The Mediating Role of Psychological Safety in the Relationship Between Paradoxical Leadership and Organizational Citizenship Behavior

2.1 PL

Paradoxes are enduring contradictions between interconnected elements that create tensions (Schad et al., 2016). Researchers have extensively examined how leaders can behave “to meet contradictory but interrelated, competing workplace demands simultaneously and over time” (Boemelburg et al., 2023). Embracing and integrating opposing demands – rather than choosing between the two ends of a tension – is argued to enhance long-term and sustainable effectiveness (Feng et al., 2022). Scholars have developed the concept of PL by blending principles from Chinese Yin-Yang philosophy with Western leadership theories (Le et al., 2020). Zhang and Han (2019) define PL behaviors as simultaneously competing and interconnected actions. They argue that paradoxical leaders engage in these behaviors over time to address conflicting demands, such as (a) ensuring both organizational stability and flexibility, (b) balancing short-term efficiency with long-term development, (c) adapting to and shaping external collective forces, and (d) attending to the needs of both shareholders and stakeholders. Zhang et al. (2015) framed the dual aspects of leadership behaviors using the “both/and” approach, identifying five key dimensions of PL: (a) retaining control over decisions while fostering autonomy, (b) upholding work standards while offering flexibility, (c) ensuring equal treatment of subordinates while accommodating individuality, (d) striking a balance between maintaining distance and building closeness, and (e) navigating between self-focus and focus on others. The core principle of PL is to address the tensions and contradictions within organizational systems (Ren & Yang, 2021). Paradoxical leaders manage and reconcile the tensions between control and autonomy over time, using this integration to create innovative, long-term strategies that ensure organizational survival (Kim, 2021).

PL is seen as a double-edged sword as it requires the simultaneous use of two opposing behaviors to meet both structural and individual needs in an organization (Niu et al., 2022). In some cases, paradoxes arise not from the leader’s own behavior but from conflicting external pressures, which can lead to subordinates ostracizing the leader, as they may lack awareness of the various factors involved in leadership (Al Hasnawi & Abbas, 2020). Yet, limited research exists on when, how, and under what conditions a leader’s knowledge of paradox can act as a double-edged sword, affecting not only individual followers but also other organizational members (Niu et al., 2022). The debate regarding the positive and negative outcomes of PL behaviors remains active (Boemelburg et al., 2023). Despite these complexities, organizations must cultivate PL to unlock employee potential by leveraging their talents, providing guidance, and addressing their concerns (Xue et al., 2020). Since the concept’s introduction, research on PL’s organizational impact has expanded rapidly, with a major focus on fostering innovative work behaviors. PL has a positive effect on employees (Wang et al., 2023), such as proactivity, creativity, and performance (Zhang et al., 2024), employee adaptability, team cognition, employee dual behavior, strategic agility (Qiang et al., 2023), job satisfaction (Ahmad & Umrani, 2019), and job dedication (Maraşlıoğlu & Özgener, 2022).

2.2 OCB

OCB is the acts that personnel display in relation to their jobs (Williams et al., 2002). OCB refers to voluntary individual actions that are not explicitly outlined in the formal reward system, are not mandated, and typically contribute to enhancing organizational effectiveness. The phrase “non-coercive” implies that these behaviors are not a necessity of the employees’ role or job description. In other words, the organizational employee goes beyond the job description or role requirements and exhibits some attitudes and behaviors that will increase organizational effectiveness by his/her own personal choice (Organ, 1988; Organ, 1997). OCB is useful and desirable behavior for the organization. However, managers are undecided about rewarding employees if such behaviors are displayed or punishing them if they are not (Williams et al., 2002).

Organ categorized OCB into five groups (Farh et al., 1990; Organ, 1988, 1997): (a) Altruism (thinking of coworkers; Podsakoff et al. (2003) called it “helping behavior”): Helping a coworker who is unable to accomplish tasks or solve a problem. (b) High Task Consciousness (conscientiousness-generalized compliance; Podsakoff et al. (2003) called it “individual initiative”): Not wasting time, being punctual. (c) Courtesy (Podsakoff et al. (2003) called it “helping behavior”): When an employee informs or informs those who will be affected by an action or behavior before doing it. (d) Membership Virtue (civic virtue; Podsakoff et al. (2003) also called it “membership virtue: civic virtue”): Showing a high level of commitment and interest in the organization as a whole. (e) Sportsmanship (Podsakoff et al. (2003) also called it “sportsmanship”): Not complaining about work-related inconveniences and impossibilities, not getting angry when coworkers do not follow one’s own suggestions, being optimistic, overriding one’s own interests for the good of the organization, putting one’s personal opinions aside when necessary.

The contributions of OCB to the organization can be expressed as follows (Cohen & Vigoda, 2000; Schnake & Dumler, 2003): (a) increases organizational performance and effectiveness; (b) increases employee productivity; (c) increases efficiency in resource use and sharing; (d) provides effective coordination between individuals and units; and (e) increases adaptability to environmental changes.

The probability of exhibiting OCB will increase depending on the behavior, purpose, environment, and timing. Similarly, employees’ exhibiting similar attitudes and behaviors in the past will contribute to their exhibiting citizenship behaviors in the organization (Williams et al., 2002). OCB is very significant for ensuring organizational effectiveness. Employees who exhibit OCB show such behaviors without expecting any financial reward from the organization they work for (Deluga, 1995). OCB is based on volunteerism and is an out-of-role behavior. Developing OCB as a requirement of organizational justice may be safer than changing behaviors based on formal job descriptions (Deluga, 1995).

2.3 PS

Kahn (1990) defines PS as a state of mind in which personnel feel “safe to assert themselves and engage without fear of negative repercussions on their self-image, status, or career.” Similarly, Edmondson (1999, p. 354) explains it as “the shared belief among team members that the team is safe for interpersonal risk-taking.” This concept reflects how employees perceive their interactions with both managers and colleagues, and it is widely recognized as essential for fostering an environment where individuals feel at ease sharing their opinions (Sun et al., 2022). This sense of safety encourages openness and emphasizes “the importance of creating a workplace where perceptions of interpersonal risk are minimized”. PS inherently involves trust, such that employees feel secure when they trust their organization and view it as supportive (Ahmad & Umrani, 2019; Yıkılmaz et al., 2023). In typical work environments, employees who feel low risk in taking actions that may question relationships or the status quo tend to experience greater PS, which bolsters their courage, confidence, and motivation to involve in behaviors they might otherwise avoid (Sumanth et al., 2024). Although PS is partly shaped by one’s direct supervisor, it is also broadly affected by various workplace factors, such as supportive organizational practices (e.g., diversity initiatives, mentoring), relational networks, and team dynamics (Sumanth et al., 2024). Consequently, interpersonal relationships, group dynamics, leadership, and organizational norms are regarded as critical elements influencing PS (Kahn, 1990).

PS is comprised of four stages: Inclusion safety, learner safety, contributor safety, and challenger safety (Liu & Keller, 2021). Inclusion safety addresses the fundamental human need for connection and belonging. Learner safety, on the other hand, fulfills the need to learn and develop, fostering a secure environment where individuals feel comfortable engaging fully in the learning process-whether by asking questions, exchanging feedback, experimenting, or inevitably making mistakes. Contributor safety meets the intrinsic need to make meaningful contributions, allowing individuals to participate as full team members, using their skills and abilities to support value creation. Finally, challenger safety supports the need for change and improvement by providing a respectful space where individuals feel permitted to dissent or express disagreement when they believe change is necessary. This stage empowers individuals to resist conformist pressures and encourages innovation and creativity.

2.4 PL and OCB

Leader effectiveness centers on the ability to navigate contradictions and crises in complex, uncertain environments (Le et al., 2020). A paradoxical approach offers a useful framework for navigating organizational tensions. Ideally, leaders adopt a paradoxical approach by embracing a “both/and” strategy, which involves accepting persistent tensions, valuing the interdependence of opposing forces, and balancing these forces to achieve optimal outcomes concurrently (Batool et al., 2023). Leaders who employ paradoxical thinking use both differentiation and integration strategies to address organizational contradictions, enabling them to select solutions that capture the essential aspects of conflicting issues within the organization (Niu et al., 2022). This approach requires the leader to show understanding to followers. The most fundamental element that determines an employee’s behavior is his/her interaction with a superior (Güleryüz et al., 2023; Wayne et al., 1997). Research has demonstrated that PL positively influences various employee attitudes and behaviors, including compliance, initiative, dualistic behaviors, team awareness, and innovation (Meng et al., 2021). In other words, PL creates harmony by taking into account seemingly opposing but interrelated behaviors, encourages communication and interaction, and enables employees to exhibit proactive behaviors (Maraşlıoğlu & Özgener, 2022). Proactive work behavior reflects individuals’ readiness to initiate changes in their situations and pursue a better future (Uri, 2017). Key elements of proactivity include self-initiative, a focus on change, and a forward-looking perspective (Kim, 2021). The development of employees’ OCB depends on a positive social and psychological environment. SET claims that people will continue their mutual relations if they think that they will be reciprocated. According to the principle of reciprocity, subordinates tend to reciprocate the support provided by their superiors (Zhong et al., 2022). Indeed, Pan (2021) found that PL has a positive effect on OCB. Similarly, Lee et al. (2023) identified a consistently positive relationship between PL behavior and OCB. Drawing from prior research, and the SET, the following hypotheses are proposed.

H1: Paradoxical leadership positively and significantly affects organizational citizenship behavior.

2.5 PL and PS

PS is defined as employees’ assurance that they can express themselves freely without concern for negative consequences (Iqbal et al., 2022). Employees who feel psychologically secure can speak up, express themselves, and ask for help from others without fear, while employees who feel psychologically insecure may take a defensive approach (Yang et al., 2021). PS can vary among individuals based on factors such as the manager’s leadership style, aspects of job design, and supportive workplace environments (Liu & Keller, 2021). Kim (2021) proposed that PL cultivates an environment of autonomy and equality, where employees feel more comfortable openly discussing ambiguous messages without fear, leading to clearer communication and encouraging “unorthodox” discussions rather than defensive responses. Conversely, when PS is low, employees feel insecure, are more defensive in response to tensions, and are less prone to share their concerns with others (Batool et al., 2023). Paradoxical leaders provide the team with a psychologically safe environment, which in turn creates positive emotions and interpersonal liking, the essence of group cohesion (Batool et al., 2023). According to SCT, people watch and evaluate the people around them and act according to this evaluation (Stajkovic & Luthans, 2002). In this context, the behaviors of the leader will ensure whether the employees feel safe or not. Drawing from prior research, and the SCT, the following hypotheses have been proposed.

H2: Paradoxical leadership positively and significantly affects psychological safety.

2.6 PS and OCB

Research indicates that employees are often hesitant to speak up when they have valuable information to share, whether with authority figures or colleagues (Morrison, 2011). However, when personnel comprehend high levels of PS in the organization, they are more inclined to communicate openly and exchange information without the fear of being judged as disruptive or weak (Liu & Keller, 2021). High PS also supports employees in overcoming the challenges of translating creative ideas into practical applications (Yang et al., 2021). It fosters open communication, encourages the sharing of suggestions and ideas, and promotes seeking feedback (Pearsall & Ellis, 2011). Additionally, PS enhances team cohesion, fosters a sense of inclusion, and motivates employees to exceed expectations (Liu & Keller, 2021). It also helps reduce anxiety, boosts self-confidence, and improves overall performance (Liu & Keller, 2021). As a result, personnel who feel psychologically safe under their leaders’ actions are more likely to engage in discretionary behaviors, such as OCB (Din et al., 2024). This suggests that PS fosters collaboration among individuals from diverse backgrounds, contributing to the achievement of team goals (Liu & Keller, 2021). Previous research has shown that PS mediates the relationship between leadership and both creative problem-solving and creative performance (Ahmad & Umrani, 2019). Furthermore, OCB is positively associated with creativity and productivity (Yaakobi & Weisberg, 2020). Studies by Salman et al. (2020) and Liu and Keller (2021) have demonstrated a positive relationship between PS and OCB, as well as between PS and task performance, knowledge sharing, and OCB. Drawing from prior research, the following hypothesis is proposed.

H3: Psychological safety positively and significantly affects organizational citizenship behavior.

2.7 The Mediating Role of PS

PL involves integrating contrasting behaviors to embrace competing demands, thereby harnessing the underlying benefits of these paradoxes (Xue et al., 2020). This leadership style fosters an environment that helps employees manage creative tensions, ultimately boosting their self-efficacy and PS at work (Walumbwa et al., 2011). Through PS, leadership behavior encourages positive work behaviors, facilitating the voluntary contribution of ideas and collective actions toward shared goals (Erkutlu et al., 2019). From a learning and adaptability standpoint, this can lead to a variety of outcomes, including enhanced learning behaviors, knowledge sharing, creativity, and OCB. Moreover, the literature suggests that PS facilitates optimal team functioning by encouraging followers to engage in proactive team discussions and participate in workgroups that foster a safe organizational climate. It also contributes to a heightened sense of security and follower well-being (Batool et al., 2023). Research has shown that PS can mediate organizational practices and work behaviors (Din et al., 2024; Sun et al., 2022), particularly in relation to voice behavior and supervisor-subordinate relationships (Salman et al., 2020). Xue et al. (2020) demonstrated that PS mediates the relationship between PL and voice behavior encouragement. Additionally, PS has been recognized as a mediating factor across various leadership styles, including inclusive leadership (Carmeli et al., 2010), transformational leadership (Yin et al., 2020), authentic leadership (Liu et al., 2015), servant leadership (Wang et al., 2022), supportive leadership (Elsaied, 2019), spiritual leadership (Chen et al., 2019), and ambidextrous leadership (Iqbal et al., 2022). Overall, PS acts as a mechanism linking PL with proactive employee behaviors, such as OCB, by shaping individuals’ decisions to share and act on their ideas. Drawing from prior research, the following hypothesis is proposed.

H4: Psychological safety mediates the effect of paradoxical leadership on organizational citizenship behavior.

The research model is presented in Figure 1.

Figure 1

Research model.

3.1 Research Procedures and Sample

The sample of this study consists of the employees of three large-scale manufacturing enterprises in Istanbul. In order to conduct a survey on employees, first of all, the managers of the relevant businesses were contacted and permission was requested for the research. After the necessary permissions were obtained, the survey was conducted. Participants were informed about the purpose of the research, told that participation was voluntary, and reminded that they could withdraw from the survey at any time. Prior to administering the survey to the full sample, a pilot study was conducted with 25 participants to test the clarity, readability, and comprehensibility of the scale items. Based on the pilot results, it was concluded that the items were appropriate, and the main survey was subsequently implemented. While face-to-face surveys were conducted in two enterprises, in one enterprise, the questionnaires were given to the HRM in sealed envelopes and received back in sealed envelopes. After two weeks of data collection, 351 questionnaires were obtained. Since 49 of the questionnaires were not scientifically usable, they were not included in the analysis.

The three enterprises have approximately 730 employees. Therefore, 750 questionnaires were prepared to collect data. Out of 750 questionnaires, 302 valid questionnaire data were obtained. The response rate of the surveys was 40.27%. This rate can be considered sufficient for scientific studies in the field of organizational behavior (Tucker et al., 2008).

Of the participants, 93 were female and 209 were male. 189 were married and 113 were single. 119 of the participants were under the age of 25, 120 were between the ages of 26–35, 25 were between the ages of 36–45, and 38 were 46 years of age or older. Of these participants, 156 had a high school degree or below, 121 had an associate’s/bachelor’s degree, and 25 had a postgraduate degree. While 55 participants had less than 1 year of work experience, 114 participants had 1–5 years, 87 participants had 6–10 years, and 46 participants had 11 years or more of work experience.

1. Informed consent: Informed consent has been obtained from all individuals included in this study.

2. Ethical approval: The research related to human use has been complied with all the relevant national regulations, institutional policies, and in accordance with the tenets of the Helsinki Declaration, and has been approved by the authors’ institutional review board or equivalent committee.

3.2 Measures

The scales used in this study are well-established instruments that have been frequently employed in recent research and have demonstrated validity and reliability (Dhir & Vallabh, 2025; Khan et al., 2025; Zawawi et al., 2025). These scales were selected due to their proven reliability in contemporary studies and their capacity to yield robust and meaningful results. Detailed information regarding the scales is presented below.

3.3 Common Method Bias (CMB)

Since the data were collected at the same time and were self-reported, we implemented precautions to reduce CMB. First, to avoid social desirability bias, questionnaires were delivered and collected in sealed envelopes. In face-to-face surveys, we ensured that the respondent was alone and prevented him/her from being influenced by his/her surroundings. Participants were also assured that their responses would remain both confidential and anonymous. Secondly, the statements in the questionnaire were randomly mixed to create psychological separation. Finally, the participants were provided with written instructions clarifying that the statements did not have right or wrong answers, and that their responses should reflect their personal opinions.

In addition to the procedural methods presented above, we used various statistical methods to confirm that our study is CMB-free. First, we applied Harman’s single-factor test, and the analysis indicate that there was no single-factor structure explaining the 50% variance threshold (maximum 39.5%). The correlation analysis results also showed that the relationship between the variables was below 0.90 (Table 3). Finally, VIF values were checked, and it was found that the VIF values were below the lower threshold value of 3.3. The analyses confirmed that CMB is not a source of concern for this research (Bagozzi et al., 1991; Kock, 2015; Sürücü et al., 2023).

3.4 Data Analysis

The research data were analyzed using SPSS 27.0 and AMOS 22.0. First, descriptive statistics were computed using SPSS to summarize the basic characteristics of the sample. Reliability analyses (Cronbach’s alpha) and correlation analyses were also conducted in SPSS to examine the internal consistency and initial relationships among the variables. Next, confirmatory factor analysis (CFA) was performed in AMOS 22.0 to assess the construct validity of the measurement model. The goodness-of-fit indices (e.g., χ²/df, CFI, GFI, TLI, and RMSEA) were evaluated to determine the adequacy of the model fit. Following the validation of the measurement model, the research hypotheses were tested using the PROCESS Macro (Model 4) developed by Hayes (2017), which was run in SPSS. This approach was chosen for its robust bootstrapping procedure in testing mediation effects, offering more accurate estimations of indirect effects compared to traditional methods. The analysis was conducted using 5,000 bootstrap resamples and a 95% confidence interval. The combined use of AMOS and PROCESS allowed for a rigorous examination of both the measurement and structural components of the model, with AMOS providing model fit diagnostics and PROCESS enabling hypothesis testing for mediation through a regression-based framework.

4.1 Preliminary Checks

In the initial stage of the analysis, the extent to which the scale items adequately represented their respective constructs was assessed. Factor analysis revealed that the factor loadings for items measuring PL ranged from 0.536 to 0.924, for PS from 0.515 to 0.811, and for OCB from 0.584 to 0.901. According to the literature, factor loadings above 0.50 indicate that the item has a strong representation of the corresponding factor (Sürücü et al., 2024).

Subsequently, the validity and reliability of the scales were evaluated. Validity was assessed through face validity, convergent validity, and discriminant validity. For reliability, composite reliability (CR) and Cronbach’s alpha coefficients, which indicate internal consistency, were taken into account. The results are presented in Table 1.

Face validity refers to the extent to which a test appears to measure what it is intended to measure, and it is considered the initial step in assessing overall validity (Sürücü & Maslakçı, 2020). Therefore, prior to administering the main survey, face validity was evaluated by collecting feedback from 10 academics (4 Assistant Professors, 5 Associate Professors, and 1 Full Professor) and 11 potential participants. The reviewers were asked to examine the measurement technique and scale items and to assess their appropriateness for measuring the study variables. The feedback indicated that the scales demonstrated strong face validity. Following this step, the main data collection was conducted, and the data were subsequently analyzed.

According to the literature, to establish convergent validity, the Average Variance Extracted (AVE) should be greater than 0.50 and less than the CR (Fornell & Larcker, 1981; Sürücü & Maslakçı, 2020). The results of the analysis confirmed that these criteria were met. The correlation analysis conducted to check discriminant validity showed that the correlation values were less than 0.85, and the √AVE value was greater than the correlation values between the variables (Table 3). Table 2 shows that the four-factor research model has better fit indices than the other models (χ ²/df = 2.82, CFI = 0.951, GFI = 0.948, TLI = 0.951, and RMSEA = 0.045). When all the findings were considered together, it was determined that the four variables in the research model demonstrated face validity, convergent validity, and discriminant validity.

The literature emphasizes that using reliable scales in research is crucial for obtaining valid results (Bekmezci & Sürücü, 2025). In the analyses conducted to determine reliability, it was determined that Cronbach’s Alpha and CR values were greater than the lower threshold value of 0.7 (Hair et al., 2017). All these findings confirm that the scales are valid and reliable.

As seen in Table 3, after control variables are restricted, PL has a positive relationship with PS (r = 0.245, p < 0.05) and OCB (r = 0.266, p < 0.05). It was also found that PS has a positive relationship with OCB (r = 0.356, p < 0.05).

4.2 Hypotheses Testing

The research hypotheses were tested using the Process Macro developed by Hayes (2017), as it allows for a robust analysis of mediation and moderation effects, which are central to the theoretical model of this study. Compared to other methods, Process Macro provides a user-friendly and statistically rigorous approach, making it the most suitable choice for this research. In particular, the traditional four-step regression approach proposed by Baron and Kenny (1986) for testing mediation effects has been criticized for its low statistical power. In contrast, the bootstrap method implemented in the Process Macro is a more contemporary and widely accepted technique among researchers. This method adopts a non-parametric resampling approach for estimating effect sizes and testing hypotheses, without making assumptions about the shape or parameters of the sampling distribution. As a result, it yields more reliable and valid inferences, especially in studies involving indirect effects (Hayes & Preacher, 2010; Preacher & Hayes, 2004). The findings of the analyses conducted with 5,000 resamples at a 95% confidence interval are presented in Table 4.

The findings in Table 4 show that PL has a significant and positive effect on PS (β = 0.126, p < 0.05, [CI = (0.089, 0.287)]) and OCB (β = 0.168, p < 0.05, [CI = (0.103, 0.209)]). In addition, the effect of PS on OCB is also significant and positive (β = 0.301, p < 0.05, [CI = (0.243, 0.462)]).

The findings confirm that PL has a significant effect on OCB through PS (β = 0.059, p < 0.05, [CI = (0.015, 0.106)]). Therefore, PS plays a mediating role between PL and OCB. In line with the findings obtained, Hypotheses 1, 2, 3, and 4, which were developed to be tested in the study, were supported.

5.1 Theoretical Implications

Firstly, PL proactively responds to the dynamic and complex business conditions of organizations, enhancing organizational performance. We found that PL serves as a significant precursor to employees’ OCB. The results confirm a positive relationship between PL and OCB, which is consistent with previous studies. Research on the relationship between PL and proactive behaviors has revealed a positive link between these two variables (Yang et al., 2021; Zhang et al., 2015). Similarly, Pan (2021) found that PL significantly impacts OCB. The findings indicate that PL significantly predicts PS, and in turn, PS significantly predicts OCB. These results support a positive relationship between PL and PS, as well as between PS and OCB, consistent with prior research (Batool et al., 2023; Liu & Keller, 2021; Salman et al., 2020). Although PS has been examined as a mediating variable in the context of various leadership styles, this study extends the literature by demonstrating its mediating role specifically in the relationship between PL and OCB. To the best of our knowledge, no previous study has empirically investigated PS as a mediator in this particular relationship. Therefore, the present study offers a novel contribution to the leadership and organizational behavior literature.

Secondly, while prior research on PL has predominantly employed direct-effect or single-path models (Niu et al., 2022), this study proposes and empirically tests a mediating mechanism through PS. By introducing PS as a mediator, the study provides a more nuanced understanding of how PL influences OCB, highlighting the psychological processes that underlie this relationship. This contributes to the literature by shifting the focus from a purely behavioral framework to one that incorporates cognitive and affective mediators. Furthermore, by applying this framework within the context of the manufacturing sector – a setting often characterized by hierarchical structures, operational rigidity, and safety concerns – this study illustrates how PL can foster a psychologically safe environment even in traditionally risk-averse industries. Our findings align with those of previous studies. For instance, Lee et al. (2023), through a meta-analysis, found that PL positively affects employees’ OCB, with this effect occurring through PS. The present study complements such findings by offering empirical evidence from a sector where such dynamics are underexplored, thereby broadening the generalizability of existing theoretical models.

Thirdly, this study contributes to the growing body of research on PL by examining its relevance within the Turkish context, which is characterized by a collectivist and high power distance culture. The findings suggest that PL, as a leadership style that balances competing demands (e.g., control and autonomy, uniformity and individuality), may be particularly effective in navigating the tensions inherent in such cultural environments. By situating the study in Türkiye, we offer contextual insights into how PL functions beyond the typically studied Western or East Asian settings, thereby enhancing the cross-cultural generalizability of PL theory.

Finally, this study advances the theoretical foundations of PL by adopting a multi-theoretical lens. While previous research has largely interpreted the effects of PL through a single framework – most commonly SET – this study integrates both SET and SCT to explain the underlying mechanisms linking PL and OCB. Our findings demonstrate that SET helps explain how relational exchanges (e.g., trust, perceived support) fostered by PL lead to reciprocal behaviors such as OCB, while SCT provides insight into the cognitive and psychological processes (e.g., self-efficacy, PS) that enable such behaviors. This dual-theoretical approach offers a more comprehensive understanding of the complex pathways through which PL influences employee outcomes, and it opens avenues for future research to adopt similarly integrative frameworks.

5.2 Practical Implications

The rapid pace of change and increasing uncertainty in the external environment have led to the coexistence of conflicting and competing demands within organizations. This dynamic often gives rise to tensions and paradoxes that leaders must navigate effectively. As a result, there is a growing need for leaders to adopt a holistic mindset and to manage competing expectations simultaneously. Moreover, it is essential for leaders to meet the dual needs of their organizations and employees, adapt to varying organizational contexts, and exhibit a flexible leadership style that can respond to these complexities. In this regard, the findings of this study offer several important practical implications.

First, PL has been shown to support OCB and PS. So this study provides important insights for managerial practice by highlighting the positive influence of PL on both OCB and PS. Leaders, particularly in manufacturing organizations where hierarchical structures and rigid procedures are common, are encouraged to adopt a PL mindset and invest in leadership development programs that promote paradoxical thinking. Traditional leadership approaches often rely on an “either/or” logic, where selecting one course of action necessitates the rejection of the other. This binary mindset can lead to rigidity and create environments in which employees may feel unsafe when navigating ambiguity or conflicting expectations. In contrast, paradoxical leaders embrace a “both/and” perspective, which acknowledges the coexistence of competing demands and the inherent tensions within organizational life. Rather than attempting to resolve all paradoxes, effective leaders learn to manage them constructively. This approach fosters a psychologically safe environment by validating complexity and promoting openness to multiple perspectives. When leaders align paradoxical behaviors – such as providing autonomy while maintaining control – with the organization’s mission, vision, and values, employees are more likely to perceive coherence and purpose in their work. Moreover, adopting a “both/and” mindset encourages creativity, adaptability, and long-term problem-solving – key capabilities in today’s volatile and uncertain business landscape. Organizations that cultivate PL are therefore better positioned to enhance employee well-being, foster innovation, and sustain high levels of performance. These findings underscore the need for leadership development practices that build cognitive complexity and emotional resilience, enabling leaders to thrive amid contradictions rather than be constrained by them.

Secondly, this study highlights the critical role of PS in enhancing the positive effects of PL on OCB, particularly within the manufacturing sector. In this context, leaders are encouraged to intentionally cultivate environments where employees feel safe to express themselves without fear of embarrassment, rejection, or punishment. When individuals perceive threats to their self-image or identity, their natural response is often defensive, which can hinder open communication, innovation, and cooperative behavior. Therefore, it is imperative for leaders to foster an atmosphere of mutual respect, empathy, and trust. Leaders can enhance PS by actively encouraging the expression of diverse viewpoints, remaining receptive to feedback, and validating employees’ contributions. This involves not only being open to new ideas but also challenging assumptions in a constructive manner that promotes collective learning rather than individual blame. Questioning should be positioned as a tool for exploration and mutual understanding, rather than criticism. Furthermore, leaders should empower employees by creating mechanisms that support idea implementation, minimize bureaucratic barriers, and promote transparent communication regarding progress and outcomes. In doing so, leaders not only strengthen employees’ sense of inclusion and belonging but also stimulate creativity, proactive behavior, and shared accountability – all of which are essential for fostering a sustainable organizational culture. Given that the manufacturing sector often operates in structured and hierarchical environments, the intentional promotion of PS can act as a catalyst for bottom-up innovation and continuous improvement. As such, the findings of this study suggest that cultivating PS is not merely a soft skill, but a strategic leadership capability that directly contributes to positive organizational behavior and long-term effectiveness.

Thirdly, given that not all employee-generated ideas can be implemented due to various practical constraints, it becomes even more critical for leaders to safeguard employees’ PS. Employees are more likely to engage in OCB when they feel mentally secure, respected, and appreciated, even if their suggestions are not always adopted. In this context, leadership practices should be designed to foster a non-threatening environment where employees are encouraged to contribute ideas without fear of failure or rejection. This includes providing constructive feedback, recognizing effort regardless of outcome, and reinforcing the value of experimentation and learning from setbacks. Furthermore, leaders should communicate the organization’s goals and strategic priorities with clarity and transparency to ensure that employees understand the broader purpose of their roles. Such alignment helps employees feel that their efforts contribute meaningfully to organizational objectives, thereby increasing motivation and initiative. Once these objectives are clearly defined, it is essential for leaders to empower employees with the autonomy to make decisions and take ownership of their work. This autonomy, when combined with PS, fosters a sense of trust and encourages employees to express their individuality and pursue innovative solutions without the fear of negative consequences.

Lastly, during periods of uncertainty, change, or organizational challenge, the presence of PS becomes even more valuable. Employees who perceive that they are trusted and valued by leadership are more likely to demonstrate discretionary behaviors, such as OCB, that go beyond formal job requirements. These behaviors, in turn, contribute to organizational resilience and sustained competitive advantage. Therefore, investing in leadership development programs that emphasize the cultivation of PS can serve as a critical lever for long-term organizational success.

5.3 Limitations and Future Directions

When interpreting the research findings, it is important to consider certain limitations. First, the relatively small sample size and the use of a convenience sampling method limit the generalizability of the results. The study is also focused on a specific cultural context (Türkiye) and sector (manufacturing), which may restrict its external validity. To support broader generalizations, future research should involve larger, more representative samples across different cultures and industries. The study was cross-sectional rather than longitudinal; thus, future studies could benefit from a longitudinal design to expand the scope and strengthen causal interpretations of the model. Third, data were gathered from employees working in three large-scale manufacturing companies in Türkiye, and future research could explore whether these findings hold across different cultural settings. Lastly, although relevant controls were applied, the possibility of unmeasured variables influencing the relationships cannot be ruled out. Future studies might also explore the indirect effects of PL on OCB by investigating other potential mediating mechanisms such as psychological empowerment, organizational trust, or emotional exhaustion. These factors are theoretically linked to both PL and OCB and may help explain how the seemingly contradictory behaviors exhibited by paradoxical leaders promote discretionary and prosocial behaviors among employees. Future research could also examine contextual variables – such as organizational culture or role clarity – that might influence the strength or direction of these mediating effects.