Introduction
Employees who contribute beyond their formal job requirements to an organization have been the subject of increasing interest among scholars and managers (Grant & Mayer, Reference Grant and Mayer2009; Hoffman, Blair, Meriac, & Woehr, Reference Hoffman, Blair, Meriac and Woehr2007; Ilies, Nahrgang, & Morgeson, Reference Ilies, Nahrgang and Morgeson2007). Given that organizational citizenship behaviour (OCB) is positively associated with task performance (Hoffman et al., Reference Hoffman, Blair, Meriac and Woehr2007), employee performance and organizational productivity, efficiency and customer satisfaction (Podsakoff, Whiting, Podsakoff, & Blume, Reference Podsakoff, Whiting, Podsakoff and Blume2009), it makes sense to identify those variables that increase this behaviour in organizational settings. The existing literature on the relationship between leadership behaviours and OCB mostly focuses on the conventional leadership contingency perspective, such as the ‘either…or…’ approach (i.e., transformational leadership, ethical leadership and servant leadership) (Ko, Ma, Kang, English, & Haney, Reference Ko, Ma, Kang, English and Haney2017; Newman, Schwarz, Cooper, & Sendjaya, Reference Newman, Schwarz, Cooper and Sendjaya2017; Nohe & Hertel, Reference Nohe and Hertel2017), which cannot effectively resolve the contradictions and conflicts caused by complex internal and dynamic external conditions (Rosing, Frese, & Bausch, Reference Rosing, Frese and Bausch2011). As Zhang, Waldman, Han, and Li (Reference Zhang, Waldman, Han and Li2015) pointed out, to maintain long-term effectiveness in an increasingly dynamic, complex and competitive organizational environment, leaders need the ability to respond to such paradoxical challenges.
Paradoxical leadership is defined as ‘seemingly competing, yet interrelated, behaviours to meet structural and follower demands simultaneously and over time’ (Zhang et al., Reference Zhang, Waldman, Han and Li2015: 538). When faced with various management paradoxes (Lavine, Reference Lavine2014), such as the balance between control and authorization, efficiency and flexibility, and individualism and collectivism, conventional leadership contingency theory holds that leaders are expected to make the best decision between the two, such as choosing authorization or control (Li, She, & Yang, Reference Li, She and Yang2018; Waldman & Bowen, Reference Waldman and Bowen2016). However, according to Smith & Lewis (Reference Smith and Lewis2011), such a decision is favourable only in the short-term. Conversely, paradoxical leadership adopts a ‘both…and…’ strategy in which leaders seek an integrative middle ground, negotiating acceptable trade-offs by which all stakeholders can abide (Rosing, Frese, & Bausch, Reference Rosing, Frese and Bausch2011).
The authors of previous studies have considered the influence of paradoxical leadership on employees' attitudes and extra-role behaviours from different theoretical perspectives, such as self-determination theory (Yang, Li, Liang, & Zhang, Reference Yang, Li, Liang and Zhang2021), social cognitive theory (Shao, Nijstad, & Täuber, Reference Shao, Nijstad and Täuber2019), leader-member exchange theory (Xue, Li, Liang, & Li, Reference Xue, Li, Liang and Li2020) and social learning theory (Franken, Plimmer, & Malinen, Reference Franken, Plimmer and Malinen2020; Ishaq, Bashir, & Khan, Reference Ishaq, Bashir and Khan2019). Very little attention has been given to the affect-based perspective as a framework for understanding the link between paradoxical leadership and employee OCB. Emotions are ubiquitous in organizations, and individual behaviour is a comprehensive consequence of rational cognitive and affective experiences (Mischel & Shoda, Reference Mischel and Shoda1995). This study aims to fill this gap by introducing affective events theory (AET) as a theoretical framework that could explain the mechanism by which employees' affect state can be triggered by a paradoxical supervisor and that such positive affect can shape their OCB. AET is an event-specific framework for understanding the role of emotions in the workplace (Spence, Brown, Keeping, & Lian, Reference Spence, Brown, Keeping and Lian2014) that emphasizes the role of work events as proximal causes of affective reactions (Weiss & Cropanzano, Reference Weiss and Cropanzano1996). The purpose of this study is to explore how paradoxical leadership, as a feature of the work environment, could stimulate employees' positive affect, which is ‘a pleasant feeling state or good mood’ (Estrada, Isen, & Young, Reference Estrada, Isen and Young1994: 286), and could motivate their OCB as a result.
Moreover, AET proposes that individual affect levels are influenced by dispositions and a variety of environmental factors (Volmer, Reference Volmer2015). Individuals are the subject of decisions virtually every day of their organizational lives, and they judge the decision-making they experience with a very critical eye by asking questions such as, ‘Was that fair?’ (Colquitt, Reference Colquitt2001). Individuals' procedural justice perceptions are idiosyncratic evaluations of employees in the workgroup regarding the fairness of organizational authorities (Greenberg, Reference Greenberg1990). Researchers in the field of leadership and fairness advocate the integration of fairness and leadership and speak to the importance of fairness in explaining leadership effectiveness (i.e., Van Knippenberg, De Cremer, & Van Knippenberg, Reference Van Knippenberg, De Cremer and Van Knippenberg2007). However, there is limited research on the boundary conditions of how employees respond to paradoxical leaders, as well as their emotional reaction and behaviours. The authors of this study focus upon the boundary condition of procedural fairness and its impact on the relationship between positive affect and OCB. We then develop a moderated mediation model of the psychological processes that links paradoxical leadership and employee OCB. The research model is shown in Figure 1.
Theoretical model.
In the present study, we aim to make three contributions. First, we seek to contribute to the emerging research on paradoxical leadership by proposing leader paradoxical behaviour as an important affective event to encourage employee positive affect. Second, we systematically examine the psychological mechanisms underlying the impact of paradoxical leadership on employee OCB by focusing on positive affect as potential mediators based on AET. Last, this study is one of the first to advance the application of AET in the context of paradoxical leadership research by examining the moderating effect of procedural fairness on the hypothesized relationship between positive affect and OCB.
Theory and hypothesis development
Mediating role of positive affect
AET suggests that affective responses play an intermediary role between work events and behavioural results (Weiss & Cropanzano, Reference Weiss and Cropanzano1996). According to Clark, Watson, and Leeka (Reference Clark, Watson and Leeka1989), a positive affect, such as happiness, joyfulness and enjoyment, is generated by the interaction between an individual and the environment. Consist with AET, positive affect at work is a state generated at the work place and through events and conditions encountered there, including leadership behaviour (Weiss & Cropanzano, Reference Weiss and Cropanzano1996). This study hypothesizes that paradoxical leadership, as an important affective event, can influence employees' affective reactions such as positive affect and then generate affect-driven behaviour (i.e., OCB). Specifically, paradoxical leadership exerts a synergistic effect of integrating contradictions through paradoxical thinking and endows subordinates with proficiency, adaptivity and proactivity, which benefits their short- and long-term development (Zhang et al., Reference Zhang, Waldman, Han and Li2015). The use of ‘both…and…’ terminology enables followers to experience work autonomy, fairness and support in the work process, which can generate their positive affect (Rousseau & Aube, Reference Rousseau and Aube2010; Wegge, Dick, Fisher, West, & Dawson, Reference Wegge, Dick, Fisher, West and Dawson2006).
In addition, affect has also been related to leadership effectiveness (Damen, Knippenberg, & Knippenberg, Reference Damen, Knippenberg and Knippenberg2008). For example, paradoxical leadership maintains both distance and closeness in terms of relationships with employees; it also establishes good relationships with employees while maintaining authority (Pan, Reference Pan2021; Zhang et al., Reference Zhang, Waldman, Han and Li2015). This harmonious supervisor–subordinate relationship can be considered a source of positive affect for employees (Haller & Hadler, Reference Haller and Hadler2006; Stephens, Heaphy, & Dutton, Reference Stephens, Heaphy, Dutton, Cameron and Spreitzer2012).
Furthermore, people with high positive affect are enthusiastic, active and alert (Watson, Clark, & Tellegen, Reference Watson, Clark and Tellegen1988); moreover, they have the potential to contribute to valued organizational outcomes (Carlson, Kacmar, Grzywacz, Tepper, & Whitten, Reference Carlson, Kacmar, Grzywacz, Tepper and Whitten2013; Mostafa, Reference Mostafa2016). According to Park, Shim, Hai, Kwon, and Kim (Reference Park, Shim, Hai, Kwon and Kim2021), positive affect enhances cognitive flexibility by expanding an individual's scope of attention and cognitive repertoire. Isen and Baron's (Reference Isen and Baron1991) study showed that employees with a higher level of positive affect are more willing to help others. Bachrach and Jex (Reference Bachrach and Jex2000) found that positive affect leads people to define their jobs more broadly, resulting in employees becoming more likely to perform OCB. Research has demonstrated that employees who are experiencing positive affect are more willing to engage in OCB as a means of maintaining or prolonging their good feelings (Ilies, Scott, & Judge, Reference Ilies, Scott and Judge2006; Organ & Ryan, Reference Organ and Ryan1995; Rioux & Penner, Reference Rioux and Penner2001; Williams & Shiaw, Reference Williams and Shiaw1999). Based on the aforementioned arguments, the following hypothesis is advanced:
Hypothesis 1: Positive affect mediates the relationship between paradoxical leadership and employee OCB.
Moderating role of procedural fairness
AET suggests that dispositions and environmental factors influence how people feel towards and react to affective events (Weiss & Cropanzano, Reference Weiss and Cropanzano1996). The fairness of treatment that people receive, or procedural fairness, communicates important relational information with respect to one's standing within the group (De Cremer & Van Knippenberg, Reference De Cremer and Van Knippenberg2002; Vermunt, Van Knippenberg, Van Knippenberg, & Blaauw, Reference Vermunt, Van Knippenberg, Van Knippenberg and Blaauw2001). The current research identifies employees' procedural fairness perceptions as an important boundary condition between positive affect and OCB.
Procedural fairness constitutes an organizational form of the ‘rule of law’ in terms of its impersonal or universalistic treatment of employees and should be distinguished from the particularistic treatment of employees (Sun, Chow, Chiu, & Pan, Reference Sun, Chow, Chiu and Pan2013). People are influenced by procedural fairness because it addresses more symbolic or social/psychological concerns, such as needs for self-esteem, self-identity and affiliation (Brockner, Reference Brockner2002). In organizations with high procedural fairness, benefits are more likely to be offered to employees on the basis of their merit-related criteria rather than nonmerit criteria (Sun et al., Reference Sun, Chow, Chiu and Pan2013). As a result of having high positive affect, employees display the ability of OCB to contribute to fulfilling social normative expectations and attaining positive rewards (Walumbwa, Hartnell, & Oke, Reference Walumbwa, Hartnell and Oke2010).
In contrast, when employees have previously experienced injustice within their organization, they tend to be highly sensitive to subsequent justice events (Loi, Yang, & Diefendorff, Reference Loi, Yang and Diefendorff2009). According to Cropanzano, Byrne, Bobocel, and Rupp (Reference Cropanzano, Byrne, Bobocel and Rupp2001), inconsistent and biased procedures lead employees to feel that they are not equally valued, which negatively impacts their attitudes and behaviours. Therefore, although they experienced positive affect evoked by paradoxical leadership, they may be less likely to perform OCB in response to experienced unfairness. Based on the above arguments, the following hypothesis is advanced:
Hypothesis 2: Procedural fairness moderates the relationship between employee positive affect and OCB, and the positive relationship will be strengthened under conditions of high procedural fairness.
This research further proposes a moderated mediation model in which paradoxical leadership influences employee OCB through its relationship with positive affect; the indirect effect will be stronger when employees' perceived procedural fairness is strong rather than weak. The behaviours that paradoxical leaders execute can be considered affective events that are conducive to evoking positive affect from employees. Positive experiences triggered by affective events will further bring about positive behaviours, such as OCB. A high level of procedural fairness strengthens employees' evaluation of leadership effectiveness and enhances the indirect effect of paradoxical leadership on employee OCB via positive affect. Conversely, employees who experience an unjust work environment might experience decreased positive affect and become unwilling to perform OCB as a negative response to paradoxical leadership. Accordingly, the following hypothesis is proposed:
Hypothesis 3: Procedural fairness moderates the mediating effect of positive affect on the relationship between paradoxical leadership and employee OCB. Thus, the indirect effect of paradoxical leadership on employee OCB via positive affect will be stronger under conditions of high procedural fairness.
Methods
Sample and data collection
The respondents of this study are employees of several small- and medium-sized companies located in Hubei province of China. To reduce the risk of social desirability bias, the cover letter to the survey explained the general research purpose, assured the participants of complete confidentiality, explicated that their participation was voluntary, and indicated that their answers were anonymous (De Clercq & Belausteguigoitia, Reference De Clercq and Belausteguigoitia2020). To avoid common method bias, this study collected data at two time points with an interval of 2 weeks, and each questionnaire was coded for two-phase matching. In the first phase, with the help of HR managers, 300 questionnaires assessing the employees' demographic information and their perceived paradoxical leadership were distributed. In the second phase, the same number of questionnaires assessing employees' perceptions of positive affect, procedural fairness and OCB were distributed to employees who completed the questionnaire in the first stage. Finally, 218 questionnaires were returned, and after deleting incomplete questionnaires, we obtained 201 valid questionnaires as evidence for our theoretical model, with a valid response rate of 67%.
Among the respondents, 64.20% were men. Regarding their age, 76.61% were under 30 years old, 16.92% were 30–40 years old and 6.47% were over 40 years old. The majority (67.16%) of the respondents held a bachelor's degree, 11.44% held a junior college or below, 19.40% held a master's degree, and 2% held a PhD degree. Regarding tenure, 67.60% had <1 year of tenure, 22.90% had 1–5 years of tenure and 9.50% had more than 5 years of tenure.
Measurement
All variables were measured from validated scales. Brislin's (Reference Brislin1980) translation and back-translation procedure was followed to prepare Chinese questionnaires.
Paradoxical leadership
This study used the scale developed by Zhang et al. (Reference Zhang, Waldman, Han and Li2015), which includes five dimensions with 22 items. The sample items are, ‘Uses a fair approach to treat all subordinates uniformly but also treats them as individuals (treating subordinates uniformly while allowing individualization, UI)’, ‘Shows a desire to lead, but allows others to share the leadership role (combining self-centeredness with other-centeredness, SO)’, ‘Controls important work issues, but allows subordinates to handle details (maintaining decision control while allowing autonomy, CA)’, ‘Stresses conformity in task performance but allows for exceptions (enforcing work requirements while allowing flexibility, RF)’, and ‘Recognizes the distinction between supervisors and subordinates but does not act superior in the leadership role (maintaining both distance and closeness, DC)’. Respondents scored these items on a 7-point Likert scale (1 = strongly disagree to 7 = strongly agree). The Cronbach's α of this scale was .94.
Positive affect
Positive affect was assessed using the positive affect subscale of the Positive and Negative Affect Schedule (PANAS) (Watson, Clark, & Tellegen, Reference Watson, Clark and Tellegen1988). The PANAS is one of the most widely used instruments to measure individual differences in mood (Boyraz & Efstathiou, Reference Boyraz and Efstathiou2011). Ding & Lin (Reference Ding and Lin2020) used this scale based on Chinese samples with a Cronbach's α of .92. The subscale assessing positive affect has 10 items capturing different positive emotions (e.g., ‘enthusiastic’, ‘interested’, ‘excited’, ‘inspired’, ‘alert’, ‘active’, ‘strong’, ‘proud’, ‘attentive’ and ‘excited’). Respondents were asked to indicate the extent to which they were experiencing 10 particular positive affective states in the last month. The 7-point Likert response scale ranged from one (1 = very slightly or not at all) to seven (7 = extremely), and higher scores indicate higher levels of positive affect. The Cronbach's α of this scale was .95.
Procedural fairness
We measured procedural fairness using seven items from Colquitt's (Reference Colquitt2001) scale. Hon and Lu (Reference Hon and Lu2010) measured this scale based on Chinese samples with a Cronbach's α of .77. The seven items were as follows: ‘The following items refer to the procedures used to arrive at your (outcome); to what extent have you experienced the following: Have you been able to express your views and feelings during those procedures? Have you had influence over the (outcome) arrived at by those procedures? Have those procedures been applied consistently? Have those procedures been free of bias? Have those procedures been based on accurate information? Have you been able to appeal the (outcome) arrived at by those procedures? Have those procedures upheld ethical and moral standards?’ The respondents scored these items on a 7-point Likert scale (1 = strongly disagree to 7 = strongly agree). The Cronbach's α of this scale was .91.
OCB
OCB was measured with the 10-item scale developed by Bachrach, Hui, Bendoly, and Zhang (Reference Bachrach, Hui, Bendoly and Zhang2007) with two dimensions. Sample items were the following: ‘Help other employees out if someone falls behind in his or her work (helping)’ and ‘Be willing to risk disapproval to express beliefs about what's best for the unit (civic virtue)’. Respondents scored these items on a 7-point Likert scale (1 = strongly disagree to 7 = strongly agree). The Cronbach's α of this scale was .92.
Control variables
Employees' demographic information was collected, including gender, age, education and tenure. Gender was coded as 1 = male and 0 = female. Age was coded as 1 = under 30 years old, 2 = 30–40 years old and 3 = more than 40 years old. Education was coded 1 = junior college or below, 2 = bachelor's degree, 3 = master's degree and 4 = PhD degree. Tenure was coded as 1 = <1 year, 2 = 1–5 years and 3 = more than 5 years.
Results
Descriptive statistics and preliminary results
The results of the means, standard deviations and correlations are presented in Table 1. Paradoxical leadership was positively correlated with positive affect (r = .54, p < .01), procedural fairness (r = .41, p < .01) and OCB (r = .34, p < .01). Positive affect was positively correlated with procedural fairness (r = .52, p < .01) and OCB (r = .50, p < .01).
Descriptive statistics and simple correlations
PL, paradoxical leadership; PA, positive affect; PF, procedural fairness; OCB, organizational citizen behaviour; *p < .05, **p < .01 (two-tailed test).
We conducted confirmatory factor analysis with Mplus 7.4 software (Muthén & Muthén, Reference Muthén and Muthén2012–2017) to assess the construct validity of the variables included in the research. As suggested by Hair, Anderson, Tatham, and Black (Reference Hair, Anderson, Tatham and Black1998), the overall model's χ2, comparative fit index (CFI), Tucker–Lewis index (TLI), root mean square error of approximation (RMSEA) and standardized root mean square residual (SRMR) were used to assess the model fit. The findings showed that the hypothesized four-factor model exhibited a better and more acceptable fit than all other alternative measurement models (χ2/df = 2.04, CFI = .93, TLI = .92, RMSEA = .07, SRMR = .06). Additionally, all factor loadings were significant, which provides evidence of convergent validity. We then conducted Harman's one-factor test, as recommended by Podsakoff, Mackenzie, and Podsakoff (Reference Podsakoff, Mackenzie and Podsakoff2012). All the items were loaded on one factor, and the amount of explained variance was examined. The factor explained only 33.87% of the variance and, thus, did not explain the majority of the total variance. Taken together, these results indicated that the data did not suffer serious common method bias.
Hypothesis testing
Hierarchical regression analysis was adopted to test hypotheses, and the results are shown in Table 2. Hypothesis 1 predicted that positive affect mediates the relationship between paradoxical leadership and employee OCB. Model 2 shows that paradoxical leadership was positively related to positive affect (b = .56, p < .001). In Model 4, paradoxical leadership was positively related to OCB (b = .24, p < .001). In Model 5, when both paradoxical leadership and positive affect were included in the regression analysis, the effect of positive affect on OCB was significant (b = .33, p < .001), while that of paradoxical leadership was not (b = .05, ns). These results demonstrated that positive affect played a fully mediating role between paradoxical leadership and employee OCB. Hence, Hypothesis 1 is supported.
Result of hierarchical regression analysis
PL, paradoxical leadership; PA, positive affect; PF, procedural fairness; OCB, organizational citizen behaviour; *p < .05, **p < .01, ***p < .001 (two-tailed test); unstandardized estimates and their associated standard errors in parentheses are reported.
In addition, to further test the mediating effect of positive affect, the bootstrapping approach was used for testing indirect effects proposed by Preacher and Hayes (Reference Preacher and Hayes2004). We used Mplus 7.4 software to analyse the mediating effect of positive affect, and 5,000 resampling options were chosen with the bootstrap technique. The results in Table 3 show that the indirect effect of positive affect on the relationship between paradoxical leadership and employee OCB was significant (estimated effect = .19, 95% CI .11 to .29). In addition, the direct effect was not significant (estimated effect = .05, 95% CI −.09 to .20), indicating that positive affect played a full mediating role between paradoxical leadership and employee OCB. Therefore, Hypothesis 1 is further verified.
Mediating effect of positive affect between paradoxical leadership and OCB
PL, paradoxical leadership; PA, positive affect; PF, procedural fairness; OCB, organizational citizen behaviour.
Hypothesis 2 assumed that procedural fairness moderates the relationship between employee positive affect and OCB, and this positive relationship will be strengthened under conditions of high procedural fairness. Model 6 in Table 2 shows that the coefficient of the interaction term (positive affect multiplied by procedural fairness, where we centred the positive affect and procedural fairness when calculating the interaction terms to avoid collinearity) was significant (b = .22, p < .001), indicating that procedural fairness moderated the relationship between positive affect and OCB. Thus, Hypothesis 2 is supported. To further illustrate the moderating effect, a simple slope analysis method was used (Aiken & West, Reference Aiken and West1991) to draw a diagram of the moderating effect of procedural fairness. Specifically, the method plots a graph showing plus and minus one standard deviation from the mean of the procedural fairness. Figure 2 shows that the positive relationship between positive affect and OCB is much more distinct in high procedural fairness than in low procedural fairness (simple slope high = .59, t = 6.79, p < .001; simple slope low = .19, t = 3.50, p < .001; difference = .40, p < .001). Therefore, Hypothesis 2 is verified.
Moderating effect of procedural fairness between positive affect and OCB.
Hypothesis 3 proposed that procedural fairness moderates the mediating effect of positive affect on the relationship between paradoxical leadership and employee OCB. Thus, the indirect effect of paradoxical leadership on employee OCB via positive affect will be stronger under conditions of high procedural fairness. We also used Mplus 7.4 software to analyse the moderated mediating model based on the bootstrapping approach with iterative resampling (n = 5,000). Table 4 demonstrates that paradoxical leadership had a stronger positive indirect impact on employee OCB when the level of procedural fairness was high (estimated effect = .35, 95% CI .19 to .57) rather than low (estimated effect = .12, 95% CI .003 to .23). In addition, the difference was significant (difference = .23, 95% CI .05 to .42). Taken together, these results reveal that procedural fairness moderates the indirect positive effect of paradoxical leadership on employee OCB. Therefore, Hypothesis 3 is supported.
Conditional indirect effect of paradoxical leadership on OCB via positive affect
Robustness testing
Procedural fairness has also been shown to be strongly associated with subordinates' perception of leadership (Folger & Konovsky, Reference Folger and Konovsky1989), trust in one's supervisor and the organization (Colquitt et al., Reference Colquitt, Scott, Rodell, Long, Zapata, Conlon and Wesson2013), and affect (Murphy & Tyler, Reference Murphy and Tyler2008). Therefore, we tested the possibility of an alternative model: procedural fairness moderating the relationship between paradoxical leadership and positive affect. We found the criteria for moderation in this model was not met, as the results show that the interaction between paradoxical leadership and procedural fairness is not significantly related to positive affect (b = .03, ns). Also, the difference between high and low level of procedural fairness (plus and minus one standard deviation from the mean of the procedural fairness) is nonsignificant (difference = .07, ns). Individuals with positive affect are generally energetic, enthusiastic, confident and active (Lyubomirsky, King, & Diener, Reference Lyubomirsky, King and Diener2005; O'Neill, Stanley, & O'Reilly, Reference O'Neill, Stanley and O'Reilly2011). Employees who experience positive affect are likely to exhibit generally positive behaviours, such as OCB under the high level of procedural fairness perceptions because they rely on their judgement of fairness as a heuristic to direct their efforts towards either serving the demands of the collective or fulfilling their self-interests (Lind, Reference Lind, Greenberg and Cropanzano2001; Van den Bos, Reference Van den Bos2001). Consequently, the possibility of this alternative model was discarded.
Discussion
This research is based on the affect-driven perspective and views AET as a vital rationale for examining the unique mechanisms and boundary conditions of paradoxical leadership. The results show that positive affect plays a full mediating role between paradoxical leadership and employee OCB. In addition, procedural fairness moderates the relationship between employee positive affect and OCB. That is, under a high level of procedural fairness, positive affect has a stronger influence on OCB. Finally, procedural fairness moderates the indirect effect of paradoxical leadership on employee OCB through positive affect. That is, paradoxical leadership has a greater effect on employee OCB through positive affect for employees with high levels of procedural fairness than for employees with low levels of procedural fairness. These findings provide insights with theoretical and practical implications.
Theoretical implications
First, this study enriches the antecedent variables of OCB and the consequences of paradoxical leadership on employees' behaviours. The existing literature on the relationship between leadership styles and employee OCB mostly focuses on the conventional leadership contingency perspective (i.e., Ko, Bashir, & Khan, 2017; Newman et al., Reference Newman, Schwarz, Cooper and Sendjaya2017; Nohe & Hertel, Reference Nohe and Hertel2017) with the ‘either…or…’ approach, which lacks the ‘both…and…’ strategy to handle organizational paradoxes (Zhang et al., Reference Zhang, Waldman, Han and Li2015). Research on the relationship between paradoxical leadership and employee OCB not only expands the antecedents of OCB but also provides more empirical evidence for researching the effectiveness of paradoxical leadership, especially in China.
Second, existing studies mainly focus on the rational perspective to explore the influence mechanism of paradoxical leadership on employees' attitudes and behaviours (i.e., Franken, Plimmer, & Malinen, Reference Franken, Plimmer and Malinen2020; Ishaq, Bashir, & Khan, Reference Ishaq, Bashir and Khan2019; Shao, Nijstad, & Täuber, Reference Shao, Nijstad and Täuber2019; Xue et al., Reference Xue, Li, Liang and Li2020; Yang et al., Reference Yang, Li, Liang and Zhang2021). Nevertheless, we still have a limited understanding of the role of affect in leadership processes (Damen, Knippenberg, & Knippenberg, Reference Damen, Knippenberg and Knippenberg2008). Positive affect is important in AET because it is expected to influence important work attitudes and behaviours (Weiss & Cropanzano, Reference Weiss and Cropanzano1996). In addition, AET proposes that individual affect levels are influenced by dispositions and a variety of environmental factors (Weiss & Cropanzano, Reference Weiss and Cropanzano1996). The present study tests a mediation model wherein affective events such as leaders' paradoxical behaviour could lead to employees' emotional reactions and eventually to important work outcomes (i.e., OCB). Therefore, we extended the understanding of the paradoxical leadership literature from an AET-based lens.
Lastly, previous studies on paradoxical leader behaviour lack attention to situational factors (Yang et al., Reference Yang, Li, Liang and Zhang2021). This study explores the moderating role of procedural fairness on the relationship between positive affect and OCB, and we highlight how positive affect will facilitate employees to perform OCB under the boundary condition of procedural fairness. We suggest that procedural fairness is an important contextual variable that has meaningful influences on individual behaviours. In other words, our findings suggest that higher levels of OCB are likely to be enhanced when positive affect is accompanied by reinforcement of fair procedures. In addition, for employees, procedural fairness may affect the effectiveness of leadership behaviours based on a specific sense of procedural fairness to interpret leadership behaviours that occur in this situation and respond accordingly. Therefore, we enriched the knowledge about the impact of situational factors on the effectiveness of paradoxical leadership.
Practical implications
First, the study's findings suggest that paradoxical leadership seems potentially instrumental in developing employees' positive affect and OCB at work. Hence, it is worthwhile for organizations to encourage mid-level managers or supervisors to develop paradoxical leadership. Specifically, organizations should coach mid-level managers or supervisors to accept paradoxical goals and behave paradoxically. Mid-level managers or supervisors need to learn about this type of leadership and have the opportunity to apply it to employees within an organization. In addition, mid-level managers or supervisors need to change the previous ‘either…or…’ strategy to the ‘both…and…’ strategy, cultivating their contradictory and integrated thinking to effectively cope with the integration of the organizational paradox with a more open and inclusive attitude.
Second, this study found that positive affect fully mediated the relationship between paradoxical leadership and employee OCB. Organizations might consider increases in positive affect to be valuable and pay close attention to the treatment of their employees. Mid-level managers or supervisors should foster employees' feelings such as enthusiasm, joy and inspiration through management interventions by job design, team building, leadership training and work climate creation. For instance, mid-level managers or supervisors could adopt open communication with employees, establish a pleasant spirit and harmonious interpersonal relationships, and try to reduce their dissatisfaction, misunderstandings and worries at work.
Finally, this finding has important implications for developing procedural fairness. Specifically, employees use a ‘filter’ of the fairness and correctness of organizational decision-making procedures to examine the effectiveness of leadership behaviour and react accordingly. Therefore, mid-level managers or supervisors need to pay attention to the fairness and legitimacy of the organizational decision-making process, develop an open, transparent, rule-based distribution system to create a fair competition environment, and establish a monitoring mechanism for the effective implementation of the distribution system.
Limitations and future research directions
The current study is subject to some limitations, but the limitations may also offer direction for future research. First, although data were collected from the respondents in two different phases and statistical tools were used to check for common method bias, the possibility of such bias cannot be ruled out. Future studies could employ a multisource design to obtain more reliable results. For example, studies would benefit from the use of a more objective measure of paradoxical leadership and OCB (e.g., supervisors' and co-workers' ratings).
Second, the time-lagged study design does not permit inferences about the causal linkages among the variables. Future researchers are encouraged to employ longitudinal designs to examine dynamic reciprocal influences between the variables in this study's model. Specifically, it would be interesting to investigate whether employees' positive reactions towards their leader or engaging in OCB encourages the leader to display more paradoxical leadership behaviours over time.
Third, this study collected data from small- and medium-sized companies in China, and the final sample in the multi-wave design was relatively small (N = 201). Future research will encourage the expansion of sample sources and verify whether the paradoxical leadership–employee OCB relationship is equally applicable in cross-cultural contexts.
Finally, this study examined only the effects of paradoxical leadership on employee OCBs at the individual level. However, evidence suggests that paradoxical leadership can also be aggregated at the team level (Li, She, & Yang, Reference Li, She and Yang2018). Since the cross-level effect may exist simultaneously, different levels may influence each other. Thus, multilevel analyses should be used in future studies.
Financial support
This research was financially supported by the National Natural Science Foundation of China (71902068); the Fundamental Research Funds for the Central Universities (CCNU20QN037; CCNU19TD009).
Conflict of interest
The authors declare that they have no conflict of interest. The authors have full control of all primary data and agree to allow the journal to review their data if requested.
Ethical standards
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Informed consent
Informed consent was obtained from all individual participants included in the study.
