Paradoxical leadership and employee voice: Superior–subordinate relationships and psychological safety as chain mediators
Main Article Content
Cite this article:
Wang, X., Wu, H., Gao, R., & Sun, Y.
(2026). Paradoxical leadership and employee voice: Superior–subordinate relationships and psychological safety as chain mediators.
Social Behavior and Personality: An international journal,
54(4),
e15115.
Abstract
Full Text
References
Tables and Figures
Acknowledgments
Author Contact
Employee voice behavior is a pivotal determinant in fostering organizational innovation and driving improvement. Leaders, who assume the role of custodians of organizational resources and are the key recipients of employee feedback, exert a substantial influence on this behavior. We constructed a chain mediation model from the dual perspectives of superior–subordinate relationships and psychological safety to explain the influence of paradoxical leadership on employee voice behavior, using data from 597 Chinese employees. The results demonstrated that paradoxical leadership had a significant positive effect on employee voice, and superior–subordinate relationships and psychological safety had both independent and chain mediating effects on this relationship. These findings provide empirical evidence for understanding how paradoxical leadership influences employee voice and offer practical insights for breaking employee silence and motivating voice behavior.
Article Highlights
The five dimensions of paradoxical leadership were found to have significant positive effects on employee voice behavior.
Superior–subordinate relationships and psychological safety acted as independent mediators of the relationship between paradoxical leadership and employee voice.
Superior–subordinate relationships and psychological safety had a chain mediating effect on the relationship between paradoxical leadership and employee voice.
Employee voice behavior refers to the voluntary and informal expression of work-related ideas, concerns, or suggestions to superiors and other organizational members, aimed at fostering organizational development (Lee et al., 2023; Morrison, 2014). This behavior exerts significant effects across various organizational levels, including the enhancement of personal vitality and well-being (Röllmann et al., 2021), improvement of innovation performance (Song et al., 2017), mitigation of negative effects associated with abusive leadership (Sun et al., 2022), reinforcement of ethical upgrading among colleagues (Chen & Treviño, 2022), and facilitation of management innovation (Guzman & Espejo, 2019). Therefore, voice behavior has received extensive attention in both academic and practical contexts (Y. Li & Sun, 2015).
The existing literature has delineated various antecedents and mechanisms underpinning employee voice behavior (Morrison, 2023). Leaders, as stewards of organizational resources and significant recipients of employee feedback, significantly influence this behavior (Morrison, 2023). Research has indicated that ethical leadership (Bai et al., 2019) and transformational leadership (Duan et al., 2017) positively affect voice behavior, whereas authoritarian leadership exerts a negative influence (Zheng et al., 2021). Among the various leadership styles, paradoxical leadership has garnered increasing attention due to its dialectical perspective on organizational competitiveness and its effective approaches to addressing organizational paradoxes (Y. Zhang et al., 2015). Rather than adopting an either/or way of thinking, paradoxical leadership adopts a both/and approach to comprehensively meet the competitive demands of organizations, employees, and other stakeholders (Volk et al., 2023; Y. Zhang et al., 2015). Studies have demonstrated that paradoxical leadership has positive impacts on employee performance (Kundi et al., 2023), organizational citizenship behavior (Pan, 2021), and team innovation (Wei et al., 2023; I. D. Zhang et al., 2025).
Scholars have examined the relationship between paradoxical leadership and employee voice behavior primarily by focusing on psychological processes or social relationships (see, e.g., X. Li et al., 2020; Xue et al., 2020). Xue et al. (2020) conducted a comprehensive study and showed that psychological safety mediated the relationship between paradoxical leadership and promotive voice behavior. X. Li et al. (2020) found that employees’ self-efficacy and psychological safety mediated the extent to which paradoxical leadership influenced subordinates’ voice behavior. However, these studies focused on a single influence mechanism, disregarding combined effects, which warrant further exploration. In addition, voice behavior is sensitive to the voice recipient; employees tend to exhibit different behaviors based on the characteristics (e.g., hierarchy and personality) of specific voice targets (W. Liu et al., 2010). The existing literature has not sufficiently addressed the differences among different voice recipients (Duan et al., 2017; Goel et al., 2024; Morrison, 2011). It remains empirically untested whether paradoxical leadership affects voice behavior toward different objects through a uniform mechanism.
Social exchange theory posits that reciprocity is a fundamental principle guiding social interactions (Cook et al., 2013). Paradoxical leadership enhances employees’ organizational identity by balancing the interests of both the organization and its employees, which bolsters job satisfaction and loyalty while encouraging proactive contributions, such as voice behavior (Kundi et al., 2023). In response, employees may offer constructive actions to express their concern for the interests and development of the organization, according to the principle of reciprocity.
Social learning theory holds that individuals amass knowledge and internalize behavioral patterns through observing and imitating influential figures (Bandura & Walters, 1977). By assimilating leaders’ paradoxical thinking, employees learn to navigate various conflicts and contribute positively to the organization, including generating improvement suggestions (She et al., 2020). Furthermore, paradoxical leadership fosters a fair, open, and respectful work environment (Y. Zhang et al., 2015), which may further incentivize employee voice. Studies have found that paradoxical leadership has a positive impact on employee voice behavior (Shehata et al., 2023). We sought to empirically explore if this impact differs according to the five dimensions of paradoxical leadership: treating subordinates uniformly while allowing individualization, combining self‐centeredness with other‐centeredness, maintaining decision control while allowing autonomy, enforcing work requirements while allowing flexibility, and maintaining both distance and closeness.
The first dimension of paradoxical leadership refers to leaders who treat employees equally while addressing individual needs; this can enhance perceived organizational fairness and support, which are critical factors in fostering employee voice (H. Kim & Kiura, 2023; Morrison, 2011; Qi et al., 2023). By meeting their differentiated needs, paradoxical leadership reflects leaders’ support and trust in employees, thereby stimulating voice behavior in accordance with the principle of social reciprocity (Gao et al., 2011).
The second dimension, engaging in shared leadership with employees while simultaneously sustaining the leader’s own influence (combining self-centeredness with other-centeredness), may augment the comprehensiveness and effectiveness of decision-making processes, and foster employees’ trust in their leaders. When leaders take employees’ needs into account, and actively encourage diversity and self-expression, this may enhance employees’ self-efficacy. This, in turn, can help to meet their needs for self-growth and self-realization (J. Liu et al., 2010), reinforce their sense of obligation to reciprocate with certain behaviors (Golden & Veiga, 2008), and ultimately stimulate them to exhibit more voice behavior.
The third dimension involves leaders maintaining appropriate decision-making control while bestowing a certain level of independence upon employees by creating a work environment characterized by the coexistence of boundaries and autonomy. This form of empowerment may cultivate a climate of trust and support among employees, eliciting a sense of reciprocal obligation that enhances the proclivity to engage in extrarole behaviors, such as voicing their opinions (Gao et al., 2011).
The fourth dimension involves leaders simultaneously furnishing employees with autonomy while upholding rigorous expectations concerning employee performance and advancement. This strategy can not only enhance employees’ efficacy and work-related capability, but also mitigates negative emotions and the deleterious impact of mental health conditions such as stress and anxiety. Such adaptability expedites the cultivation of employees’ organizational identity and their commitment to the organization, thereby encouraging them to proactively manifest their opinions (W. Liu et al., 2010).
The fifth dimension entails leaders maintaining authority while nurturing amicable connections with employees through harmonious integration of power distance and interpersonal warmth. This balance not only safeguards their leadership authority but also engenders a sense of camaraderie within the employee cohort. The authority of leaders may stimulate social learning effects, and the warmth of interpersonal relationships may encourage employees to perceive themselves as respected and, thus, engage in reciprocal behavior. Ultimately, this dynamic stimulates voice behavior among employees (S. Kim et al., 2022). Therefore, we proposed the following hypothesis:
Hypothesis 1: The five dimensions of paradoxical leadership will be positively related to employees’ voice behavior toward superiors and peers.
Leadership style and behavior significantly influence superior–subordinate relationships, which, in turn, affect employee interests, psychological feelings, and organizational behavior (Young et al., 2021). Paradoxical leaders, characterized by their balanced values and behaviors, are particularly adept at fostering psychological resonance and identifying with their employees, thereby promoting a sense of belonging and enhancing interpersonal relationships (X. Li et al., 2020). Through demonstration of respect, support, and trust, these leaders can strengthen social interactions with their subordinates, ultimately developing a positive superior–subordinate relationship (Miao et al., 2020; Y. Zhang et al., 2015).
Superior–subordinate relationships play a crucial role in shaping employee behavior choices (Epitropaki et al., 2020; Miao et al., 2020). A close superior–subordinate relationship facilitates employees’ access to resources and fosters trust, leading to more positive attitudes and behaviors, such as organizational citizenship behavior, organizational commitment, and a sense of fairness (Miao et al., 2020). Hence, this relationship can stimulate employees’ voice behavior (Chong et al., 2015; Morrison, 2023). Moreover, a strong superior–subordinate relationship deepens emotional connection through frequent interaction, which can foster mutual tolerance and encouragement (Golden & Veiga, 2008). This supportive environment may encourage employees to actively express their thoughts and opinions. In addition, the superior–subordinate relationship enhances employees’ job satisfaction and organizational commitment, thereby improving their work engagement and voice behavior (Bao et al., 2021; Chong et al., 2015; Pei et al., 2018). Given these dynamics, superior–subordinate relationships may serve as an important mediator between paradoxical leadership and employee voice behavior (Xue et al., 2020). Therefore, we proposed the following hypothesis:
Hypothesis 2: The superior–subordinate relationship will mediate the relationships between the five dimensions of paradoxical leadership and employees’ voice behavior toward superiors and peers.
A meta-analysis demonstrated that transformational leadership and trust in leaders significantly impact employees’ sense of psychological safety (Frazier et al., 2017). Paradoxical leaders contribute to the development of an inclusive and safe organizational climate by promoting shared values, encouraging open expression, building trust, and providing constructive feedback through the social learning process (Y. Zhang et al., 2015). This approach enables subordinates to clarify behavioral norms and enhances their sense of psychological safety (Yang et al., 2024).
Psychological safety refers to an individual’s belief regarding the safety of engaging in risky behaviors within the workplace (Lee et al., 2023; Newman et al., 2017), and it serves as a significant antecedent for individuals to undertake challenging and potentially risky actions (Frazier et al., 2017). When individuals experience psychological safety, they are more likely to feel accepted, adopt a positive outlook, take risks, and engage in constructive conflict (Newman et al., 2017), including voice behavior (Sherf et al., 2021).
Employees are more likely to offer suggestions when they feel secure in expressing their opinions within the organization, and free from concerns about potential career or personal image repercussions (Morrison, 2011). This sense of security also stimulates employees’ prosocial motivation, leading them to advocate for organization benefits, and facilitating effective organizational communication and problem solving (Newman et al., 2017). Thus, paradoxical leadership may reinforce employees’ psychological safety through the social learning process, which, in turn, may stimulate employees’ voice behavior, thus creating a positive feedback loop that contributes to organizational development and innovation. Therefore, we proposed the following hypothesis:
Hypothesis 3: Psychological safety will mediate the relationships between the five dimensions of paradoxical leadership and employees’ voice behavior toward superiors and peers.
When employees consider engaging in voice behavior, they first evaluate the consequences and safety of their actions (Liang et al., 2012; Morrison, 2023). A good superior–subordinate relationship enables employees to gain more psychological and cognitive resources and organizational support (Frazier et al., 2017), and it may lead to a stronger psychological sense of security, which prompts proactive voice behavior. Paradoxical leaders aim to reduce internal conflicts, making employees feel united and respected, thereby reducing employees’ concerns about voice (X. Li et al., 2020). In other words, when employees have a strong sense of psychological security, they may perform more voice behavior. Psychological security is a key condition for employees’ voice behavior (Morrison, 2023), which directly affects employees’ voice willingness and voice behavior, and plays a mediating role in the link between the superior–subordinate relationship and voice behavior (Liang et al., 2012; Pei et al., 2018). Therefore, we proposed the following hypothesis:
Hypothesis 4: The five dimensions of paradoxical leadership will be positively related to employees’ voice behavior toward superiors and peers through the chain mediators of superior–subordinate relationships and psychological safety.
The theoretical model of the study is depicted in Figure 1.
Figure 1. Theoretical Model
Method
Participants and Procedure
This research was approved by the College Ethics Committee of Chengdu University of Technology. We explained the study purpose and confidentiality commitment to the participants at the beginning of the questionnaire, which was presented on the Chinese survey platform, Credamo (https://www.credamo.world). We recruited 701 participants; after excluding invalid surveys with short response times or the same response given to all items, the final sample consisted of 597 respondents. Among them, 334 (55.95%) were men and 263 (44.05%) were women; 96 were aged under 25 years (16.08%), 246 were aged 25–35 years (41.21%); 198 were aged 35–45 years (33.16%), and 57 were aged over 45 years (9.55%); 126 (21.11%) had a master’s degree or higher, 343 (57.45%) had a bachelor’s degree, and 128 (21.44%) had a high school education or lower. The mean number of working years was 2.32 (SD = 1.03).
Measures
We measured the core concepts with scales drawn from the literature, which were translated through a standard back-translation procedure involving scholars who are fluent in both Chinese and English and have extensive experience in the field of organizational behavior. We tested all scales in a pilot study (N = 109) to assess their reliability. All items were rated on a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree).
Paradoxical Leadership
We assessed paradoxical leadership with a scale from Y. Zhang et al. (2015), in which the employees evaluate the leadership style of their direct supervisors. The scale contains 22 items spread over five dimensions: treating subordinates uniformly while allowing individualization (five items, e.g., “My supervisor manages us uniformly, but considers our individualized needs”), combining self-centeredness with other-centeredness (five items, e.g., “My supervisor insists on getting respect, but also shows respect toward us”), maintaining decision control while allowing autonomy (five items, e.g., “My supervisor maintains overall control, but gives us appropriate autonomy”), enforcing work requirements while allowing flexibility (four items, e.g., “My supervisor clarifies work requirements, but does not micromanage work”), and maintaining both distance and closeness (four items, e.g., “My supervisor keeps distance from us, but does not remain aloof”). Cronbach’s alpha values for each dimension were .89, .85, .84, .81, and .86, respectively.
Employee Voice Behavior
We measured employee voice behavior with the Speaking Out and Speaking Up Scale from W. Liu et al. (2010). Employees rate their own voice behavior using 15 items spread over two dimensions: speaking up toward the supervisor (nine items, e.g., “I will speak to my supervisor with new ideas for projects or changes in procedures”) and six items for speaking out toward peers (six items, e.g., “I will provide constructive suggestions to my colleagues to improve their work”). Cronbach’s alpha values were .91 and .90, indicating the scale is reliable.
Superior–Subordinate Relationships
We measured superior–subordinate relationships with a Chinese version (Guo & Li, 2015) of the Affective Attachment Scale (Chen et al., 2009) to ensure alignment with the cultural context of our study. Specifically, we retained the core construct of the original scale while revising individual items to better capture the nuanced expression of reciprocal support in Chinese workplace interactions; for instance, adapting the original item “If my supervisor has problems with their personal life, I will do my best to help them out” to read “My supervisor and I will try our best to deal with each other’s difficulties.” This revision emphasizes the mutual nature of support that is salient in Chinese relational norms. Employees evaluated their relationship with their immediate supervisor using four items. Cronbach’s alpha coefficient was .92, indicating the scale is reliable.
Psychological Safety
We used the Psychological Safety Scale by Liang et al. (2012) to test the sense of psychological safety perceived by employees. The scale has five items, such as “In the workplace, I can express my thoughts freely.” Cronbach’s alpha was .92, indicating the scale is reliable.
Control Variables
With reference to the literature on employee voice behavior, we took employees’ age, gender, education level, and working years as control variables.
Data Analysis
We first conducted a confirmatory factor analysis using Mplus 7.4 to validate the data measurement model, then used Harman’s single-factor test to check for common method bias. We conducted path analyses to examine the hypothesized direct effects and performed path analyses to test the independent and chain mediation effects.
Results
Model Fit Analysis
We constructed six competing models by combining different factors and comparing their fit indices. The results indicated that the proposed nine-factor model fitted the data best, χ2/df = 2.635, comparative fit index = .917, Tucker–Lewis index = .981, standardized root-mean-square residual = .042, root-mean-square error of approximation = .050, and was significantly better than other models. Harman’s single-factor test revealed there were eight factors with eigenvalues greater than 1, and the amount of variance explained by the first factor was 40%. Further, the single-factor model showed the poorest fit indices. These results indicate that there was no serious common method bias in the data. The highest observed variance inflation factor in the benchmark model was 3.09, indicating that there was no serious multicollinearity.
Descriptive Statistics and Correlation Analysis
The means, standard deviations, and correlation coefficients of the variables are shown in Table 1. The mean value of speaking-up voice behavior was relatively smaller than the mean of speaking-out voice behavior, which indicates that the voice behaviors depend on the speaker’s objectives, and it is reasonable to distinguish the two kinds of voice. The five dimensions of paradoxical leadership were significantly and positively correlated with superior–subordinate relationships, psychological safety, speaking-up voice behavior, and speaking-out voice behavior, thus supporting Hypothesis 1. Superior–subordinate relationships were significantly and positively correlated with psychological safety, speaking-up voice behavior, and speaking-out voice behavior. Psychological safety was significantly and positively correlated with speaking-up voice behavior and speaking-out voice behavior. These results offered preliminary support for Hypotheses 2, 3, and 4.
Table 1. Descriptive Statistics and Correlation Analysis
Note. N = 597. UI = treating subordinates uniformly while allowing individualization; SO = combining self-centeredness with other-centeredness; CA: = maintaining decision control while allowing autonomy; RF = enforcing work requirements while allowing flexibility; DC = maintaining both distance and closeness; SSR = superior–subordinate relationships.
* p < .05. ** p < .01.
Hypothesis Testing
We used path analysis (1,000 bootstrapped resamples) via Mplus 7.4 to test the hypothesized relationships. The results revealed four key findings (see Table 2). First, all five paradoxical leadership dimensions significantly predicted both speaking-up voice behavior and speaking-out voice behavior, thus supporting Hypothesis 1. Second, superior–subordinate relationships mediated the effect of all paradoxical leadership dimensions on both speaking-up voice behavior and speaking-out voice behavior, thus supporting Hypothesis 2. Third, psychological safety similarly served as a significant mediator of both speaking-up voice behavior and speaking-out voice behavior, thus supporting Hypothesis 3. Last, sequential mediation through superior–subordinate relationships followed by psychological safety emerged for both speaking-up voice behavior and speaking-out voice behavior, thereby supporting Hypothesis 4.
Table 2. Hypothesis Testing Results
Note. N = 597. SPU = speaking-up voice behavior; SPO = speaking-out voice behavior; UI = treating subordinates uniformly while allowing individualization; SO = combining self-centeredness with other-centeredness; CA = maintaining decision control while allowing autonomy; RF = enforcing work requirements while allowing flexibility; DC = maintaining both distance and closeness; SSR = superior–subordinate relationships; PS = psychological safety.
** p < .01. *** p < .001.
All five paradoxical leadership dimensions significantly predicted both speaking-up voice behavior and speaking-out voice behavior, supporting the proposal that paradoxical leadership plays a critical role in stimulating multidirectional voice behaviors. Notably, the associations were consistently stronger for speaking-up voice behavior than for speaking-out voice behavior, with a medium-sized difference observed according to Cohen’s (2013) guidelines. This divergence may reflect China’s high power-distance culture (Hofstede, 1980), where vertical communication with authorities may bring greater psychological safety than lateral peer interactions. Regarding mediation mechanisms, superior–subordinate relationships demonstrated significantly larger indirect effects than psychological safety for both speaking-up voice behavior and speaking-out voice behavior. This pattern aligns with social exchange theory (Blau, 1986), suggesting that high-quality superior–subordinate relationships create stronger reciprocal obligations for voice contribution compared to individual psychological states. The chain mediation effect we observed through superior–subordinate relationships followed by psychological safety further indicates that relational foundations preceded psychological safety in the voice–motivation sequence.
Discussion
Paradoxical leadership, operationalized through Y. Zhang et al.’s (2015) five-dimensional model, has emerged as a critical lens for understanding how leaders navigate organizational tensions. Grounded in social exchange theory (Blau, 1986) and social learning theory (Bandura & Walters, 1977), this study has advanced prior work by simultaneously investigating two distinct voice targets (speaking-up voice behavior and speaking-out voice behavior) and introducing a dual mediation model that integrates social relationships (superior–subordinate relationships) and psychological processes (psychological safety). The findings revealed three key advancements relating to the dimensional specificity of paradoxical leadership’s effects, the interplay of relational and psychological mechanisms, and the differential impact on vertical versus lateral voice.
Theoretical Implications
Our results extend the literature in three ways. First, while prior studies predominantly treated paradoxical leadership as a unitary construct (e.g., Kundi et al., 2023; X. Li et al., 2020), we demonstrated that all five dimensions independently predicted voice behavior. For instance, maintaining decision control while allowing autonomy (control–autonomy balance) showed weaker effects when modeled holistically, suggesting dimension-specific thresholds for empowering employees to voice. Second, we reconciled conflicting findings on mediation mechanisms: Unlike Xue et al. (2020), who focused solely on psychological safety, our dual mediation model revealed that superior–subordinate relationships and psychological safety sequentially transmitted the effect of paradoxical leadership. This aligns with social exchange theory’s emphasis on reciprocity (with superior–subordinate relationships as relational currency) preceding risk appraisal from social learning (with psychological safety as cognitive buffer). Third, we empirically validated the influence of voice-target sensitivity, a theoretical proposition proposed by W. Liu et al. (2010) that previously lacked systematic testing. The stronger association between paradoxical leadership and speaking-up voice behavior (vs. speaking-out voice behavior) may reflect China’s high power-distance cultural setting (Hofstede, 1980), where employees perceive vertical voice as culturally normative yet psychologically riskier without leader endorsement.
By deconstructing paradoxical leadership into its five dimensions, we have responded to Y. Zhang et al.’s (2015) call for granular investigations. The differential predictive power of these dimensions suggests that leaders should tailor specific paradoxical behaviors to channel voice toward intended recipients. Second, the chain mediation model bridges social exchange and social learning theories. Superior–subordinate relationships foster obligation-driven voice, while psychological safety enables risk-tolerant voice. This sequential mechanism challenges the prevailing assumption that relational and psychological pathways operate in parallel (X. Li et al., 2020). Third, by validating voice-target sensitivity in a high power-distance cultural context, we have extended Morrison’s (2014) voice framework to non-Western settings, highlighting how cultural norms shape the translation of leadership into voice channels.
Practical Inspirations
Organizations can implement three evidence-based strategies drawn from our findings. First, implement paradoxical leadership training through scenario-based workshops to balance opposing behaviors. Second, conduct audits of the link between superior–subordinate relationships and psychological safety via pulse surveys, identifying gaps through anonymized team feedback. Third, leaders should strive to create distinct platforms for upward and peer voice types, leveraging distance–closeness norms, thereby encouraging active participation in communication with superiors and peers.
Limitations and Future Research
Three limitations to this study merit attention. First, the cross-sectional data preclude drawing causal claims. In future work, longitudinal designs could be used to test whether superior–subordinate relationships and psychological safety reciprocally reinforce paradoxical leaderships over time. Second, monosource data increase the risk common method bias; thus, future studies could collect multiwave, multisource data. Third, our focus on a Chinese sample may limit generalizability, so replication with samples from other cultures (e.g., those with low power-distance settings) could clarify whether speaking-up voice behavior and speaking-out voice behavior differences persist.
Bai, Y., Lin, L., & Liu, J. T. (2019). Leveraging the employee voice: A multi-level social learning perspective of ethical leadership. The International Journal of Human Resource Management, 30(12), 1869–1901. https://doi.org/10.1080/09585192.2017.1308414
Bandura, A., & Walters, R. H. (1977). Social learning theory. Prentice Hall.
Bao, Y., Han, P., Liao, S., Liao, J., & Deng, C. J. (2021). To speak up or stay silent? How employee–supervisor value differences affect speaking up behavior in the hospitality industry. International Journal of Hospitality Management, 96, Article 102946. https://doi.org/10.1016/j.ijhm.2021.102946
Blau, P. (1986). Exchange and power in social life (2nd ed.). Routledge. https://doi.org/10.4324/9780203792643
Chen, A., & Treviño, L. K. (2022). Promotive and prohibitive ethical voice: Coworker emotions and support for the voice. Journal of Applied Psychology, 107(11), 1973–1994. https://doi.org/10.1037/apl0001003
Chen, Y., Friedman, R., Yu, E., Fang, W., & Lu, X. (2009). Supervisor–subordinate guanxi: Developing a three-dimensional model and scale. Management and Organization Review, 5(3), 375–399. https://doi.org/10.1111/j.1740-8784.2009.00153.x
Chong, M. P. M., Peng, T., Fu, P. P., Richards, M., Muethel, M., Caldas, M. P., & Shang, Y. F. (2015). Relational perspectives on leaders’ influence behavior: The mediation of Western leader–member exchange and Chinese guanxi. Journal of Cross-Cultural Psychology, 46(1), 71–87. https://doi.org/10.1177/0022022114554035
Cohen, J. (2013). Statistical power analysis for the behavioral sciences (2nd ed.). Routledge. https://doi.org/10.4324/9780203771587
Cook, K. S., Cheshire, C., Rice, E. R. W., & Nakagawa, S. (2013). Social exchange theory. In J. DeLamater & A. Ward (Eds.), Handbook of social psychology (pp. 61–88). Springer. https://doi.org/10.1007/978-94-007-6772-0_3
Duan, J., Li, C., Xu, Y., & Wu, C. (2017). Transformational leadership and employee voice behavior: A Pygmalion mechanism. Journal of Organizational Behavior, 38(5), 650–670. https://doi.org/10.1002/job.2157
Epitropaki, O., Radulovic, A. B., Ete, Z., Thomas, G., & Martin, R. (2020). Leader–follower transgressions, relationship repair strategies and outcomes: A state-of-the-science review and a way forward. The Leadership Quarterly, 31(1), Article 101376. https://doi.org/10.1016/j.leaqua.2019.101376
Frazier, M. L., Fainshmidt, S., Klinger, R. L., Pezeshkan, A., & Vracheva, V. (2017). Psychological safety: A meta-analytic review and extension. Personnel Psychology, 70(1), 113–165. https://doi.org/10.1111/peps.12183
Gao, L., Janssen, O., & Shi, K. (2011). Leader trust and employee voice: The moderating role of empowering leader behaviors. The Leadership Quarterly, 22(4), 787–798. https://doi.org/10.1016/j.leaqua.2011.05.015
Goel, A., Khan, N., & Dyaram, L. (2024). The yin and yang of employee voice: An exploratory study. Personnel Review, 53(2), 473–487. https://doi.org/10.1108/PR-06-2022-0453
Golden, T. D., & Veiga, J. F. (2008). The impact of superior–subordinate relationships on the commitment, job satisfaction, and performance of virtual workers. The Leadership Quarterly, 19(1), 77–88. https://doi.org/10.1016/j.leaqua.2007.12.009
Guo, X. W., & Li, C. Y. (2015). Supervisor–subordinate guanxi in China: Construction and preliminary test of an integrated construct [In Chinese]. Chinese Journal of Management, 12(2), 167–177. https://doi.org/10.3969/j.issn.1672-884x.2015.02.002
Guzman, F. A., & Espejo, A. (2019). Introducing changes at work: How voice behavior relates to management innovation. Journal of Organizational Behavior, 40(1), 73–90. https://doi.org/10.1002/job.2319
Hofstede, G. (1980). Culture and organizations. International Studies of Management & Organization, 10(4), 15–41. https://doi.org/10.1080/00208825.1980.11656300
Kim, H., & Kiura, M. (2023). The influences of social status and organizational justice on employee voice: A case of customer care workers. International Journal of Business Communication, 60(3), 802–822. https://doi.org/10.1177/2329488420969776
Kim, S., Noh, G., & Miao, S. (2022). Authentic leadership and employee voice behavior: The effect of Chinese employees’ zhongyong thinking. International Journal of Organization Theory & Behavior, 25(3/4), 167–185. https://doi.org/10.1108/IJOTB-01-2022-0016
Kundi, Y. M., Aboramadan, M., & Abualigah, A. (2023). Linking paradoxical leadership and individual in-role and extra-role performance: A multilevel examination. Management Decision, 61(10), 2851–2871. https://doi.org/10.1108/MD-10-2022-1353
Lee, J., Kim, S. L., & Yun, S. (2023). Encouraging employee voice: Coworker knowledge sharing, psychological safety, and promotion focus. The International Journal of Human Resource Management, 34(5), 1044–1069. https://doi.org/10.1080/09585192.2021.2018014
Li, X., Xue, Y., Liang, H., & Yan, D. (2020). The impact of paradoxical leadership on employee voice behavior: A moderated mediation model. Frontiers in Psychology, 11, Article 2408. https://doi.org/10.3389/fpsyg.2020.537756
Li, Y., & Sun, J.-M. (2015). Traditional Chinese leadership and employee voice behavior: A cross-level examination. The Leadership Quarterly, 26(2), 172–189. https://doi.org/10.1016/j.leaqua. 2014.08.001
Liang, J., Farh, C. I. C., & Farh, J.-L. (2012). Psychological antecedents of promotive and prohibitive voice: A two-wave examination. Academy of Management Journal, 55(1), 71–92. https://doi.org/10.5465/amj.2010.0176
Liu, J., Siu, O.-L., & Shi, K. (2010). Transformational leadership and employee well-being: The mediating role of trust in the leader and self-efficacy. Applied Psychology, 59(3), 454–479. https://doi.org/10.1111/j.1464-0597.2009.00407.x
Liu, W., Zhu, R., & Yang, Y. (2010). I warn you because I like you: Voice behavior, employee identifications, and transformational leadership. The Leadership Quarterly, 21(1), 189–202. https://doi.org/10.1016/j.leaqua.2009.10.014
Miao, C., Qian, S., Banks, G. C., & Seers, A. (2020). Supervisor-subordinate guanxi: A meta-analytic review and future research agenda. Human Resource Management Review, 30(2), Article 100702. https://doi.org/10.1016/j.hrmr.2019.100702
Morrison, E. W. (2011). Employee voice behavior: Integration and directions for future research. Academy of Management Annals, 5(1), 373–412. https://doi.org/10.5465/19416520.2011.574506
Morrison, E. W. (2014). Employee voice and silence. Annual Review of Organizational Psychology and Organizational Behavior, 1(1), 173–197. https://doi.org/10.1146/annurev-orgpsych-031413-091328
Morrison, E. W. (2023). Employee voice and silence: Taking stock a decade later. Annual Review of Organizational Psychology and Organizational Behavior, 10(1), 79–107. https://doi.org/10.1146/annurev-orgpsych-120920-054654
Newman, A., Donohue, R., & Eva, N. (2017). Psychological safety: A systematic review of the literature. Human Resource Management Review, 27(3), 521–535. https://doi.org/10.1016/j.hrmr.2017.01.001
Pan, Z. (2021). Paradoxical leadership and organizational citizenship behaviour: The serial mediating effect of a paradoxical mindset and personal service orientation. Leadership & Organization Development Journal, 42(6), 869–881. https://doi.org/10.1108/LODJ-08-2020-0351
Pei, Z., Pan, Y., Skitmore, M., & Feng, T. (2018). Leader–member exchange and prohibitive voice in nonprofit organizations in China: The moderating roles of superior–subordinate tenure matching. Nonprofit Management and Leadership, 29(2), 203–222. https://doi.org/10.1002/nml.21330
Qi, L., Xu, Y., & Liu, B. (2023). Does justice matter in voice? Inclusive leadership and employee voice: The moderating role of organizational justice perception. Frontiers in Psychology, 14, Article 1313922. https://doi.org/10.3389/fpsyg.2023.1313922
Röllmann, L. F., Weiss, M., & Zacher, H. (2021). Does voice benefit or harm occupational well-being? The role of job insecurity. British Journal of Management, 32(3), 708–724. https://doi.org/10.1111/1467-8551.12471
She, Z., Li, Q., Yang, B., & Yang, B. (2020). Paradoxical leadership and hospitality employees’ service performance: The role of leader identification and need for cognitive closure. International Journal of Hospitality Management, 89, Article 102524. https://doi.org/10.1016/j.ijhm.2020.102524
Shehata, A. E., Khan, M. A., Khalid, R., Raza, M., & Selem, K. M. (2023). Consequences of paradoxical leadership in the hotel setting: Moderating role of work environment. Journal of Hospitality Marketing & Management, 32(5), 670–693. https://doi.org/10.1080/19368623.2023.2190318
Sherf, E. N., Parke, M. R., & Isaakyan, S. (2021). Distinguishing voice and silence at work: Unique relationships with perceived impact, psychological safety, and burnout. Academy of Management Journal, 64(1), 114–148. https://doi.org/10.5465/amj.2018.1428
Song, J., Wu, J., & Gu, J. (2017). Voice behavior and creative performance moderated by stressors. Journal of Managerial Psychology, 32(2), 177–192. https://doi.org/10.1108/JMP-03-2016-0078
Sun, S., Burke, M., Chen, H., Tan, Y., Zhang, J., & Hou, L. (2022). Mitigating the psychologically detrimental effects of supervisor undermining: Joint effects of voice and political skill. Human Relations, 75(1), 87–112. https://doi.org/10.1177/0018726721992849
Volk, S., Waldman, D. A., & Barnes, C. M. (2023). A circadian theory of paradoxical leadership. Academy of Management Review, 48(4), 611–638. https://doi.org/10.5465/amr.2020.0468
Wei, W., Zhou, Y., & Wang, D. (2023). Learning to integrate conflicts: Paradoxical leadership fosters team innovation. Journal of Business Research, 165, Article 114076. https://doi.org/10.1016/j.jbusres.2023.114076
Xue, Y., Li, X., Liang, H., & Li, Y. (2020). How does paradoxical leadership affect employees’ Voice behaviors in workplace? A leader–member exchange perspective. International Journal of Environmental Research and Public Health, 17(4), Article 1162. https://doi.org/10.3390/ijerph17041162
Yang, N., Chen, H., & Wang, X.-H. (2024). Stealth innovation: The dance of paradoxical leadership behavior, leader trustworthiness, and psychological safety in fueling employee bootlegging behavior. European Management Journal, 243(2), 286–296. https://doi.org/10.1016/j.emj.2024.03.007
Young, H. R., Glerum, D. R., Joseph, D. L., & McCord, M. A. (2021). A meta-analysis of transactional leadership and follower performance: Double-edged effects of LMX and empowerment. Journal of Management, 47(5), 1255–1280. https://doi.org/10.1177/0149206320908646
Zhang, I. D., Lam, L. W., Zhu, J. N. Y., & Lee, J. (2025). Why do employees perform better under paradoxical leaders? The mediating role of group harmony. Journal of Business and Psychology, 40(1), 113–127. https://doi.org/10.1007/s10869-024-09942-x
Zhang, Y., Waldman, D. A., Han, Y.-L., & Li, X.-B. (2015). Paradoxical leader behaviors in people management: Antecedents and consequences. Academy of Management Journal, 58(2), 538–566. https://doi.org/10.5465/amj.2012.0995
Zheng, Y., Graham, L., Farh, J.-L., & Huang, X. (2021). The impact of authoritarian leadership on ethical voice: A moderated mediation model of felt uncertainty and leader benevolence. Journal of Business Ethics, 170(1), 133–146. https://doi.org/10.1007/s10551-019-04261-1
Figure 1. Theoretical Model
Table 1. Descriptive Statistics and Correlation Analysis
Note. N = 597. UI = treating subordinates uniformly while allowing individualization; SO = combining self-centeredness with other-centeredness; CA: = maintaining decision control while allowing autonomy; RF = enforcing work requirements while allowing flexibility; DC = maintaining both distance and closeness; SSR = superior–subordinate relationships.
* p < .05. ** p < .01.
Table 2. Hypothesis Testing Results
Note. N = 597. SPU = speaking-up voice behavior; SPO = speaking-out voice behavior; UI = treating subordinates uniformly while allowing individualization; SO = combining self-centeredness with other-centeredness; CA = maintaining decision control while allowing autonomy; RF = enforcing work requirements while allowing flexibility; DC = maintaining both distance and closeness; SSR = superior–subordinate relationships; PS = psychological safety.
** p < .01. *** p < .001.
The authors thank all participants who completed the survey.
The data that support the findings of this study are available on request from the corresponding author.
This research was funded by the National Social Science Fund of China (19BGL123 and 22BGL296) and the Society Foundation of Chengdu University of Technology (YJ2023-PY005).
Ruyi Gao, College of Management Science, Chengdu University of Technology, 1 East 3rd Section of Erxianqiao Road, Chenghua District, Chengdu, Sichuan Province, 610059, People’s Republic of China. Email: [email protected]
Article Details
© 2026 Scientific Journal Publishers Limited. All Rights Reserved.