For the past decade, employee voice behavior has been studied as an important means by which organizations can gain the advantage of being more innovative and adaptive to today’s dynamic business environment (e.g., Liu, Zhu, & Yang, 2010; Morrison, 2011; Mowbray, Wilkinson, & Tse, 2015). The leader is one of the most significant factors that influence employee voice behavior (Mowbray et al., 2015); thus, it is important to gain a deeper understanding of the relationship between a leader and the voice behavior of his or her employees.

Employee voice behavior refers to employees’ expressions of opinions or concerns about work tasks or issues within an organization. Scholars have reported finding positive organizational outcomes associated with voice behavior, such as innovation capability, group learning, and crisis prevention (e.g., Schwartz & Wald, 2003; Van Dyne, Ang, & Botero, 2003). Scholars have also explored antecedents and contextual factors of voice behavior (e.g., Kish-Gephart, Detert, Treviño, & Edmondson, 2009; Morrison, 2011). However, some scholars consider voice behavior as a double-edged sword in regard to the possibility of benefits and risks for the person who voices his or her ideas, so that voice behavior requires careful consideration before use (Liang, Farh, & Farh, 2012). When voicing constructive ideas, the employee may expect to achieve favorable social outcomes, such as favorable evaluations and promotion opportunities, but doing so could increase the risk of the employee being misunderstood by his or her supervisor, as well as bringing forth other undesirable social consequences, such as abusive behavior by their supervisor.

Leader impression management behavior is a series of behaviors performed by a leader that is intended to help maintain that leader’s self-image. The study of leader impression management behavior can be traced back to the work of Gardner and Avolio (1998) on the self-monitoring behavior of charismatic leaders. These authors proposed that leaders with high self-monitoring abilities are predisposed to constructing a charismatic image and those leaders value their identity image more than do those with less self-monitoring behavior. According to the classification used in the study by Gardner and Avolio (1998), there are five strategies of impression management for a highly self-monitored leader to build and maintain a self-image: these are behaviors of exemplification, ingratiation, self-promotion, intimidation, and supplication. In the present study, we focused on the impression management style of prosocial behavior, that is, exemplification and ingratiation, in comparison with the self-serving behaviors of self-promotion and intimidation. For the two types of prosocial behavior, exemplification refers to “present[ing] oneself as a worthy role model” and ingratiation refers to “mak[ing] oneself more attractive or likable to others” (Sosik, Avolio, & Jung, 2002, p. 220). For the self-serving behavior, self-promotion is defined as the leader presenting himself or herself as “highly competent” at certain skills and intimidation is presenting himself or herself as “a dangerous potent person who is able and willing to challenge others” (Sosik et al., 2002, p. 220).

According to findings reported in previous studies, it is clear that supervisors and their leadership behaviors play an important role in affecting employees’ voice behavior in organizations (Mowbray et al., 2015). In our study, we argued that when employees perceive their supervisor as self-serving-oriented rather than being oriented toward serving others’ interests, this would decrease the chance of voice behavior, because employees’ voicing ideas in this situation would be perceived by the supervisor as being more like challenging the status quo than making a constructive contribution to the leader. Although challenging the status quo in organizations is the very aim and concept of voice behavior, employees must evaluate the risk of being perceived and responded to negatively by their supervisor (Liu et al., 2010).

Scholars have found that employees’ trust in their leader and his or her leadership is closely related to employee voice behaviors and employees’ propensity to engage in such behaviors (Boxall & Purcell, 2011; Gao, Janssen, & Shi, 2011). A well-accepted definition of trust is that of Mayer, Davis, and Schoorman (1995), who describe it as “the willingness of a party to be vulnerable to the actions of another party” (p. 712) when that party cannot be controlled or monitored. This definition of trust is drawn from within the framework of social exchange theory (Mayer et al., 1995; Schoorman, Mayer, & Davis, 2007). According to Mayer et al., the trust construct changes according to the individual and the relationship, as well as the time. In the current study, we adopted this definition and focused on the employee’s trust in the leader, based on a social exchange relationship and considered as the result of social interaction between dyadic individuals. In this context, dyadic trust would be an important factor within the relationship of the supervisor and the employee.

Seigele (2014) described the current era as one of globalization and information explosion. Business relationships are developing in an environment that is more dynamic and with a much quicker rate of change than was the case in earlier times. There is not much time for building mutual trust, or even for filtering and finding important information for trust (Bobko, Barelka, & Hirshfield, 2014). This situation has led to some scholars looking at the creation and growth of suspicion (Lyons, Stokes, Eschleman, Alarcon, & Barelka, 2011; Olson, 2009). However, the literature on suspicion is limited to fields such as marketing, communication, human factors, and psychology, and has not been extended into the fields of organizational behavior and human resource management (Hoffmann, 2007). There are two streams of research on suspicion: suspicion as a temperament (e.g., Buss & Durkee, 1957; Levine & McCornack, 1991) and suspicion as a psychological state (Fein, 1996; Lyons et al., 2011). In our study, we have focused on suspicion as a psychological state. There are several definitions of state suspicion. We judged that the definition of Bobko and colleagues (2014) to be conclusive and accurate. These authors describe state suspicion as “a person’s simultaneous state of cognitive activity, uncertainty, and perceived malintent about underlying information” (p. 503). When an individual is in the psychological state of suspicion, he or she must suspend his or her judgment as to “whether or not the other entity is trustworthy” (Bobko et al., 2014, p. 503). The social exchange-based definition of trust offered by Mayer and colleagues (1995) is like a decision or choice to be vulnerable to others, whereas suspicion is more about a psychological learning process in which the individual goes through several stages. Thus, some scholars suggest that trust and suspicion are different constructs, and that they can coexist during an event or in a dyadic relationship. For example, Kleiman, Sher, Elster, and Mayo (2015) argued that there are many situations in which individuals are neither fully trusting nor fully distrusting and a suspicion mindset considers both scenarios of trust and distrust.

As already outlined, from the social exchange perspective, the assumptions underlying the definition of trust include two contradictory aspects. First, trusting or distrusting other people is considered to be a stable inclination and decision. Second, trust is still assumed to be a situation-specific variable (Schoorman et al., 2007). According to social cognitive theory (Bandura & Walters, 1963), inclination and decision (personal context) and situation-specific variables such as one’s psychological state (social context) should both be taken into consideration when understanding others’ behaviors. Therefore, an employee’s trust in and suspicion about their supervisor’s goodwill or behavior could be critical mediators between leader impression management and employee voice behavior. If the employee perceives the leader as performing prosocial conduct more than other types of behavior, the employee would feel less threatened when giving a different, challenging opinion and, vice versa, if the leader’s behavior is perceived as being more self-serving, then the employee would feel more threatened when voicing a challenging opinion. Other than that, we suspected that an employee’s perception of a leader’s prosocial behavior would be related more closely to employee trust rather than to suspicion, whereas a perception of a leader’s self-serving behavior would be related more closely to employee suspicion than trust. This is because when employees think about the prosocial side of their supervisor’s behavior, their memory of the good things, such as reciprocity or benevolence, would be activated, and reciprocity and benevolence are related to trust. Similarly, when employees think about the self-serving side of their supervisor’s behavior, evaluations that are negative, such as uncertainty and malicious conduct of the supervisor toward employees, would be activated in their cognitive process.

According to the reasoning we have outlined, our aim in the current study was therefore, to expand the literature on employee voice behavior, leader impression management, trust, and suspicion. Thus, we proposed the following hypotheses:
Hypothesis 1a: Leader’s prosocial behavior will be positively related to employees’ voice behavior.
Hypothesis 1b: Leader’s self-serving behavior will be negatively related to employees’ voice behavior.
Hypothesis 2: The relationship between employees’ trust and suspicion will be nonsignificant.
Hypothesis 3a: An employee’s perception of his or her leader’s trustworthiness will mediate the relationship between the leader’s prosocial behavior and the employee’s voice behavior.
Hypothesis 3b: An employee’s perception of his or her leader’s trustworthiness will mediate the relationship between the leader’s self-serving behavior and the employee’s voice behavior.
Hypothesis 4a: An employee’s perception in regard to suspicion about his or her leader will mediate the relationship between the leader’s prosocial behavior and the employee’s voice behavior.
Hypothesis 4b: An employee’s perception in regard to suspicion about his or her leader will mediate the relationship between the leader’s self-serving behavior and the employee’s voice behavior.

Method

Participants and Procedure

We collected the data for this study from supervisors and their subordinates who were employed in 98 work teams by high-tech companies in central and southern China. Participants completed a supervisor–subordinate paired survey in order to reduce systemic errors, such as common method variance. Subordinates rated their perceptions of their supervisor’s impression management, trust-worthiness, and the level of their suspicion about the supervisor. The supervisors separately rated their subordinates’ voice behavior. For each team, the matched survey forms were put into a single envelope. We sent out 584 survey forms and the forms gathered by a business consulting agency, whose staff have a long-term, high-quality relationship with the participants, comprised responses from 108 supervisors and 476 subordinates. After excluding nonrespondent and inappropriate samples, data from 71 supervisor–subordinate dyads were able to be used for analysis. The sample for analysis comprised 241 employees (47.1% women and 52.9% men) and 71 leaders (28.2% women and 71.8% men). In regard to age, 51.4% of the participants were younger than 30 years, and 48.6% of the participants were aged between 30 and 55 years. In regard to academic level, the highest qualification of 24.3% of the participants was a high school education or below, 20.3% had a junior college education, 45.5% had a bachelor’s degree, and 9.9% had a master’s degree or other postgraduate qualification. The average number of years employed by the company was 10.45 years and for tenure in the current position, the average period was 3.25 years.

Of the supervisors, the average age was 37.26 years and in terms of academic level, 19.9% had a master’s degree or other postgraduate qualification, 54.1% had a bachelor’s degree, 14.8% had a junior college education, and 11.2% had a high school education or below. The leaders’ average employment length was 14.84 years and their average tenure was 6.32 years. Of the subordinates, the average age was 32.26 years and in terms of academic level, 19.5% had a master’s degree or other postgraduate qualification, 43.1% had a bachelor’s degree, 21.9% had a junior college education, and 15.5% had a high school education or below. The subordinates’ average employment length was 9.13 years and average tenure was 2.33 years.

Measures

We adopted all the scales we used in this study from existing studies conducted in the Western context. To adapt these to the Chinese context, we invited a PhD candidate and a professor who have published relevant papers in an English language journal to perform a translation and back-translation procedure as is done in most cross-cultural studies (Brislin, 1980).

Perceived prosocial and self-serving behavior. All items were adopted from the Leader Impression Management Questionnaire developed by Gardner and Cleavenger (1998). Items are rated on a 5-point Likert scale as to how often a leader displays each impression management behavior (1 = not at all, 5 = frequently, if not always). Prosocial impression management consists of 15 items (α = .89), of which there are six items measuring exemplification and nine items measuring ingratiation. Sample items are “Leads by example whenever possible” and “Praises others’ ideas of work.” Self-serving impression management consists of 10 items (α = .89), of which five items measure intimidation and five items measure self-promotion. Sample items are “Demands respect from his or her subordinates” and “Readily takes credit for past and current successes.”

Trust and suspicion. We adopted Mayer and Gavin’s (2005) method of measuring general trust. There are 22 items, comprising 17 items to measure trustworthiness and five items to measure trust (α = .89). Sample items are “I am afraid of what my supervisor might do to me at work” and “If I had my way, I wouldn’t let my supervisor have any influence over issues that are important to me.”

We used the suspicion scale developed by Bobko and his colleagues (2014) to measure employee suspicion about the leader. There are 20 items (α = .91), with five items measuring general suspicion, four items measuring cognitive activity, six items measuring perception of malicious intent, and five items measuring uncertainty. Sample items include “I kept thinking that some behaviors were unusual” and “I felt like I was being taken advantage of.” All items are rated on a 5-point Likert scale (1 = disagree strongly, 5 = agree strongly).

Employee voice behavior. We used the six-item scale developed by Van Dyne and LePine (1998) to measure employee voice behavior (α = .87). Items are rated on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree). Sample items are “Develops and makes recommendations concerning issues that affect this workgroup” and “Speaks up and encourages others in this group to get involved in issues that affect the group.”

Control variables. Demographic variables of participants’ age, gender, tenure, years employed by the company, and education level were controlled for according to accepted empirical study methods ( e.g., Detert & Burris, 2007). Leader–member exchange was also controlled for because it has been identified as a vital variable that influences the trust and suspicion levels of employees (Deluga, 1994).

Results

Structural Relationship Analysis of Trust and Suspicion

First, we split the sample into two halves (156 in each) and used these for the factor analysis. We used one half of our sample for the exploratory factor analysis. The result of a Kaiser-Meyer-Olkin test for sampling adequacy was .932, which is higher than the .70 criterion, χ^² = 5929.282 (df = 861) at a significant level (p < .001). Therefore, the sample was suitable for principal component analysis. From this analysis, and using the varimax rotation method, we found that we had extracted eight factors and that the items of trust and suspicion were separated into different factors, which means that trust and suspicion are separate measures. Then we set trust and suspicion as two independent factors for a confirmatory factor analysis in which we calculated chi square (χ^²), degrees of freedom (df), probability (χ^²/df), goodness-of-fit index (GFI), comparative fit index (CFI), Tucker–Lewis index (TLI), incremental fit index (IFI), and root mean square error of approximation (RMSEA). The results were χ^² = 1960.897, df = 737, χ^²/df = 2.661 (p < .001), GFI = .86, CFI = .83, TLI = .88, IFI = .91, RMSEA = 0.083, which indicates a well-fitting model. Combining the two-factor analysis we were able to conclude that trust and suspicion are two independent constructs. Thus, Hypothesis 2 was supported.

Multiple Mediation Analysis

We adopted a widely used bootstrapping method as set out by Preacher and Hayes (2008). By using this bootstrapping method, the researcher can effectively avoid Type 1 error. We adopted the macro code provided by Preacher and Hayes to process the data in SPSS 20.0. The descriptive statistics and intercorrelations of the variables are shown in Table 1. Following our hypotheses, we set leader prosocial behavior as the independent variable in our first model, with trust and suspicion as mediators, and employee voice behavior as the dependent variable. The results are shown in Table 2. In our second model, we changed the independent variable from prosocial behavior to self-serving behavior and the other variables remained the same. In the two models, the bootstrapping sample was set at 5,000, as this is a common setting in previous studies, and the confidence interval (CI) was set to 95%.

Normal regression was used in the early stage of our analysis in order to set the parameters for the next step of our analysis. According to the normal regression analysis, as shown in Model 1, the leader’s prosocial behavior positively predicted employee voice behavior (β = .108, p < .05), and, as shown in Model 2, leader’s self-serving behavior negatively predicted employee voice behavior. (β = -.134, p < .01). Thus, Hypotheses 1a and 1b were supported.

After Step 1 of the regression analysis, we used the bootstrapping method for multiple mediators. From the results set out in Table 2, it can be seen (Model 1) that the null hypothesis that the total indirect effect of prosocial behavior on voice through trust and suspicion would be zero was not supported (p < .05). As for the contrast effects, because zero is contained in the 95% CI (see Table 2), the two indirect effects cannot be distinguished in terms of magnitude, which means there is no difference between the two mediators, that is, they are equally important. From the results shown in Table 2, we are able to conclude that both trust and suspicion are important mediators between leader prosocial behavior and employee voice behavior. Therefore, Hypothesis 3a and 4a were supported. In Table 3, the results are shown of our second model of the effect of leader self-serving behavior on employee voice, as mediated by trust and suspicion. The results (Model 2) show that, again, the null hypothesis was not supported, and that there is a difference between the two mediators. In this case, we were able to conclude that suspicion was the only significant mediator between leader self-serving behavior and employee voice behavior. Thus, Hypothesis 3b was not supported, and Hypothesis 4b was supported. As the p values of the direct paths from the independent variable to the dependent variable are .08 and .07, respectively, in the two models, this indicates that the direct path from the independent variable (leader prosocial/self-serving behavior) to the dependent variable was nonsignificant in the multimediation model; thus, leader prosocial/self-serving behavior was completely mediated by trust and suspicion in both models.

Discussion

In this study, our focus was on a trust-based psychological process within the relationship between employee perception of leader impression management and employee voice behavior. In the empirical test, we first examined the independence of the two similar constructs of trust and suspicion. The result that trust and suspicion are two different constructs was one that we found counterintuitive. This result provides empirical support for the theoretical assumptions made by Bobko et al. (2014) about the relationship between the concepts of suspicion and trust. Through our further analysis, we also found empirical evidence to suggest that trust and suspicion each plays a different role as a mediator in the relationship between leader impression management and employee voice behaviors. For the employees’ perception of a leader’s self-serving behavior, we found that suspicion plays a more important role than trust as a mediator in the relationship between leader self-serving behavior and employee voice behavior. However, for leader prosocial behavior, according to our results, trust and suspicion both play an important mediating role in the relationship between leader behavior and employee voice behavior. Previously, researchers have explored the relationship between trust and employee voice behavior only in the context of certain leader behaviors (e.g., empowering behavior; Gao et al., 2011). In our research, the findings indicate that suspicion is likely to have a stronger effect on employee voice behavior than trust does when employees are trying to deal with their leaders’ impression management behavior.

The managerial implications are as follows: First, employees’ perceptions of leader behavior will actively affect their voice behavior. Leader behaviors and decisions that are self-centered rather than prosocial will increase an employee’s suspicion of the leader’s underlying motive. This may suggest that, if they want their subordinates to make more suggestions and to be constructive, leaders need to make the effort to build and protect a prosocial image. Second, we believe that suspicion will prove to be an important factor influencing not only employees’ voice behavior but also employees’ performance. If employees are bothered by state suspicion, they would need time to generate alternative explanations for the issue they are dealing with, and that requires significant cognitive consumption. If their state suspicion cannot be controlled, the overall performance of employees would be harmed. Even worse, if the whole team climate or organizational climate is filled with suspicion, more damage would be done to organizational performance. In order to reduce employees’ state suspicion, our suggestion for leaders is that they pay attention to timely information disclosure (e.g., information about change of personnel or strategic change in the organization) and be authentic to their followers (e.g., frankly talk about some of the difficulties and challenges the management team face). This would help employees to form fewer alternative explanations of their leaders’ behavior and about the organization’s potential conduct toward them, so that the level of state suspicion among employees would be lower (Bobko et al., 2014; Seigele, 2014).

Despite our contributions, there are limitations in this study. First, we used cross-sectional data. To better verify the causal relationships, we suggest that a longitudinal study be conducted in the future. Second, the data for this study were obtained mainly from one industrial sector in China, so the issue of external validity cannot be excluded. Third, trust and suspicion are two major constructs in organizational research, but more control variables, such as organizational climate and other leadership styles, should be added to test further the validity of the relationships that we found. Further research is suggested not only to avoid these limitations, but also to take consideration of the following: voice behavior should be specified by measuring prohibitive and promotive types of voice behavior; other outcome variables, such as creativity and unethical pro- organizational behavior, should be examined; and other potential mechanisms should be sought in which trust and suspicion might be implicated.

Bandura, A., & Walters, R. H. (1963). Social learning and personality development. New York, NY: Holt, Rinehart, & Winston.

Bobko, P., Barelka, A. J., & Hirshfield, L. M. (2014). The construct of state-level suspicion: A model and research agenda for automated and information technology contexts. Human Factors, 56, 489–508. https://doi.org/f5zgp7

Boxall, P. F., & Purcell, J. (2011). Strategy and human resource management (3rd ed.). Basingstoke, UK: Palgrave Macmillan.

Brislin, R. W. (1980). Cross-cultural research methods: Strategies, problems, applications. In I. Altman & J. F. Wohlwill (Eds.), Human behavior and environment: Advances in theory and research. Vol. 4: Environment and culture (pp. 47–82). New York, NY: Plenum Press. https://doi.org/b54t

Buss, A. H., & Durkee, A. (1957). An inventory for assessing different kinds of hostility. Journal of Consulting Psychology, 21, 343–349. https://doi.org/btc79s

Deluga, R. J. (1994). Supervisor trust building, leader-member exchange and organizational citizenship behavior. Journal of Occupational and Organizational Psychology, 67, 315–326. https://doi.org/dmg92k

Detert, J. R., & Burris, E. R. (2007). Leadership behavior and employee voice: Is the door really open? Academy of Management Journal, 50, 869–884. https://doi.org/czqk92

Fein, S. (1996). Effects of suspicion on attributional thinking and the correspondence bias. Journal of Personality and Social Psychology, 70, 1164–1184. https://doi.org/fndffd

Gao, L., Janssen, O., & Shi, K. (2011). Leader trust and employee voice: The moderating role of empowering leader behaviors. The Leadership Quarterly, 22, 787–798. https://doi.org/b2bjqt

Gardner, W. L., & Avolio, B. J. (1998). The charismatic relationship: A dramaturgical perspective. The Academy of Management Review, 23, 32–58. https://doi.org/dc3n69

Gardner, W. L., & Cleavenger, D. (1998). The impression management strategies associated with transformational leadership at the world-class level: A psycho-historical assessment. Management Communication Quarterly, 12, 3–41. https://doi.org/dchjrx

Hoffmann, W. H. (2007). Strategies for managing a portfolio of alliances. Strategic Management Journal, 28, 827–856. https://doi.org/d74qnh

Kish-Gephart, J. J., Detert, J. R., Treviño, L. K., & Edmondson, A. C. (2009). Silenced by fear: The nature, sources, and consequences of fear at work. Research in Organizational Behavior, 29, 163–193. https://doi.org/brzjq4

Kleiman, T., Sher, N., Elster, A., & Mayo, R. (2015). Accessibility is a matter of trust: Dispositional and contextual distrust blocks accessibility effects. Cognition, 142, 333–344. https://doi.org/f7mfbn

Levine, T. R., & McCornack, S. A. (1991). The dark side of trust: Conceptualizing and measuring types of communicative suspicion. Communication Quarterly, 39, 325–340. https://doi.org/bbpz8q

Liang, J., Farh, C. I. C., & Farh, J.-L. (2012). Psychological antecedents of promotive and prohibitive voice: A two-wave examination. Academy of Management Journal, 55, 71–92. https://doi.org/szf

Liu, W., Zhu, R., & Yang, Y. (2010). I warn you because I like you: Voice behavior, employee iden- tifications, and transformational leadership. The Leadership Quarterly, 21, 189–202. https://doi.org/bx7rxq

Lyons, J. B., Stokes, C. K., Eschleman, K. J., Alarcon, G. M., & Barelka, A. J. (2011). Trustworthiness and IT suspicion: An evaluation of the nomological network. Human Factors, 53, 219–229. https://doi.org/dwj4h3

Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. The Academy of Management Review, 20, 709–734. https://doi.org/fs6wzz

Mayer, R. C., & Gavin, M. B. (2005). Trust in management and performance: Who minds the shop while the employees watch the boss? Academy of Management Journal, 48, 874–888. https://doi.org/dzjjtn

Morrison, E. W. (2011). Employee voice behavior: Integration and directions for future research. The Academy of Management Annals, 5, 373–412. https://doi.org/chtgwq

Mowbray, P. K., Wilkinson, A., & Tse, H. H. M. (2015). An integrative review of employee voice: Identifying a common conceptualization and research agenda. International Journal of Management Reviews, 17, 382–400. https://doi.org/bhvd

Olson, M. T. (2009). The development of IT suspicion as a construct and subsequent measure (Master’s thesis). Retrieved from Defense Technical Information Centre. (Accession No. ADA502469.)

Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891. https://doi.org/b9b2k3

Schoorman, F. D., Mayer, R. C., & Davis, J. H. (2007). An integrative model of organizational trust: Past, present, and future. Academy of Management Review, 32, 344–354. https://doi.org/bwcntq

Schwartz, J., & Wald, M. L. (2003, March 9). The Nation: NASA’s curse?: Groupthink is 30 years old, and still going strong. New York Times. Retrieved from https://nyti.ms/2zb0mgW

Seigele, L. (2014, January 17). Tech startups: A Cambrian moment. The Economist. Retrieved from https://econ.st/2AgeGlP

Sosik, J. J., Avolio, B. J., & Jung, D. I. (2002). Beneath the mask: Examining the relationship of self- presentation attributes and impression management to charismatic leadership. The Leadership Quarterly, 13, 217–242. https://doi.org/c52jbs

Van Dyne, L., Ang, S., & Botero, I. C. (2003). Conceptualizing employee silence and employee voice as multidimensional constructs. Journal of Management Studies, 40, 1359–1392. https://doi.org/dv5qmm

Van Dyne, L., & LePine, J. A. (1998). Helping and voice extra-role behaviors: Evidence of construct and predictive validity. Academy of Management Journal, 41, 108–119. https://doi.org/b2bw8v

Shudi Liao is now at the School of Business

Hubei University.

Shudi Liao, School of Business, Hubei University, Youyi Dadao 368, Wuchang District, Wuhan, Hubei 430062, People’s Republic of China. Email: [email protected]