Walumbwa 2010

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/44605960
Servant Leadership, Procedural Justice Climate, Service Climate, Employee

Attitudes, and Organizational Citizenship Behavior: A Cross-Level
Investigation
Article in Journal of Applied Psychology · May 2010

DOI: 10.1037/a0018867 · Source: PubMed
CITATIONS READS
448 911
3 authors, including:
Chad Hartnell Adegoke Oke

Georgia State University Arizona State University
21 PUBLICATIONS 1,668 CITATIONS 38 PUBLICATIONS 2,767 CITATIONS
SEE PROFILE SEE PROFILE
All content following this page was uploaded by Chad Hartnell on 20 March 2020.
The user has requested enhancement of the downloaded file.

Journal of Applied Psychology © 2010 American Psychological Association
2010, Vol. 95, No. 3, 517–529 0021-9010/10/$12.00 DOI: 10.1037/a0018867
Servant Leadership, Procedural Justice Climate, Service Climate,

Employee Attitudes, and Organizational Citizenship Behavior:
A Cross-Level Investigation
Fred O. Walumbwa, Chad A. Hartnell, and Adegoke Oke

Arizona State University
This study tests the influence of servant leadership on 2 group climates, employee attitudes, and
organizational citizenship behavior. Results from a sample of 815 employees and 123 immediate
supervisors revealed that commitment to the supervisor, self-efficacy, procedural justice climate, and
service climate partially mediated the relationship between servant leadership and organizational citi-
zenship behavior. Cross-level interaction results revealed that procedural justice climate and positive
service climate amplified the influence of commitment to the supervisor on organizational citizenship
behavior. Implications of these results for theory and practice and directions for future research are
discussed.
Keywords: servant leadership, leadership, climate, organizational citizenship behavior, cross-level

research
Employees who contribute to an organization beyond their although research has demonstrated that specific climate can affect
formal job requirements have been the subject of increasing inter- employee attitudes (Liao & Chuang, 2007; Schulte, Ostroff, Shmu-
est among scholars and managers alike (Grant & Mayer, 2009; lyian, & Kinicki, 2009; Walumbwa, Wu, & Orwa, 2008), research
Hoffman, Blair, Meriac, & Woehr, 2007; Ilies, Nahrgang, & has yet to systematically consider the interactive effects of em-
Morgeson, 2007). One reason why employees’ discretionary work ployee attitudes and specific climates on employee behaviors, such
behavior has garnered such attention is that it is strongly associated as citizenship behavior.
with task performance (␳ ⫽ .74; Hoffman et al., 2007). Indeed, a Servant leadership is an emerging perspective in which to ex-
recent meta-analysis indicated that organizational citizenship be- tend theory in OCB research. Hale and Fields (2007) defined
haviors (OCBs) are positively associated with employee perfor- servant leadership as “an understanding and practice of leadership
mance and organizational productivity, efficiency, and customer that places the good of those led over the self-interest of the leader,
satisfaction (N. P. Podsakoff, Whiting, Podsakoff, & Blume, emphasizing leader behaviors that focus on follower development,
2009). These outcomes reiterate the importance of developing and and de-emphasizing glorification of the leader” (p. 397). It em-
fostering cooperative and helpful behaviors in organizational set- phasizes leaders’ moral behavior, protecting followers from self-
tings. interested leaders pursuing ends for their own selfish gain (Gra-
Although individuals’ citizenship behaviors are a well- ham, 1991; Liden, Wayne, Zhao, & Henderson, 2008). Servant
researched domain in organizational behavior, relatively few stud- leaders also recognize their moral responsibility to the success of
ies have addressed the group-level factors that facilitate their the organization as well as to the success of their subordinates, the
emergence (Bommer, Dierdorff, & Rubin, 2007; Kirkman, Chen, organization’s customers, and other stakeholders (Ehrhart, 2004).
Farh, Chen, & Lowe, 2009). Schnake and Dumler (2003, p. 286) Consequently, such leaders imbue the importance of service within
cogently noted, “Despite the potential for group-level and mixed-
as well as outside of the organization. Thus, servant leaders’
or cross-level theory in OCB research, relatively few studies have
pervasive focus on developing others is a cogent domain in which
examined the effects of group and organizational phenomena on
to investigate leadership’s association with employee OCB.
individual-level OCB.” This gap in the literature is surprising
Accordingly, we had two objectives for this study. First, al-
given contextual factors’ importance in influencing individual
though previous research has demonstrated a positive relationship
attitudes and behaviors in organizations (Johns, 2006). Further,
between servant leadership and employee OCB (Ehrhart, 2004;
Liden et al., 2008; Neubert, Kacmar, Carlson, Chonko, & Roberts,
2008), relatively few studies have focused on understanding how
Fred O. Walumbwa and Chad A. Hartnell, Department of Management, servant leadership is related to employee OCB. Further, research
W. P. Carey School of Business, Arizona State University; Adegoke Oke, has yet to consider both individual and situational mechanisms
Department of Supply Chain Management, W. P. Carey School of Busi-
through which servant leadership influences employee OCB. To
ness, Arizona State University.
Correspondence concerning this article should be addressed to Fred O. address these theoretically and practically relevant research issues,
Walumbwa, Department of Management, W. P. Carey School of Business, our first main objective was to examine the extent to which
Arizona State University, Tempe, AZ 85287-4006. E-mail: fred.walumbwa@ employee attitudes (i.e., affective commitment to the supervisor
asu.edu and self-efficacy) and two specific group climates (i.e., procedural
517
518 WALUMBWA, HARTNELL, AND OKE
justice climate and service climate) mediate the relationship be- vant leadership and climate, affect between-group differences in
tween servant leadership and OCB. In particular, we developed employee OCB, as opposed to within-group differences. This is
and tested a cross-level model in which servant leadership influ- not to say that OCB, as conceptualized and operationalized in this
ences two employee attitudes and two group climates simulta- study, is a group-level construct; instead, it is an individual-level
neously. This model also explains how these individual and situ- construct that varies both within and between groups.
ational variables, in turn, influence employee OCB.
Our second objective was to examine the possible moderating Theory and Hypothesis Development
roles of procedural justice climate and service climate on the
relationship between employee attitudes and citizenship behavior.
Servant Leadership and Related Leadership Theories
Although previous research has demonstrated a cross-level link
between group processes (e.g., procedural justice climate) and Servant leadership represents a model of leadership that is both
individual OCB (Naumann & Bennett, 2000; Walumbwa, Wu, & inspirational and contains moral safeguards (Graham, 1991;
Orwa, 2008), group process variables may also serve as cross-level Greenleaf, 1977). Spears (2004) delineated 10 characteristics
moderators of the relationship between employee attitudes and common among servant leaders: listening, empathy, healing,
OCB at the individual level (Ehrhart, 2004). Thus, we extend awareness, persuasion, conceptualization, foresight, steward-
previous OCB research by highlighting the importance of group ship, growth, and building community. Intuitively, servant leader-
climates’ cross-level moderating effects in the relationship be- ship overlaps with other idealized notions of leadership (e.g.,
tween employee attitudes and OCB and, in doing so, help point out charismatic/transformational, ethical, authentic, and spiritual lead-
ways in which organizations can foster higher levels of citizenship ership) by exhibiting the following behaviors: role modeling, in-
behaviors. Figure 1 summarizes our theoretical model. spirational communication, and altruism (Brown & Treviño,
In the present study and consistent with past research (e.g., 2006). However, servant leadership is theoretically distinct in
Ehrhart, 2004; Liden et al., 2008), we conceptualize servant lead- several important ways.
ership as ambient behavior directed toward the leader’s entire First, servant leadership includes a moral component, a concept
work unit that is a common stimulus shared among group members lacking from popular leadership theories, such as charismatic and
(Chen & Bliese, 2002; Kirkman et al., 2009; Liao & Chuang, transformational leadership (Bass, 1985). Second, although ethical
2007). In contrast to considering servant leadership at the group level, and authentic leadership contain moral dimensions (Brown &
we examined OCB exclusively at the individual level of analysis. To Treviño, 2006; Walumbwa, Avolio, Gardner, Wernsing, & Peter-
be clear, OCB involves individual behavior that can vary as a function son, 2008), servant leadership is uniquely concerned with the
of both individual-level differences (e.g., employee attitudes and success of all organizational stakeholders. Third, servant leaders
perceptions) and contextual/group differences (e.g., leadership and act in the best interest of the follower, precluding manipulative,
climate). An important goal of our study was to identify both self-interested behavior. In other words, servant leadership is fo-
individual- and group-level mechanisms through which servant cused on followers’ individual growth and development (Smith,
leadership influences employee OCB. Whereas individual-level Montagno, & Kuzmenko, 2004). Conversely, other leadership
mechanisms (i.e., commitment to the supervisor and self-efficacy) behaviors, such as charismatic and transformational leadership
are likely to explain individual-level differences in employee behaviors, focus on inspiring and engaging followers as the means
OCB, group-level mechanisms (i.e., procedural justice climate and to attain mission-focused ends through connecting the goals to
service climate) are likely to explain group-level differences in valued aspects of the followers’ self-concept (Bass, 1985). Finally,
employee OCB. Because employees are exposed to similar lead- servant leaders engage in self-reflection to attenuate the leader’s
ership behavior and similar policies, procedures, and practices hubris (Graham, 1991), a characteristic lacking in authentic, ethi-
within a unit, they share perceptions of the unit’s contextual cal, and transformational leadership behaviors.
characteristics. Consequently, group-level variables, such as ser- Some empirical evidence supports the distinctiveness of servant
leadership from related leadership theories. For example, Ehrhart
(2004) reported that servant leadership significantly predicted an
additional 5% of the variance in employee commitment, 7% of the
variance in satisfaction with supervisor, 4% of the variance in
perceived supervisor support, and 8% of the variance in procedural
justice above and beyond that of both leader–member exchange
and transformational leadership. Similarly, Liden et al. (2008)
reported that servant leadership behavior explained variance in
citizenship behavior and in-role performance beyond that pre-
dicted by leader–member exchange and transformational leader-
ship. The evidence regarding the incremental validity of servant
leadership in explaining variance in employee attitudes and behav-
iors beyond leader–member exchange and transformational lead-
ership is important given how strongly leader–member exchange
and transformational leadership behaviors are associated with pos-
itive employee attitudes and behaviors (Ilies et al., 2007; Judge &
Piccolo, 2004). Thus, theory and empirical data to date support the
Figure 1. Hypothesized model. notion that servant leadership is a unique leadership theory that can
SERVANT LEADERSHIP, CLIMATE, AND EMPLOYEE OUTCOMES 519
extend researchers’ knowledge about leadership processes and Servant Leadership, Self-Efficacy, Commitment, and
outcomes. Organizational Citizenship Behavior
Servant leadership emphasizes listening, empathy, stewardship,

Servant Leadership and Organizational
and awareness to develop followers holistically as an end in itself.
Citizenship Behavior We draw on social learning theory (Bandura, 1977) and social
Organizational citizenship behavior is defined as “individual exchange theory (Blau, 1964) to illuminate how servant leadership
behavior that is discretionary, not directly or explicitly recognized relates to employee attitudes and consequently to OCB. According
by the formal reward system, and that in the aggregate promotes to social learning theory, individuals learn by “paying attention to
the effective functioning of the organization” (Organ, 1988, p. 4). and emulating the attitudes, values and behaviors of attractive and
More recently, citizenship behaviors have been articulated as vol- credible models” (Brown & Treviño, 2006, p. 597). Followers
untary behaviors that transcend an employee’s specified role re- perceive servant leaders as attractive because servant leaders em-
quirements and are not formally rewarded by the organization phasize the development of others and place the good of those led
(Organ, Podsakoff, & MacKenzie, 2006). Although researchers over their own interests (Hale & Fields, 2007). This notion is
have distinguished between several types of OCB (e.g., Lavelle, supported by the individual-differences perspective of leadership,
Rupp, & Brockner, 2007; Lee & Allen, 2002), recent studies which suggests that employees’ attitudes are influenced by their
support a single-factor model of OCB (Bommer et al., 2007; differential perceptions of and cognitive categorizations of leader-
Hoffman et al., 2007; LePine, Erez, & Johnson, 2002). Specifi- ship behaviors (Hall & Lord, 1995). Drawing on these perspec-
cally, these studies suggested that treating OCB as a global con- tives, we expect servant leadership behavior to serve as a salient
struct “represents a more accurate estimate of the relationship context that ameliorates followers’ positive attitudes, resulting in
between OCB and other variables” (Hoffman et al., 2007, p. 562). more OCB.
Consistent with these recent empirical summaries of the OCB In the present study, we consider two types of employee atti-
literature, we conceptualize and operationalize OCB as a global tudes as particularly important in explaining the influence of
construct, because it captures both OCB directed toward individ- servant leadership on employee OCB: self-efficacy and commit-
uals (i.e., assisting coworkers with individual issues, such as ment to the supervisor. We chose to focus on self-efficacy and
helping a coworker run a special computer program or sending commitment to the supervisor because they are complementary
birthday greetings to coworkers in the office) and OCB directed to attitudes that are especially salient in service organizations. More-
the group (i.e., behaviors that are meant to improve the workings over, self-efficacy and commitment to the supervisor have both
of the group and attendance at functions that help the group image, been found to enhance employee behavior across a number of
such as favorably speaking about the group to outsiders or offering different performance domains, including citizenship behavior
new innovative ideas to improve the functioning of the group). (Becker & Kernan, 2003; Todd & Kent, 2006). Self-efficacy
Servant leadership behavior creates a pervasive social context engages an employee’s cognitive appraisal of being able to ac-
that positively affects employees’ attitudes and behavior. Such complish a given task (Bandura, 1997). Commitment to the super-
leader behavior provides situational cues from which followers’ visor, on the other hand, represents how an employee feels about
interpret and understand their environment (Takeuchi, Chen, & his or her supervisor. Analyzing these two attitudes in concert
Lepak, 2009), thus influencing employee attitudes and behaviors. illuminates how complementary employee attitudes (tapping cog-
Although these perspectives indicate how leaders’ behavior toward nitive and affective attitudes) mediate the effect of servant lead-
the group influences followers, social exchange theory (Blau, ership.
1964) explicates why followers respond to their leaders’ behavior. Employee self-efficacy. Self-efficacy is defined as the “belief
According to social exchange theory (Blau, 1964), the quality of in one’s capabilities to organize and execute the courses of action
social interactions induces unspoken obligations to return favors to required to produce given attainments” (Bandura, 1997, p. 3).
those who have acted in one’s interest. This relationship is often Servant leaders develop followers’ self-efficacy through enhanc-
depicted as a form of unspoken reciprocity in which people seek to ing their followers’ technical competence. They also demonstrate
repay favors until a perceived balance of exchanges exists (Blau, and disseminate the knowledge required to find successful solu-
1964). Servant leaders transcend self-interest, express genuine care tions to work problems. Furthermore, they are attentive to their
and concern, and act in the best interest of their followers (i.e., followers’ personal development through understanding their ex-
developing their skills, knowledge, and abilities), thereby creating isting skills, knowledge, needs, goals, and abilities. Servant leaders
a social context in which followers reciprocate by engaging in also provide opportunities for followers to develop new skills and
extrarole prosocial behavior. When leaders engage in selfless, assist them in attaining their career goals. Therefore, such leaders
supporting, and developmental behavior across all followers in are more likely to enable their followers to be successful by
their work group, followers reciprocate. More specifically, follow- developing and improving their repertoire of skills, knowledge,
ers engage in OCB toward the leader as a specific form of ex- and abilities, thereby increasing self-efficacy. In support, Chen and
change, and followers perform OCB as a general form of exchange Bliese (2002) argued and found that leadership plays an important
to support and sustain the positive social environment in response role in enhancing self-efficacy because leaders help clarify roles
to the leader’s behavior. On the basis of this reasoning, we propose and provide social support to employees.
the following: From a social learning perspective, servant leaders’ investment
in developing their followers, by ameliorating their self-efficacy,
Hypothesis 1: Servant leadership positively relates to organizational prompts followers to emulate their leaders’ selfless, prosocial
citizenship behavior. behavior. Moreover, highly efficacious individuals are more likely
to engage in OCB because they are confident in their competence. source of the behavioral data on which employees base their views
Consistent with this argument, Todd and Kent (2006) found that of organizational objectives and policies” (Mayer, Nishii, Schnei-
self-efficacy was positively associated with an employee’s propen- der, & Goldstein, 2007, p. 931).
sity to engage in OCB. Thus, we expect employee self-efficacy to Although early research analyzed organizational climate’s in-
mediate the relationship between servant leadership and OCB. fluence on organizational and employee outcomes, multiple cli-
However, because we argue later that the influence of servant mates can operate concomitantly (Schneider, 1990). Hence, cli-
leadership on OCB may also be carried through other mediators, mates can be defined more specifically and can refer to more
such as commitment to the supervisor, we propose partial media- specific outcomes. The literature delineating specific climates has
tion rather than complete mediation. enhanced the knowledge of contextual influences on outcomes at
Hypothesis 2A: Employee self-efficacy partially mediates the relation-
the individual and work group levels (cf. Kuenzi & Schminke,
ship between servant leadership and organizational citizenship behavior. 2009). Similar to Schneider’s (1990) argument that multiple cli-
mates can exist in concert, we contend that servant leadership
Employee commitment to the supervisor. Although com- behavior can promote multiple climates simultaneously. Servant
mitment has historically been separated into different foci (Allen leadership advocates building trust through altruistically serving
& Meyer, 1990), affective commitment seems to be the most all stakeholders: employees, the organization, their community,
robust predictor of behavioral criteria (Lavelle et al., 2007) and is and society (Liden et al., 2008). As such, two climates become
the focus in the present study. Affective commitment is defined as apparent from servant leadership’s core tenants: procedural justice
an emotional attachment to, identification with, and involvement climate and service climate.
in the organization (Allen & Meyer, 1990). Servant leaders affect Procedural justice climate. Procedural justice climate is de-
followers’ affective commitment by investing, without a hidden fined as “distinct group-level cognition about how a work group as
agenda, in developing their followers and by considering employ- a whole is treated” (Naumann & Bennett, 2000, p. 882). Three
ees’ input before making important decisions. Consistent with important criteria undergirding procedural justice perceptions in-
social exchange theory (Blau, 1964), leader behavior that is con- clude the extent to which the process (a) is ethical and moral, (b)
cerned with follower development and involvement imbues posi- is consistently applied, and (c) includes the opportunity for em-
tive psychological and affective benefits within followers (e.g., ployees to voice their opinions and influence the outcome (Lev-
gratitude and trust), thereby fostering commitment to the supervi- enthal, 1980). Accordingly, servant leaders may be especially
sor (Liden et al., 2008). Servant leadership thus creates a positive adept at influencing a group’s collective perception of procedural
social exchange through affective commitment to the supervisor, justice because they maintain consistently high ethical standards
compelling followers to reciprocate through engaging in citizen- across group members and seek input from and attempt to reach
ship behavior. consensus among employees on important decisions (Ehrhart,
Research has also documented that affective commitment is 2004; Liden et al., 2008). Servant leaders also influence group
associated with several important outcomes, including OCB members’ perceptions of fair treatment when they apply their
(Lavelle et al., 2007). For example, Becker and Kernan (2003) values and beliefs—in terms of setting ethical standards, seeking
found that commitment to the supervisor was positively related to employee input, and developing quality relationships— consis-
employees’ prosocial behaviors. Thus, affective commitment to tently among all group members. When leaders set high ethical
the supervisor may be an important mechanism through which
standards, they encourage followers to voice opinions and sugges-
servant leadership influences OCB. Consistent with Hypothesis
tions about work-related processes and work context (Walumbwa
2A, we propose partial mediation rather than complete mediation.
& Schaubroeck, 2009), thereby enhancing procedural justice cli-
Hypothesis 2B: Employee commitment to the supervisor partially mate. Previous research supports this by suggesting that leader
mediates the relationship between servant leadership and organiza- personality and servant leadership behavior are positively related
tional citizenship behavior. to employees’ perceptions of procedural justice climate (e.g.,
Ehrhart, 2004; Mayer et al., 2007).
Servant Leadership, Justice Climate, Service Climate, When employees’ perceive that they are being treated fairly,
and Organizational Citizenship Behavior they are more likely to engage in extrarole behaviors (Naumann &
In addition to affecting OCB through influencing employee Bennett, 2000; Walumbwa, Wu, & Orwa, 2008). According to the
attitudes, servant leadership may also influence OCB through group-value and relational models of procedural justice, fair pro-
fostering specific work group climates. Kozlowski and Doherty cedures signal to employees that they are valued by their leaders
(1989) maintained that climate formation is an “implicit aspect of (Tyler & Blader, 2003). Employees thus engage in OCB to en-
leadership processes” (p. 547). Naumann and Bennett (2000) ex- hance and maintain the positive psychological benefits that ema-
tended Kozlowski and Doherty’s argument by postulating, nate from identification with their leader and coworkers (Tyler &
Blader, 2003). Lind and Earley (1992) reported that fair proce-
Climates are likely to develop in work groups composed of employees dures induce feelings of harmony that subsequently influence
sharing a supervisor who exposes them to the same policies, practices, employees to exhibit positive perceptions toward coworkers. Con-
and procedures. In other words, supervisors as “climate engineers” are
sistent with this justice research, we posit that procedural justice
likely to shape the meaning employees attribute to these organiza-
climate is related to OCB. Thus, we expect procedural justice
tional characteristics. (p. 883)
climate to act as a mediator in the relationship between servant
This notion is based on the idea that “leaders play an important leadership and OCB. However, because servant leadership may
role in the development of climate, for they are the immediate influence employee OCB through other contextual mechanisms,
such as service climate (as we discuss later), we propose partial reciprocate, whereas the climate delineates the currency, or the
mediation rather than complete mediation. valued behavior, for the exchange. Similarly, we argue that em-
ployees who are highly efficacious or committed to their supervi-
Hypothesis 3A: Procedural justice climate partially mediates the re- sor are likely to be motivated to engage in prosocial behavior, with
lationship between servant leadership and organizational citizenship this positive relationship enhanced by procedural justice climate
behavior.
and positive service climate.
Service climate. Service climate is defined as “employees’ Procedural justice climate indicates both the correct ways of
shared perceptions of the policies, practices, and procedures that doing things and the valued behavior in a work group in terms of
are rewarded, supported, and expected concerning customer ser- high ethical standards. A high procedural justice climate cultivates
vice” (Schneider, Salvaggio, & Subirats, 2002, p. 222). The rela- a positive affective environment by fostering respectful relation-
tionship between servant leadership and service climate has not ships, thus increasing employees’ perceptions that OCB will be
been articulated in previous research. However, drawing on related positively received and reciprocated by others (Tangirala & Ra-
research positively linking service and transformational leadership manujam, 2008). Employees who are committed to their supervi-
behaviors to service climate (e.g., Liao & Chuang, 2007; Schnei- sor and have high self-efficacy thus engage in OCB to contribute
der, Ehrhart, Mayer, Saltz, & Niles-Jolly, 2005), we expect servant to and reinforce the positive affective tone propagated by a climate
leadership to positively influence group service climate. with a high level of procedural justice. Similarly, a climate that
Spears (1998) noted that servant leadership “emphasizes in- places a high value on service influences employee instrumentality
creased service to others, a holistic approach to work, promoting a and subsequent behavior by articulating service behaviors as the
sense of community, and the sharing of power in decision-making” valued currency within the social context (Schneider, 1990).
(p. 3). Servant leaders mold the group’s service climate by (a) Therefore, positive service climate is also likely to promulgate a
imbuing service values to the group (i.e., personal integrity, trust, positive and cooperative context by encouraging positive ex-
building relationships, and helping others grow and succeed; Liden changes among group members. As a result, employees engage in
et al., 2008) and (b) accentuating aspects of the organization’s OCB to fulfill social normative expectations and to attain positive
existing policies, practices, and procedures that most reinforce the rewards.
servant leaders’ beliefs. Furthermore, servant leaders accentuate In contrast, inconsistent and biased procedures lead employees
the importance of serving multiple stakeholders, thus expanding to perceive that they are not equally valued, which has a negative
service climate’s traditionally singular focus on the external cus- impact on their attitudes (Cropanzano, Byrne, Bobocel, & Rupp,
tomer to organizational colleagues, the community, and society 2001). A climate with a low level of procedural justice thus
(Liden et al., 2008). attenuates the extent to which employees express concern and
Much of the service climate research supports the notion that courtesy for other employees who are perceived to be treated more
employees who experience positive service climate are more likely favorably. Said differently, employees view OCB as an ineffective
to provide positive customer service experiences to their customers currency in work groups with a low procedural justice climate
(Liao & Chuang, 2007; Schneider, White, & Paul, 1998). Individ- because it will not be reciprocated or induce more positive social
uals who are subjected to the same norms, policies, and procedures interaction. Moreover, employees in work groups with a low
that emphasize and reward positive customer service experiences procedural justice climate are more likely to feel that their opinions
may be inclined to translate that experience within the context of are not sought or valued in shaping group outcomes (Colquitt,
the group. Furthermore, servant leaders model positive service- 2001). Hence, a low procedural justice climate mitigates employ-
oriented behavior to multiple organizational stakeholders, provid- ees’ willingness to perform OCB because they believe that their
ing a salient context for employees to deliver prosocial behavior. behavior is not a valued input that will improve their work group’s
Indeed, Schneider et al. (2005) reported that service climate is inequitable social context. Likewise, committed and highly effica-
associated with OCB. Thus, we expect service climate to at least cious employees are less likely to perform OCB in a low service
partially mediate the relationship between servant leadership and climate because they are less certain that prosocial behavior is a
OCB. valued currency that will result in desired outcomes. As such, they
are less likely to engage in OCB; instead, they may direct their
Hypothesis 3B: Service climate partially mediates the relationship attention to other behaviors associated with more explicit social
between servant leadership and organizational citizenship behavior. and financial rewards.
Some empirical evidence supports the potential moderating role
Climate as a Moderator of Individual-Level of procedural justice climate and service climate. Demonstrating
Relationships procedural justice climate’s positive normative effects, Tangirala
and Ramanujam (2008) examined how procedural justice climate
Ehrhart (2004) suggested that group-level process variables may influences employees’ decisions to intentionally withhold critical
act as cross-level moderators in the relationship between employee work information (i.e., employee silence). They found that a high
attitudes and OCB. Because work group climate directs behavior procedural justice climate strengthened the negative relationship
by signaling to employees what the group values, desires, expects, between work group identification and employee silence as well as
and rewards (Ostroff, Kinicki, & Tamkins, 2003; Schneider, 1990; between professional commitment and employee silence. Liao and
Schneider et al., 2005), climates that foster positive affective Chuang (2007) examined service climate as a moderator in the
environments are more likely to amplify the positive influence of relationship between transformational leader behavior and em-
employee attitudes on OCB. Hofmann, Morgeson, and Gerras ployee service performance and reported that positive service
(2003) posited that social exchange creates the motivation to climate enhanced the relationship between transformational leader
behavior and employee service performance. More recently, work experience of 11 years. All supervisors had an equivalent of
Walumbwa, Peterson, Avolio, and Hartnell (in press) found that a U.S. community college or a college degree. Employee data for
the relationship between follower psychological capital (a core this study were collected at two times separated by about 3 weeks
construct comprising hope, optimism, resiliency, and self-efficacy) to allow us to reduce common method bias (P. M. Podsakoff,
and employee performance was stronger when service climate was MacKenzie, Lee, & Podsakoff, 2003).
high than when it was low. These findings support the notion that Time 1. We initially distributed 1,300 envelopes to employ-
procedural justice climate and service climate may constitute po- ees. Each envelope included an introductory letter from Fred O.
tent moderators of the relationship between employee attitudes and Walumbwa and endorsement of the project from senior manage-
OCB. In view of this discussion, we expect procedural justice ment. The introductory letter explained that participation in this
climate and service climate to serve as a situational impetus to study was strictly voluntary, that only aggregated data would be
strengthen the likelihood that employees who are committed to the reported, and that all identifying information would be removed
supervisor and who are highly efficacious will engage in OCB. once data were coded. The introductory letter also explained the
Therefore, we propose the following: purpose of the study, informed participants that they would be
receiving another short survey in approximately 3 weeks, and
Hypothesis 4A: Procedural justice climate moderates the influence of
stressed the importance of responding to both surveys. Participants
employee commitment to the supervisor on organizational citizenship
also were asked to provide their names so that we could match
behavior, such that the influence of commitment to the supervisor is
more positive when procedural justice climate is high than when it is their responses at Time 1 and Time 2. Each employee was pro-
low. vided an envelope and had 3 days to complete the survey distrib-
uted on-site. We set up a central collection box for survey drop-off
Hypothesis 4B: Procedural justice climate moderates the influence of in each facility. To further heighten participants’ motivation to
employee self-efficacy on organizational citizenship behavior, such participate in this study, employees were told that a raffle would be
that the influence of self-efficacy is more positive when procedural held where three participants would be randomly selected to re-
justice climate is high than when it is low. ceive $80 (first prize), $50 (second prize), and $20 (third prize).
Hypothesis 5A: Service climate moderates the influence of employee
Participants completed personal information and rated their group
commitment to the supervisor on organizational citizenship behavior, leaders’ leadership behavior. Participants returned 877 complete
such that the influence of commitment to the supervisor is more surveys at Time 1, representing a response rate of 68%.
positive when service climate is high than when it is low. Time 2. Approximately 3 weeks later, the 877 respondents
who participated at Time 1 were asked to complete a survey
Hypothesis 5B: Service climate moderates the influence of employee measuring commitment to the supervisor, self-efficacy, procedural
self-efficacy on organizational citizenship behavior, such that the justice climate, and service climate. Once again, employees pro-
influence of self-efficacy is more positive when service climate is
vided their names. A total of 815 matched usable surveys were
high than when it is low.
returned (an effective overall response rate of 63%). We compared
data for the final usable cases and the cases that dropped after
Methods Time 1. No significant differences between these respondents in
terms of age, gender, and tenure were detected.
Context, Sample, and Procedures Approximately 2 weeks after Time 2, we identified supervisors
of each work group with the help of the Human Resources De-
We used data collected from seven multinational companies partment in each organization. Each supervisor provided OCB
operating in Kenya to test the proposed theoretical framework. ratings for their respective direct reports. All 123 supervisors
Kenya is a member of the World Trade Organization and has completed and returned their surveys to Fred O. Walumbwa.
increasingly become more attractive to foreign investors, including Supervisors were guaranteed anonymity, told that participation in
American and other multinational companies around the world the study was strictly voluntary, and told that all identifying
(Walumbwa, Orwa, Wang, & Lawler, 2005). information would be removed.
Our participants consisted of 815 full-time employees, consti-
tuting 123 work groups who completed surveys during regular
work hours. We considered employees to be members of a work Measures
group when they had the same supervisor. As a result, some
departments had multiple work groups (ranging from 5 to 8). The Unless otherwise indicated, all measures were answered on a
survey was written in English, Kenya’s official language. 5-point Likert scale (1 ⫽ strongly disagree to 5 ⫽ strongly agree).
The average employee’s age was 34 years, with mean work Servant leadership. We asked individual employees to use
experience of 5 years across the seven organizations. Of the par- Ehrhart’s (2004) 14-item Servant Leadership Scale (␣ ⫽ .91) to
ticipants, 52% were male, and 48% were female; they were relatively rate their work group supervisor. Sample items are “My work
well educated on average (more than 90% had an equivalent of a U.S. group supervisor spends the time to form quality relationships with
community college diploma or a college degree). Participants were employees” and “My work group supervisor creates a sense of
from a broad cross-section of job types. Across organizations, the community among employees.” Because our interest in the present
most common job categories were administrative/clerical support study is on the overall pattern of servant leadership behavior
(24%), customer service (18%), managerial/professional (15%), displayed at the work group as a whole and to be consistent with
human resource/training (13%), and technical/computer program- Ehrhart, we averaged within work groups employees’ evaluations
ming (9%). The average supervisor’s age was 38 years, with mean of the group supervisor’s leadership behavior.
To assess the appropriateness of aggregating individual scores As with procedural justice climate, we assessed the appropri-
to the work group, we calculated within-group agreement (rwg; ateness of aggregating individual scores to the work group level,
James, Demaree, & Wolf, 1984), the intraclass correlations and results revealed that the average rwg was .76 (Mdn ⫽ .83), the
(ICC[1]), and the reliability of the means (ICC[2]; Bliese, 2000). ICC(1) was .37, and the ICC(2) was .79. Further, the ANOVA
The average rwg was .86 (Mdn ⫽ .90), the ICC(1) was .28, and the indicated that the group effect was significant ( p ⬍ .01), suggest-
ICC(2) was .72. In addition, the analysis of variance (ANOVA) on ing that it was statistically appropriate to aggregate service climate
which the ICC(1) value was based indicated that the group effect to the work group.
was significant ( p ⬍ .01), providing empirical justification for Organizational citizenship behavior. Supervisors completed
aggregating servant leadership. the 16-item OCB measure (developed and validated by Lee &
Self-efficacy. We used 10 items developed and validated by Allen, 2002) to indicate the extent to which they agreed with the
Riggs, Warka, Babasa, Betancourt, and Hooker (1994) to measure statements about their direct reports’ OCB. Although Lee and
self-efficacy (␣ ⫽ .89). Sample items are “I have confidence in my Allen’s (2002) scale was designed to capture OCB directed toward
ability to do my job” and “I have all the skills needed to perform individuals and OCB directed toward the organization, respec-
my job very well” (1 ⫽ very inaccurate to 5 ⫽ very accurate). tively, we combined these two scales to measure overall OCB. We
Commitment to the supervisor. We used a six-item revised conducted a principal factor analysis of the 16 items and found
Affective Commitment Scale (e.g., Meyer, Allen, & Smith, 1993) only one factor with an eigenvalue greater than 1.00, with a
to measure commitment to the supervisor (␣ ⫽ .75). This revision composite internal reliability (␣) of .91. Therefore, we used the 16
was based on the original eight-item measure developed by Allen items to create an index of OCB. Sample items are “This employee
and Meyer (1990) to increase item clarity and scale brevity. In shows genuine concern and courtesy toward coworkers, even
adapting this scale to the supervisor, we reworded the items to under the most trying business or personal situations” and “This
specify supervisor as the target of commitment (see also Becker & employee offers ideas to improve the functioning of the work
Kernan, 2003). Sample items are “I would be very happy to spend unit.”
the rest of my career with this supervisor” and “I do not feel
‘emotionally attached’ to this supervisor” (reverse coded). Analytic Strategy
Procedural justice climate. Procedural justice climate (␣ ⫽
.72) was measured with a seven-item scale (e.g., Colquitt, 2001). Our theoretical model is multilevel in nature, consisting of
We asked employees to use a 5-point scale (1 ⫽ to a small extent variables at both the individual (i.e., commitment to the supervisor,
to 5 ⫽ to a large extent) to evaluate the procedures used in their self-efficacy, and organizational citizenship behavior) and group
work groups. Sample items are “To what extent are procedures in (i.e., servant leadership, procedural justice climate, and service
your work group applied consistently?” and “To what extent are climate) level of analysis. Employees in this study were also nested
procedures in your work group based on accurate information?” within supervisors (each supervisor provided ratings for at least
Walumbwa, Wu, and Orwa (2008) modified this scale to measure five employees, and there was one supervisor per group). To
procedural justice at the unit level of analysis. Therefore, consis- account for nonindependence, we used hierarchical linear model-
tent with the level of theory and past research that has theorized ing (Raudenbush & Bryk, 2002) and followed Hofmann, Griffin,
and tested procedural justice climate at the unit level (e.g., Walum- and Gavin’s (2000) procedures for testing cross-level effects.
bwa, Wu, & Orwa, 2008), we aggregated individuals’ procedural
justice climate to the work group to form the measure of proce- Results
dural justice climate.
We assessed the appropriateness of aggregating individual
scores to the work group level, and results revealed that the
Measurement Issues
average rwg was .91 (Mdn ⫽ .95), the ICC(1) was .34, and the We conducted a confirmatory factor analysis to assess the
ICC(2) was .77. The ANOVA indicated that the group effect was psychometric properties of our measures at the individual level of
significant ( p ⬍ .01), empirically supporting aggregating proce- analysis, using all items as indicators for each variable in the study.
dural justice climate to the work group. First, we examined a measurement model that included all six
Service climate. We used a seven-item scale developed by measures (a six-factor model, including servant leadership, the two
Schneider et al. (1998) to measure service climate (␣ ⫽ .86). The climates, self-efficacy, commitment, and OCB). We then tested
items asked employees to respond to each question on the basis of this six-factor measurement model against an alternative model in
what they have personally observed in their work groups (1 ⫽ which all the measures collected from employees (servant leader-
poor to 5 ⫽ excellent). Sample items are “How would you rate the ship, the two climates, self-efficacy, and commitment) were set to
recognition and rewards employees receive for the delivery of correlate at 1.0, and OCB freely estimated.
superior work and service?” and “How would you rate the tools, Results showed that the hypothesized six-factor model fit the
technology, and other resources provided to employees to support data well, ␹2(1663) ⫽ 4,318.47, p ⬍ .01; comparative fit index ⫽
the delivery of superior quality work and service?” Because prior .93, root-mean-square residual ⫽ .05, root-mean-square error of
theory and research (e.g., Liao & Chuang, 2007; Schneider et al., approximation ⫽ .04. Relative to the hypothesized model, the fit
1998) have suggested that service climate is formed through for an alternative model in which all measures collected from
bottom-up emergent processes, we aggregated individuals’ service employees were set to correlate at 1.0 and OCB freely estimated
climate ratings to the work group level to form the service climate was significantly worse, ␹2(1673) ⫽ 5358.34, p ⬍ .01; compara-
measure. tive fit index ⫽ .83, root-mean-square residual ⫽ .49, root-mean-
square error of approximation ⫽ .09. These results support our Step 3 requires that both commitment to the supervisor and self-
measures’ discriminant validity. efficacy be related to OCB, whereas Step 4 requires that the
significant relationship between servant leadership and OCB in
Hypothesis Tests Step 1 is reduced (partial mediation) or eliminated (complete
mediation) when commitment to the supervisor and self-efficacy
Table 1 shows descriptive statistics among all the study vari- are included in the same model. Results revealed that both com-
ables at the individual level. Because data for this study came from mitment to the supervisor and self-efficacy were significantly
seven organizations, we used ANOVAs to examine whether there related to OCB (commitment: ŷ ⫽ .18, p ⬍ .01; self-efficacy: ŷ ⫽
were significant differences between these organizations in terms .24, p ⬍ .01; Model 4) and that with their inclusion in the model,
of the main variables. No significant differences were detected the effect of servant leadership remained significant but was re-
between organizations on the main variables. duced in magnitude (ŷ ⫽ .19, p ⬍ .05; Model 4) when compared
with the effect in Model 1. Of interest, this was true even after we
Hierarchical Linear Modeling Results: Main and controlled for procedural justice climate and service climate. These
Mediating Effects results suggest that commitment to the supervisor and self-efficacy
Before conducting our analyses, we first examined the degree of partially mediated the relationship between servant leadership and
between-group variance in OCB. Results of a null model revealed OCB. The Sobel (1982) tests confirmed that the indirect effect
that 36% of the variance in OCB resides between groups. The (commitment to the supervisor: z ⫽ 2.35, p ⬍ .01; self-efficacy:
chi-square test revealed that the between-group variance was sig- z ⫽ 4.56, p ⬍ .01) was significant. Therefore, Hypotheses 2A and
nificant; that is, the intercept term significantly varied across 2B received support.
groups. Hypotheses 3A and 3B suggested that procedural justice climate
Hypothesis 1 predicted that servant leadership would be posi- and service climate, respectively, would partially mediate the
tively related to OCB. Table 2 shows the results. Hierarchical relationship between servant leadership and OCB. Table 2 shows
linear modeling results revealed that servant leadership signifi- the results. We followed the same four-step test procedure enu-
cantly predicted OCB (ŷ ⫽ .45, p ⬍ .01; Model 1). Thus, Hypoth- merated earlier to test for the proposed mediation in Hypothesis 3.
esis 1 is supported. We controlled for commitment to the supervisor and self-efficacy
Hypotheses 2A and 2B predicted that employee self-efficacy at the individual level of analysis because previous research sug-
and commitment to the supervisor, respectively, would partially gests they are related to employee behavior (Becker & Kernan,
mediate the relationship between servant leadership and OCB. The 2003; Liao & Chuang, 2007). As with Hypothesis 2, we included
results are also shown in Table 2. We followed the four-step both procedural justice climate and service climate in the same
procedures for mediation described by Kenny, Kashy, and Bolger model so that each could control for the effect of the other.
(1998) and controlled for procedural justice climate and service Results of Hypothesis 1 regarding the relationship between
climate at the group level of analysis because previous research servant leadership and OCB provide support for Step 1 (see Model
has suggested that they are related to employee attitudes (Liao & 1). Because servant leadership, procedural justice climate, and
Chuang, 2007; Mayer et al., 2007). In addition, to ensure that we service climate were all at the group level, we used a regular
obtained unbiased parameter estimates, we included both self- ordinary least squares analysis to test the effect of servant leader-
efficacy and commitment in the same model so that each could ship on procedural justice climate and service climate in Step 2.
control for the effect of the other. The results revealed that servant leadership positively predicted
In Step 1, servant leadership needs to be related to OCB. This procedural justice climate (␤ ⫽ .27, p ⬍ .01; adjusted R2 ⫽ .07)
requirement was supported by the results of Hypothesis 1 (see and service climate (␤ ⫽ .48, p ⬍ .01; adjusted R2 ⫽ .39). In Step
Model 1). Step 2 requires that servant leadership be related to both 3 and Step 4, we included procedural justice climate and service
commitment to the supervisor and self-efficacy. Results revealed climate as Level 2 predictors in hierarchical linear modeling to-
that servant leadership was significantly related to commitment to gether with servant leadership. As shown in Table 2, results
the supervisor (ŷ ⫽ .28, p ⬍ .01; Model 2) and self-efficacy (ŷ ⫽ revealed that both procedural justice climate (ŷ ⫽ .15, p ⬍ .01;
.42, p ⬍ .01; Model 3). Thus, the second requirement was met. Model 4) and service climate (ŷ ⫽ .20, p ⬍ .01; Model 4) were
Table 1
Means, Standard Deviations, and Correlations Among the Variables
Variable M SD 1 2 3 4 5 6
1. Servant leadership 3.15 0.74 .91

2. Commitment to the supervisor 3.61 0.71 .19ⴱⴱ .75
3. Self-efficacy 3.45 0.75 .39ⴱⴱ .44ⴱⴱ .89
4. Procedural justice climate 3.69 1.22 .17ⴱⴱ .22ⴱⴱ .27ⴱⴱ .72
5. Service climate 3.79 1.28 .45ⴱⴱ .22ⴱⴱ .37ⴱⴱ .13ⴱⴱ .86
6. Organizational citizenship behavior 3.53 0.52 .45ⴱⴱ .31ⴱⴱ .40ⴱⴱ .20ⴱⴱ .38ⴱⴱ .91
Note. N ⫽ 815. All correlations are at the individual level, with group-level variables (i.e., servant leadership and climates) assigned down to individuals.
Reliability coefficients are reported on the diagonal.
ⴱⴱ
p ⬍ .01 (two-tailed test).
Table 2
Hierarchical Linear Modeling Results: Effects of Servant Leadership on Citizenship Behavior
OCB Commitment Self-efficacy OCB

Level and variable (Model 1) (Model 2) (Model 3) (Model 4)
Level 1 (n ⫽ 815)
Intercept 2.99 (0.08)ⴱⴱ 3.42 (0.11)ⴱⴱ 3.29 (0.10)ⴱⴱ 3.30 (0.10)ⴱⴱ
Commitment to the supervisor 0.18 (0.05)ⴱⴱ
Self-efficacy 0.24 (0.04)ⴱⴱ
Level 2 (n ⫽ 123)
Servant leadership 0.45 (0.07)ⴱⴱ 0.28 (0.09)ⴱⴱ 0.42 (0.06)ⴱⴱ 0.19 (0.09)ⴱ
Procedural justice climate 0.16 (0.05)ⴱⴱ
Service climate 0.20 (0.06)ⴱⴱ
Note. Values in parenthesis are standard errors. Level 1 variables are grand-mean centered. Entries corre-
sponding to the predicting variables are estimations of the fixed effects, ␥s, with robust standard errors. OCB ⫽
organizational citizenship behavior.
ⴱ
p ⬍ .05 (two-tailed test). ⴱⴱ p ⬍ .01 (two-tailed test).
significantly related to OCB; the effect of servant leadership ing OCB were significant. These results suggest that positive
remained significant but was reduced in magnitude (ŷ ⫽ .19, p ⬍ procedural justice climate and service climate enhanced the
.05; Model 4) when compared with the effect in Model 1. These individual-level influence of commitment to the supervisor on
results confirmed Hypotheses 3A and 3B that procedural justice OCB, providing support for Hypotheses 4A and 5A. To the con-
climate and service climate partially mediated the effect of servant trary, Hypotheses 4B and 5B concerning the moderating effects of
leadership on OCB, even after we controlled for the effects of procedural justice climate and service climate, respectively, in the
commitment to the supervisor and self-efficacy. The Sobel (1982) relationship between self-efficacy and OCB were not supported by
tests confirmed that the indirect effect (procedural justice climate: our data.
z ⫽ 2.61, p ⬍ .01; service climate: z ⫽ 2.83, p ⬍ .01) was To determine the nature and form of the significant interactions,
significant. we plotted interactions, using a cut value of one standard deviation
below the mean and one standard deviation above the mean of
Cross-Level Interaction Results procedural justice climate and service climate, respectively (Aiken
Hypothesis 4A posited a positive cross-level Commitment to the & West, 1991). These results are depicted in Figure 2. The cross-
Supervisor ⫻ Procedural Justice Climate interaction in predicting level interactions represent how the within-group relationship be-
OCB, whereas Hypothesis 4B suggested a positive cross-level tween commitment to the supervisor and OCB changes as a
Self-Efficacy ⫻ Procedural Justice Climate interaction in predict- function of both procedural justice climate and service climate.
ing OCB. Following Hofmann and Gavin’s (1998) suggestion that Further, using a moderated regression, we found that this block of
spurious cross-level interactions may be found if between-group product variables accounted for an additional 5% of the variance in
interactions are not controlled for, we aggregated commitment and OCB.
self-efficacy to the group level to calculate group-level interac-
tions. We then included the Group-Level Commitment ⫻ Proce-
dural Justice (Service) Climate interaction and the Group-Level Additional Analyses
Self-Efficacy ⫻ Procedural Justice (Service) Climate interaction
Given the significant results we obtained for Hypotheses 4A and
as Level 2 predictors.
5A regarding the cross-level moderating effects of procedural
As shown in Table 3, the results revealed that after we con-
justice climate and service climate in predicting OCB, we explored
trolled for main effects of commitment to the supervisor, self-
the possibility of a three-way Procedural Justice Climate ⫻ Ser-
efficacy, procedural justice climate, and service climate, the
between-group Commitment to the Supervisor ⫻ Procedural Jus- vice Climate ⫻ Commitment to the Supervisor interaction in
tice Climate interaction (ŷ ⫽ .11, ns), the Commitment to the predicting OCB. To do this, we followed the procedure recom-
Supervisor ⫻ Service Climate interaction (ŷ ⫽ .08, ns), the Self- mended by Aiken and West (1991) by first entering main effects
Efficacy ⫻ Procedural Justice Climate interaction (ŷ ⫽ .06, ns), and the two-way interaction terms as predictors of OCB, followed
and the Self-Efficacy ⫻ Service Climate interaction (ŷ ⫽ .10, ns) by the three-interaction term (i.e., Procedural Justice Climate ⫻
were not significant. Similarly, Table 3 also shows that, after Service Climate ⫻ Commitment to the Supervisor). We found that
controlling for group-level interactions, the cross-level Self- the three-way interaction between procedural justice climate, ser-
Efficacy ⫻ Procedural Justice Climate interaction (ŷ ⫽ .10, ns) as vice climate, and commitment to the supervisor was significantly
well as the Self-Efficacy ⫻ Service Climate interaction (ŷ ⫽ .14, related to OCB. Specifically, when both procedural justice climate
ns) in predicting OCB were not significant. However, the cross- and service climate are high, there is a strong, positive relationship
level Commitment to the Supervisor ⫻ Procedural Justice Climate between commitment to the supervisor and OCB (ŷ ⫽ .21, p ⬍
interaction (ŷ ⫽ .21, p ⬍ .01) and the Commitment to the Super- .01). This three-way interaction term accounted for an additional
visor ⫻ Service Climate interaction (ŷ ⫽ .24, p ⬍ .01) in predict- 2% of the variance in OCB.
Table 3
Hierarchical Linear Modeling Results for Moderation Hypotheses
Organizational citizenship
behavior
Variable Estimate Standard error
Level 1
Intercept 3.52ⴱⴱ 0.09
Commitment to the supervisor 0.20ⴱⴱ 0.06
Self-efficacy 0.25ⴱⴱ 0.05
Level 2
Group-level commitment to the supervisor 0.04 0.04
Group-level self-efficacy 0.06 0.08
Procedural justice climate 0.14ⴱⴱ 0.04
Service climate 0.18ⴱⴱ 0.06
Group-Level Commitment ⫻ Procedural Justice Climate 0.11 0.07
Group-Level Commitment ⫻ Service Climate 0.08 0.07
Group-Level Self-Efficacy ⫻ Procedural Justice Climate 0.06 0.05
Group-Level Self-Efficacy ⫻ Service Climate 0.10 0.07
Cross-level interactions
Commitment to the Supervisor ⫻ Procedural Justice Climate 0.21ⴱⴱ 0.05
Commitment to the Supervisor ⫻ Service Climate 0.24ⴱⴱ 0.06
Self-Efficacy ⫻ Procedural Justice Climate 0.10 0.06
Self-Efficacy ⫻ Service Climate 0.14 0.08
Note. Level 1 variable: n ⫽ 815; Level 2 variable: n ⫽ 123. In Level 1 analyses, commitment to the supervisor
and self-efficacy were group-mean centered. All entries corresponding to the predicting variables are estimations
of the fixed effects, ␥s, with robust standard errors.
ⴱⴱ
p ⬍ .01 (two-tailed test).
Discussion Second, we elucidated the contextual mechanisms through

which servant leadership is associated with OCB. Here again, we
Utilizing social exchange (Blau, 1964) and social learning (Ban- found that procedural justice climate and service climate partially
dura, 1977) theories, we sought to examine how leaders amplify
mediated the relationship between servant leadership and OCB. In
their leadership effectiveness through influencing individual-level
explicating the theory of servant leadership, Greenleaf (1977)
attitudes and situational contexts. We found that servant leadership
suggested that servant leaders recognize their moral responsibility
influences OCB through different mechanisms. Furthermore, in
to their organization, their followers, and the organization’s stake-
response to a call for research to examine multiple climates simul-
holders. The positive association between servant leadership and
taneously as cross-level moderators of the relationship between
procedural justice climate may be representative of the leaders’
employee attitudes and citizenship behaviors at the individual level
recognized moral responsibility to their followers. The positive
of analysis (Ehrhart, 2004; Kuenzi & Schminke, 2009), we found
relationship between servant leaders and service climate, on the
evidence for procedural justice climate and service climate as
moderators that amplify the influence of commitment to the su- other hand, may be indicative of the leaders’ focus on the organi-
pervisor on OCB. These findings contribute to the servant leader- zation’s omnibus stakeholders. Hence, moral leadership theories
ship, climate, and OCB literatures in several ways. that encapsulate multiple stakeholders may serve to propitiate
First, we enumerated the employee attitudes through which multiple positive climates simultaneously. However, the fact that
servant leadership is related to OCB. Our findings indicated that we found partial as opposed to complete mediation suggests that
servant leadership was positively associated with OCB, with this future research may consider including other specific climates,
relationship partially mediated by employee perceptions of self- such as safety climate (e.g., Hofmann et al., 2003; Zohar & Luria,
efficacy and commitment to the supervisor. These findings suggest 2005), to further explain the servant leadership–OCB relationship.
that there may be other mediation mechanisms involved in addi- Finally, we tested climates’ (e.g., procedural justice climate and
tion to self-efficacy and commitment to the supervisor. Future service climate) cross-level interactions with commitment to the
research may consider other employee attitudes that may serve as supervisor and self-efficacy in predicting OCB. We found that
mediating mechanisms. For example, it is possible that leader– procedural justice climate and service climate positively moder-
member exchanges and individual attitudes, such as trust, work ated the relationship between commitment to the supervisor and
satisfaction, identification with the supervisor and psychological OCB. Of interest, however, neither procedural justice climate nor
empowerment, may serve as additional potential intervening vari- service climate moderated the relationship between employee self-
ables in explaining the effect of servant leadership on employee efficacy and OCB. Although we were surprised by the lack of
OCB. Notably, psychological empowerment encompasses both significant interactions between self-efficacy and procedural jus-
self-efficacy and intrinsic motivation (Chen, Kanfer, Kirkman, tice (service climate), we suspect that highly efficacious individ-
Allen, & Rosen, 2007) and thus may be a powerful mediator. uals are already intrinsically motivated to support coworkers. As
citizenship behaviors. Research suggests that when leaders are fair,

A 5
employees are likely to learn more and to engage in extrarole
4.5
behaviors that benefit other individuals within the organizations
4 (Walumbwa, Cropanzano, & Hartnell, 2009).
3.5
OCB
3
2.5 Low Procedural Justice Climate Limitations and Conclusions
2
High Procedural Justice Climate This study has some limitations. First, it is important to recog-
1.5 nize that our study relied on short survey designs. Therefore, we
1 cannot make any definitive inferences about causality. Future
Low Commitment to High Commitment to research may consider using experimental designs in which data
Supervisor Supervisor are collected over repeated observations to provide evidence of
causality between servant leadership and OCB. Moreover, we used
B 5 only follower self-reported evaluations of climates. Future work
4.5 should include a more complete assessment of climate from both
4 followers and their supervisors.
3.5
Second, servant leadership, employee attitudes, and climate
ratings were supplied by the employees, suggesting potential
OCB
3
common-method bias. P. M. Podsakoff et al. (2003) suggested
2.5 Low Climate for Service
several strategies, such as collecting data at different times. In this
2 study, we attempted to mitigate common-method bias by collect-
High Climate for Service
1.5 ing servant leadership ratings at Time 1 and employee commit-
1 ment to the supervisor, self-efficacy, procedural justice, and ser-
Low Commitment to High Commitment to vice climate ratings at Time 2. In addition to the temporal
Supervisor Supervisor separation, we aggregated servant leadership and the two climates
to the group level. These design strategies, plus the fact that some
Figure 2. A: Interactive effects of commitment to the supervisor and
of our hypotheses included cross-level moderation, suggest that
procedural justice climate in predicting organizational citizenship behavior
(OCB). B: Interactive effects of commitment to the supervisor and service
common method bias was not completely responsible for our
climate in predicting OCB. findings—pointing to a methodological strength of the present
study. Nevertheless, future research should engage longitudinal
designs wherein both qualitative and quantitative data are collected
such, procedural justice climate and service climate may be less over repeated observations to provide greater insights into the
influential in accentuating highly efficacious employees’ citizen- dynamic relationships among servant leadership, employee atti-
ship behavior (Todd & Kent, 2006). Regardless, the fact that we tudes, and specific climates.
detected cross-level interactions for procedural justice climate and Third, although our study lends initial evidence of servant
service climate with commitment to the supervisor is an important leadership’s generalizability to other cultures, in this case Kenya,
extension to the climate and OCB literatures. More important, we encourage future research to replicate and extend our findings
when we explored the three-way interaction (i.e., Procedural Jus- in additional cultures before making firm conclusions. Future
tice Climate ⫻ Service Climate ⫻ Commitment to the Supervisor), research may also consider cross-cultural comparative studies of
we found that there was a strong and positive relationship between servant leadership. Such context-rich studies will be valuable for
commitment to the supervisor and OCB. These findings suggest understanding how servant leadership influences specific climates
that procedural justice climate and service climate are important and employee attitudes and behaviors in distinct cultures.
contextual variables that have meaningful influences on individual Fourth, research is needed to clarify the unique nature and
behavior. In other words, our findings suggest that higher levels of function of servant leadership. For example, research should con-
OCB are likely to be enhanced when commitment to the supervisor sider the conditions under which servant leadership is more or less
is accompanied by reinforcement of fair procedures and positive effective. Values, such as those identified by Hofstede (1980) and
service climate. This is an important area for future research. House and his colleagues in the Global Leadership and Organiza-
From a practical point of view, our results provide insight into tional Behavior Effectiveness project (e.g., House, Hanges, Javi-
how servant leadership can be used to enhance employee citizen- dan, Dorfman, & Gupta, 2004), or individual differences, such as
ship behavior. Our findings suggest that servant leadership is personality traits, may constitute powerful moderators of follow-
instrumental in developing positive climates that can then be used ers’ reactions to servant leadership. Focusing on such individual
to enhance employee citizenship behavior in organizations. It values or personality traits may help researchers to understand
follows that the effectiveness of leadership programs aimed at more fully how individual differences affect the relationship be-
enhancing procedural justice climate and service climate can be tween servant leadership and important work-related attitudes and
improved further by incorporating training in servant leadership behaviors. Further research is also needed to theoretically integrate
skills. Our results also highlight the importance of teaching man- servant leadership and other related leadership theories to deter-
agers/supervisors both to be fair in their decision processes and to mine the extent to which this stream of research makes a unique
explicitly discuss work related practices and policies to enhance contribution to the literature.
In conclusion, this study offers evidence concerning the impor- Hall, R. J., & Lord, R. G. (1995). Multi-level information processing
tance of servant leadership to organizations by providing fairly explanations of followers’ leadership perceptions. Leadership Quarterly,
strong support for the hypothesized model. Our findings suggest 6, 265–287.
that servant leadership ameliorates positive employee attitudes as Hoffman, B. J., Blair, C. A., Meriac, J. P., & Woehr, D. J. (2007).
well as climates that yield important behaviors that directly benefit Expanding the criterion domain? A quantitative review of the OCB
literature. Journal of Applied Psychology, 92, 555–566.
individuals and the work group. If further supported by additional
Hofmann, D. A., & Gavin, M. B. (1998). Centering decisions in hierar-
empirical research in other organizational settings and cultural chical linear models: Implications for research in organizations. Journal
contexts, our findings can significantly extend researchers’ under- of Management, 24, 623– 641.
standing of the processes and conditions under which servant Hofmann, D. A., Griffin, M. A., & Gavin, M. B. (2000). The application
leadership is more (or less) effective. of hierarchical linear modeling to organizational research. In K. J. Klein
& S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods
References in organizations: Foundations, extensions, and new directions (pp. 467–
511). San Francisco, CA: Jossey-Bass.
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and Hofmann, D. A., Morgeson, F. P., & Gerras, S. J. (2003). Climate as a
interpreting interactions. Newbury Park, CA: Sage. moderator of the relationship between leader–member exchange and
Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents of content specific citizenship: Safety climate as an exemplar. Journal of
affective, continuance, and normative commitment to the organization. Applied Psychology, 88, 170 –178.
Journal of Occupational Psychology, 63, 1–18. Hofstede, G. (1980). Culture’s consequences. Beverly Hills, CA: Sage.
Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ:
House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W., & Gupta, V.
Prentice–Hall.
(2004). Culture, leadership, and organizations: The GLOBE study of 62
Bandura, A. (1997). Self-efficacy: The exercise of control. New York, NY:
societies. Thousand Oaks, CA: Sage.
Freeman.
Ilies, R., Nahrgang, J. D., & Morgeson, F. P. (2007). Leader–member
Bass, B. M. (1985). Leadership and performance beyond expectations.
exchange and citizenship behaviors: A meta-analysis. Journal of Applied
New York, NY: Free Press.
Psychology, 92, 269 –277.
Becker, T. E., & Kernan, M. C. (2003). Matching commitment to super-
James, L. R., Demaree, R. G., & Wolf, G. (1984). Estimating within-group
visors and organization to in-role and extra-role performance. Human
inter-rater reliability with and without response bias. Journal of Applied
Performance, 16, 327–348.
Psychology, 69, 85–98.
Blau, P. M. (1964). Exchange and power in social life. New York, NY:
Johns, G. (2006). The essential impact of context on organizational behav-
Wiley.
ior. Academy of Management Review, 31, 386 – 408.
Bliese, P. D. (2000). Within-group agreement, non-independence, and
Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional
reliability: Implications for data aggregation and analysis. In K. J. Klein
leadership: A meta-analytic test of their relative validity. Journal of
& S. W. J. Kozlowski (Eds.), Multilevel theory, research and methods in
Applied Psychology, 89, 755–768.
organizations: Foundations, extensions, and new directions (pp. 349 –
Kenny, D. A., Kashy, D. A., & Bolger, N. (1998). Data analysis in social
381). San Francisco, CA: Jossey-Bass.
Bommer, W. H., Dierdorff, E. C., & Rubin, R. S. (2007). Does prevalence psychology. In D. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), Handbook
mitigate relevance? The moderating effect of group-level OCB on em- of social psychology (Vol. 1, pp. 233–265). New York, NY: McGraw-
ployee performance. Academy of Management Journal, 50, 1481–1494. Hill.
Brown, M. E., & Treviño, L. K. (2006). Ethical leadership: A review and Kirkman, B. L., Chen, G., Farh, J., Chen, Z. X., & Lowe, K. B. (2009).
future directions. Leadership Quarterly, 17, 595– 616. Individual power distance orientation and follower reactions to transfor-
Chen, G., & Bliese, P. D. (2002). The role of different levels of leadership mational leaders: A cross-level, cross-cultural examination. Academy of
in predicting self- and collective efficacy: Evidence for discontinuity. Management Journal, 52, 744 –764.
Journal of Applied Psychology, 87, 549 –556. Kozlowski, S. W., & Doherty, M. L. (1989). Integration of climate and
Chen, G., Kanfer, R., Kirkman, B. L., Allen, D., & Rosen, B. (2007). A leadership: Examination of a neglected issue. Journal of Applied Psy-
multilevel study of leadership, empowerment, and performance in teams. chology, 74, 546 –553.
Journal of Applied Psychology, 92, 331–346. Kuenzi, M., & Schminke, M. (2009). Assembling fragments into a lens: A
Colquitt, J. A. (2001). On the dimensionality of organizational justice: A review, critique, and proposed research agenda for the organizational
construct validation of a measure. Journal of Applied Psychology, 86, work climate literature. Journal of Management, 35, 634 –717.
386 – 400. Lavelle, J. J., Rupp, D. E., & Brockner, J. (2007). Taking a multifoci
Cropanzano, R., Byrne, Z. S., Bobocel, D. R., & Rupp, D. E. (2001). Moral approach to the study of justice, social exchange, and citizenship behav-
virtues, fairness heuristics, social entities, and other denizens of organi- ior: The target similarity model. Journal of Management, 33, 841– 866.
zational justice. Journal of Vocational Behavior, 58, 164 –209. Lee, K., & Allen, N. J. (2002). Organizational citizenship behavior and
Ehrhart, M. G. (2004). Leadership and procedural justice climate as ante- workplace deviance: The role of affect and cognitions. Journal of
cedents of unit-level organizational citizenship behavior. Personnel Psy- Applied Psychology, 87, 131–142.
chology, 57, 61–94. LePine, J. A., Erez, A., & Johnson, D. E. (2002). The nature and dimen-
Graham, J. W. (1991). Servant-leadership in organizations: Inspirational sionality of organizational citizenship behavior: A critical review and
and moral. Leadership Quarterly, 2, 105–119. meta-analysis. Journal of Applied Psychology, 87, 52– 65.
Grant, A. M., & Mayer, D. M. (2009). Good soldiers and good actors: Leventhal, G. S. (1980). What should be done with equity theory? New
Prosocial and impression management motives as interactive predictors approaches to the study of fairness in social relationships. In K. Gergen,
of affiliative citizenship behaviors. Journal of Applied Psychology, 94, M. Greenberg, & R. Willis (Eds.), Social exchange: Advances in theory
900 –912. and research (pp. 27–55). New York, NY: Plenum Press.
Greenleaf, R. K. (1977). Servant leadership. New York, NY: Paulist Press. Liao, H., & Chuang, A. (2007). Transforming service employees and
Hale, J. R., & Fields, D. L. (2007). Exploring servant leadership across climate: A multilevel, multisource examination of transformational lead-
cultures: A study of followers in Ghana and the USA. Leadership, 3, ership in building long-term service relationships. Journal of Applied
397– 417. Psychology, 92, 1006 –1119.
Liden, R. C., Wayne, S. J., Zhao, H., & Henderson, D. (2008). Servant tional climate configurations: Relationships to collective attitudes, cus-
leadership: Development of a multidimensional measure and multi-level tomer satisfaction, and financial performance. Journal of Applied Psy-
assessment. Leadership Quarterly, 19, 161–177. chology, 94, 618 – 634.
Lind, E. A., & Earley, P. C. (1992). Procedural justice and culture. Smith, B. N., Montagno, R. V., & Kuzmenko, T. N. (2004). Transforma-
International Journal of Psychology, 27, 227–242. tional and servant leadership: Content and contextual comparisons.
Mayer, D., Nishii, L., Schneider, B., & Goldstein, H. (2007). The precur- Journal of Leadership and Organizational Studies, 10, 80 –92.
sors and products of justice climates: Group leader antecedents and Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects in
employee attitudinal consequences. Personnel Psychology, 60, 929 – structural equation models. In S. Leinhardt (Ed.), Sociological method-
963. ology (pp. 290 –312). Washington, DC: American Sociological Associ-
Meyer, J. P., Allen, N. J., & Smith, C. A. (1993). Commitment to orga- ation.
nizations and occupations: Extension and test of a three-component Spears, L. C. (1998). Tracing the growing impact of servant-leadership. In
conceptualization. Journal of Applied Psychology, 78, 538 –551. L. C. Spears (Eds.), Insights on leadership: Service, stewardship, spirit,
Naumann, S. E., & Bennett, N. (2000). A case for procedural justice and servant-leadership (pp. 1–12). New York, NY: Wiley.
climate: Development and test of a multilevel model. Academy of Spears, L. C. (2004). The understanding and practice of servant leadership.
Management Journal, 43, 881– 889. In L. C. Spears & M. Lawrence (Eds.), Practicing servant-leadership:
Neubert, M. J., Kacmar, K. M., Carlson, D. S., Chonko, L. B., & Roberts, Succeeding through trust, bravery, and forgiveness (pp. 9 –24). San
J. A. (2008). Regulatory focus as a mediator of the influence of initiating Francisco, CA: Jossey-Bass.
structure and servant leadership on employee behavior. Journal of Ap- Takeuchi, R., Chen, G., & Lepak, D. P. (2009). Through the looking glass
plied Psychology, 93, 1220 –1233. of a social system: Cross-level effects of high-performance work sys-
Organ, D. W. (1988). Organizational citizenship behavior: The good tems on employees’ attitudes. Personnel Psychology, 62, 1–29.
soldier syndrome. Lexington, MA: Lexington Books. Tangirala, S., & Ramanujam, R. (2008). Employee silence on critical work
Organ, D. W., Podsakoff, P. M., & MacKenzie, S. B. (2006). Organiza- issues: The cross level effects of procedural justice climate. Personnel
tional citizenship behavior: Its nature, antecedents, and consequences. Psychology, 61, 37– 68.
Thousand Oaks, CA: Sage. Todd, S. Y., & Kent, A. (2006). Direct and indirect effects of task
Ostroff, C., Kinicki, A. J., & Tamkins, M. M. (2003). Organizational characteristics on organizational citizenship behavior. North American
culture and climate. In W. C. Borman, D. R. Ilgen, R. J. Klimoski, & I. Journal of Psychology, 8, 253–268.
Weiner (Eds.), Handbook of psychology (pp. 565–593). Hoboken, NJ: Tyler, R. R., & Blader, S. L. (2003). The group engagement model:
Wiley. Procedural justice, social identity, and cooperative behavior. Personality
Podsakoff, N. P., Whiting, S. W., Podsakoff, P. M., & Blume, B. D. (2009). and Social Psychology Review, 7, 349 –361.
Individual- and organizational-level consequences of organizational cit- Walumbwa, F. O., Avolio, B. J., Gardner, W. L., Wernsing, T. S., &
izenship behaviors: A meta-analysis. Journal of Applied Psychology, 94, Peterson, S. J. (2008). Authentic leadership: Development and validation
122–141. of a theory-based measure. Journal of Management, 34, 89 –126.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Walumbwa, F. O., Cropanzano, R., & Hartnell, C. A. (2009). Organiza-
Common method biases in behavioral research: A critical review of the tional justice, voluntary learning behavior, and job performance. A test
literature and recommended remedies. Journal of Applied Psychology, 5, of the mediating effects of identification and leader–member exchange.
879 –903. Journal of Organizational Behavior, 30, 1103–1126.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Walumbwa, F. O., Orwa, B., Wang, P., & Lawler, J. J. (2005). Transfor-
Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: mational leadership, organizational commitment, and job satisfaction: A
Sage. comparative study of Kenyan and U.S. financial firms. Human Resource
Riggs, M. L., Warka, J., Babasa, B., Betancourt, R., & Hooker, S. (1994). Development Quarterly, 6, 235–256.
Development and validation of self-efficacy and outcome expectancy Walumbwa, F. O., Peterson, S. J., Avolio, B. J., & Hartnell, C. A. (in
scales for job-related applications. Educational and Psychological Mea- press). An investigation of the relationships between leader and follower
surement, 58, 1017–1034. psychological capital, service climate and job performance. Personnel
Schnake, M. E., & Dumler, M. P. (2003). Levels of measurement and Psychology.
analysis issues in organizational citizenship behaviour research. Journal Walumbwa, F. O., & Schaubroeck, J. (2009). Leader personality traits and
of Occupational and Organizational Psychology, 76, 283–301. employee voice behavior: Mediating roles of ethical leadership and work
Schneider, B. (1990). The climate for service: An application of the climate group psychological safety. Journal of Applied Psychology, 94, 1275–
construct. In B. Schneider (Ed.), Organizational climate and culture (pp. 1286.
383– 412). San Francisco, CA: Jossey-Bass. Walumbwa, F. O., Wu, C., & Orwa, B. (2008). Contingent reward trans-
Schneider, B., Ehrhart, M. G., Mayer, D. M., Saltz, J. L., & Niles-Jolly, K. actional leadership, work attitudes, and organizational citizenship be-
(2005). Understanding organization– customer links in service settings. havior: The role of procedural justice climate perceptions and strength.
Academy of Management Journal, 48, 1017–1032. Leadership Quarterly, 19, 251–265.
Schneider, B., Salvaggio, A. N., & Subirats, M. (2002). Climate strength: Zohar, D., & Luria, G. (2005). A multilevel model of safety climate:
A new direction for climate research. Journal of Applied Psychology, 87, Cross-level relationships between organization and group-level climates.
220 –229. Journal of Applied Psychology, 90, 616 – 628.
Schneider, B., White, S. S., & Paul, M. C. (1998). Linking service climate
and customer perceptions of service quality: Test of a causal model. Received December 9, 2008
Journal of Applied Psychology, 83, 150 –163. Revision received December 23, 2009
Schulte, M., Ostroff, C., Shmulyian, S., & Kinicki, A. (2009). Organiza- Accepted December 29, 2009 䡲
View publication stats

Walumbwa 2010

Uploaded by

Copyright:

Available Formats

Walumbwa 2010

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Walumbwa 2010

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Servant Leadership, Procedural Justice Climate, Service Climate, Employee

Article in Journal of Applied Psychology · May 2010

Chad Hartnell Adegoke Oke

SEE PROFILE SEE PROFILE

The user has requested enhancement of the downloaded file.

Servant Leadership, Procedural Justice Climate, Service Climate,

Fred O. Walumbwa, Chad A. Hartnell, and Adegoke Oke

Keywords: servant leadership, leadership, climate, organizational citizenship behavior, cross-level

Servant leadership emphasizes listening, empathy, stewardship,

1. Servant leadership 3.15 0.74 .91

OCB Commitment Self-efficacy OCB

Variable Estimate Standard error

Discussion Second, we elucidated the contextual mechanisms through

citizenship behaviors. Research suggests that when leaders are fair,

View publication stats

You might also like