’ innovativeness by triggering cognitive appraisals that alter stress perceptions, and to examine the boundary role of psychological resilience in this process, as illustrated in Figure 1. This research contributes to the literature in three key aspects: First, it reinterprets the relationship between supervisory styles and graduate students’ innovativeness from the perspective of the cognitive appraisal theory of stress, thereby expanding existing research through a stress-cognitive lens. Second, by conceptualizing supportive supervisory style as a stressor, it extends stress-related research and refines the identification and definition of stressors within the cognitive appraisal theory of stress. Third, integrating insights from the Job Demands-Resources model, it proposes and tests the moderating role of psychological resilience, exploring the theoretical boundaries of the cognitive appraisal theory of stress.

Figure 1.

Framework of the study.

Theoretical Analysis and Research Hypotheses

The Cognitive Appraisal Theory of Stress and JDR Model

The cognitive appraisal theory of stress, proposed by Lazarus and Folkman (1984), serves as a foundational theoretical framework in stress research. This theory posits that stress does not arise directly from external events but rather through an individual’s cognitive appraisal of those events. Grounded in this perspective, our study examines how external contexts or behaviors trigger individuals’ cognitive appraisal processes, thereby elevating their stress levels, and treats stress states as a critical mediating mechanism influencing behavioral performance (Lee et al., 2018; D. Liu et al., 2021). The Job Demands-Resources (JD-R) model offers another essential lens for understanding stress mechanisms in environmental contexts (Bakker & Demerouti, 2007), emphasizing that resources play a pivotal role in helping individuals cope with demands, mitigating the negative effects of high demands, and fostering positive behavioral outcomes (Bani-Melhem et al., 2020). Integrating the cognitive appraisal theory of stress with the JD-R model provides a more comprehensive understanding of stress mechanisms in graduate education environments. Based on this integration, we propose the following theoretical framework: supervisor supervisory styles influence graduate students’ innovativeness by triggering cognitive appraisals that induce stress, and psychological resilience serves as a key moderating factor in this process.

Impact of Supervisory Style on Graduate Students’ Stress

Based on the cognitive appraisal theory of stress, different types of supervisory styles can elicit distinct cognitive appraisals from graduate students regarding their abilities, research tasks, and environments, consequently leading to either challenge stress or hindrance stress. Supportive supervisory style is often closely associated with the formation of challenge stress. On one hand, supportive supervisors typically exhibit high levels of care, encouragement, and resource provision, fostering an open and trusting research atmosphere (Hamid et al., 2021). In such an environment, students are more willing to undertake challenging research tasks and perceive them as opportunities for growth and development (Ahsan, 2025), thereby enhancing challenge stress. On the other hand, through clear goal-setting and positive feedback, supervisors convey expectations, trust, and recognition (Penning et al., 2023), which motivates students to transform external demands into intrinsic motivation and hold themselves to higher academic standards, further elevating challenge stress.

Conversely, controlling supervisory style is more likely to trigger hindrance stress. First, controlling supervisors often exhibit highly directive behaviors, strict supervision, and limited autonomy granting (Basten et al., 2014). Such an approach tends to make students feel constrained, lacking autonomy, and diminishes their self-assessment of value and competence, thereby increasing hindrance stress. Second, controlling supervision is frequently accompanied by unclear task expectations and one-way communication (Gu et al., 2017), making it difficult for students to accurately understand their supervisor’s expectations and receive effective feedback, further elevating hindrance stress. Third, the critical feedback characteristic of controlling style may exacerbate students’ psychological burden and deplete their psychological resources (Wichmann-Hansen & Schmidt Nielsen, 2024), leading them to perceive research tasks as threats, thereby intensifying hindrance stress. Based on this, the following hypotheses are proposed:

H1a: Supportive supervisory style is positively associated with challenge stress.

H1b: Controlling supervisory style is positively associated with hindrance stress.

Mediating Role of Stress

The content and nature of challenge stress and hindrance stress differ, leading individuals to exhibit varying levels of innovativeness when confronted with these two types of stress. Challenge stress typically arises from research tasks that are goal-oriented, demanding yet achievable, and it tends to stimulate positive cognitive and proactive behaviors in graduate students. On one hand, high-challenge tasks often involve significant complexity and uncertainty, requiring students to break away from conventional thinking and explore novel methods and approaches (Ohly, 2019), thereby directly fostering innovative thinking and creative problem-solving abilities (Y. Xu et al., 2024). On the other hand, challenge stress is often accompanied by high anticipated rewards (e.g., academic achievements, supervisor recognition, personal growth), motivating students to invest time, effort, and cognitive resources when they perceive potential benefits (Byron et al., 2018), and sustaining their engagement in innovative activities. In contrast, hindrance stress often stems from situations characterized by role ambiguity, unclear tasks, lack of autonomy, or excessive control. Such stress depletes students’ cognitive and behavioral resources. On one hand, hindrance stress typically does not emphasize innovative requirements and may even shift students’ focus toward coping with the stress itself (e.g., avoiding criticism, completing directive tasks), thereby reducing their attention and investment in innovative activities (Y. Liu & Ren, 2022). On the other hand, this type of stress can easily trigger anxiety, self-doubt, and emotional exhaustion (D. Bao et al., 2024), significantly diminishing the psychological and cognitive resources available for deep thinking, experimentation, and risk-taking, ultimately suppressing their innovative willingness and performance. Building on the arguments presented in H1a and H1b, the following hypotheses are proposed:

H2a: Challenge stress mediates the relationship between supportive supervisory style and graduate students’ innovativeness.

H2b: Hindrance stress mediates the relationship between controlling supervisory style and graduate students’ innovativeness.

Moderating Role of Psychological Resilience

Psychological resilience, as a key psychological resource for individuals to cope with stress, adapt to challenges, and overcome obstacles, plays a critical moderating role in the relationship between stress and graduate students’ innovativeness. According to the Job Demands-Resources (JD-R) model (Bakker & Demerouti, 2007, 2024), psychological resilience can enhance students’ learning engagement and motivation when facing stress, thereby contributing to improved innovativeness. When confronted with challenge stress, students with high psychological resilience typically exhibit greater self-confidence, optimism, and emotional regulation abilities. They are more likely to perceive challenging tasks as opportunities for growth and learning (Yeager & Dweck, 2012). This positive cognitive appraisal strengthens their intrinsic motivation, encouraging them to proactively engage in highly creative and complex research tasks. Additionally, highly resilient students are enthusiastic about pursuing new experiences and novel ideas in their learning. When faced with challenging academic demands, they tend to avoid conventional or established approaches and instead explore innovative solutions (De Clercq & Pereira, 2019). In the context of hindrance stress, students with high psychological resilience demonstrate greater psychological adaptability and recovery capabilities. They employ strategies such as emotional regulation and positive reframing to mitigate the negative emotions and cognitive load caused by role ambiguity, interpersonal conflicts, or excessive control (Hartmann et al., 2020), thereby reducing the depletion of psychological resources induced by hindrance stress. Furthermore, psychological resilience helps students maintain focus on their personal learning and development goals, enhancing their proactivity and flexibility in overcoming adversity, which in turn supports their engagement in innovative activities (Z. Xu et al., 2025). Based on this, the following hypotheses are proposed:

H3a: Psychological resilience moderates the relationship between challenge stress and graduate students’ innovativeness (i.e., the greater the psychological resilience, the stronger the positive effect of challenge stress on graduate students’ innovativeness).

H3b: Psychological resilience moderates the relationship between hindrance stress and graduate students’ innovativeness (i.e., the greater the psychological resilience, the weaker the negative effect of hindrance stress on graduate students’ innovativeness).

To synthesize these hypotheses regarding mediating and moderating effects, we constructed a moderated mediation model in the present study. Specifically, psychological resilience moderates the mediating effect of graduate students’ stress between supervisory styles and students’ innovativeness. The following hypotheses are proposed:

H4a: Psychological resilience positively moderates the mediating role of challenge stress between supportive supervisory style and graduate students’ innovativeness (i.e., the greater the psychological resilience, the stronger the mediating role of challenge stress).

H4b: Psychological resilience negatively moderates the mediating role of hindrance stress between controlling supervisory style and graduate students’ innovativeness (i.e., the greater the psychological resilience, the weaker the mediating role of hindrance stress).

Research Design

Data Sources

Given that both master’s and doctoral students generally require innovative capabilities, a convenience sampling method was adopted to collect data more efficiently and effectively (S. Li et al., 2025; B. Yang et al., 2022). This study targeted graduate students (including both master’s and doctoral students) from six universities in Zhejiang Province. These institutions encompass comprehensive, science and engineering, and finance and economics universities, providing regional representativeness and disciplinary diversity. First, the research team obtained potential participant lists from graduate schools or academic affairs offices of these institutions and contacted supervisors to explain the research purpose and procedures. Subsequently, an online survey platform was used to generate electronic questionnaire links, and data were collected in three waves with a 3-month time lag (C. Yang et al., 2021). This approach was chosen because the impact of supervisory styles on student’s innovativeness manifests over time, and multi-wave data can effectively capture this lagged effect. In the first wave, questionnaires assessing supportive and controlling supervisory styles were distributed to graduate students, resulting in 830 valid responses from 182 research groups. In the second wave, the same cohort of students received questionnaires measuring challenge stress, hindrance stress, and psychological resilience, yielding 812 valid responses from 180 research groups. In the third wave, 178 supervisors evaluated the innovativeness of 805 graduate students. After matching and excluding invalid responses, 755 valid paired questionnaires from 170 supervisors and 755 students were retained, representing a valid response rate of 94%. Based on our analysis of the sample’s basic characteristics, we obtained several key insights. In terms of gender, 57.6% of the respondents were men and 42.4% were women. Regarding education level, most respondents were master’s students (79.2 %), with doctoral students comprising the remaining 20.8%. In terms of discipline distribution, 45.7% of the respondents were from the natural sciences; 38.8%, from the social sciences; and 15.5%, from the humanities. Team sizes varied, with 85.9% of the respondents belonging to teams with fewer than five members, 10.6% on teams with 6 to 10 members, and 3.5% on teams with more than 11 members.

During the data collection process, this study strictly adhered to scientific research ethics norms and implemented the following measures to safeguard the rights and safety of the participants: First, this study employed anonymous questionnaire surveys that did not involve sensitive personal information or physiological interventions. The content focused on academic guidance behaviors and psychological perceptions, ensuring that no physical or psychological harm would be caused to participants. Data were collected through an encrypted online questionnaire platform, accessible only to the research team. During storage and analysis, all data were de-identified to ensure security and privacy protection. Second, prior to distributing the questionnaires, all potential participants (both graduate students and their supervisors) received a written informed consent statement. This document clearly explained the research purpose, procedures, data usage, confidentiality measures, and the principle of voluntary participation. Participants were required to check the box “I have read and agree to participate in this study” on the first page of the questionnaire before proceeding. Those who did not provide consent were unable to participate in the survey. Third, this study provides valuable insights into how supervisory styles influence graduate students’ innovation capabilities, with findings offering evidence-based support for improving mentor training and graduate education policies. Participation also encouraged self-reflection on academic growth and stress management.

Variable Measurement

The questionnaire used in this study was scored on a seven-point Likert scale ranging from 1: completely disagree to 7: completely agree.

Supervisory Style

Supervisory style aims to measure the behavioral patterns and interaction approaches adopted by supervisors in their daily guidance of graduate students. A revised version of the Supervisory Style Scale developed by Oldham and Cummings (1996) was used. The scale includes seven items for the supportive supervisory style (Cronbach’s α = .949), such as “My supervisor helps me solve problems in my research work,” and seven items for the controlling supervisory style (Cronbach’s α = .927), such as “My supervisor often communicates with me by assigning tasks.”

Stress

Stress aims to measure the level and nature of stress perceived by graduate students in their research activities. We adopted the stressor scale developed by Cavanaugh et al. (2000) and adapted it into a stress scale based on the approach of Geng et al. (2014), with further revisions to align with the graduate student context. The challenge stress scale includes six items (Cronbach’s α = .910), such as “I have a large number of projects and tasks to handle.” The hindrance stress scale includes four items (Cronbach’s α = .924), such as “I often cannot clearly understand my supervisor’s expectations for my research.”

Psychological Resilience

Psychological resilience aims to measure graduate students’ capacity to mobilize positive adaptation and recovery in the face of research-related adversity and setbacks. The scale developed by Connor and Davidson (2003) was revised to include 10 items (Cronbach’s α = .959), such as “I will do my best regardless of the outcome.”

Innovativeness

Innovativeness aims to measure the behavior and outcomes of graduate students in generating novel and valuable ideas in their research work. The scale developed by J. Zhou and George (2001) was revised and includes 13 items (Cronbach’s α = .896), such as “He/She often comes up with creative solutions to problems”.

Control Variables

This study incorporated graduate students’ gender, academic level, discipline, and research team size as control variables. Gender was coded as a binary variable (0 = male, 1 = female). Academic level was represented numerically, with 1 indicating master’s students and 2 representing doctoral students. Discipline was captured using dummy variables for humanities, social sciences, and natural sciences. Research team size was treated as a continuous variable.

Research Results

Common Method Bias

Although this study employed supervisor-student paired questionnaires and included reverse-scored items for ex ante control, all variables except the dependent variable were self-reported by students, potentially introducing common method bias. To address this concern, Harman’s single-factor test was conducted to assess common method variance. All measurement items were subjected to an unrotated exploratory factor analysis, which identified nine factors with eigenvalues greater than 1, collectively accounting for 67.98% of the total variance. Among these, the first factor with an eigenvalue greater 1 explained 21.44% of the variance, which is below the critical threshold of 40% for common method bias concerns. This indicates that common method bias does not pose a significant issue in this study.

Confirmatory Factor Analysis

This study employed confirmatory factor analysis (CFA) to examine the discriminant validity among latent variables and constructed models ranging from a single-factor to a six-factor structure. The six-factor model demonstrated excellent fit indices (χ²/df = 3.512, TLI = 0.900, CFI = 0.906, SRMR = 0.043, RMSEA = 0.058) and was significantly superior to the other models. This confirms the distinctiveness of the six conceptual constructs—supportive supervisory style, controlling supervisory style, challenge stress, hindrance stress, psychological resilience, and innovativeness—and supports their use in subsequent analyses.

Descriptive Statistics and a Correlation Analysis

This study calculated the means, standard deviations, and correlation coefficients for all variables, with the results presented in Table 1. Supportive supervisory style showed significant positive correlations with challenge stress (r = .327, p < .01) and innovativeness (r = .427, p < .01). Controlling supervisory style was significantly positively correlated with hindrance stress (r = .362, p < .01) and significantly negatively correlated with innovativeness (r = −.109, p < .01). Challenge stress demonstrated a significant positive correlation with innovativeness (r = .408, p < .01), while hindrance stress exhibited a significant negative correlation with innovativeness (r = −.150, p < .01). These findings provide preliminary support for the hypotheses proposed in this study.

Table 1.

Descriptive Statistics and Correlation Analysis.

Variable	1	2	3	4	5	6	7	8	9	10	11
1. Gender	1
2. Education qualifications	−.100**	1
3. Humanities	−.057	.182**	1
4. Social sciences	−.217**	.381**	−.036	1
5. Team members	−.224**	−.013**	.052	.428**	1
6. Supportive supervisory style	−.037	.049	−.029	.048	.088*	1
7. Controlling supervisory style	.105**	−.109**	−.040	−.235**	−.118**	.172**	1
8. Challenge stress	.023	.011	.004	.001	.064	.327**	.249**	1
9. Hindrance stress	−.017	−.021	.044	−.106**	−.083*	−.264**	.362**	.214**	1
10. Psychological resilience	.128**	.064	.003	−.024	.039	.285**	−.053	.127**	−.273**	1
11. Innovativeness	−.013	.095**	−.005	.069	.061	.427**	−.109**	.408**	−.150**	.195**	1
Mean	0.580	1.080	0.010	0.120	1.180	6.088	4.836	4.718	3.681	5.295	4.715
SD	0.494	0.264	0.096	0.329	0.465	1.036	1.236	1.139	1.500	1.133	0.824

Note. n = 755.

p < .05. **p < .01. ***p < .001 (applies to all tables).

Hypothesis Testing

Mediating Effect Testing

The results of the mediation effects test are presented in Table 2. Model 1 illustrates the significant and positive effect of supportive supervisory style on challenge stress (β = .324, p < .001), thus validating H1a. Model 3 illustrates the significant and positive effect of supportive supervisory style on innovativeness (β = .421, p < .001). The results from Model 4 showed that the effect of challenge stress on innovativeness was significantly positive (β = .302, p < .01), while the effect of supportive supervisory style on innovativeness remained positive but was attenuated (β = .323, p < .001). This indicates that challenge stress partially mediates the relationship between supportive supervisory style and innovativeness, thus supporting H2a.

Table 2.

Hierarchical Regression Analysis.

Variable	Challenge stress	Hindrance stress	Innovativeness
Variable	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
Gender	.042	−.066	.018	.005	.016	.008	−.026	−.026
Education qualifications	.011	.006	.071	.068	.098*	.099*	.078*	.076*
Humanities	.009	.055	−.005	−.008	−.030	−.023	−.016	−.010
Social sciences	−.035	−.013	.016	.027	−.015	−.017	.009	.006
Team members	.060	−.052	.022	.004	.063	.056	.005	−.001
Supportive supervisory style	.324***		.421***	.323***
Controlling supervisory style		.363***			−.097**	−.052
Challenge stress				.302**			.458***	.456***
Hindrance stress						−.125***	−.218***	−.227***
Psychological resilience							.072*	.077*
Challenge stress × Psychological resilience							.066*
Hindrance stress × Psychological resilience								.104***
R ²	.111	.140	.189	.270	.023	.036	.243	.249
F	15.532***	20.360***	29.028***	39.427***	2.880***	3.985***	26.537***	27.483***

Similarly, results from Model 2 indicated that controlling supervisory style had a significant and positive effect on hindrance stress (β = .363, p < .001), thus validating H1b. Model 5 illustrates the significant and negative effect of controlling supervisory style on innovativeness (β = −.097, p < .01). Results from Model 6 showed that hindrance stress had a significant and negative effect on innovativeness (β = −.125, p < .05) while the effect of controlling supervisory style on innovativeness became statistically insignificant (β = −.052, p > .05). This indicates that hindrance stress fully mediates the relationship between controlling supervisory style and innovativeness, thereby supporting H2b.

Furthermore, the Bootstrap method was employed to validate the mediating roles of challenge stress and hindrance stress. The results demonstrated that the indirect effect of supportive supervisory style on innovativeness through challenge stress was statistically significant (Indirect effect = 0.078, 95% CI [0.050, 0.106]). Similarly, the indirect effect of controlling supervisory style on innovativeness through hindrance stress was also significant (Indirect effect = −0.030, 95% CI [−0.054, −0.008]). These findings are consistent with the results obtained from the hierarchical regression analysis, further confirming the robustness of the mediating mechanisms.

Moderation Effect Testing

Challenge stress, hindrance stress, and psychological resilience were centered on constructing an interaction term between the independent and moderating variables to determine whether there was a moderating effect. Models 7 and 8 showed that the interaction term between challenge stress and psychological resilience had a significant and positive effect on innovativeness (β = .066, p < .05). This finding indicates that the greater the psychological resilience, the stronger the positive effect of challenge stress on innovativeness, thus validating H4a. The interaction term between hindrance stress and psychological resilience had a significant and positive effect on innovativeness (β = .104, p < .001), indicating that the greater the psychological resilience, the weaker the negative effect of hindrance stress on innovativeness, thus validating H4b. The moderating effects are illustrated in Figures 2 and 3.

Figure 2.

The interaction between challenge stress and psychological resilience on innovativeness.

Figure 3.

The interaction between hindrance stress and psychological resilience on innovativeness.

Moderated Mediation Effects Testing

In this study, the bootstrap method was used to test for moderated mediation effects. Regarding challenge stress, when psychological resilience was at a low level, the indirect effect of supportive supervisory style on innovativeness through challenge stress was significant (95% CI [0.057, 0.134]). When psychological resilience was high, the indirect effect remained significant (95% CI [0.099, 0.188]). The index of moderated mediation was significant (Index = 0.021, 95% CI [0.001, 0.039]). This indicates that psychological resilience positively moderates the indirect relationship between supportive supervisory style and innovativeness through challenge stress, thus supporting H4a. Similarly, for hindrance stress, when psychological resilience was low, the indirect effect of controlling supervisory style on innovativeness through hindrance stress was significant (95% CI [−0.094, −0.027]). However, when psychological resilience was high, the indirect effect became non-significant (95% CI [−0.040, 0.008]). The index of moderated mediation was significant (Index = 0.018, 95% CI [0.002, 0.035]). This demonstrates that psychological resilience negatively moderates the indirect relationship between controlling supervisory style and innovativeness through hindrance stress, confirming H4b.

Discussions

Based on an analysis of 755 supervisor-graduate student paired questionnaires in China, the study found that supportive supervisory style enhances graduate students’ innovativeness by triggering challenge stress, while controlling supervisory style undermines innovativeness by inducing hindrance stress. Furthermore, psychological resilience strengthens both the positive impact of challenge stress on innovativeness and the indirect effect of supportive supervisory style on innovativeness through challenge stress. It also mitigates the negative influence of hindrance stress on innovativeness and weakens the indirect effect of controlling supervisory style on innovativeness through hindrance stress.

Theoretical Implication

First, this study systematically unveils the internal mechanism through which supervisor supervisory styles influence graduate students’ innovativeness by introducing the cognitive appraisal theory of stress from a cognitive perspective, thereby extending existing research. While previous studies have predominantly explored the relationship between supervisory styles and innovativeness from resource-based (B. Yang et al., 2022; Y. Zhang et al., 2024) or motivational perspectives (Gu et al., 2017; Han et al., 2022; S. Li et al., 2025), few have conducted in-depth analyses from a cognitive lens. This study innovatively conceptualizes supportive and controlling supervisory styles as distinct types of stressors and elucidates how they affect innovativeness through students’ cognitive appraisal processes via a dual-path mediation model involving challenge and hindrance stress. This approach expands the mechanistic understanding of the relationship between supervisory styles and innovativeness through the lens of stress cognition.

Second, this study enriches the understanding of stressors and extends stress research as well as the identification and definition of stressors within the cognitive appraisal theory of stress. Previous stress research and the cognitive appraisal theory have primarily focused on negative events or behaviors as stressors, such as exploitative leadership (Syed et al., 2021) and leader incivility (Kabat-Farr et al., 2019). However, they have overlooked the fact that events or behaviors widely perceived as positive and constructive, such as transformational leadership (X. Xu et al., 2021) and empowering leadership (Ni, 2020), can also act as stressors. By identifying supportive supervisory style as a stressor and discussing how it enhances graduate students’ innovativeness through challenge stress, this study expands stress-related research and contributes to a more comprehensive understanding of the cognitive appraisal theory of stress.

Third, this study extends the application of the cognitive appraisal theory of stress to the intersection of organizational and educational contexts and clarifies its theoretical boundaries. By incorporating challenge and hindrance stress, this research reveals the complete mechanism through which supervisory styles are cognitively appraised by students and transformed into different types of stress, subsequently influencing innovative behavior. This deepens the understanding of the “cognition-stress-behavior” chain reaction. Furthermore, by integrating the Job Demands-Resources (JD-R) model into the framework of the cognitive appraisal theory, this study empirically examines the moderating role of psychological resilience as a critical individual resource in the stress transmission pathway, thereby exploring the theoretical boundaries of the cognitive appraisal theory of stress.

Practical Implications

First, supervisors should actively develop and practice supportive supervisory style as a core strategy to enhance graduate students’ innovativeness. Empirical findings from this study demonstrate that supportive supervisory style effectively stimulates challenge stress, transforming it into a driver of innovation. Therefore, supervisors should move away from traditional “strict mentor” approaches and controlling management models, and instead foster a safe, trusting, and challenging academic environment by providing emotional support, resource facilitation, autonomy, and constructive feedback. Specifically, supervisors should respect students’ agency, encourage independent exploration of research directions, offer scaffolded assistance rather than one-way directives when students encounter difficulties, and reinforce students’ innovative confidence and intrinsic motivation through recognition and encouraging feedback. This requires universities to prioritize supportive mentoring competencies in supervisor training and evaluation, guiding the transition of mentor–student relationships from “command-obedience” to “inspiration-collaboration,” thereby fundamentally empowering the innovation process.

Second, universities and supervisors should collaboratively establish stress transformation mechanisms to systematically convert inevitable research-related pressures into innovation momentum. The results indicate that challenge stress and hindrance stress play distinct roles in influencing innovativeness. Thus, scientifically managing and transforming stress is essential for improving the quality of graduate education. Universities should optimize training programs and evaluation systems by clarifying phased goals and graduation criteria to reduce hindrance stress stemming from role ambiguity and institutional uncertainty, thereby providing students with clear and stable research expectations. They should also integrate internal and external resources to offer abundant academic exchange platforms, interdisciplinary projects, and seed funding, embedding high-level academic challenges within ample resource support. This shifts the source of stress from supervisory control to task challenges, thereby catalyzing challenge stress. Additionally, efforts should be made to create an academic culture that encourages exploration and tolerates failure, enabling students to propose novel ideas and experiment with new methods in a psychologically safe environment, thus mitigating the breeding ground for hindrance stress.

Third, psychological resilience training should be incorporated into graduate education systems to provide students with internal buffers and growth resources for coping with research pressures. This study shows that psychological resilience enhances students’ ability to leverage challenge stress while buffering the negative effects of hindrance stress. Therefore, universities and supervisors should systematically cultivate psychological resilience as a core competency equally important as academic skills. Specifically, this can be achieved by offering courses, workshops, and group counseling sessions on mental health and stress management to help students develop strategies for emotional regulation and cognitive reframing. In daily supervision, supervisors should adopt a growth mindset in their feedback, emphasizing the value of effort and improvement rather than focusing solely on outcomes, thereby implicitly fostering students’ resilience and self-efficacy. Ultimately, high psychological resilience will serve as critical psychological capital for graduate students to navigate uncertainty and undertake high-challenge innovation tasks, enabling a fundamental shift from passively enduring stress to actively harnessing challenges.

Limitations and Future Research

First, the study’s sample representation and generalizability are limited. Although convenience sampling was employed across multiple universities in Zhejiang Province, covering comprehensive, science and engineering, and finance and economics institutions to ensure certain disciplinary diversity, the sample was primarily concentrated in one economically developed eastern province. It lacked adequate representation of central and western regions, as well as diverse types and tiers of higher education institutions. This may expose the findings to potential confounding factors such as regional socioeconomic contexts, institutional policies, and academic climates, thereby constraining the external validity and generalizability of the research model (S. Li et al., 2025). Future studies could adopt more rigorous stratified random sampling or large-scale cross-regional longitudinal surveys to verify the robustness of the conclusions across different geographical, institutional, and disciplinary contexts.

Second, although the use of supervisor ratings to measure graduate students’ innovativeness aligns with mainstream research practices, this method has inherent limitations. Innovative activities are often implicit, process-oriented, and incomplete, making it difficult for supervisors, as external observers, to fully capture students’ internal cognitive breakthroughs or unrealized innovative attempts. Consequently, relying solely on other-report measures may not comprehensively reflect the true level of students’ innovativeness, particularly for non-explicit or exploratory innovative behaviors. Future research could incorporate multi-source assessments—such as self-reports, peer evaluations, and objective innovative outputs—to construct a more comprehensive and reliable innovativeness indicator system, thereby enhancing measurement validity and the robustness of the findings.

Third, although the theoretical model was tested using three-wave time-lagged data to partially capture temporal sequences, the study remains fundamentally based on cross-sectional data, which limits its ability to fully reveal the complex dynamic interactions and long-term evolutionary mechanisms among supervisory styles, stress, and innovativeness. For instance, challenge stress and innovativeness may exhibit an inverted U-shaped relationship, or the protective role of psychological resilience may vary with the duration of stress exposure—dynamic effects that cannot be thoroughly examined with the current methodology. Future research should employ longitudinal designs, experience sampling methods, or in-depth case studies, collecting multi-wave and multi-point data to more accurately delineate causal pathways and dynamic patterns, thereby deepening the understanding of supervisor-student interactions and the processes of innovation cultivation.

Footnotes

Acknowledgements

We would like to thank Editage () for its linguistic assistance during the preparation of this manuscript. We would like to thank the editor and the anonymous referees for their suggestions.

ORCID iDs

Zhenhao Xu

Peifu Chen

Ethical Considerations

Formal ethical approval has been waived instate this study adhered to the principles of the Declaration of Helsinki following strict ethical standards. Formal ethical approval was waived due to participation was anonymous, confidential, and voluntary, with informed consent obtained from all participants. There were no biomarkers or tissue samples collected for analysis. Participants had the freedom to withdraw from the study at any point.

Consent to Participate

Oral consent was obtained from all individuals involved in this study.

Consent for Publication

All authors have read and agreed to the published version of the manuscript.

Author Contributions

Zhenhao Xu: Resources, investigation, project management. Peifu Chen: Data analysis, validation, investigation, methods, theoretical support.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Zhejiang Province Philosophy and Social Science Planning Program (No. 26NDJC016YB). Graduate Education Reform Project of Zhejiang University of Technology (No. 2023310).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets used and/or analyzed during the current study are available from the corresponding author* on reasonable request.

References

Ahsan

M. J.

(2025). Cultivating a culture of learning: The role of leadership in fostering lifelong development. The Learning Organization, 32(2), 282–306. https://doi.org/10.1108/TLO-03-2024-0099

Antwi

C. O.

Fan

C. J.

Aboagye

M. O.

Brobbey

Jababu

Affum-Osei

Avornyo

(2019). Job demand stressors and employees' creativity: A within-person approach to dealing with hindrance and challenge stressors at the airport environment. The Service Industries Journal, 39(3–4), 250–278.

Awais

Jones

(2024). Developing students’ resilience during the crisis: A moderated model linking college support, study demands, student resilience, and students’ change-oriented behaviours. Higher Education Quarterly, 78(3), 565–585. https://doi.org/10.1111/hequ.12468

Bakker

A. B.

Demerouti

(2007). The job demands-resources model: State of the art. Journal of Managerial Psychology, 22(3), 309–328. https://doi.org/10.1108/02683940710733115

Bakker

A. B.

Demerouti

(2024). Job demands–resources theory: Frequently asked questions. Journal of Occupational Health Psychology, 29(3), 188.

Bani-Melhem

Abukhait

R. M.

Mohd. Shamsudin

(2020). Does job stress affect innovative behaviors? Evidence from Dubai five-star hotels. Journal of Human Resources in Hospitality & Tourism, 19(3), 344–367. https://doi.org/10.1080/15332845.2020.1737769

Bao

Mydin

Surat

Lyu

Pan

Cheng

(2024). Challenge-hindrance stressors and academic engagement among medical postgraduates in China: A moderated mediation model. Psychology Research and Behavior Management, 17, 1115–1128. https://doi.org/10.2147/PRBM.S448844

Bao

Yang

(2021). What guidance abilities should graduate mentors possess. Research in Higher Education of Engineering, 1, 108–114.

Basten

Meyer-Ahrens

I. N. G. A.

Fries

Wilde

(2014). The effects of autonomy-supportive vs. controlling guidance on learners’ motivational and cognitive achievement in a structured field trip. Science Education, 98(6), 1033–1053. https://doi.org/10.1002/sce.21125

10.

Byron

Peterson

S. J.

Zhang

LePine

J. A.

(2018). Realizing challenges and guarding against threats: Interactive effects of regulatory focus and stress on performance. Journal of Management, 44(8), 3011–3037. https://doi.org/10.1177/0149206316658349

11.

Cavanaugh

M. A.

Boswell

W. R.

Roehling

M. V.

Boudreau

J. W.

(2000). An empirical examination of self-reported work stress among US managers. Journal of Applied Psychology, 85(1), 65–74. https://doi.org/10.1037/0021-9010.85.1.65

12.

Connor

K. M.

Davidson

J. R.

(2003). Development of a new resilience scale: The Connor-Davidson resilience scale (CD-RISC). Depression and Anxiety, 18(2), 76–82. https://doi.org/10.1002/da.10113

13.

De Clercq

Pereira

(2019). Resilient employees are creative employees, when the workplace forces them to be. Creativity and Innovation Management, 28(3), 329–342. https://doi.org/10.1111/caim.12328

14.

Geng

Liu

Feng

(2014). The effects of emotional labor on frontline employee creativity. International Journal of Contemporary Hospitality Management, 26(7), 1046–1064. https://doi.org/10.1108/IJCHM-12-2012-0244

15.

Liu

(2017). Supervisory styles and graduate student creativity: The mediating roles of creative self-efficacy and intrinsic motivation. Studies in Higher Education, 42(4), 721–742. https://doi.org/10.1080/03075079.2015.1072149

16.

Hamid

N. A. A.

Abd Rahman

Hamidin

N. M.

(2021). Nurturing positive atmosphere in supervisor-supervisee relationship throughout postgraduate studies. International Journal of Advanced Research in Education and Society, 3(1), 21–33.

17.

Han

Liu

Wang

(2022). Graduate students’ perceived supervisor support and innovative behavior in research: The mediation effect of creative self-efficacy. Frontiers in Psychology, 13, Article 875266. https://doi.org/10.3389/fpsyg.2022.875266

18.

Hartmann

Weiss

Newman

Hoegl

(2020). Resilience in the workplace: A multilevel review and synthesis. Applied Psychology, 69(3), 913–959. https://doi.org/10.1111/apps.12191.

19.

Kabat-Farr

Walsh

B. M.

McGonagle

A. K.

(2019). Uncivil supervisors and perceived work ability: The joint moderating roles of job involvement and grit. Journal of Business Ethics, 156(4), 971–985. https://doi.org/10.1007/s10551-017-3604-5

20.

Lazarus

R. S.

Folkman

(1984). Stress, appraisal, and coping. Springer Publishing Company.

21.

Lee

Duffy

M. K.

Scott

K. L.

Schippers

M. C.

(2018). The experience of being envied at work: How being envied shapes employee feelings and motivation. Personnel Psychology, 71(2), 181–200. https://doi.org/10.1111/peps.12251

22.

(2023). High quality development of postgraduate education in industry-characterized universities in the new era: The Connotation of the Times, Realistic Dilemmas, and Practical Approaches. Journal of Graduate Education, 6, 21–28. https://doi.org/10.19834/j.cnki.yjsjy2011.2023.06.03

23.

Huang

Hussain

Dong

(2025). How does supervisor support impact Chinese graduate students’ research creativity through research self-efficacy and intrinsic motivation?—A multi-group analysis. Thinking Skills and Creativity, 55, Article 101700. https://doi.org/10.1016/j.tsc.2024.101700

24.

Liu

Chen

(2021). Tackling the negative impact of COVID-19 on work engagement and taking charge: A multi-study investigation of frontline health workers. Journal of Applied Psychology, 106(2), 185.

25.

Liu

Ren

(2022). Challenge-hindrance stressors and career initiative: A moderated mediation model. Journal of Managerial Psychology, 37(5), 467–479. https://doi.org/10.1108/JMP-05-2020-0227

26.

Mavrogalou-Foti

A. P.

Kambouri

M. A.

Çili

(2024). The supervisory relationship as a predictor of mental health outcomes in doctoral students in the United Kingdom. Frontiers in Psychology, 15, Article 1437819. https://doi.org/10.3389/fpsyg.2024.1437819

27.

(2020). Does authorization have to be omnipotent? The “double-edged sword” effect of empowering leadership on employee’s behavior. Open Journal of Social Sciences, 8(3), 62.

28.

Ohly

(2019). Characteristics of challenging situations: two policy-capturing studies. Journal of Managerial Psychology, 34(3), 170–183. https://doi.org/10.1108/JMP-06-2018-0232

29.

Oldham

G. R.

Cummings

(1996). Employee creativity: Personal and contextual factors at work. Academy of Management Journal, 39(3), 607–634. https://doi.org/10.5465/256657.

30.

Penning de Vries

Knies

(2023). “I expected more from you”: The effect of expectation-disconfirmation on employees’ satisfaction with supervisory support. Review of Public Personnel Administration, 43(2), 309–335. https://doi.org/10.1177/0734371X211065348

31.

Runco

M. A.

Chand

(1995). Cognition and creativity. Educational Psychology Review, 7(3), 243–267. https://doi.org/10.1007/BF02213373

32.

Shin

Taylor

M. S.

Seo

M. G.

(2012). Resources for change: The relationships of organizational inducements and psychological resilience to employees’ attitudes and behaviors toward organizational change. Academy of Management Journal, 55(3), 727–748. https://doi.org/10.5465/amj.2010.0325.

33.

Syed

Naseer

Akhtar

M. W.

Husnain

Kashif

(2021). Frogs in boiling water: A moderated-mediation model of exploitative leadership, fear of negative evaluation and knowledge hiding behaviors. Journal of Knowledge Management, 25(8), 2067–2087. https://doi.org/10.1108/JKM-11-2019-0611

34.

Tariq

Almashayekhi

Ali

Burhan

Butt

H. P.

(2023). All stressors are not bad: An affect-based model of role overload—The supervisor-level antecedent of abusive supervision. International Journal of Conflict Management, 34(3), 545–569. https://doi.org/10.1108/IJCMA-10-2022-0159

35.

Wichmann-Hansen

Schmidt Nielsen

K. J.

(2024). Can hands-on supervision get out of hand? The correlation between directive supervision and doctoral student independence in a Danish study context. Scandinavian Journal of Educational Research, 68(6), 1121–1136. https://doi.org/10.1080/00313831.2023.2204113

36.

T. Y.

Yang

C. C.

(2013). Abusive supervision and workload demands from supervisors: Exploring two types of supervisor-related stressors and their association with strain. Stress and Health, 29(3), 190–198. https://doi.org/10.1002/smi.2440

37.

Jiang

Hong

P. Y.

Roche

(2021). Will mindful employees benefit from positive work reflection triggered by transformational leadership? A two-study examination. International Journal of Stress Management, 28(1), 61. https://doi.org/10.1037/str0000222

38.

Liao

Huang

Xiao

Pan

(2024). The positive side of stress: Investigating the impact of challenge stressors on innovative behavior in higher education. Acta Psychologica, 246, Article 104255.

39.

Zhao

Liu

Xie

(2025). The influence of college students’ psychological resilience on problem solving: Based on eye tracking technology. Frontiers in Psychology, 16, Article 1616452. https://doi.org/10.3389/fpsyg.2025.1616452

40.

Yang

Bao

(2022). Supervisory styles and graduate student innovation performance: The mediating role of psychological capital and the moderating role of harmonious academic passion. Frontiers in Psychology, 13, Article 1034216. https://doi.org/10.3389/fpsyg.2022.1034216

41.

Yang

Ding

(2021). Linking leader humor to employee creativity: The roles of relational energy and traditionality. Journal of Managerial Psychology, 36(7), 548–561. https://doi.org/10.1108/JMP-06-2020-0300

42.

Yeager

D. S.

Dweck

C. S.

(2012). Mindsets that promote resilience: When students believe that personal characteristics can be developed. Educational Psychologist, 47(4), 302–314. https://doi.org/10.1080/00461520.2012.722805

43.

Zhang

Bartol

K. M.

(2010). Linking empowering leadership and employee creativity: The influence of psychological empowerment, intrinsic motivation, and creative process engagement. Academy of Management Journal, 53(1), 107–128. https://doi.org/10.5465/amj.2010.48037118

44.

Zhang

Shen

Shi

Liu

(2024). Supportive supervision and doctoral student creativity: The double-edged sword of family support. Studies in Higher Education, 49(7), 1253–1268. https://doi.org/10.1080/03075079.2023.2263035

45.

Zhou

George

J. M.

(2001). When job dissatisfaction leads to creativity: Encouraging the expression of voice. Academy of Management Journal, 44(4), 682–696. https://doi.org/10.5465/3069410

46.

Zhou

Shalley

C. E.

(2003). Research on employee creativity: A critical review and directions for future research. In Martocchio

J. J.

Ferris

G. R.

(Eds.), Research in personnel and human resources management (Vol. 22, pp. 165–217). Elsevier Science Ltd. https://doi.org/10.1016/S0742-7301(03)22004-1

47.

Zhou

(2023). How supportive and controlling supervision influence employees’ innovative behavior. Social Behavior and Personality: An International Journal, 51(10), 1–18. https://doi.org/10.2224/sbp.12253