This is the authors’ final peered reviewed (post print) version of the item published as: Newman, Alexander, Allen, Belinda and Miao, Qing 2015, I can see clearly now the moderating effects of role clarity on subordinate responses to ethical leadership, Personnel Review, vol. 44, no. 4, pp. 611-628.

Available from Deakin Research Online: http://hdl.handle.net/10536/DRO/DU:30079953 Reproduced with the kind permission of the copyright owner. Copyright : 2015, Emerald Insight

I can see clearly now The moderating effects of role clarity on subordinate responses to ethical leadership

Alexander Newman and Belinda Allen

Department of Management, Monash University, Caulfield East, Australia, and Qing Miao

College of Public Administration, Zhejiang University, Hangzhou, China

Abstract Purpose – Although there is growing research on the relationship between ethical leadership and subordinate work behaviors, limited research has examined the boundary conditions under which ethical leadership is more or less effective. The purpose of this paper is to investigate whether subordinate perceptions of role clarity in their job role influence the relationship between ethical leadership and subordinate work behaviors. Drawing on both social exchange and social learning theories, the authors predict that in contexts where subordinates perceive low levels of role clarity, the relationship between ethical leadership behavior and subordinate helping and deviant behaviors will be weaker. Design/methodology/approach – In total, 239 employees in the Chinese public sector completed surveys across three separate time points. Confirmatory factor analysis and hierarchical regression analysis were used to analyze the data. Findings – Analyses provided support for the hypothesized relationships. When subordinates perceived higher levels of role clarity the positive relationship between ethical leadership and helping behavior was stronger, and the negative relationship between ethical leadership and deviant behavior was stronger. Research limitations/implications – As with all research the findings of this study need to be viewed in light of its limitations. First, the use of data from a single set of respondents opens up the possibility of common method bias. Second, given the study used of a sample of public sector employees from one part of China, there would be value in future research examining whether the findings from the present study are generalizable to other industrial and cultural contexts. Practical implications – This research has a number of practical implications. Given that the authors found a significant positive relationship between ethical leadership and helping behavior, and a significant negative relationship between ethical leadership and deviant behavior, it is crucial for organizations to include ethical training as an essential part of leadership development programs. However, the findings also suggest at the same time as facilitating the development of ethical leadership behaviors amongst supervisory employees, it is important for organizations to also provide employees with clarity over what is expected of them in their jobs, and the means they should employ to facilitate goal achievement. Originality/value – This study responds to recent calls for more research to identify factors which may strengthen or mitigate the influence of ethical leadership in the workplace. Keywords Deviant behavior, Ethical leadership, Quantitative, Role clarity, Helping behavior Paper type Research paper

1

Introduction Over the last ten years, increasing research has examined the role played by leaders in influencing the ethical conduct of subordinates, and facilitating behaviors that contribute to the proper functioning of the organization. Ethical leadership has been shown to have a positive influence on the extra-role behaviors of subordinates such as helping behavior (Avey et al., 2011; Kacmar et al., 2011; Kalshoven et al., 2011; Mayer et al., 2009; Ruiz-Palomino et al., 2011), and a mitigating influence on their deviant or unethical behavior (Avey et al., 2011; Mayer et al., 2009, 2010; Miao et al., 2013; Stouten et al., 2010). In addition, growing research examines the underlying mechanisms through which ethical leadership transmits its effects on desirable and undesirable workplace behaviors (Avey et al., 2012; Mayer et al., 2010; Newman et al., 2014; Piccolo et al., 2010; Walumbwa and Schaubroeck, 2009; Walumbwa et al., 2011). Despite this, comparatively little research has examined the boundary conditions under which ethical leadership is more or less effective (Avey et al., 2011). Although researchers have begun to investigate the moderating effects of individual differences between subordinates on outcomes of ethical leadership, (Avey et al., 2011; Kalshoven et al., 2013), and subordinate perceptions of organizational or team-level factors (Kacmar et al., 2011; Kalshoven and Boon, 2012), we have limited knowledge as to how subordinate perceptions of their job role may influence the relationship between ethical leadership and subordinate work outcomes. In order to address this gap in the literature, the present study examines whether subordinate perceptions of role clarity m oderate the influence of their supervisor’s ethical leadership behavior on subordinates’ helping and deviant behaviors. Role clarity refers to the extent to which role expectations are clear and fully understood by the employee in their job (Rizzo et al., 1970). In the present study we argue that for subordinates who perceive high levels of role clarity in their job, the exhibition of ethical leadership by the supervisor will reduce the likelihood that they engage in deviant behavior, and increase the likelihood that they will engage in helping behavior. We use social exchange theory (Blau, 1964), social learning theory (Bandura, 1977, 1986), and conservation of resources (COR) theory (Hobfoll, 1989) to support the proposed relationships. More specifically, based on social exchange theory, we argue that high levels of role clarity will engender a greater sense of felt obligation by subordinates to reciprocate the positive context created by their supervisor, and to pay this back by engaging in discretionary helping behavior. In contrast, where there are low levels of role clarity, subordinates will be less likely to reciprocate ethical leadership in the form of helping behavior. In addition, consistent with social learning theory and COR theory, we argue that in situations where there is low role clarity, ethical leadership of the supervisor will be less salient, with subordinates’ limited energy focused on trying to understand the key tasks and responsibilities of their job rather than listen to and act upon their supervisor’s guidance. As a consequence, we assert the capacity of ethical leadership to reduce deviant behaviors and promote helping behaviors will be more effective when there are higher levels of role clarity. In examining these issues we contribute to the literature by responding to recent calls for more research to identify factors which may strengthen or mitigate the influence of ethical leadership in the workplace (Avey et al., 2011). By simply focussing on the direct effects of ethical leadership, and ignoring the context in which leadership behavior is enacted, previous research may have led to incomplete or incorrect conclusions. On a more practical

2

note, our findings should also enable managers to tailor jobs in order to maximize the utility of ethical leadership in their organizations. The remainder of this paper is structured as follows. First, the existing literature on ethical leadership, with a particular emphasis on research in relation to its effects on subordinate behavior, will be reviewed. Second, drawing on social exchange theory and social learning theory the potential moderating influence of role clarity on the relationship between ethical leadership and both subordinate helping and deviant behaviors will then be explored and hypothesized. Following this the method employed by the study will be detailed. The data analysis techniques employed to test the proposed relationships will then be described, with the results of analysis presented. The paper will then conclude with a discussion of the contributions of the study’s findings for both theory and practice and consideration also given to the study’s limitations and areas for future research. Literature review Ethical leadership Ethical leadership has been defined as a style of leadership in which the leader exhibits normatively appropriate conduct, and stresses the importance of such conduct to their subordinates through two-way communication, reinforcement, and decision making (Brown et al., 2005). This definition encompasses both the personal traits and ethical behavior of the leader, and the behavior of the leader that promotes follower ethical behavior (Brown and Trevino, 2006). According to Brown et al. (2005) ethical leaders are characterized by four main features. First, through exercising self-discipline and responsibility, ethical leaders act as ethical role models to their followers. Second, ethical leaders make it clear to followers what they consider to be ethical, and seek their feedback on ethical issues. Third, ethical leaders establish clear ethical standards, and make followers follow these standards through the adoption of appropriate rewards and punishment. Finally, ethical leaders make decisions with relation to ethical principles and ensure followers observe the process of ethical decision-making. In order to provide an empirical basis with which to examine the influence of ethical leadership on follower work outcomes, Brown et al. (2005) developed the Ethical Leadership Scale (ELS). Based on this scale, ethical leadership has been shown to be distinguishable from similar constructs such as interactional justice, leader-member exchange, and the idealized influence dimension of transformational leadership (Brown et al., 2005; Kalshoven et al., 2011; Mayer et al., 2012; Toor and Ofori, 2009). Ethical leadership and subordinate behavior Over the course of the last decade, a number of research studies have examined the relationship between the ethical leadership behavior of the immediate supervisor and subordinate work outcomes (Brown and Mitchell, 2011; Hunter, 2012). Prior research has established that ethical leadership is positively related to subordinate in-role performance (Piccolo et al., 2010; Walumbwa et al., 2011), and extra-role behaviors that contribute to organizational effectiveness (Avey et al., 2011; Kacmar et al., 2011; Piccolo et al., 2010; Ruiz- Palomino et al., 2011). In addition, there is growing empirical evidence which indicates that ethical leadership is effective in reducing subordinate misconduct and deviant behavior in the workplace (Avey et al., 2011; Mayer et al., 2009, 2010; Stouten et al., 2010). In the present study we examine whether ethical leadership is positively related to one measure of subordinates’ extra-role behavior, namely their helping behavior, and negatively

3

related to their deviant behavior. Helping behavior is the positive, discretionary behavior employees engage in, which is focussed on assisting co-workers with work-related tasks or problems or helping to ensure problems do not occur (Organ, 1998). Helping behaviors are distinguishable from other citizenship behaviors such as conscientiousness, civic virtue, and sportsmanship since they are directed towards co-workers and not the organization directly (Podsakoff et al., 2000). Importantly, helping behavior does not tend to be recognized by job descriptions or formal reward systems (Organ, 1988). In contrast, deviant behaviors are voluntary counterproductive behaviors which violate organizational norms and potentially harm the organization (Bennett and Robinson, 2000). Two main theoretical explanations have been proposed to explain why ethical leadership facilitates subordinates to engage in extra-role behavior and reduces their propensity to exhibit deviant behavior. The first, social exchange theory (Blau, 1964), explains why subordinates working under ethical leaders may exhibit higher levels of desired extra-role behavior such as helping behavior. In addition to being viewed as moral persons who can be trusted, ethical leaders are seen as principled individuals who make decisions fairly and take care of their subordinates (Brown and Trevino, 2006; Brown et al., 2005). When subordinates perceive that their leaders are taking their interests into account and treating them well and fairly, they will feel a sense of obligation to respond positively by returning the favorable treatment, and exert effort on behalf of their leader (Trevino et al., 2006). This is a key principle of social exchange theory known as the “norm of reciprocity” (Blau, 1964). In addition, through encouraging subordinates’ opinions and soliciting their ideas, ethical leaders are able to develop meaningful interpersonal relationships with their subordinates. This is further likely to strengthen the emotional ties between supervisor and subordinate, and encourage the subordinate to reciprocate in the form of desired behaviors such as helping behavior (Walumbwa and Schaubroeck, 2009; Walumbwa et al., 2011). The second theoretical explanation, social learning theory (Bandura, 1977, 1986), has typically been forwarded to explain why ethical leadership may reduce subordinates propensity to engage in unethical or deviant behaviors. Social learning theory proposes that individuals learn from observing the behavior of others which they then seek to emulate (Bandura, 1977). In the organizational context, leaders act as a major source of information as to what constitutes appropriate behavior given their organizational status and power over subordinates (Bandura, 1986). In addition to setting down what behaviors are expected, rewarded, and punished in the organization, ethical leaders are likely to be seen as legitimate role models due to high levels of trustworthiness and the credibility of their ethical conduct (Walumbwa and Schaubroeck, 2009). As a result subordinates working under an ethical leader will be less likely to exhibit deviant behaviors which go against the interests of the organization. This leads us to the following hypotheses:

H1. Ethical leadership will be positively related to subordinates’ helping behavior. H2. Ethical leadership will be negatively related to subordinates’ deviant behavior.

Moderating influence of role clarity The work context is known to press upon and shape workplace behavior, and to this end, leadership behavior does not occur in a vacuum, but is interpreted by employees in light of their broader work context. Indeed, Johns (2006) has argued that studying the interaction effects of context is an important but neglected area of research, as context is often a key impinging force on organizational phenomena.

4

One way in which work contexts can vary is in the extent to which jobs are clearly defined and employees are provided with adequate information to perform their roles effectively (Rizzo et al., 1970). Role clarity and role ambiguity are used interchangeably in the literature and are thought to represent opposite ends of a continuum. In situations of high role clarity or low role ambiguity, employees understand what is expected of them in their job, and have knowledge on the available means to carry out their job tasks. In contrast, in situations of low role clarity or high role ambiguity, employees lack an understanding of what is expected of them in their job and the processes they should employ for goal attainment. Ambiguous contexts such as this, where employees have limited understanding in relation to core aspects of their job, have been shown to limit the capacities of employees to match appropriate behaviors with task specific role requirements resulting in lower levels of performance (Tubre and Collins, 2000). A lack of role clarity has also been shown to represent a situational stressor which can result in employees experiencing stress, tension, and anxiety as they struggle to understand the most effective and desired behaviors to engage in (Gilboa et al., 2008; Jackson and Schuler, 1985; Jex et al., 2003). Consistent with the COR stress framework (Hobfoll, 1989), we would therefore argue that a lack of role clarity is likely to act to deplete the resources or energy of subordinates. In order to cope with and offset this loss in resources, subordinates will seek to conserve their remaining resources or energy by reducing discretionary behaviors such as helping behavior. This is consistent with existing research which has demonstrated that lower levels of role clarity are associated with lower levels of helping behaviour (Eatough et al., 2011). Under contexts of low role clarity subordinates’ energy is likely to be focussed on coping with the source of the stress they are experiencing, and taken up with trying to understand their basic role, and its key tasks and responsibilities. In this way contexts which are characterised by low levels of role clarity are likely to ameliorate the positive effect of supervisor ethical leadership on helping behaviour. In contexts of low role clarity ethical leadership displayed by a subordinate’s supervisor may also not be salient. Bandura (1977, 1986) argues that salience of observed behaviors will influence the extent to which an individual performs those same behaviors. In situations where subordinates have low role clarity, leadership behaviors which are more directive, and assist them in clarifying their tasks, are likely to be more salient than behaviors which focus on ethical issues. This line of reasoning is consistent with contingency leadership theories (e.g. see Fiedler, 1967; Hersey and Blanchard, 1969; House, 1971), which have long argued that the effectiveness of different leadership styles would be influenced by, among other things, the structure of tasks and subordinates’ job contexts. Job context characteristics are therefore recognized as potentially important moderators in the relationship between leadership styles and follower behaviors (e.g. see House, 1996). Situations of low role clarity may indicate that there has been insufficient directive, task-oriented leadership to clarify subordinates’ performance goals, the means by which subordinates can effectively carry out tasks, and clarify standards against which subordinates’ performance will be judged. Since it has been demonstrated that leader behavior is seen as effective by subordinates to the extent that it facilitates their goal attainment (House, 1996), in job contexts where there are low levels of role clarity, a more task-oriented leadership style, as opposed to ethical leadership, may be more motivating for subordinates, as this style is more likely to help them resolve role clarity issues. This is consistent with the findings of O’Driscoll and Beehr (1994) who found that in situations of high uncertainty, subordinates were predominantly looking for their supervisor to initiate

5

structure, set goals, assist with problem solving, provide\social and material support, and give feedback on job performance, with these behaviors being associated with employees experiencing less strain and having more positive attitudes. We argue therefore that a lack of role clarity constrains the capacity for ethical leadership to be effective and consequently acts as a moderator of the relationship between ethical leadership and both subordinate helping and deviant behavior. Our next line of reasoning for the proposed moderating role of role clarity results from evidence which suggests that through their interactions with subordinates, supervisors’ behavior is critical for alleviating the unpredictability which subordinates feel about their job tasks, about management, and about the social and political dynamics of the organization (O’Driscoll and Beehr, 1994). Indeed, supervisors, as opposed to organizations more broadly, are likely to be the most important provider of role clarity, as many aspects of an employee’s role (e.g. goals, responsibilities, rules of conduct) are, to a large degree, determined by their supervisor (Chen et al., 2002; Panaccio and Vandenberghe, 2011; Stinglhamber and Vandenberghe, 2004). Supervisors can also play a vital role in interpreting rules and procedures that may have been determined by the organization, and in doing so reduce the levels of role ambiguity experienced by subordinates (O’Driscoll and Beehr, 1994). In situations where an employee perceives low levels of role clarity, their supervisor has potentially failed to perform this important role, and may therefore be indicative of a poor supervisory relationship where the supervisor has provided inadequate feedback (Podsakoff et al., 1996). Indeed, there is evidence to suggest that low role clarity is often interpreted by employees as a signal that their supervisor is either unwilling or unable to provide support (Kahn et al., 1964). Further, research also indicates that in situations of high role clarity, subordinates perceive greater levels of support from their supervisor, with this in turn resulting in subordinates being more conscientious about carrying out their work responsibilities (Eisenberger et al., 1990; Stinglhamber and Vandenberghe, 2004). Additional evidence for this line of reasoning can also be found in the leadership literature more broadly, where subordinates have been found to payback their leaders by engaging in citizenship (i.e. discretionary) behaviors that benefit the leader and others in the work setting, in contexts where they perceive high levels of support from their supervisor (Liden et al., 1997; Settoon et al., 1996). It therefore follows that the positive effects of ethical leadership on discretionary behaviors like helping are likely to be strengthened in contexts where subordinates perceive high levels of role clarity. Conversely, in contexts where low role clarity is perceived, attempts by the supervisor to engage in ethical leadership may be met with skepticism, and so therefore not have as strong an influence on the discretionary behaviors of subordinates. Indeed under these circumstances subordinates may not feel that the ethical leadership behaviors of their supervisors are credible or genuine. This lack of credibility as an ethical role model has the potential to weaken the positive effect of displays of ethical leadership by supervisors on helping behaviour and also open up the possibility that employees will display more deviant behaviors. In these contexts subordinates may feel that since they have not received the positive benefits of role clarity they are not compelled to reciprocate their leader’s ethical behavior with desired behaviors such as helping behavior. This line of reasoning is consistent with the key tenets of social exchange theory (Blau, 1964), which as indicated previously, has at its core the notion of reciprocation, or the payback of positive behaviors, where subordinates perceive there to be a positive high-quality relationship with their supervisor. To this end,

6

low levels of role clarity might be seen by employees as indicative of the low quality exchange they have with their supervisor resulting in them being less motivated, despite the ethical leadership of their supervisor, to put collective interests over their own and engage in prosocial as opposed to deviant behaviors. There is evidence to support this with research findings suggesting that one reason employees engage in OCBs is out of gratitude or a desire to reciprocate positive feelings of satisfaction (MacKenzie et al., 1998). Given also that there is substantial evidence linking a lack of role clarity to job dissatisfaction and more negative emotions (e.g. see Abramis, 1994; De Ruyter et al., 2001 Jackson and Schuler, 1985; Kahn et al., 1964; Kelloway and Barling, 1990; Quah and Campbell, 1994; Von Emster and Harrison, 1998) it is unlikely that in contexts of low role clarity subordinates would have a positive mindset. Evidence also indicates that situations of low role clarity tend to be viewed as hindering employees’ abilities to attain personal and professional goals leading to less positive work-related emotions and attitudes which are known predictors of deviant behaviour (Fox et al., 2001; Judge et al., 2006; Spector and Fox, 2002). Additionally, research evidence indicates that in contexts where subordinates feel they lack support from their supervisor and there message is not seen as genuine or credible, subordinates are more likely to display deviant behaviors (Dalal, 2005; Dineen et al., 2006). Overall therefore in situations of low role clarity, the negative attributions made by subordinates in relation to the quality of the leadership provided by their supervisor, together with the stress and frustration ambiguous roles are known to cause, make displays of positive discretionary behaviors unlikely. Simultaneously, contexts where there is a lack of role clarity would also seem to, increase the likelihood of deviant behaviors, with subordinates more likely to engage in such behaviors as a form of retribution for the lack of role clarity and support provided by their supervisor. We therefore predict that the capacity of ethical leadership to enhance helping behavior and diminish deviant behavior will be weaker in situations of low role clarity, and enhanced in situations of high role clarity. Consistent with this the following two hypotheses are proposed:

H3. The relationship between ethical leadership and helping behavior will be moderated by role clarity such that the positive relationship between ethical leadership and helping behavior will be stronger when role clarity is higher. H4. The relationship between ethical leadership and deviant behavior will be moderated by role clarity such that the negative relationship between ethical leadership and deviant behavior will be stronger when role clarity is higher.

Method Participants and procedure The study participants were all full-time government employees from Zhejiang Province in China. They were solicited to participate in the study by e-mail, using contact details obtained from a Master of Public Administration alumni database from the College of Public Administration, Zhejiang University, China. A total of 1,000 alumni, chosen randomly from the database, were invited to participate and guaranteed confidentiality. Those that agreed to participate were provided links to three waves of surveys in two week intervals in the period from April to May 2011. This was done to limit the possibility of common method variance and reduce respondent fatigue. In the first wave employees were asked to provide demographic data and rate the ethical leadership of their supervisor. In the second wave, employee data on role clarity was obtained. In the third and final wave, employees rated

7

the extent to which they engaged in deviant or helping behavior. Prior to survey distribution the back translation procedure (Brislin, 1993) was used to translate the questionnaire into Chinese. The back translation was similar to the original version of the questionnaire, so highlighted no problems for concern. Out of the 362 participants who agreed to participate in the study, 239 completed all three waves of surveys, resulting in a response rate of around 24 percent. We compared the demographic characteristics of those who had responded fully, those who had responded partially and non-respondents in order to establish the existence of non-response bias. As no differences between the three groups were identified, non-response bias did not seem to be present in our study. Male employees accounted for 63 percent of participants. In total, 93 percent of participants were younger than 40 years of age and around 59 percent were in a managerial position. In order to ascertain whether our sample was representative of civil servants in the Chinese public sector we compared the age and gender demographics of our with that of the general population of civil servants in Zhejiang province, and found no significant differences. Measures Independent variable: ethical leadership. Brown et al.’s (2005) ten-item ELS was used to measure ethical leadership. Subordinates rated the ethical leadership behavior of their supervisor using a five-point Likert scale (1 = strongly disagree, 5 = strongly agree) with higher scores indicating greater ethical leadership behavior. This scale has been widely validated in a wide variety of industrial and cultural settings both in China and overseas (Avey et al., 2011; Kalshoven et al., 2011; Miao et al., 2013; Piccolo et al., 2010). Given that the data from the present study was not nested in teams, ethical leadership was measured at the individual not team-level. Sample items included “My supervisor disciplines employees who violate ethical standards” and “My supervisor sets an example of how to do things the right way in terms of ethics”. The Cronbach’s α for this scale was 0.87. Moderating variable: role clarity. Role clarity was assessed using a five-item scale taken from Rizzo et al. (1970). This scale has been widely validated in both Chinese and non-Chinese settings (Malhotra et al., 2007; Newman and Sheikh, 2012). Participants responded using a five-point Likert scale (1 = strongly disagree, 5 = strongly agree) and sample items included “I know exactly what is expected of me in my job”. The Cronbach’s α for this scale was 0.87. Dependent variables: helping behavior and deviant behavior. Helping behavior was measured using a seven-item scale taken from Van Dyne and Le Pine (1998). This scale has been widely validated in both Chinese and non-Chinese settings (Bao and Wang, 2011; Ng and Van Dyne, 2005). Employees rated their own helping behavior using a five-point Likert scale (1 = strongly disagree, 5 = strongly agree). Sample items included “I volunteer to do things for my work group” and “I attend functions that help this work group”. The Cronbach’s α for this scale was 0.92. Bennett and Robinson’s (2000) 12-item scale was used to measure deviant behavior. This scale has been widely validated in both Chinese and non-Chinese settings (Chiu and Peng, 2008). Employees rated their own deviant behavior using a five-point Likert scale (1 = never, 5= daily). Sample items included “I come in late to work without permission” And “I put little effort into my work”. The Cronbach’s α for this scale was 0.88.

8

Control variables. A number of demographic characteristics of participants were controlled for in the analyses. Both age and tenure with supervisor were measured as categorical variables. The following categories were utilized to measure age: 1 = “26-30”, 2 = “31-35”, 3 = “36-40”, 4 = “41-45”, 5 = “46-50”, and 6 = “51-55”. Similarly, the following categories were utilized to measure tenure with supervisor: 1 = “less than 2 years”, 2= “2-5years”, 3 = “6-9 years”, 4 = “10-13 years”, 5 = “14-16 years”, and 6 = “17-19 years”. A dichotomous variable was used to measure gender (0 = female, 1=ale). Position in organization was coded using a categorical variable of 1 through 4 based on the hierarchy of ranks in Chinese government departments (1= non-managerial employees, 2= section managers, 3= department managers, and 4= senior managers). The means, standard deviations, reliability coefficients, and correlations between all of the main study variables are presented in Table I. Data analysis and results Data analysis was conducted in two steps. In the first step confirmatory factor analysis was undertaken using LISREL 8.80 to examine the discriminant validity of the study’s variables. The goodness-of-fit statistics of a four-factor model where items were forced to load on their respective constructs of ethical leadership, helping behavior, deviant behavior, and role clarity were compared to a series of nested models. The results of the analyses are presented in Table II. As can be seen, the goodness-of-fit indices indicate that the proposed four-factor model fitted the data extremely well (χ2= 857.29; df= 521, RMSEA= 0.05, IFI= 0.98, CFI= 0.98), whereas alternative nested models showed poorer fit. Based on the fit index recommendations of Hu and Bentler (1999), it was concluded that the measurement model was good enough to proceed with hypothesis testing. To test for common method bias, we also performed a Harman’s one-factor test. A χ2 difference test indicated that the one-factor model was significantly poorer than the five-factor model (χ2 difference (df= 7) = 3867.85, p<0.01), suggesting that common method bias is not a significant problem in this study. Table 1. Descriptive statistics and correlations amongst main study variables.

Second, the study’s hypotheses were tested using hierarchical regression analysis. The findings are presented in Table III. In order to deal with potential multicollinearity between study variables all independent, moderating, and control variables were mean- centered prior to analysis, following the recommendations of Aiken and West (1991). Initially we investigated the relationship between ethical leadership and helping behavior, and the moderating influence of role clarity on this relationship. In Model 1 the control variables were entered into the first step of the regression. Two control variables, managerial position (β= 0.18, p<0.05) and gender (β= −0.14, p<0.05) were significantly but weakly related to helping behavior. This suggests that managerial level and female employees are more likely to engage in helping behavior. The independent variable, ethical leadership, and the moderating variable, role clarity, were then entered into the second step of the regression (Model 2). Ethical leadership was positively related to helping behavior (β=0.31, p<0.01) in line with H1. In addition, role clarity was positively related to

9

helping behavior (β=0.18, p<0.01). Finally, in the third step the interaction term, ethical leadership × role clarity, was entered into the regression (Model 3). This interaction was significantly and positively related to helping behavior (β=0.14, p<0.05), indicating that the positive relationship between ethical leadership and helping behavior was stronger when subordinates perceived role clarity to be higher. H3 was therefore supported. Table II. Results of confirmatory factor analysis

Table III. Results of hierarchical regression analyses

Following this we examined the relationship between ethical leadership and deviant behavior, and the moderating influence of role clarity on this relationship. In Model 4 the control variables were entered into the first step of the regression. None of the control variables were significantly related to deviant behavior. The independent variable, ethical leadership, and the moderating variable, role clarity, were then entered into the second step of the regression (Model 5). Ethical leadership was significantly negatively related to deviant behavior (β= −0.16, p<0.01) in line with H2. In addition, role clarity was negatively related to deviant behavior (β= −0.24, p<0.01). Finally, in the third step the interaction term, ethical leadership x role clarity, was entered into the regression (Model 6). This interaction

10

was significantly negatively related to deviant behavior (β= −0.14, p<0.05), indicating that the negative relationship between ethical leadership and deviant behavior was stronger when subordinates perceived role clarity to be higher. Support for H4 was therefore also found. Figure 1. Interaction between ethical leadership and role clarity on helping behavior

Figure 2. Interaction between ethical leadership and role clarity on deviant behavior

Following the recommendations of Aiken and West (1991) regression equations were plotted at different levels of role clarity (i.e. one standard deviation above and below the mean) to better interpret the moderating effects of role clarity on the relationships between ethical leadership and both helping behavior and deviant behavior. These are presented graphically in Figures 1 and 2. Figure 1 shows the moderating influence of role clarity on the relationship between ethical leadership and helping behavior. As can be seen, when role clarity was high the relationship between ethical leadership and helping behavior was stronger than when role clarity was low. Figure 2 shows the moderating influence of role clarity on the relationship between ethical leadership and deviant behavior. As can be seen, when role clarity was high the attenuating influence of ethical leadership on deviant behavior was stronger than when role clarity was low. In addition to plotting the significant interactions, simple slope analyses were undertaken to provide further evidence of the significant moderation effects. The findings showed that the relationship between ethical leadership and helping behavior under conditions of high role clarity was significantly different from zero (simple slope = 0.29, t= 5.08, p<0.01). In contrast, under conditions

11

of low role clarity the relationship between ethical leadership and helping behavior was not significant (simple slope = 0.11, t= 1.83, ns). A second simple slope analysis also showed that under conditions of high role clarity the relationship between ethical leadership and deviant behavior was significantly different from zero (simple slope = −0.20, t= −3.45, p<0.01). In contrast, under conditions of low role clarity the relationship between ethical leadership and deviant behavior was not significant (simple slope = −0.02, t= −0.31, ns). Next, these findings will be discussed in light of the existing literature and research on ethical leadership. Discussion As predicted, ethical leadership was associated with higher levels of self-reported helping behavior and lower levels of self-reported deviant behavior by subordinates. Importantly, the findings also indicated that the relationship between ethical leadership and both helping and deviant behavior was moderated by role clarity. Overall this study makes an important theoretical contribution to the extant literature by answering the calls of researchers to identify boundary conditions under which ethical leadership is more or less effective (Avey et al., 2011). More specifically, the findings of this research indicate that jobs need to be defined clearly by organizations and supervisors to maximize role clarity, in order for the positive effects of ethical leadership on subordinate behavior to be realized. These findings are in line with the tenets of social exchange (Blau, 1964), COR (Hobfoll, 1989), and social learning theories (Bandura, 1977, 1986). First, in line with social exchange theory (Blau, 1964), our findings suggest that high levels of role clarity lead subordinates to reciprocate the positive context created by ethical leaders in the form of discretionary helping behavior. In contrast, by not providing them with clear direction as to their role, the supervisor may lead subordinates to feel that they are receiving sub-standard supervision. In such a situation the subordinate is likely to perceive the ethical leadership of their supervisor with skepticism, and less willing to reciprocate in the form of helping behavior. Our findings are supportive of previous work which suggests that employees may blame their supervisors for a lack of role clarity, given that supervisors are typically viewed as agents of the organization whose role it is to interpret rules and procedures that have been determined by the organization (O’ Driscoll and Beehr, 1994). In situations where an employee perceives low levels of role clarity, their supervisor has potentially failed to perform this important role, and may therefore be indicative of a poor supervisory relationship where the supervisor has provided inadequate feedback (Podsakoff et al., 1996). Second, in line with social learning theory (Bandura, 1977, 1986) and COR theory (Hobfoll, 1989) our findings indicate a lack of role clarity makes ethical leadership behavior less salient for subordinates, with their focus being on using their limited resources to try to better understand the key tasks and responsibilities associated with their job, rather than listen to and act upon the ethical guidance provided by their supervisor. As a consequence subordinates are less likely to feel obligated to engage in discretionary behaviors such as helping behavior, and at the same time make them more likely to engage in deviant behaviors that do not benefit the organization. In examining the moderating effects of role clarity on the relationship between ethical leadership and work outcomes, our research makes an important contribution to the job characteristics literature. By showing that followers are more likely to respond positively to ethical leadership when their jobs roles are clear, it highlights a need for researchers to investigate the importance of job characteristics in determining how followers respond to their leader’s behavior. This is supportive of prior work which highlights job context as a key

12

impinging force on organizational behavior (Johns, 2006). Our findings are also supportive of contingency leadership theories (e.g. see Fiedler, 1967; Hersey and Blanchard, 1969; House, 1971), which highlight the need to consider the structure of tasks and subordinates’ job contexts on the effectiveness of different leadership styles. This research also has a number of practical implications. Given that we found a significant positive relationship between ethical leadership and helping behavior, and a significant negative relationship between ethical leadership and deviant behavior, it would be prudential for organizations to include ethical training as an essential part of leadership development programs. However, our findings also suggest at the same time as facilitating the development of ethical leadership behaviors amongst supervisory employees, it is important for organizations to also provide employees with clarity over what is expected of them in their jobs, and the means they should employ to facilitate goal achievement. As the results of our study indicate, without ensuring that basic task-oriented leadership responsibilities are first taken care of, the demonstration of ethical leadership behaviors executed by supervisors in unlikely to bestow significant positive effects in terms of employee behavior. Supervisors should therefore aim to provide their employees with role clarity, particularly in the absence of organizational mechanisms which do so, in order to enhance the credibility and therefore effectiveness of their ethical leadership. Providing subordinates with informal guidance in relation to what is expected of them in their job role, and how their job-related goals may be achieved, is likely to engender feelings of reciprocation, in turn increasing the likelihood that they will also display helping behavior and reducing the likelihood of deviant behavior. Limitations and suggestions for future research As with all research the findings of this study need to be viewed in light of its limitations. First, the use of data from a single set of respondents opens up the possibility of common method bias (Podsakoff et al., 2012). However, as Podsakoff et al. (2003, pp. 564-565) note, “if a study is designed to test hypotheses about [...] interaction effects, rather than main effects, then method bias would not be able to account for any statistically significant effects observed.” We also followed the recommendations of Podsakoff et al. (2003) for reducing the possibility of common method bias by administering the survey over time, assuring confidentiality of responses and randomly ordering the items within each survey. We also performed a Harman’s one factor test to check that common method bias had not significantly influenced the results of the study. We do, however, acknowledge that in order to definitively rule out the potential problems caused by common method bias, future research should aim to utilize other rated measures of employee behaviors. Second, since the study used a sample of public sector employees from one part of China, there would be value in future research examining whether the findings from the present study are generalizable to other industrial and cultural contexts. More specifically, there would be value in comparative research being conducted in western cultures, which are more individualistic and less relationship oriented than the Chinese culture, to gauge to what extent these findings are culturally influenced. Third, the fact that study participants were recruited from an alumni database brings into question the extent to which their views represent those of others in the organizations participants were employed in. However, given the sensitive setting of our research, the Chinese public sector, and the sensitive nature of the questions related to ethical leadership and deviant behavior, we feel contacting the participants directly allowed us to reduce social desirability bias, given the organizations in which the participants were employed were not involved in the process of data collection. Finally, although the focus of our study was on ethical leadership, we might also expect role clarity to accentuate the positive influence of other leadership styles such as transformational and entrepreneurial leadership (Bass, 1985; Renko et al., 2015) on follower behaviors, given that

13

such leadership styles also involve supervisors’ role modeling expected behavior. Exploring the extent to which aspects of the work context, such as role clarity, also attenuate the positive effects of other leadership styles would be another fruitful area for future research. Conclusion The present study examined role clarity as a moderator of the relationship between ethical leadership and subordinate work behavior, specifically deviant and helping behaviors. We found that when role clarity was higher, the positive relationship between ethical leadership and helping behavior was stronger and the negative relationship between ethical leadership and deviant behavior was stronger. These findings highlight the importance of creating a context in which subordinates are clear as to what is expected of them in their job and the processes they should employ for goal attainment, in order for ethical leadership to enhance constructive and minimize destructive employee behaviors. References Abramis, D.J. (1994), “Work role ambiguity, job satisfaction, and job performance: meta-analysis and review”, Psychological Reports, Vol. 75 No. 3f, pp. 1411-1433. Aiken, L.S. and West, S.G. (1991), Multiple Regression Testing and Interpreting Interactions, Sage, Newbury Park, CA. Avey, J.B., Palanski, M.E. and Walumbwa, F.O. (2011), “When leadership goes unnoticed: the moderating role of follower self-esteem on the relationship between ethical leadership and follower behavior”, Journal of Business Ethics, Vol. 98 No. 4, pp. 573-582. Avey, J.B., Wernsing, T.S. and Palanski, M.E. (2012), “Exploring the process of ethical leadership: the mediating role of employee voice and psychological ownership”, Journal of Business Ethics, Vol. 107 No. 1, pp. 21-34. Bandura, A. (1977), Social Learning Theory, Prentice-Hall, Englewood Cliffs, NJ. Bandura, A. (1986), Social Foundations of Thought and Action: A Social Cognitive Theory, Prentice-Hall, Englewood Cliffs, NJ. Bao, L.L. and Wang, T. (2011), “The impact of employment relationship modes on employee helping behavior in transitional period”, Chinese Journal of Management, Vol. 8 No. 11, pp. 1646-1654. Bass, B.M. (1985), Leadership and Performance Beyond Expectations, The Free Press, New York, NY. Bennett, R.J. and Robinson, S.L. (2000), “Development of a measure of workplace deviance”, Journal of Applied Psychology, Vol. 85 No. 3, pp. 349-360. Blau, P.M. (1964), Exchange and Power in Social Life, Academic Press, New York, NY. Brislin, R. (1993), Understanding Culture’s Influence on Behavior, Harcourt Brace College Publishers, Fort Worth, TX. Brown, M.E. and Mitchell, M.S. (2011), “Ethical and unethical leadership: exploring new avenues for future research”, Business Ethics Quarterly, Vol. 20 No. 4, pp. 583-616. Brown, M.E. and Trevino, L.K. (2006), “Ethical leadership: a review and future directions”, Leadership Quarterly, Vol. 17 No. 16, pp. 595-616. Brown, M.E., Trevino, L.K. and Harrison, D.A. (2005), “Ethical leadership: a social learning perspective for construct development and testing”, Organizational Behavior and Human Decision Processes, Vol. 97 No. 2, pp. 117-134. Chen, Z.X., Tsui, A.S. and Farh, J.L. (2002), “Loyalty to supervisor vs. organizational commitment: relationships to employee performance in China”, Journal of Occupational and

14

Organizational Psychology, Vol. 75 No. 3, pp. 339-356. Chiu, S.F. and Peng, J.C. (2008), “The relationship between psychological contract breach and employee deviance: the moderating role of hostile attributional style”, Journal of Vocational Behavior, Vol. 73 No. 3, pp. 426-433. Dalal, R.S. (2005), “A meta-analysis of the relationship between organizational citizenship behavior and counterproductive work behaviour”, Journal of Applied Psychology, Vol. 90 No. 6, pp. 1241-1255. De Ruyter, K., Wetzels, M. and Feinberg, R. (2001), “Role stress in call centres: its effects on employee performance and satisfaction”, Journal of Interactive Marketing, Vol. 15 No. 2, pp. 23-35. Dineen, B.R., Lewicki, R.J. and Tomlinson, E.C. (2006), “Supervisory guidance and behavioral integrity: relationships with employee citizenship and deviant behaviour”, Journal of Applied Psychology, Vol. 91 No. 3, pp. 622-635. Eatough, E.M., Chang, C.H., Miloslavic, S.A. and Johnson, R.E. (2011), “Relationships of role stressors with organizational citizenship behavior: a meta-analysis”, Journal of Applied Psychology, Vol. 96 No. 3, pp. 619-632. Eisenberger, R., Fasolo, P. and Davis-LaMastro, V. (1990), “Perceived organizational support and employee diligence, commitment, and innovation”, Journal of Applied Psychology, Vol. 75 No. 1, pp. 51-59. Fiedler, F.E. (1967), A Theory of Leadership Effectiveness, McGraw-Hill, New York, NY. Fox, S., Spector, P.E. and Miles, D. (2001), “Counterproductive work behavior (CWB) in response to job stressors and organizational justice: some mediator and moderator tests for autonomy and emotions”, Journal of Vocational Behavior, Vol. 59 No. 3, pp. 291-309. Gilboa, S., Shirom, A., Fried, Y. and Cooper, C. (2008), “A meta-analysis of work demand stressors and job performance: examining main and moderating effects”, Personnel Psychology, Vol. 61 No. 2, pp. 227-271. Hersey, P. and Blanchard, K.H. (1969), Management of Organizational Behavior - Utilizing Human Resources, Prentice Hall, Upper Saddle River, NJ. Hobföll, S. (1989), “Conservation of resources: a new attempt at conceptualizing stress”, Applied Psychology, Vol. 44 No. 3, pp. 513-524. House, R.J. (1971), “A path-goal theory of leader effectiveness”, Administrative Science Quarterly, Vol. 16 No. 3, pp. 321-339. House, R.J. (1996), “Path-goal theory of leadership: lessons, legacy, and a reformulated theory”, Leadership Quarterly, Vol. 7 No. 3, pp. 323-352. Hu, L.T. and Bentler, P.M. (1999), “Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives”, Structural Equation Modeling, Vol. 6 No. 1, pp. 1-55. Hunter, S.T. (2012), “(Un)ethical leadership and identity: what did we learn and where do we go from here?”, Journal of Business Ethics, Vol. 107 No. 1, pp. 79-87. Jackson, S.E. and Schuler, R.S. (1985), “A meta-analysis and conceptual critique of research on role ambiguity and role conflict in work settings”, Organizational Behavior and Human Decision Processes, Vol. 36 No. 1, pp. 16-78. Jex, S.M., Adams, G.A., Bachrach, D.G. and Sorenson, S. (2003), “The impact of situational constraints, role stressors, and commitment on employee altruism”, Journal of Occupational Heath Psychology, Vol. 8 No. 3, pp. 171-180. Johns, G. (2006), “The essential impact of context on organizational behavior”, Academy of Management Review, Vol. 31 No. 2, pp. 386-408.

15

Judge, T.A., Scott, B.A. and Ilies, R. (2006), “Hostility, job attitudes, and workplace deviance: test of a multilevel model”, Journal of Applied Psychology, Vol. 91 No. 1, pp. 126-138. Kacmar, K.M., Bachrach, D.G., Harris, K.J. and Zivnuska, S. (2011), “Fostering good citizenship through ethical leadership: exploring the moderating role of gender and organizational politics”, Journal of Applied Psychology, Vol. 96 No. 3, pp. 633-642. Kahn, R., Wolfe, D., Quinn, R., Snoek, J. and Rosentbal, R. (1964), Organizational Stress: Studies in Role Conflict and Ambiguity, Wiley, New York, NY. Kalshoven, K. and Boon, C.T. (2012), “Ethical leadership, employee well-being, and helping: the moderating role of human resource management”, Journal of Personnel Psychology, Vol. 11 No. 1, pp. 60-68. Kalshoven, K., Den Hartog, D.N. and De Hoogh, A.H.B. (2011), “Ethical leadership at work questionnaire (ELW): development and validation of a multidimensional measure”, Leadership Quarterly, Vol. 22 No. 1, pp. 51-69. Kalshoven, K., Den Hartog, D.N. and De Hoogh, A.H.B. (2013), “Ethical leadership and follower helping and courtesy: moral awareness and empathic concern as moderators”, Applied Psychology: An International Review, Vol. 62 No. 2, pp. 211-235. Kelloway, E.K. and Barling, J. (1990), “Item content vs. item wording: disentangling role conflict and role ambiguity”, Journal of Applied Psychology, Vol. 75 No. 6, pp. 738-742. Liden, R.C., Sparrowe, R.T. and Wayne, S.J. (1997), “Leader-member exchange theory: the past and potential for the future”, Research in Personnel and Human Resources Management, Vol. 15 No. 1, pp. 47-119. MacKenzie, S.B., Podsakoff, P.M. and Ahearne, M. (1998), “Some possible antecedents and consequences of in-role and extra-role salesperson performance”, Journal of Marketing, Vol. 62 No. 3, pp. 87-98. Malhotra, N., Budhwar, P. and Prowse, P. (2007), “Linking rewards to commitment: an empirical investigation of four UK call centres”, International Journal of Human Resource Management, Vol. 18 No. 12, pp. 2095-2128. Mayer, D.M., Aquino, K., Greenbaum, R.L. and Kuenzi, M. (2012), “Who displays ethical leadership and why does it matter: an examination of antecedents and consequences of ethical leadership”, Academy of Management Journal, Vol. 55 No. 1, pp. 151-171. Mayer, D.M., Kuenzi, M. and Greenbaum, R.L. (2010), “Examining the link between ethical leadership and employee misconduct: the mediating role of ethical climate”, Journal of Business Ethics, Vol. 95 No. 1, pp. 7-16.Mayer, D.M., Kuenzi, M., Greenbaum, R., Bardes, M. and Salvador, R. (2009), “How low does ethical leadership flow? Test of a trickle-down model”, Organizational Behavior and Human Decision Processes, Vol. 108 No. 1, pp. 1-13. Miao, Q., Newman, A., Yu, J. and Xu, L. (2013), “The relationship between ethical leadership and unethical pro-organizational behavior: Linear or curvilinear effects?”, Journal of Business Ethics, Vol. 116 No. 3, pp. 641-653. Newman, A. and Sheikh, A.Z. (2012), “Organizational commitment in Chinese small- and medium-sized enterprises: the role of extrinsic, intrinsic and social rewards”, International Journal of Human Resource Management, Vol. 23 No. 2, pp. 349-367. Newman, A., Kiazad, K., Miao, Q. and Cooper, B. (2014), “Examining the cognitive and affective trust-based mechanisms underlying the relationship between ethical leadership and organizational citizenship: a case of head leading the heart?”, Journal of Business Ethics, Vol. 123, pp. 113-123. Ng, K.Y. and Van Dyne, L. (2005), “Antecedents and performance consequences of helping behavior in work groups”, Group and Organization Management, Vol. 30 No. 5, pp. 514-540.

16

O’Driscoll, M.P. and Beehr, T.A. (1994), “Supervisor behaviors, role stressors, and uncertainty as predictors of personal outcomes for subordinates”, Journal of Organizational Behavior, Vol. 15 No. 2, pp. 141-155. Organ, D.W. (1988), Citizenship Behavior: The Good Soldier Syndrome, Lexington Books,Lexington, MA. Panaccio, A. and Vandenberghe, C. (2011), “The relationships of role clarity and organization-based self-esteem to commitment to supervisors and organizations and turnover intentions”, Journal of Applied Social Psychology, Vol. 41 No. 6, pp. 1455-1485. Piccolo, R.F., Greenbaum, R., Den Hartog, D.N. and Folger, R. (2010), “The relationship between ethical leadership and core job characteristics”, Journal of Organizational Behavior, Vol. 31 Nos 2-3, pp. 259-278. Podsakoff, P.M., MacKenzie, S.B. and Bommer, W.H. (1996), “Transformational leadership behaviors and substitutes for leadership as determinants of employee satisfaction, commitment, trust, organizational citizenship behaviors”, Journal of Management, Vol. 22 No. 2, pp. 259-298. Podsakoff, P.M., MacKenzie, S.B. and Podsakoff, N.P. (2012), “Sources of method bias in social science research and recommendations on how to control it”, Annual Review of Psychology, Vol. 65 No. 1, pp. 539-569. Podsakoff, P.M., MacKenzie, S.B., Lee, J.Y. and Podsakoff, N.P. (2003), “Common method biases in behavioral research: a critical review of the literature and recommended remedies”, Journal of Applied Psychology, Vol. 88 No. 5, pp. 879-903. Podsakoff, P.M., MacKenzie, S.B., Paine, J.B. and Bachrach, D.G. (2000), “Organizational citizenship behaviors: a critical review of the theoretical and empirical literature and suggestions for future research”, Journal of Management, Vol. 26 No. 3, pp. 513-563. Quah, J. and Campbell, K.M. (1994), “Role conflict and role ambiguity as factors in work stress among managers in Singapore: some moderator variables”, Research and Practice in Human Resource Management, Vol. 2 No. 1, pp. 21-33. Renko, M., Tarabishy, A.E., Carsrud, A. and Brannback, M. (2015), “Understanding and measuring entrepreneurial leadership style”, Journal of Small Business Management, Vol. 53 No. 1, pp. 54-74. Rizzo, J.R., House, R.J. and Lirtzman, S.I. (1970), “Role conflict and ambiguity in complex organizations”, Administrative Science Quarterly, Vol. 15 No. 2, pp. 150-163. Ruiz-Palomino, P., Ruiz-Amaya, C. and Knorr, H. (2011), “Employee organizational citizenship behavior: the direct and indirect impact of ethical leadership”, Canadian Journal of Administrative Sciences, Vol. 28 No. 3, pp. 244-258. Settoon, R.P., Bennett, N. and Liden, R.C. (1996), “Social exchange in organizations: perceived organizational support, leader-member exchange, employee reciprocity”, Journal of Applied Psychology, Vol. 81 No. 3, pp. 219-227. Spector, P.E. and Fox, S. (2002), “An emotion-centered model of voluntary work behavior: some parallels between counterproductive work behavior and organizational citizenship behaviour”, Human Resource Management Review, Vol. 12 No. 2, pp. 269-292. Stinglhamber, F. and Vandenberghe, C. (2004), “Favorable job conditions and perceived support: the role of organizations and supervisors”, Journal of Applied Social Psychology, Vol. 34 No. 7, pp. 1470-1493. Stouten, J., Baillien, E., Van den Broeck, A., Camps, J., De Witte, H. and Euwema, M. (2010), “Discouraging bullying: the role of ethical leadership and its effects on the work environment”, Journal of Business Ethics, Vol. 95 No. 1, pp. 17-27.

17

Toor, S. and Ofori, G. (2009), “Ethical leadership: examining the relationships with full range leadership model employee outcomes, and organizational culture”, Journal of Business Ethics, Vol. 90 No. 4, pp. 533-547. Trevino, L.K., Weaver, G.R. and Reynolds, S.J. (2006), “Behavioral ethics in organizations: a review”, Journal of Management, Vol. 32 No. 6, pp. 951-990. Tubre, T.C. and Collins, J.M. (2000), “Jackson and Schuler (1985) revisited: a meta-analysis of the relationships between role ambiguity, role conflict, and job performance”, Journal of Management, Vol. 26 No. 1, pp. 155-169. Van Dyne, L. and Le Pine, J. (1998), “Helping and voice extra-role behaviors: evidence of construct and predictive validity”, Academy of Management Journal, Vol. 41 No. 1, pp. 108-119. Von Emster, G.R. and Harrison, A.A. (1998), “Role ambiguity, spheres of control, burnout, and work-related attitudes of teleservice professionals”, Journal of Social Behavior and Personality, Vol. 13 No. 2, pp. 375-385. Walumbwa, F.O. and Schaubroeck, J. (2009), “Leader personality traits and employee voice behavior: mediating roles of ethical leadership and work group psychological safety”, Journal of Applied Psychology, Vol. 94 No. 5, pp. 1275-1286. Walumbwa, F.O., Mayer, D.M., Wang, P., Wang, H., Workman, K. and Christensen, A.L. (2011), “Linking ethical leadership to employee performance: the rules of leader-member exchange, self-efficacy, and organizational identification”, Organizational Behavior and Human Decision Processes, Vol. 115 No. 2, pp. 204-213. Further reading Dierdorff, E.C., Rubin, R.S. and Bachrach, D.G. (2012), “Role expectations as antecedents of citizenship and the moderating effects of work context”, Journal of Management, Vol. 38 No. 2, pp. 573-598. Corresponding author Qing Miao can be contacted at: mqok@163.com

18