The impact of leader inclusiveness on employees' turnover intention and organizational citizenship behavior in Chinese work context

(1)

Master Thesis

The impact of leader inclusiveness on employees’ turnover intention

and organizational citizenship behavior in Chinese work context

University of Amsterdam

Faculty of Economics and Business

MSc in Business Administration – Leadership and Management track

Student: Fu Yixuan

Student number: 11390654

Thesis supervisor: Prof. Dr. Deanne Den Hartog Date of submission: January 26, 2018

(2)

Statement of originality

This document is written by Fu Yixuan who declares to take full responsibility for the contents of this document. Fu Yixuan declares that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

Abstract

The current study examined the indirect impact of leader inclusiveness on employees’ turnover intention through the mediation of job satisfaction. Also, it investigated the positive relationship between leader inclusiveness and Organizational Citizenship Behavior (OCB) with two moderations of Role-Breadth Self-Efficacy (RBSE) and task interdependence. After conducting a survey of 204 Chinese employees and running the regression analyses, the results supported the positive relationship between leader inclusiveness and job satisfaction, the negative relationship between job satisfaction, the full mediation of job satisfaction in the relationship of leader inclusiveness and turnover intention, as well as the positive relationship between leader inclusiveness and OCB. However, there was no evidence about the moderating effect of RBSE or task interdependence on the relationship between leader inclusiveness and OCB. These findings expand the understanding of the strong impact of leader inclusiveness on Chinese employees’ attitude and behavior. Accordingly, the study provides the strategies regarding promoting leader inclusiveness in the workplace.

Key Words: leader inclusiveness, job satisfaction, turnover intention, Organizational Citizenship Behavior, Role-Breadth Self-Efficacy, task interdependence

(3)

In the nowadays’ volatile work environment characterized by rapid human resource mobilization, the problem of high turnover rate poses risks to organizations in a wide range of industries. Aon Hewitt (2016) reported that in spite of salary increase, the overall employee turnover rate in China remains high at 20.8%, with the voluntary turnover rate at 14.9%; in some burgeoning industries such as hospitality and internet, the voluntary turnover rates almost reached 40%. Since the heterogeneous human capital pool of an organization is comprised of every employee’s valuable knowledge, skills, and abilities, which improves the organizational performance, high incidents of voluntary turnover are detrimental to the collective human capital (Dess & Shaw, 2001). Moreover, the average cost of replacing an employee can be as high as 50%-60% of the employee’s annual salary, and the total loss to the organization varies from 90% to 200% of the annual salary (Neese, 2016). Hence, organizations should not underestimate the necessity of reducing the potential turnovers. Employees’ turnover intention has been proved to be a proxy indicator of actual turnover (Griffeth, Hom, & Gertner, 2000), so it is practically useful to retain employees through decreasing their turnover intentions.

Looking at the reasons for turnover intention, job satisfaction plays a critical role in many turnover theories (Trevor, 2001, p.622). Rosin and Korabik (1991) directly described turnover intention as indicative of job dissatisfaction. Leadership style predicts employees’ attitude towards their jobs, and the way employees view their leaders is one of the important determinants of job satisfaction (Hogan & Kaiser, 2005). If employees have negative perceptions about their leaders, they are less prone to obtain job satisfaction, because such negative perceptions can cause detrimental impacts on employees’ work experience and expectations. The type of leaders that make employees tend to leave their jobs commonly have little understanding about subordinates’ need for encouragement or appreciation, as well as very few practices of sharing ideas and open discussions; instead, they overemphasize the mechanistic aspects of work performance, and cause much misery on the well-being of followers (Higginbottom, 2015). Under such kind of management style, employees are monotonously required to work and do not have opportunities to speak up, recognitions for their contributions and cares for psychological fulfillments. Gradually, they are inclined to

(5)

feel unsatisfied about their jobs. If the level of job satisfaction is sufficiently low, the employee would probably develop a behavioral intention to leave (Spector, 1997, p63). Therefore, effective leadership is needed to increase employees’ job satisfaction and reduce the turnover propensity.

Direct supervisors have a strong influence on subordinates’ experience in the workplace, and they have the capacity of creating a sense of inclusiveness among subordinates (Shore, Randel, Chung, Dean, Ehrhart, & Sing, 2011, p1277) through conveying the values of responsiveness, respect, recognition, and responsibility (Hollander, 2009). Leader inclusiveness is the essential core of relational leadership, which emphasizes inviting diverse voices and appreciating contributions (Nembhard & Edmondson, 2006). According to Shore et al. (2011, p1276), if employees perceive leaders as inclusive when their unique perspectives are valued and their necessary needs are supported, they would be likely to develop job satisfaction. Compared with those leadership styles (e.g., transformational leadership, ethical leadership) that have yielded rich amounts of studies, inclusive leadership is still in its infancy, but its significance is rapidly emerging since today’s work groups are increasingly diverse. Recent literature has substantiated several positive outcomes of inclusive leadership, such as psychological safety (Nembhard & Edmondson, 2006; Carmeli, Reiter-Palmon, & Ziv, 2010), and team identity (Mitchell, Boyle, Parker, Giles, Chiang, & Joyce, 2015). But to my knowledge, the scholars have not fully investigated whether inclusive leadership can increase employees’ job satisfaction and therefore diminish their turnover intentions, although Mor Barak, Levin, Nissly, and Lane (2006) had a suggestion of such indirect path. Therefore, this thesis would explore the relationship between inclusive leadership and employees’ turnover intention, with job satisfaction as a mediator.

Besides, Shore et al. (2011, p1279) further argued that leader inclusiveness could promote organizational citizenship behavior (OCB). Because on the basis of the norm of reciprocation, this type of supervisory behaviors could foster an obligation to reciprocate the fair treatment and avoid detrimental actions among employees. But the relationship has no supports yet. OCB, defined as discretionary behaviors in the workplace that go beyond employees’ enforceable job requirements and in the aggregate promote organizational effectiveness (Organ, 1988, p4), has been confirmed by Podsakoff, Whiting, Podsakoff, and

(6)

Blume (2009) to have favorable consequences on the individual- (e.g., job performance ratings) and organizational-level (e.g., unit efficiency, profitability). OCB helps to enhance the effective functioning of organizations and cope with many unexpected contingencies; without the spontaneous behaviors, the social system within an organization would become very fragile (Smith, Organ, & Near, 1983). Podsakoff, MacKenzie, Paine, and Bachrach (2000) proved leader behavior as one of the main antecedents of OCB. Combined with Shore et al. (2011, p.1276)’s argument, this thesis would examine whether the specific supportive leader behavior – leader inclusiveness is positively related to OCB.

In addition to that, on the basis of Kerr and Jermier’s (1978) study, employees’ ability and knowledge as well as cohesive, interdependent work groups may form the substitutes for leadership, and thus moderate the relationship between leader behavior and subordinate performance. It means that if employees are aware that they have skills and knowledge needed to perform a task successfully, or if employees are in the context of highly interdependent groups, the impact of leader behavior on employees would reduce. According to Kerr and Jermier’s (1978) argument, the thesis would explore two possible moderators, namely, role breadth self-efficacy and task interdependence, in the relationship between leader inclusiveness and OCB.

Role breadth self-efficacy (RBSE) refers to the extent to which people view themselves as capable of undertaking broader work roles (Parker, 1998, p.835). The impact of leadership is not same for everyone, and individual attributions make difference. Leader behavior could have less impacts on employees who have confidence in their ability and knowledge (Kerr & Jermier, 1978). Translated to RBSE and OCB, the theory of leadership substitutes suggests that OCB of high-RBSE employees is less affected by leader inclusiveness than that of employees low on RBSE. That is, because highly efficacious employees are already confident to take a broader range of roles and to proactively engage in OCB, leader behavior would have less effect on them. Consistent with this, Speier and Frese (1997) pointed out that employees with high self-efficacy did not need sufficient situational support to show personal initiatives. On the contrary, those with low RBSE may rely on the strong encouragement and support from leaders to conduct OCB, so they are likely further motivated by leaders. Therefore, it can be rationally assumed that employees’ RBSE could moderate the

(7)

relationship between leader inclusiveness and OCB.

The other potential moderator, task interdependence, describes the case of high levels of communication, cooperation, and coordination among the members (Gully, Devine, &Whitney, 1995). Mitchell and her colleagues (2015) suggested that it is possible that leader inclusiveness has differing effects when subordinates are engaged in interdependent tasks; future research could extend their study by exploring the moderating effect of task interdependence on the impact of leader’s behavior on employees. Yet, this area still remains unexplored. The extent of task interdependence ought to make a great difference for studying leadership styles: in low interdependence, leader’s considerate behaviors likely magnify subordinate’s performance; while in high interdependence, the impact of leader consideration on employees is less strong (Fry, Kerr, & Lee, 1986). The argument is consistent with Kerr and Jermier’s (1978, p.379) theory of leadership substitutes, which proposes that interdependent work group could render leadership less important or even inconsequential. Inclusive leadership is supportive and considerate, as it values each employee’s contribution and appreciate everyone’s uniqueness (Shore et al., 2011). Within the context of highly interdependent tasks, in accordance with the suggestions aforementioned, leader inclusiveness would have a weaker impact on motivating OCB because task interdependence may work as a substitute for inclusive leaders. Therefore, employees will not need leader inclusiveness to stimulate OCB very much. Hence, taking into account that the different levels of task interdependence can influence the relationship between leader inclusiveness and OCB, this thesis will also examine the moderating effect of task interdependence.

To sum up, the objectivity of this thesis is to study the mediating role of job satisfaction in the relationship between leader inclusiveness and turnover intention. Moreover, it will investigate whether leader inclusiveness is positively related with OCB, with RBSE and task interdependence as two moderators. Altogether, they lead to the conceptual model (see Figure 1). In order to examine these sets of relationships, this thesis contains following sections. First, it provides a literature review on the main variables, together with the theoretical framework to form the hypotheses. Next is the data and method section, followed by a result section once the data has been processed properly. Penultimate, based on the results, the discussion section provides the main findings, theoretical and managerial implications, as well as the limitations

(8)

and future research recommendations. Lastly, the thesis ends with a conclusion.

Figure 1: Conceptual model

2. Literature review

2.1. Leader inclusiveness

The term leader inclusiveness was firstly brought forward by Nembhard and Edmondson (2006, p947), and the definition encompasses words and deeds of a leader that show the invitation and appreciation to the team members’ contributions, which convey a signal to the members that their diverse opinions are genuinely valued. The research done by Nembhard and Edmondson (2006) focused on health care teams, and the results indicated that inclusive leaders could help the subordinates to mitigate the restraints of status differences, to develop psychological safety, and thus to better engage in quality improvement work. Later, Hollander (2009) viewed leader inclusiveness as an essential characteristic of leader – follower relationship, and coined the concept inclusive leadership, emphasizing on how followers can be actively included in leadership for mutual benefits. In the present study, the terms inclusive leadership and leader inclusiveness are used interchangeably, because they both refer to the relational operations between leaders and followers.

During the next years, a series of studies have been conducted to explore the effects of leader inclusiveness on employees’ positive behaviors, namely involvement in creative tasks (Carmeli et al., 2010), learning from failures (Hirak, Peng, Carmeli, & Schaubroeck, 2012),

(9)

and unit performance (Mitchell et al., 2015). Generally speaking, there are two common points among these articles. Firstly, leader inclusiveness is principally measured from the employees’ perception, since employees as the recipient of inclusion are in a better position to claim whether they are included or not. Secondly, psychological safety is a frequently-used mediator between leader inclusiveness and the employees’ behavior. When supervised by an inclusive leader, the employees would perceive more supports and benevolence, and less threats of negative interpersonal risks and status differences. Therefore, the employees are likely to vigorously show the proactive behaviors in the workplace.

Drawing on the salient characteristics of inclusive leadership, respect, recognition, responsiveness, and responsibility are the substantial components of two-way practices between leaders and followers (Hollander, 2009). Carmeli et al. (2010) concluded three dimensions of inclusive leadership: openness to new ideas, accessibility and availability to members. Specifically, inclusive leaders invite employees to participate in discussions and decision-making, listen carefully to their opinions, provide clear feedback and support for their activities that benefit the whole organization, and appreciate the employees’ contributions (Nembhard & Edmondson, 2006). As a consequence, every employee is treated as an insider, and meanwhile, allowed to keep uniqueness (Shore et al., 2011). It is noteworthy that some traits of inclusive leadership overlap with participative leadership, which underlies shared decision-making power and consulting with followers (Nembhard & Edmondson, 2006), as well as transformational leadership, which stimulates new ideas and enhances morale (Mitchell et al., 2015). However, inclusive leadership is distinct from the other two leadership styles. It explicitly highlights situations featured by power differences (Nembhard & Edmondson, 2006), and it emphasizes leader’s behavior that creates a reassuring environment for evoking diverse opinions to better accomplish the collective goals (Mitchell et al., 2015).

When leaders exhibit their inclusiveness and give supports physically and psychologically, employees are more likely to develop a series of positive attitudinal and behavioral outcomes, such as high quality relationship with colleagues and supervisors, job satisfaction, improved performance, and organizational citizenship behavior (Shore et al., 2011). Inclusive leaders not only promise a safe workplace where employees are encouraged

(10)

to speak up and share diverse ideas, but also empower employees to utilize their knowledge, skills, and autonomy for work, instead of giving autocratic instructions. Therefore, it is clear for employees that their contributive ideas make sense and they have freedom to carry out ideas as they see fit. As a result, employees would find jobs enjoyable and satisfying. Besides, the high-quality social exchanges are developed between inclusive leaders and employees, because leaders promote norms of inclusion and signal acceptance of employees regardless of their various backgrounds (Nishii & Mayer, 2009).

2.2. Job satisfaction

Job satisfaction is one of the core constructs in the management area. Spector (1997) indicated two reasons why this construct draws lots of attention: from the humanitarian perspective, job satisfaction highlights the importance of fair treatment and respect, and reflects employees’ psychological well-being; from the utilitarian perspective, employees who are satisfied with their jobs tend to behave in favor of organizational functioning. Job satisfaction generally refers to “a pleasurable or positive emotional state resulting from the appraisal of one’s job or job experiences” (Locke, 1976, p1300). Kalleberg (1977) provided a more specific definition of job satisfaction as employees overall emotional position towards work roles, and their feelings about components for which employees have certain preferential values, such as career opportunities, working conditions, and relationship with their fellow workers and leader. Consistently, there are two most common ways of measuring job satisfaction. One measures satisfaction as a global feeling about the job, and the other assesses it by facet (Brown & Peterson, 1993). Commonly measured facets include appreciation, communication, recognition, supervision, job condition, et cetera (Spector, 1997).

In the empirical literature, as Spector (1997) stated, the research of the antecedents and consequences of job satisfaction takes a major place in the organizational psychology and organizational behavior domain. As Saari and Judge (2004) put forward, there are basically three categories of antecedents of job satisfaction: work situations (such as the nature of job,

(11)

supervisory behaviors, and promotion opportunities), personal dispositions (such as extraversion and conscientiousness), and cultural influences. What organizations attempt to do are to select persons appropriate for jobs, to design jobs to be interesting and challenging, to supervise employees in a more effective way, and to adjust human resource practices in line with culture differences (Saari & Judge, 2004). The potential consequences of job satisfaction, according to Spector’s (1997) summary, can be classified into two sorts: work variables (e.g., job performance, withdrawal behavior) and non-work variables (e.g., physical and psychological health). It should be noted that the relations between job satisfaction and some variables (for instance, job performance) have yet to be established as causal, but most theories recognize turnover intention as the result of job dissatisfaction. In a word, job satisfaction is widely believed to cause many positive work and non-work outcomes.

2.2.1. Leader inclusiveness and Job satisfaction

The extant literature has provided some useful clues for logically connecting leader inclusiveness and job satisfaction. In Shore and her colleagues’ (2011) framework of inclusion, employees who perceive inclusion in the work contexts likely become satisfied with their jobs, and inclusive leadership is one of the contextual antecedents. Previous studies also supported that the perception of inclusion significantly predicts job satisfaction (Acquavita, Pittman, Gibbons, & Castellanos-Brown, 2009; Mor Barak & Levin, 2002), stating that inclusion addresses the conflicts between belonging to specific social categories and finding their own social identity within the categories. Employees with a sense of inclusion likely find it comfortable and satisfactory to do jobs, as their need for belongingness is guaranteed, their uniqueness is welcomed, and their value is recognized. In such situations, employees do not have to be faced with much pressures of conformity with the majority’s opinion, and they have freedom to speak up to let others know their different ideas.

Furthermore, leader behavior plays an important role in affecting employees’ job satisfaction (Mohammad Mosadegh Rad & Hossein Yarmohammadian, 2006). The meta-analysis done by Brown and Peterson (1993) also reported that supervisory behavior is

(12)

associated with employees’ job satisfaction, mainly in terms of consideration, feedback, and communication. Referring to Braun, Peus, Weisweiler and Frey (2013)’s research on the relationship between transformational leadership and job satisfaction, transformational leaders can influence the various facets of employees’ job satisfaction concerning with leaders, tasks, working conditions, support of development, and general job satisfaction.

Under the circumstance where leaders are dedicated to promoting psychological well-being by exhibiting inclusiveness, employees are able to freely express ideas, have initiatives at work, and try their own ways of doing jobs (Nishii & Mayer, 2009), instead of being solely instructed to implement orders. Hence, employees are likely to own a sense of responsibility and meaningfulness, and are able to utilize their capabilities and make advancements when doing work, so their job satisfaction accrues. Furthermore, the ideas and contributions of employees are highly appreciated by inclusive leaders, so employees tend to obtain the feeling of recognitions and accomplishments from doing jobs. Additionally, inclusive leaders invite employees to participate in discussions and decision making, and give timely developmental feedback and constructive communication to employees (Nembhard & Edmondson, 2006), which provide considerate supports to employees and help to guide what employees should do to fulfill their duties. Moreover, inclusive leaders encourage egalitarian working conditions and mutual respects across hierarchies (Nembhard & Edmonson, 2006), so employees would possibly feel like having an equal and pleasant relationship with their leaders, and perceive the high level of job satisfaction. Based on the abovementioned arguments, the hypothesis is formed as:

Hypothesis 1: leader inclusiveness is positively related with employees’ job satisfaction.

2.3. Employees’ turnover intention

For a number of years, turnover has drawn lots of attention of studies in several disciplines, such as psychology and sociology. Much effort has been devoted into the development of conceptual models of turnover process, and subsequent empirical studies advanced the research by validating these models (Williams & Hazer, 1986). Mobley, Griffeth,

(13)

Hand, and Meglino’s (1979) heuristic model hypothesized that turnover intention is a cognitive intermediary between the experienced job satisfaction and actual turnover, and such linkages were later supported by Michaels and Spector’s (1982) test. Actually, several researchers (Porter & Steers, 1973; Lee & Mowday, 1987; Griffeth et al., 2000) proved that turnover intention is the strongest precursor of actual turnover. According to Tett and Meyer (1993, p.262)’s definition, individual turnover intention is conceived as “a conscious and deliberate willfulness to leave the organization”, and it is always listed as one of the withdrawal behaviors. As Johnsrud and Rosser (2002) noted, actual turnover is not easy to record in scientific studies, because once employees have left the current organizations, it is difficult to locate them, and their response rate to surveys about prior workplaces is low. Therefore, intention to leave is deemed as an appropriate indicator of actual turnover, and such turnover is normally voluntary termination of the employment contract with the present organizations (Tett & Meyer, 1993).

In the extant literature, some scholars have attempted to establish a comprehensive theoretical model for explaining the reasons for turnover intentions. For example, Muchinsky and Morrow (1980) posited three sets of factors: economic opportunity factors (e.g., the unemployment rate); individual factors (e.g., age, educational level); and work-related factors (e.g., experienced job satisfaction, salary). Good, Sisler, and Gentry (1988) proposed five determinants of turnover intention: role ambiguity, role conflict, work-family conflict, job satisfaction, and organizational commitment. Despite of the diverse origins, job satisfaction has been postulated to be one of the principal antecedents of turnover intention (Williams & Hazer, 1986).

2.3.1. Job satisfaction and Employees’ turnover intention

Of different determinants, job satisfaction seems to play a critical role in all turnover theories (Trevor, 2001). Substantiated by Lambert, Hogan, and Barton (2001), job satisfaction has the largest direct effect on turnover intention, compared with individual variables (i.e., tenure, age), financial rewards, and employment availability. A meta-analysis done by Tett

(14)

and Meyer (1993) demonstrated a strong and negative relationship between job satisfaction and intention to leave both in the global and sum-of-facet measures. In a similar vein, Trevor (2001) confirmed that the more employees are satisfied with their jobs, the less the likelihood of their voluntary withdrawal from their existing organizations within two years. Trevor (2001) also found that job satisfaction matters more for reducing turnover inclinations when there are other external career opportunities. Additionally, when employees possess a satisfying cognitive appraisal of their job and the general work environment, such as satisfaction with their supervisors, they are willing to remain in the organizations and keep devoting efforts for work (Miao, Humphrey, & Qian, 2017; Wells & Peachey, 2011). Consistent with the extant literature, the hypothesis is developed as:

Hypothesis 2: Job satisfaction is negatively related to turnover intention.

2.3.2. Leader inclusiveness, Job satisfaction, Turnover intention

There is an indirect path proposed by Lambert and his colleagues (2001) that work environment can strongly shape employees’ job satisfaction, which is a highly salient determinant of turnover intention. Some scholars (Brown & Peterson, 1993; Mohammad Mosadegh Rad & Hossein Yarmohammadian, 2006) further proved that leader behavior is one of the pivotal characteristics of work environment in influencing job satisfaction. Past research also suggested that when leaders pay attention to employees’ needs, and contribute to their job satisfaction, employees are less likely to have turnover intentions (Wells & Peachey, 2011). Furthermore, as Mor Barak et al. (2006) proposed, perception of inclusion could increase employees’ well-being and reduce stress, and thus enhance job satisfaction, which leads to lower turnover intention.

These findings imply an association between leader inclusiveness and turnover intention, with the mediation of job satisfaction. Combining with the argumentations in the previous parts, it is known that leader inclusiveness is characterized by leaders’ words and deeds that exhibit invitation and appreciation for subordinates’ contributions (Nembhard & Edmondson, 2006), and is embodied by openness to novel ideas, accessibility and availability in

(15)

interactions with subordinates (Carmeli et al., 2010). This type of supervisory behavior is manifested by consideration, constructive feedback and communication, which has positive impact on job satisfaction (Brown & Peterson, 1993). Job satisfaction, in turn, is negatively related with turnover intention (Tett & Meyer, 1993; Lambert et al., 2001). Therefore, it is expected that these findings would further support the expectation of a mediator role of job satisfaction between leader inclusiveness and employees’ turnover intention. Hereby, the hypothesis is formed as:

Hypothesis 3: Job satisfaction mediates the negative relationship between leader

inclusiveness and turnover intention.

2.4. Organizational citizenship behavior

Organ (1988) firstly introduced the construct of organizational citizenship behavior (OCB), and defined it as “individual behavior that is discretionary, not directly or explicitly recognized by the formal reward system, and that in the aggregate promotes the effective functioning of the organization” (p.4). The definition emphasizes two points: first, by ‘discretionary’, it means that OCB is not compulsorily required of the job descriptions, but a matter of personal choice; second, by ‘not directly or formally recognized by the reward system’, it means that returns of OCB are not contractually guaranteed, but OCB can increase the likelihood of being rewarded (Organ, 1997, p.86, 87). Organ (1997) then redefined OCB as “performance that maintains and enhances the social and psychological context that supports task performance” (p.95). It differentiated OCB from task performance, and again highlighted that OCB should contribute to organizational effectiveness.

Despite of various ways in which the dimensions of OCB have been proposed in the past decade (Organ, 1988; Williams & Anderson, 1991; Moorman & Blakely, 1995), the most widely-accepted conceptualization is the five-dimension OCB model established by Organ (1988). This model comprises altruism (i.e., offering help to the co-workers with work tasks), conscientiousness (i.e., exceeding the minimum requirements of compliance with rational regulations such as punctuality, attendance, task completion), sportsmanship (i.e., willingness

(16)

to tolerate the less ideal working conditions and inevitable inconveniences without complaints) , courtesy (i.e., taking actions before the work-related problems occur), and civic virtue (i.e., interests in the whole organization, active participations in its governance, and expressions of constructive opinions) (Organ, 1988). All the dimensions have been proved to be positively correlated with leader supportiveness (Podsakoff et al., 2000).

Turning attention to the antecedents of OCB, Smith et al. (1983) brought forward that OCB arises from work environment (i.e., leader supportiveness and task interdependence), personal attributes (i.e., extraversion and neuroticism), and satisfaction; illustrated by the results, leader supportiveness affects altruism indirectly through its impact on satisfaction, but influence compliance (namely, conscientiousness) directly. In the following years, many researchers have continuously expended efforts to investigate this area (e.g., Konovsky & Pugh, 1994; Moorman, Blakely, & Niehoff, 1998). Podsakoff et al. (2000) made a meta-analytic review of the empirical literature on OCB: OCB stems from employee characteristics, task characteristics, organizational characteristics, and leader behaviors. The results demonstrated that leader supportiveness is significantly correlated with all the five dimensions of OCB, because leader supportiveness functions formidably in the social exchange process when leaders interact with employees, and therefore promotes OCB. The large body of relevant researches has also shown the outcomes of OCB. Podsakoff et al. (2009) did a meta-analysis on individual- and organizational-level consequences of OCB: at the organizational level, OCB is positively related to customer satisfaction and several organizational performance measures; at the individual level, OCB is positively associated with managers’ ratings of employee performance and reward decisions.

2.4.1. Leader inclusiveness and Organizational Citizenship Behavior

According to Shore et al. (2011)’s conceptual framework, when employees regard their leaders as inclusive, they are prone to exhibit OCB, because the perceptions of being included and accepted enhance the reciprocation to leaders. Podsakoff et al. (2000) substantiated the significant correlations between leader supportive behaviors and all five dimensions of OCB,

(17)

namely, altruism, courtesy, conscientiousness, sportsmanship, and civic virtue. So in this thesis, it is argued that inclusive leadership, which shows support and appreciation to employees, is related with all the dimensions of OCB. Additionally, an inclusive leader is able to increase the employees’ willingness to become good organizational citizens and take broader responsibilities for the whole organization (Nishii & Mayer, 2009). Besides, leader supportiveness is one of the determinants of OCB based on two reasons: firstly, leader supportiveness is a type of OCB in itself, so it could serve as a role model and inspire prosocial behaviors among subordinates; secondly, this type of leaders likely initiate positive social exchanges with subordinates and subordinates in turn show OCB to reciprocate their leaders (Smith et al., 1983).

If a leader presents inclusiveness by paying attention to the followers’ needs, listening to their ideas, and always providing necessary supports to them, the followers would probably conceive the leader as a role model and learn about these supportive behaviors, and then conduct them in the form of OCB in the workplace, such as volunteering to help co-workers. Moreover, supervised by inclusive leaders, subordinates likely feel respected and valued, as such leaders are willing to include subordinates into discussions and decision-making processes, and equalize the value associated with members’ contributions, rather than authorities or status (Nembhard & Edmondson, 2006). This type of considerate leaders are capable of establishing positive social exchanges with employees, which leads to broader reciprocal norms (Smith et al., 1983). Employees could be enticed to present OCB as a signal of having appreciation for leader’s inclusiveness. Hence, based on the aforementioned arguments, the following hypothesis is formed as:

Hypothesis 4: leader inclusiveness is positively related to OCB.

2.5. Role Breadth Self-Efficacy

According to Parker’s (1998) study, Role Breadth Self-efficacy (RBSE) refers to “the extent to which people feel confident that they are able to carry out a broader and more proactive role, beyond traditional prescribed technical requirements” (p.835). This kind of

(18)

efficacy is composed of two generic competencies: one requires people to be proactive, and the other requires interpersonal skills, such as experiences of group decision making (Parker, 1998). RBSE has consistency with the conception of self-efficacy, because 1) both of them focus on people’s self –belief in their capabilities of undertaking tasks, instead of what they are allowed to do; 2) both of two dynamically change over time in response to the external environment; 3) both of them can be enhanced through past performance accomplishments, vicarious experience, verbal persuasion, and judgements of physiological states (Parker, 1998, p.836). But RBSE does differentiate from self-efficacy, as the former emphasizes a range of proactive, interpersonal, and integrative activities that enlarge employees’ role, while the latter underlines one’s self-perception of competencies of performing a specific task to meet given situational demands (Axtell & Parker, 2003).

In the today’s fast-changing business world, organizations are looking for employees who are high on RBSE. This trend is reflected by the increasing prevalence of some concepts, such as citizenship behavior, and personal initiative, which stress the broader aspects of work performance (Parker, 1998). The extant literature has proved that RBSE is positively associated with employees’ positive outcomes, such as proactivity (Griffin, Neal, & Parker, 2007), and employability orientation (Nauta, Vianen, Heijden, Dam, & Willemsen, 2009). In addition, RBSE has been used as a moderator in the studies of investigating leadership’s impact on employees’ work-related behaviors (e.g., Den Hartog & Belschak, 2012). Suggested by some studies (e.g., Speier & Frese, 1997), employees’ RBSE may act as a substitute for leadership. The relationship between leadership and employee behavior weakens if employee are highly efficacious.

2.5.1. Leader inclusiveness, RBSE, Organizational citizenship behavior

The study of Kerr and Jermier (1978) proposed that, ability and knowledge of employees could act as a substitute for leadership such that leadership would become less effective or even unnecessary on employees’ behavior when employees consider themselves as capable of carrying out a task. In accordance to the argument, high-RBSE people are already confident

(19)

about their competence at taking broader roles in the workplace. So they do not need to live up to the leaders’ role model about how to contribute their citizenship behavior, or they do not have to rely on the social exchanges with leaders to engage in OCB. Therefore, these high-RBSE employees are less sensitive to leader inclusiveness, and less motivated by inclusive leaders to exhibit OCB.

On the contrary, low self-efficacy employees themselves are less likely to make extra effort beyond formal job requirements, namely, to engage in OCB (Bogler & Somech, 2004). Employees with low RBSE do not perceive themselves able to perform extra roles successfully, and may not show OCB on their own initiative. Thus, they would probably need the additional encouragement and supportiveness from inclusive leaders, because such leaders are characterized by welcoming new ideas, providing subordinates with necessary support and developmental feedback, as well as involving subordinates with discussion and decision-making process (Nembhard & Edmondson, 2006; Mitchell el al., 2015). Compared with high-RBSE employees, those low on RBSE are expected to benefit more from the impact of leader behavior, as they do not show OCB initiatively. Low-RBSE employees may rely on learning from the example set by inclusive leaders and reciprocating citizenship behaviors to show appreciation to leaders. Therefore, leader inclusiveness may further stimulate low-RBSE employees’ OCB.

Hypothesis 5: RBSE moderates the positive relationship between leader inclusiveness

and OCB, such that leader inclusiveness is more positively related to OCB when the employee

has low RBSE.

2.6. Task interdependence

Task interdependence refers to the degree to which the task requires high levels of communication, cooperation, and coordination among the members (Gully et al., 1995). It describes work “on a continuum of independence, sequentially dependence, reciprocal dependence, and team work” (Smith et al., 1983, p.658). High interdependence occurs when employees cooperate interactively and rely on each other for “information, material, and

(20)

reciprocal inputs” (Steward & Barrick, 2000, p.137). The research concerning with interdependence generally draws attention on two perspectives: one comes from organizational theory, and conceives interdependence as a product of technological development; the other one follows social psychology, and stresses that interdependence is generated by cooperation needs (Steward & Barrick, 2000). Interdependence usually provides a prerequisite for forming groups, because it may augment the motivational attributes of the tasks, and thus group members are devoted to pursuing the collective goals; or interdependence can increase the efficiency of doing work, because it highlights the necessity of coordinating with others’ work (Campion, Medsker, & Higgs, 1993).

Task interdependence is deemed as one of the most important variables that influence performance and has always been used in the scientific research as a critical moderator, to investigate its indirect impacts on performance by moderating the relationship between other variables and performance (Langfred, 2005). Most literature is contributed to investigating how the interdependence nature of work interacts with team process in relation to performance. For example, the moderating role of task interdependence is expected to influence the positive relationship between knowledge sharing and team effectiveness (Sandy Staples & Webster, 2006), as well as the relationship between maintenance behavior (such as interpersonal conflict resolving) and performance (Gladstein, 1984). In addition, interdependence is argued to substitute for leadership (Kerr & Jermier, 1978). But seemingly relatively fewer studies pay attention to the interactive effects of leadership and task interdependence on employees’ behavior. The studies exploring the moderator of task interdependence in the relation between leadership and OCB are even more limited.

2.6.1. Leader inclusiveness, Task interdependence, Organizational citizenship behavior

Suggested by Mitchell et al. (2015), leader inclusiveness would probably has varying degree impacts in the context of different levels of task interdependence, and future research could be directed towards extending the study by exploring the moderating role of task interdependence of leader’s impact. Also, Fry et al. (1986) asserted that the degree of task

(21)

interdependence is supposed to make pretty large difference for investigating the effects of leader behavior on employees. Specifically, in low interdependent contexts, leader consideration regarding the comfort, egalitarian treatment, and contribution of employees, would lead to higher employees’ well-being and performance, while leader consideration under the situation of high interdependence assumes less significance because high task interdependence already necessitates frequent cooperation, coordination and communication among employees in an effective manner, and they may not respond much to leader consideration. Such argument seems to imply the possibility of task interdependence acting as a substitute for leader consideration. That is to say, when the task interdependence is very high so that cooperation and communication are highly required, the impact of leadership on employee behavior reduces. This is in line with Kerr and Jermier’s (1978) viewpoint, which argues that interdependent groups have the capacity of rendering leadership unnecessary. Moreover, a variable that is qualified as a substitute must have an important impact on the dependent variable, and should be supported by a rational reason why it weakens the effect of leader behavior on the dependent variable (Howell, Dorfman, & Kerr, 1986).

Applying the theory of leadership substitutes, leader inclusiveness is expected to have differing effects on OCB depending on the level of task interdependence. In the case of high level of task interdependence, employees do their work reciprocally dependently with frequent interactions and mutual adjustment for better cooperation and coordination, so high task interdependence induces the social norms of helping, taking responsibility, and caring about others’ need (Smith et al., 1987). Thus, in the process of working closely with others, sharing information, material, and expertise to reach the common goals, employees are spontaneously engaged with OCB. In this condition, they will not need the inspirations of the role model of inclusive leaders, or they may not be quite sensitive to social exchanges with leaders, since they are already placed in the situation where OCB is fostered. Therefore, even though employees might still feel motivated and appreciated for leader inclusiveness, in highly interdependent tasks, an inclusive leader is not likely to further inspire employees to show increased OCB.

On the contrary, in low-interdependence contexts, employees work separately and independently without much assistance from others. When leaders treat employees with

(22)

openness, availability, and accessibility to facilitate inclusion, the function of leader inclusiveness would be presumably salient in serving as a role model and encouraging the reciprocal norms among employees, because such leader supportiveness and consideration is the important environmental source of OCB (Smith et al., 1983; Podsakoff et al., 2000). Thus, employees are much motivated by the exemplary role of inclusive leaders and likely live up to the social exchanges with leaders to show OCB, as they have fewer opportunities of interacting with other colleagues in work, while leaders are always available for them and care about their needs (Carmeli et al., 2010). In this instance, leader inclusiveness could have a stronger impact on OCB in low task interdependence compared with high interdependence. Hypothesis 6: task interdependence moderates the positive relationship between leader

inclusiveness and OCB, such that leader inclusiveness is more positively related to OCB when

(23)

3. Research method

This section presents the research method applied to conduct the current study. It describes the procedure of data collection, the respondent sample, and the instruments used to operationalize constructs.

3.1. Procedure

The study took a quantitative approach, using questionnaires to conduct a cross-sectional survey study, which is helpful as a preliminary test of the proposed research model. All the questionnaires were distributed online and were returned in an anonymous way. At the beginning of the questionnaire, we concisely informed the participants about the purpose, the length, and the confidential nature of the survey. Participation was voluntary and no incentives were offered. Furthermore, the first part of the questions were used for collecting demographic information (age, gender, education level, etc). The subsequent part consisted of total 71 items related to the six variables.

The survey was only carried out among Chinese employees to improve the reliability and validity of the questionnaire, and to further comprehend the research model in the Chinese working context. The scales were originally cited from English literature, so in order to increase the response rate and accurate understanding of the questions, we translated the whole questionnaire into Chinese. Another Chinese master student checked the translation. Both of us are native Chinese speakers, and fluent in English.

Data collection started at October 8, 2017, and ended at October 16, 2017. The questionnaires were available in Chinese and English, depending on linguistic habits of the participants. The participants were approached through my personal networks, and they were comprised of full-time employees in China. Therefore, this study was based on the convenience sampling method, considering the limited time and cost effectiveness. Due to the personal distribution of the online questionnaires, the response rate was as high as 82%, and it took nine days to finish data collection.

(24)

3.2. Sample

250 Chinese employees were invited to participate into the survey, and in total 204 employees completed the questionnaires, corresponding to the aforementioned response rate of 0.82. Among all the participants, 196 filled out the Chinese version, and eight finished the English one. There were no missing data found in this survey, because the validation option of every question was set as “force response”, and thus the participants could not skip answering any question.

The sample consisted of 77 men and 127 women, and as these participants reported, 147 of their supervisors were male, and 57 were female. Organizations in which the participants work operated in different fields of industries, including government sectors, financial institutions, Information Technology. Age ranged from 21 to 61 years, with an average of 37.36 years, and a standard deviation of 10.58. The highest education level was divided as 15.6% less than bachelor, 66.7% bachelor, 17.2% master, and 0.5% doctorate degree and above. Employee tenure varied greatly, from less than one year to 35.75 years (mean=10.62; SD=9.99). The duration of the participants working with their supervisors varied from 0.08 to 30.25 years (M =4.02; SD =4.62).

3.3. Measures

In this study, we used the questionnaire for employees. All variables were operationalized by using scales derived from the extent literature. All items, except the question on field of industry, were quantitative. The items related to the six variables were scored on a 5-point Likert scale.

Leader inclusiveness

In accordance with Carmeli et al. (2010)’s study, the scale depicting leader inclusiveness consists of 9 items, which are measured from the three dimensions of openness, availability, and accessibility, and rated on a five-point scale ranging from 1= ‘not at all’ to 5= ‘to a large extent’. The example questions are: ‘The manager is open to hearing new ideas’ and ‘The manager is accessible for discussing emerging problems’. However, the scale does not include

(25)

participation in decision-making, which is argued to be an indispensable factor of leader inclusiveness (Nembhard & Edmondson, 2006). Thus we decided to add relevant questions on this. Because empowering leadership consists of the dimension of participative decision making (Zhang & Bartol, 2010), we included the three items of this dimension into the measurement of leader inclusiveness, and every item is rated on a five-point scale. The example question is: ‘My manager makes many decisions with me’. Cronbach’s α of the whole scale with the 12 items is 0.927.

Job satisfaction

For job satisfaction, the thesis used the short version of Minnesota Satisfaction Questionnaire (MSQ) developed by Weiss, Dawis, England, and Lofquist (1967). It has been very popular among researchers. All the items were rated on a five-point scale, from 1= ‘very dissatisfied’ to 5= ‘very satisfied’. Because salary and organizational policies are usually determined by the management and the administrative department, direct supervisors are unable to directly decide the way these two issues influence employees. Thus we removed the relevant two items and remained 18 items that are relevant with supervision, ability utilization, advancement, recognition, achievement, and so on (Spector, 1997, p.16). The Cronbach’s alpha is 0.923.

Turnover Intention

Shore and Martin (1989) brought forward four items to measure turnover intention, and each item was rated on a five-point scale. Cronbach’s α is 0.775. The example question is: ‘How do you feel about leaving this organization?’

Organizational citizenship behavior

As Carpenter, Berry, and Houston (2013) suggested, employees’ self-ratings of OCB are preferred when the research emphasizes on employees’ self-perceptions of their behaviors. This thesis aims at exploring the effect of leader inclusiveness on employees’ OCB, so employees themselves are in the better place to claim the extent of such effect. The OCB scale with 24 items was established by Podsakoff, MacKenzie, Moorman and Fetter (1990). It was validly developed on the basis of Organ (1988)’s five-dimension model of OCB. Cronbach’s α of the scale is 0.861. The representative questions are: ‘Do not take extra break’ and ‘Willingly help others who have work-related problems’.

(26)

Role Breadth Self-Efficacy

According to Parker’s (1998) study, RBSE scale was used to rate how confident individuals would feel on carrying out broader-role tasks, and all ten items were measured on a five-point Likert scale, ranging from 1= ‘Not at all confident’ to 5= ‘Very confident’ (α= 0.923). The example question is: ‘How confident would you feel about presenting information to a group of colleagues?’

Task interdependence

The scale developed by Pearce and Gregersen (1991) initially contained 11 items, which were assessed on five-point scales from 1=‘Strongly disagree’ to 5= ‘Strongly agree. But the reliability was not very high. Later, Liden, Wayne, and Bradway (1997) modified Pearce and Gregersen’s (1991) scale into three items, and the example question is: ‘I frequently must coordinate my efforts with others in my work group’. The Cronbach’s α of this measurement is 0.755.

Control variables

In several studies, employees with higher education level were found to show more OCB (Gregersen, 1993), and tenure of working for the current employer also affects OCB (Cohen & Avrahami, 2006). Chou and Pearson (2011) reported that age and gender were significantly related to OCB. Besides, organizational tenure (Kim & Stoner, 2008), age (Trimble, 2006), and gender (Moura, Abrams, Retter, Gunnarsdottir, & Ando, 2009) all influence turnover intention. In addition, some researchers did studies on turnover intention in different industries and reported completely different findings, so industries might influence turnover intention as well. Besides, we added the duration of working with the current supervisor and the gender of the current supervisor as control variables. To sum up, pertaining to the demographic variables, the respondent’s age, gender, level of education, field of industry, organizational tenure, the duration of employment with the current manager, and the gender of the current manager were collected on the questionnaire, and coded into dummy variables.

(27)

4. Results

This section describes the statistical procedure taken to analyze the raw data and to test the aforementioned hypotheses. The results of the preliminary data analysis (such as correlations) and hypothesis testing are described in details.

4.1. Preliminary data analysis

First of all, a check of frequencies has been performed to assure no errors in data entry. Also, the frequency tables display that there are no missing values of any variable, as the valid number of each item is 204. Besides, we recoded the five counter-indicative items of OCB, and coded seven control variables into dummy variables.

After this, the six scale reliabilities were computed, and all Cronbach’s alpha coefficients are above 0.75 (see Table 1). It indicates a good internal consistency of the instruments (DeVellis, 2003), so no changes need to be made to the existing scale compositions. Then, scale means and standard deviations were calculated and these are listed in Table 1.

Subsequently, we did a normality check of all scales by using Kolmogorov-Smirnov test and Shapiro-Wilk test, and none of the scales have a fully normal distribution. In terms of Skewness and Kurtosis, only RBSE has a normal distribution, with Skewness and Kurtosis quite close to 0. For leader inclusiveness, job satisfaction, and task interdependence, Skewness is below 0 and Kurtosis is above 0, which demonstrates many scores clustering in the high end and pointy distribution. For turnover intention, Skewness is above 0 and Kurtosis is below 0, showing that many scores are grouped to the low end and the distribution is relatively flat. For OCB, both Skewness and Kurtosis are above 0, so the distribution is moderately positive and peaked. As suggested by Pallant (2010, p57), it is common to get a non-normal distribution with reasonably large samples (i.e., 200+ cases), and she recommended to assess the shape of the distribution. Therefore, we performed Q-Q plot for each variable, and it indicates that data is approximately normally distributed. Next, outliers were identified by Boxplot, but we did not remove any of them, because the 5% trimmed mean and mean value of every variable are very similar, and the outliers would not greatly influence the analysis (Pallant, 2010).

(28)

relationships between all the variables. The values were reported in Table 1. Leader inclusiveness correlates positively and significantly with job satisfaction and OCB, and the correlation between leader inclusiveness and job satisfaction (r=.770, p<.01) is much stronger than that between leader inclusiveness and OCB (r=.406, p<.01). Turnover intention correlates negatively and significantly with leader inclusiveness (r= -.378, p<.01), and job satisfaction (r= -.498, p<.01). RBSE correlates positively with leader inclusiveness (r=.319, p<.01) and OCB (r=.440, p<.01). Similarly, task interdependence has a positive correlation with leader inclusiveness (r=.289, p<.01) and OCB (r=.528, p<.01). There is no multicollinearity problem for any variable, with the tolerance level higher than 0.1, and VIF lower than 10.

To sum up, the preliminary analysis provides a satisfactory prerequisite for proceeding to the hypothesis testing.

(29)

Variable M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 1.Leader inclusiveness 3.64 0.63 (.927) 2.Job satisfaction 3.72 0.52 .770** (.923) 3.Turnover intention 2.28 0.87 -.378** -.498** (.775) 4.OCB 3.79 0.36 .406** .418** -.314** (.861) 5.RBSE 3.53 0.64 .319** .394** -.200** .440** (.923) 6.Task interdependence 3.85 0.57 .289** .300** -.166* .528** .468** (.755) 7.Employee gender - - .096 .132 -.047 .053 .243** .188** - 8.Employee age 3.80 2.14 -.049 -.016 -.289** .069 .101 -.078 .163* - 9.Education - - -.118 -.098 .208** -.013 .099 .079 .036 -.218** - 10.Industry - - -.011 .069 .132 -.070 .085 .017 .038 .030 -.020 - 11.Organizational tenure 6.43 5.01 -.028 -.005 -.247** .013 -.004 -.106 .107 .716** -.210** -.092 - 12.Employment with current manager 3.13 2.32 .105 .122 -.236** .090 .057 -.055 .082 .427** -.225** .246** .416** - 13.Leader gender - - .027 .140* -.136 .037 .077 .027 .260** .064 -.011 -.034 .211** .050 -

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

(30)

4.2. Hypothesis testing - Hierarchical regression

In this part, we used hierarchical multiple regression to explore the ability of the independent variable to predict the dependent variable. Firstly, we conducted hierarchical regression to see how much of the variance in job satisfaction is explained by leader inclusiveness, after controlling for the control variables. The first step was to enter seven controls: employee gender, employee age, education, industry, organizational tenure, employment with current manager, and leader gender. As presented in Table 2, none of the predictors has a significant relationship with job satisfaction. The model is insignificant, with F(7, 196)= 1.849; p>.05, and it only explains 6.2% variance of job satisfaction. In the second step, after the entry of leader inclusiveness as a predictor, the whole model is statistically significant, with F(8, 195)= 38.907; p<.001, and explains 61.5% variance of job satisfaction. Therefore, the introduction of leader inclusiveness explains an additional 55.3% variance of job satisfaction (R2_{Change=.553; F(1, 195)= 279.90; p<.001), after controlling for the seven}

demographic variables. In the final model, two predictors are significantly related with job satisfaction, with leader inclusiveness recording a much higher β value (β=.763, p<.001) than leader gender (β=.119, p<.05). The finding therefore supports Hypothesis 1, stating that leader inclusiveness is positively related to job satisfaction.

(31)

Table 2. Hierarchical Regression Model of Leader Inclusiveness on Job Satisfaction R R2 _R2 Change B SE β t Step 1 .249 .062 .062 Employee gender .119 .078 .111 1.524 Employee age -.019 .025 -.079 -.764 Education -.088 .064 -.100 -1.383 Industry .003 .007 .030 .414 Organizational tenure -.006 .011 -.061 -.570 Employment with current manager .031 .018 .136 1.655 Leader gender .142 .086 .123 1.657 Step 2 .784 .615*** .553*** Employee gender .026 .051 .024 .511 Employee age .004 .016 .017 .251 Education -.003 .041 -.004 -.085 Industry .007 .005 .073 1.551 Organizational tenure -.003 .007 -.025 -.362 Employment with current manager .004 .012 .018 .342 Leader gender .138 .055 .119* 2.501 Leader inclusiveness .634 .038 .763*** 16.730 Statistical significance: * p<.05, ** p<.01, ***p<.001

Secondly, hierarchical multiple regression was computed to investigate the effects of the control variables and leader inclusiveness on turnover intention. As summarized in Table 3, in the first step, we entered the seven control variables, and the model is significant, with F(7, 196)= 5.337; p<.001. There are significant relationships between turnover intention and employee age (β= -.227, p<.05), industry (β= .184, p<.01), as well as employment with current manager (β= -.173, p<.05). After entering leader inclusiveness in the second step, the whole model is significant and explains 28.9% variance of turnover intention, with F(8, 195)= 9.914; p<.001. Thus, leader inclusiveness explains an additional 12.9% variance of turnover intention, with F(1, 195)= 35.395; p<.001). In the final model, turnover intention is

(32)

significantly related with employee age (β= -.273, p<.01), industry (β=.163, p<.05), and leader inclusiveness (β= -.369, p<.001).

Table 3. Hierarchical Regression Model of Leader Inclusiveness on Turnover Intention

R R2 _R2 Change B SE β t Step 1 .400 .160*** .160 Employee gender .033 .123 .019 .269 Employee age -.092 .040 -.227* -2.306 Education .195 .100 .133 1.946 Industry .030 .011 .184** 2.648 Organizational tenure .010 .017 .055 .550 Employment with current manager -.065 .029 -.173* -2.230 Leader gender -.235 .135 -.122 -1.738 Step 2 .538 .289*** .129*** Employee gender .108 .114 .061 .946 Employee age -.111 .037 -.273** -3.002 Education .127 .093 .086 1.364 Industry .027 .011 .163* 2.544 Organizational tenure .007 .016 .038 .410 Employment with current manager -.044 .027 -.116 -1.610 Leader gender -.231 .125 -.120 -1.857 Leader inclusiveness -.509 .086 -.369*** -5.949 Statistical significance: * p<.05, ** p<.01, ***p<.001

Thirdly, the same process of hierarchical regression was performed to test how the control variables and job satisfaction predict turnover intention respectively. As illustrated in Table 4, the first step of this regression produces the same results as the first step of the last hierarchical regression of the control variables, leader inclusiveness, and turnover intention. After the entry of job satisfaction in the second step, the whole model explains 39.1% variance of turnover intention, with F(8, 195)= 15.676; p<.001. Therefore, job satisfaction explains an additional 23.1% variance of turnover intention, with F(1, 195)= 74.112; p<.001.

(33)

In the final model, turnover intention is significantly related with employee age (β= -.266, p<.01), industry (β=.199, p<.001), and job satisfaction (β= -.497, p<.001). The finding thus supports Hypothesis 2, stating that job satisfaction and turnover intention are negatively related.

Table 4. Hierarchical Regression Model of Job Satisfaction on Turnover Intention

R R2 _R2 Change B SE β t Step 1 .400 .160*** .160 Employee gender .033 .123 .019 .269 Employee age -.092 .040 -.227* -2.306 Education .195 .100 .133 1.946 Industry .030 .011 .184** 2.648 Organizational tenure .010 .017 .055 .550 Employment with current manager -.065 .029 -.173* -2.230 Leader gender -.235 .135 -.122 -1.738 Step 2 .626 .391*** .231*** Employee gender .132 .106 .074 1.245 Employee age -.108 .034 -.266** -3.168 Education .122 .086 .083 1.423 Industry .033 .010 .199*** 3.356 Organizational tenure .004 .015 .025 .294 Employment with current manager -.040 .025 -.106 -1.584 Leader gender -.117 .116 -.061 -1.011 Leader inclusiveness -.825 .096 -.497*** -8.609 Statistical significance: * p<.05, ** p<.01, ***p<.001

Finally, we conducted hierarchical regression to examine the impact of leader inclusiveness on OCB, after controlling the control variables. As shown in Table 5, in Step 1, none of the seven control variables has a significant relationship with OCB. The model is insignificant, with F(7, 196)= .827; p>.05. In the second step, leader inclusiveness was

(34)

entered as a predictor, the whole model is statistically significant, with F(8, 195)= 5.660; p<.001, and explains 18.8% variance of OCB. Therefore, the introduction of leader inclusiveness explains an additional 16.0% variance of job satisfaction, with F(1, 195)= 38.392; p<.001. In the final model, only leader inclusiveness is significantly related with OCB (β=.410, p<.001). The finding supports Hypothesis 4, stating that leader inclusiveness is positively related to OCB.

Table 5. Hierarchical Regression Model of Leader Inclusiveness on OCB

R R2 _R2 Change B SE β t Step 1 .169 .029 .029 Employee gender .024 .055 .033 .443 Employee age .019 .018 .115 1.085 Education .005 .045 .008 .103 Industry -.008 .005 -.118 -1.573 Organizational tenure -.010 .008 -.142 -1.312 Employment with current manager .020 .013 .126 1.505 Leader gender .033 .061 .041 .538 Step 2 .434 .188*** .160*** Employee gender -.010 .051 -.014 -.203 Employee age .028 .016 .166 1.711 Education .036 .041 .059 .871 Industry -.006 .005 -.095 -1.378 Organizational tenure -.009 .007 -.123 -1.237 Employment with current manager .010 .012 .062 .808 Leader gender .031 .055 .039 .559 Leader inclusiveness .236 .038 .410*** 6.196 Statistical significance: * p<.05, ** p<.01, ***p<.001

The impact of leader inclusiveness on employees' turnover intention and organizational citizenship behavior in Chinese work context

Master Thesis