The Leadership Quarterly xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

The Leadership Quarterly

journal homepage: www.elsevier.com/locate/leaqua

The servant leadership advantage: When perceiving low

differentiation in leader-member relationship quality influences
team cohesion, team task performance and service OCB
Myriam Chiniaraa, Kathleen Benteinb,⁎
a
Department of Psychology, University of Quebec at Montreal (UQAM), Case Postale 8888, succ. Centre-ville, H3C3P8 Montreal, Quebec, Canada
b
School of Business Administration (ESG), University of Quebec at Montreal (UQAM), Case Postale 8888, succ. Centre-ville, H3C3P8 Montreal,
Quebec, Canada

AR TI CLE I NF O AB S T R A CT

Keywords: How does servant leaders' unique ability to place each follower's needs above their own influence
Servant leadership relationships between followers and impact their collective performance? In a study that in-
Perceived leader-member LMX differentiation tegrates principles of servant leadership with the social comparison theoretical framework, we
Team cohesion tested a group-level model to examine how servant leadership induces low perceived differ-
Team task performance
entiation in leader-member relationship quality (perceived LMX differentiation) within a group,
Service organizational citizenship behaviors
which strengthens team cohesion and in turn positively influences team task performance and
service-oriented organizational citizenship behaviors (service OCB). Our sample comprised 229
employees nested in 67 work teams. Structural equation modeling results indicate that servant
leadership significantly predicts low perceived LMX differentiation; perceived LMX differentia-
tion is strongly related to team cohesion such that the lower the perceived differentiation, the
stronger the team's cohesiveness. And, team cohesion is also strongly related to both the team's
task performance and service OCB. Perceived LMX differentiation and team cohesion mediate the
effect of servant leadership on both team task performance and service OCB.

Introduction

In recent decades, organizations worldwide have restructured work around teams to enable more rapid, flexible and adaptive
responses to turbulent and complex work environments (Kozlowski & Bell, 2013; Kozlowski & Ilgen, 2006). As a result, leaders must
mobilize not only individuals, but each team as an entity. Leadership is a complex influential process that plays a central role in
enabling the integration of individual contributions into a cooperative group effort (Hogg, 2006; Northouse, 2007). Among the
numerous leadership styles, Servant leadership has drawn more recently significant attention among both academics (e.g., Graham,
1991; Liden, Panaccio, Meuser, Hu, & Wayne, 2014; van Dierendonck, 2011) and practitioners (Spears, 1995) in response to a
growing interest in a more ethical, pro-social, and people-centered management leadership style. Servant leadership is characterized
by a focus on followers' growth and empowerment, and on leaders' altruism, empathy, sense of ethics and community stewardship
(Greenleaf, 1977, 1998; Liden, Wayne, Zhao, & Henderson, 2008). Compared with related leadership styles whose primary focus is
the well-being of the organization, servant leadership is unique in that the leader is viewed as a ‘servant’ attending to followers' needs
above his or her own (van Dierendonck, 2011). Servant leadership's central premise is that servant leaders influence organizational

⁎
Corresponding author at: School of Business Administration (ESG), University of Quebec at Montreal (UQAM), Case Postale 8888, succ. Centre-ville, H3C3P8
Montreal, Quebec, Canada
E-mail addresses: myriamchiniara@gmail.com (M. Chiniara), bentein.kathleen@uqam.ca (K. Bentein).

http://dx.doi.org/10.1016/j.leaqua.2017.05.002
Received 28 January 2016; Received in revised form 4 April 2017; Accepted 16 May 2017
1048-9843/ © 2017 Elsevier Inc. All rights reserved.

Please cite this article as: Chiniara, M., The Leadership Quarterly (2017), http://dx.doi.org/10.1016/j.leaqua.2017.05.002
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

outcomes by fostering followers' growth and well-being, specifically through the process of satisfying followers' needs (Liden et al.,
2008; Mayer, 2010). Servant leadership research is still in its early stages, and although empirical results show evidence of servant
leadership's influence on individual performance (e.g., Chiniara & Bentein, 2015; Liden et al., 2008), research has only begun to
establish a relationship between servant leadership and team performance (e.g., Ehrhart, 2004; Hu & Liden, 2011; Liden, Wayne,
Liao, & Meuser, 2014). For servant leadership to gain further legitimacy as a mainstream leadership theory, research must advance
the identification of processes explaining how a leadership style with such an explicit focus on the individual can help achieve team
objectives (Mayer, 2010). While mechanisms such as team potency (Hu & Liden, 2011), service culture (Liden, Wayne et al., 2014),
procedural justice (Ehrhart, 2004) or trust in leaders (Schaubroeck, Lam, & Peng, 2011) have been found to play a role in explaining
the influence of servant leadership on team performance, the effect of servant leadership on relational dynamics between team
members within this context needs to be understood. The objective of this study is to take on this challenge.
To address this theoretical gap in the servant leadership literature, we first draw on Festinger's (1954) social comparison theory that
stipulates that human beings possess the innate need to evaluate themselves by comparing their skills, opinions, or possessions with
relevant others, which ultimately influences their attitudes and behaviors. In organizational settings, this social comparison process is
viewed as inevitable in work groups. Given the considerable importance of the relationship between leaders and followers (Ferris
et al., 2009), employees naturally tend to observe and use social comparison information on the quality of treatment received from
their authority figures. Research has demonstrated how work group members are aware of differentiated relationships their leaders
form with team members (e.g. Duchon, Green, & Taber, 1986; van Breukelen, Schyns, & Le Blanc, 2006).
Leader-member exchange quality differentiation (LMX differentiation) is a construct derived from Leader-member theory (LMX;
Dansereau, Graen, & Haga, 1975), that is used to study the variability in relationship quality between leaders and team members (for
reviews see Anand, Vidyarthi, & Park, 2016 and Henderson, Liden, Glibkowski, & Chaudhry, 2009). LMX differentiation is defined as
the degree of variation that exists when a leader forms relationships of different quality with different members within a same group
(Erdogan & Bauer, 2010). LMX differentiation is rooted in the notion that members compare their relationship quality with that of
their teammates and social comparison theory is thus considered to be the main driving force behind LMX differentiation (Anand
et al., 2016). For the past decade, LMX differentiation has been operationalized in almost all studies by calculating a dispersion
measure (mostly variance or standard deviation) of individual LMXs within a group (e.g., Harris, Li, & Kirkman, 2014; Liden,
Erdogan, Wayne, & Sparrowe, 2006), and in a few rare studies by the perception of differentiation in relationship quality (Epitropaki
et al., 2016; Hooper & Martin, 2008). Research on LMX differentiation has mainly focused on its direct and moderating effects on
followers' individual-level outcomes (e.g., Epitropaki et al., 2016; Hooper & Martin, 2008); much less attention has been paid to the
direct effects of LMX differentiation on group-level outcomes (e.g., Le Blanc & González-Romá, 2012; Liden et al., 2006). We aim to
investigate, at the group level, how servant leadership affects perceived LMX differentiation, and to examine LMX differentiation's
effect on teams' relational dynamics.
To do so, we mobilize a second body of literature from the extensive domain of group research to investigate a group-level
relational mechanism, team cohesion (Festinger, 1950), which may shed light on the process by which servant leadership affects team
performance. Cohesiveness is an important team feature that defines the level of forces that bond team members to one another and
to the purpose of the team (Festinger, 1950; Carron, 1982). Team cohesion has been studied extensively in several contexts including
work environments, and has been repeatedly found to be positively linked to work team performance (e.g., Beal, Cohen,
Burke, & McLendon, 2003; Castaño, Watts, & Tekleab, 2013; Mullen & Copper, 1994).
In the present study, building upon social comparison theory, we propose that servant leadership impacts team performance by
influencing perceptions of low LMX differentiation among team members, which in turn strengthens team cohesion. Specifically, our
model rests on the premise that servant leaders are characterized by their strong capability to build quality relationships and support
the growth of all followers, which conveys to followers how strongly they are valued and respected (Greenleaf, 1998). Given that
teammates naturally observe and rely on social comparisons with work peers to evaluate themselves (Festinger, 1954), because all
followers feel appreciated and respected by their leader, they would not perceive a difference of quality in the leader-follower
relationships between themselves, even though they might observe differences in the nature and content of leader-member re-
lationships because of differences in individual characteristics and needs. Hence, these comparisons would engender low levels of
perceived LMX differentiation (Mayer, Erdogan, & Piccolo, 2008); therefore we first postulate that servant leadership produces low
perceived LMX differentiation. A reduced level of differentiation between team members diminishes the tendency of teammates to
socially categorize themselves into subgroups and on the contrary, enhances the strength of shared attraction toward the team. It also
reinforces the ties among team members, thus positively influencing team cohesion. Therefore, our model further proposes that low
perceived LMX differentiation drives team cohesion upwards (Festinger, 1954). When team members feel cohesive, they work to-
gether not only toward achievement of team task objectives (i.e., team task performance) but also toward increased support for other
teams (i.e., service OCB). Notably, we chose other teams to be recipients of OCBs, hereafter named service OCBs, to reflect the context
of a serving-other culture that servant leadership establishes by promoting and engaging the collective to help others (Liden et al.,
2008). Hence, we finally postulate in our model that team cohesion predicts team performance as conceptualized by both team task
performance and service OCB. Fig. 1 depicts our hypothesized model.
This research makes several significant contributions to the servant leadership and LMX differentiation literatures. First, the
present study serves to substantiate servant leadership's theoretical premise that servant leaders' distinctive focus on serving fol-
lowers' individual needs can improve team-level performance. Indeed, by drawing on social comparison theory, we are able to extend
the servant leadership domain by explicitly demonstrating the specific team relational processes through which servant leadership
increases not only team task performance, but also service OCB. Specifically, by investigating servant leadership's influence on LMX
differentiation and team cohesion we shed new light on the range of influence the servant leader possesses to enhance collective

2
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

Fig. 1. Hypothesized SEM Servant Leadership mediation model on team performance.

Note: Mediation Hypotheses 4 and 5 are not represented.

performance. Second, we also significantly advance the LMX differentiation literature and in doing so, respond to proposals from
Henderson et al. (2009) and Anand et al. (2016), by examining servant leadership as an antecedent to LMX differentiation and by
investigating the effect of perceived LMX differentiation instead of that of actual LMX differentiation, which is based on a statistical
calculation of the variation in actual leader-member exchange quality (LMX), because perceived LMX differentiation is believed to
provide a more powerful angle of observation than the actual LMX differentiation (Anand et al., 2016). We further extend the LMX
differentiation literature by testing a group-level model including a perceptual LMX differentiation measure and predicting group-
level outcomes.

Theoretical background and hypotheses

Servant leadership

Greenleaf (1977) describes servant leadership not as a management technique but as a way of life, which starts with a “natural
feeling that one wants to serve, to serve first” (p.14), then aspire to lead. Greenleaf construes servant leadership as an other-oriented
approach to leadership that emphasizes serving others, sharing power, promoting teamwork and building a sense of community both
within the work group and outside the walls of the organization. Greenleaf depicts servant leaders as embodying ethical principles
and showing a deep commitment to placing the needs of their followers at the center of their efforts, before their own needs, creating
a climate in which each follower feels important, committed and empowered to do and create more. Explicitly, servant leaders are
described as communicating on a one-to-one basis to understand followers' abilities, needs, goals and desires. Servant leaders use that
knowledge to actively bring out the best in their followers and assist them in achieving their potential by helping them develop
greater task effectiveness, community stewardship and servant leadership capabilities. Servant leaders purportedly build self-con-
fidence in their followers by providing direction and challenging responsibilities, while offering empathy, emotional support,
feedback and resources (Page & Wong, 2000). Hence, servant leaders would naturally tend to build strong, positive and long-term
relationships with each follower (Liden et al., 2008). Followers in turn would view servant leaders as role models and engage in
appropriate behaviors not by obligation but by their own will, and would demonstrate the desire to become servant leaders them-
selves (Greenleaf, 1977). Through the long-term transformation of followers into servant leaders, Greenleaf proposes that a new life
and work culture can emerge and grow, building the foundation for a more caring, cohesive and creative organization.
Liden et al. (2008) distinguish seven dimensions that describe the essential characteristics of a servant leader: emotional healing,
empowering, helping subordinates grow and succeed, putting subordinates first, creating value for the community, having conceptual
skills, and behaving ethically. Emotional healing entails showing care and sensitivity to followers' personal well-being. Empowering
involves encouraging and facilitating others' ability to take on responsibilities and to manage difficult situations in their own way.
Servant leaders help subordinates grow and succeed by exhibiting a genuine interest in their subordinates' career development and goal
achievement and by offering them opportunities to enhance their skills. Servant leaders who put subordinates first demonstrate that
they place subordinates' best interests and success ahead of their own. Creating value for the community suggests that servant leaders
are involved in contributing to the community outside the organization, and that they encourage others to do the same. Servant
leaders possess and demonstrate conceptual skills because they are well informed about the organization, its goals and the task at
hand, and can therefore provide effective support, resources and direction to others. Finally, servant leaders behave ethically by acting
and interacting openly, fairly and honestly with others.

3
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

While servant leadership resembles other major leadership theories in some respects, there is empirical evidence that servant
leadership is distinct and has incremental predictive validity after controlling for transformational leadership (Hoch, Bommer,
Dulebohn, & Wu, 2016; Parolini, Patterson, & Winston, 2009) as well as leader-member exchange models (LMX) (Ehrhart, 2004;
Liden et al., 2008). For instance, transformational leadership was empirically found to place organizational values in top priority,
whereas servant leaders place the needs of their followers first (Parolini et al., 2009). Servant leadership also differs from ethical and
authentic leadership in that servant leadership's prime target of behavior is on followers (Brown, Treviño, & Harrison, 2005; Neubert,
Kacmar, Carlson, Chonko, & Roberts, 2008), whereas ethical and authentic leadership is focused on the nature of leaders' behavior as
such. Lastly, while servant leadership theory and spiritual leadership theory (Fry's, 2003; Benefiel, Fry, & Geigle, 2014) both possess
an altruistic concern for followers' needs and well-being, and they both value involvement in the community, they differ in the
orientation of their practice and related outcomes. Spiritual leadership is centered on developing a spiritual practice at work
(Rothausen, 2016; van Dierendonck, 2011), whereas servant leadership theory is positioned as explicitly secular (van Dierendonck,
2011). Hence, servant leadership is unique in its explicit focus on serving followers' needs and development to their fullest potential,
and on establishing and maintaining strong long-term interpersonal relationships with each employee, which is instrumental in
helping employees achieve their potential (Hoch et al., 2016; van Dierendonck, 2011).
Empirical research finds that servant leadership has been significantly and positively correlated to leader-member quality ex-
changes (LMX; Liden et al., 2008) and that servant leadership is also significantly related to satisfaction of followers' basic psy-
chological work needs for autonomy, competence and relatedness (Van Dierendonck, Stam, Boersma, de Windt, & Alkema, 2014;
Chiniara & Bentein, 2015). Moreover, consistent with Greenleaf's essays, servant leadership studies argue that servant leadership
operates not only at the individual level, but also at the team level (Ehrhart, 2004; Hu & Liden, 2011; Hunter et al., 2013; Liden et al.,
2008; Liden, Wayne et al., 2014; Peterson, Galvin, & Lange, 2012; Schaubroeck et al., 2011). At the team level, servant leadership
may be seen as a climate or ‘ambient stimulus’ (Hackman, 1992) in which an overall leadership pattern of altruistic and ethical
behaviors creates a context that should positively influence group-level processes. Empirical research has found support for a re-
lationship between group-level servant leadership and group-level constructs such as affect-based trust in leaders (Schaubroeck et al.,
2011), team potency (Hu & Liden, 2011), procedural justice climate (Ehrhart, 2004), and service climate (Hunter et al., 2013; Liden,
Wayne et al., 2014). Some of these studies investigated the influence of these group-level constructs in explaining the link between
servant leadership and team performance, but results were somewhat inconsistent. For example, both Hunter et al. and Liden, Wayne
et al. examined the relationship between servant leadership and store performance through the mediating effect of service culture and
found contrasting results: Liden, Wayne et al. found that servant leadership at the team level predicts a service climate that positively
influences store performance, and Hunter et al. found that servant leadership predicts service climate at the team level, but the
relationship between service climate and store performance was not significant. Although researchers have begun to recognize the
influence of servant leadership on specific intermediate group-level mechanisms, they have yet to understand how these mechanisms
mediate the relationship between servant leadership and team performance. By drawing on social comparison theory (Festinger,
1954), we develop and test in this study, a model that examines the influence of servant leadership on two uninvestigated team-level
dynamic and relational processes in the prediction of team performance. Specifically, we propose to investigate the influence of
servant leadership on both task performance and service OCB through the effect of perceived LMX differentiation and team cohesion.

Perceived LMX differentiation and the theory of social comparison

The theory of social comparison processes (Festinger, 1954; Wood, 1996) provides a framework for understanding the effects of
leadership characteristics on employees' perceptions of leader-follower relationships in a work group context. Social comparison
theory explains how in a social entity of interconnected and interdependent individuals such as a work team, individuals naturally
make social comparisons and categorizations. Team members inevitably scan their environment observing information on their
relative position to other team members both consciously and subconsciously (cf. Anand et al., 2016). Because they entail the
observation of conversations and interactions between leader and other followers, leader-member relationships, which are inherently
social exchanges, become salient objects of social comparisons. These social comparisons feed member's self-evaluation information
on how well one is being treated, respected and valued, information by which one can derive a status and standing evaluation.
Positive affect, loyalty, respect, support and mutual influence are what constitute a quality interpersonal treatment and relationship
with the leader (Graen & Uhl-Bien, 1995). Research also demonstrates that interpersonal treatment experience with the leader is
important because it conveys a message to the team members about their capabilities and social status, and fuel positive feelings of
self-worth and self-esteem (Hogg & Abrams, 1988; Sedikides & Brewer, 2001). Studies show that group members are aware of leaders'
differential exchanges with and treatment of other group members (Duchon et al., 1986; Sias & Jablin, 1995).
LMX differentiation is a construct defined as the degree of variability of leader-member exchange quality (LMX) (Dansereau et al.,
1975) between members of a team. LMX differentiation originates from the leader-member exchange theory (LMX; Dansereau et al.,
1975), which was rooted in the principle that leaders differentiate in the quality of their exchanges with followers. LMX theory dates
back over forty years. This fruitful field of research has extensively investigated the effects on employees of having a high or low
quality relationship with their leader (e.g., Graen & Uhl-Bien, 1995). More recently, because these relationships do not exist in a
vacuum but in a context of other relationships, research interests have shifted toward understanding the implications of variability in
the quality of leader-member relationships within the context of a team (Anand et al., 2016). LMX differentiation has been studied in
three distinct ways (Anand et al., 2016; Henderson, Wayne, Shore, Bommer, & Tetrick, 2008): as individual perceptions of within-
group variation (e.g. van Breukelen et al., 2006); as individuals' relative position within their own group (e.g., Hu & Liden, 2013;
Vidyarthi, Liden, Anand, Erdogan, & Ghosh, 2010), and as within-group variation across groups (e.g., Liden et al., 2006).

4
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

A decade of research shows that most of the work on LMX differentiation has focused on its direct and moderating effects on
followers' individual-level outcomes (e.g., Epitropaki et al., 2016; Hooper & Martin, 2008) and much less attention has been paid to
LMX differentiation's direct effects on group-level outcomes (e.g., Le Blanc & González-Romá, 2012; Liden et al., 2006). Although
many questions are still left unanswered, the few studies on the influence of LMX differentiation on group-level outcomes explain
how LMX differentiation can have a positive impact on team commitment and team performance (e.g., Le Blanc & González-Romá,
2012; Liden et al., 2006), while others report evidence of the negative role of differentiation on team performance (e.g., Naidoo,
Scherbaum, Goldstein, & Graen, 2011; Schyns, 2006) and on team outcomes such as supportive management climate or group conflict
(e.g., Boies & Howell, 2006; Ford & Seers, 2006; Mayer et al., 2008). Note that all group-level studies investigating LMX differ-
entiation effect on team performance were performed with actual LMX differentiation rather than perceived LMX differentiation,
which may explain these divergent results.
Relatedly, Anand et al. (2016) report a measurement issue with LMX differentiation at group-level: the vast majority of studies use
actual LMX differentiation (a statistical calculation of variation of LMX within a group) rather than perceived LMX differentiation.
Specifically because “perceptions are more powerful than reality” (p.273), they call for scholars to develop and validate a perceived
LMX differentiation construct at group-level because no such construct had yet been published, with the one exception of the one
item perceived LMX distribution measure developed and validated by Hooper and Martin (2008), also used recently in Epitropaki et al.
(2016). However, neither Hooper & Martin, nor Epitropaki et al., have examined the influence of perceived LMX distribution on team-
level outcomes.
Servant leaders are depicted as providing support for the development and well-being of all followers (through emotional healing
and helping subordinates grow and succeed), as behaving ethically and as possessing the capability to build quality relationships
because each relationship is tailored to the specific needs of the individual (Graham, 1991; Greenleaf, 1998; Liden, Panaccio et al.,
2014). Servant leaders' interest in followers makes each follower feel respected and makes them perceive how they are recognized as
valuable, which diminishes the natural need to engage in comparison processes and differentiate. Still, given that employees each
have different needs, the nature and content of the exchanges they experience with the servant leaders are bound to vary from one
team member to another (e.g., Dansereau et al., 1975; Erdogan & Bauer, 2010). By engaging in a process of social comparison, team
members are likely to notice variations in terms of the nature of the relationship between team members and the leader, however they
would not perceive variability in terms of the quality of leader-member relationships throughout the team.
Theoretical support for this contention can be found in a review by Henderson et al. (2009), which elaborates on the proposal that
servant leadership should induce low LMX differentiation. Empirical findings also indirectly support these contentions in that servant
leadership has been significantly and positively related to satisfaction of followers' basic psychological work needs for autonomy,
competence and relatedness (van Dierendonck et al., 2014; Chiniara & Bentein, 2015). Therefore, building on both servant leadership
and LMX differentiation we contend that the more leaders display servant leadership behaviors, the less employees perceive dif-
ferentiation in relationship quality between followers and leaders within a team. Hence, we predict that:
Hypothesis 1. At the group level, servant leadership is negatively related to perceived differentiation of leader-member relationship
quality (perceived LMX differentiation).

Perceived LMX differentiation and team cohesion

Drawing on the social categorization process that results from social comparisons (Festinger, 1954; Hogg, 2006; Tajfel, 1978), we
propose that perceiving an absence of differentiation in relationship quality with the leader will influence team relational processes
such that this absence of differentiation positively impacts team cohesion. Team cohesion has been the most widely studied team-
level characteristic for the past sixty years (Kozlowski & Chao, 2012). Group researchers have offered multiple definitions of cohesion,
all of which derive from Festinger's (1950) definition: “Cohesion is the resultant of all forces acting on the members to remain in the
group,” with the later addition of the notion of “remaining united to reach a common goal” (Carron, 1982). We suggest that one of
these forces results from the perception of differentiation of relationship quality with the leader, which is an important indicator of
the extent to which the followers' contribution to the group is valued relative to others. Building upon the social categorization
process, we propose that perceiving differentiation in relationship quality between a leader and a follower affects team cohesion
because of the establishment of co-workers relationship boundaries.
Engaging in social comparisons, team members' differentiation of leader-member relationship quality within the team will likely
induce a social categorization process based on leader-member relationship quality, which will introduce relational boundaries
within teams, dividing team members and fragmenting the group into two sub-groups: in-group and out-group (Anand, Hu,
Liden, & Vidyarthi, 2011). Team members who maintain a high quality relationship with the leader seem to enjoy a better standing
and in-group status, while others may be seen as out-group members. Research has demonstrated that in-groups collectively benefit
and protect valuable team resources, such as promotion opportunities or important assignments, by leveraging their high-quality
relationships with the leader (Dansereau et al., 1975). Such self-serving attitudes and behaviors may threaten the interests of out-
group members, build tension between the two sub-groups, foment negative interpersonal feelings and undermine team processes
such as team cooperation and helping. Support for this rationale can be found in the study by Sherony and Green (2002), which
reports that when team members share similar relationship quality with their leader, they also report better relationship quality with
each other, than do co-workers with different relationship quality with their leader. Sias and Jablin (1995) also advance that
variability in relationship quality with the leader produces negative emotions such as distrust and aversion between differentiated
coworkers, be they in-group or out-group. Further, recent research finds that variability in relationship quality with the leader is

5
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

positively related to emotional hostility and contempt among team members (Tse, Lam, Lawrence, & Huang, 2013) and perceived
conflict between members (Hooper & Martin, 2008; Mayer et al., 2008), and is negatively related to team coordination (Li & Liao,
2014) and team help (Tse et al., 2013). Hence, relational boundaries between team members impede the development of team
cohesion.
To summarize, although no prior research has investigated how perceiving variability in relationship quality with the leader can
affect team cohesion, we contend, based on the lines of arguments and findings above, that perceived LMX differentiation tends to
generate a relational categorization that impedes development and maintenance of the field of forces uniting a team, and creates sub-
groups, which disrupts team members' harmony and solidarity and thus negatively influences team cohesion. Hence we postulate
that:
Hypothesis 2. At the group level, perceived LMX differentiation is negatively related to team cohesion.

Team cohesion and team performance

We posit in this study that team cohesion leads to increases in team task performance and service OCB. Team cohesion has been
studied extensively, appearing in eight published meta-analytic reviews (e.g., Beal et al., 2003; Castaño et al., 2013; Mullen & Copper,
1994) conducted over the last 20 years. In their conceptualization of multilevel system functioning, Kozlowski and Chao (2012)
consider team cohesion as a bottom-up process, meaning that lower level phenomena combine, coalesce and manifest at a collective
level. Cohesion studies have demonstrated significant correlations with a variety of outcomes such as team performance, retention
and viability, as well as positive member attitudes (Greer, 2012). In the group literature, group cohesion has been referred to as being
proximal in the causal chain to critical team outcomes such as team effectiveness (Kozlowski & Ilgen, 2006). Although researchers
acknowledge that there are still inconsistencies of definition and operationalization (e.g., dimensionality of the construct or level of
measurement), and that contextual factors may influence the strength of these correlations (e.g., Chiocchio & Essiembre, 2009;
Mathieu, Kukenberger, D'Innocenzo, & Reilly, 2015), studies have consistently reported significant and positive correlations between
average team-level cohesion and team performance, of small to moderate magnitude. The predominant logic among researchers has
been that when teams are more cohesive, team members share a stronger mutual commitment to achieving tasks and team goals, and
are more willing to work together cooperatively. These forces prompt the development of work strategies, fuel members' motivation,
increase their communication and direct their attention toward accomplishing tasks and achieving team goals (e.g., Beal et al., 2003;
Casey-Campbell & Martens, 2009; Gully, Devine, & Whitney, 1995; Mathieu et al., 2015). Following this logic and the supporting
meta-analytical results, we postulate that:
Hypothesis 3a. At group level, team cohesion is positively related to team task performance.
We also expect to find that team cohesion will be positively related to service OCB (e.g., Podsakoff, MacKenzie, & Ahearne, 1997).
Service OCB refers to the set of citizenship behaviors oriented toward targets outside the team,-whether employees or groups-, as
opposed to team OCB, which refers to the same type of behaviors, but that occur between members of a team. We chose to work with
service OCB instead of team OCB, to anchor our model theoretically in the principles of servant leadership. These principles suggests
that the service orientation promoted by servant leaders would be transmitted toward their followers, and thus translated into helpful
and respectful behaviors from these employees to individuals outside the team, as they themselves become servant leaders. Hence,
service OCB is specifically conceptualized here as a set of helping and courteous-type discretionary behaviors performed by team
members, which promote effective functioning of the organization (e.g., Podsakoff et al., 1997), and that target recipients outside the
team.
In highly cohesive teams, members tend to show greater enthusiasm, engage in more frequent and positive interactions
(Schriesheim, 1980) and sustain more positive psychological states than do members in less cohesive teams (Chen, Tang, & Wang,
2009). Research has reported that individuals who experience positive psychological states tend to perceive things more positively
and to be more prosocial (George & Brief, 1992). Therefore, both positive affect and increased interactions cultivate a context of
social exchanges that extends into exchanges with individuals outside the team and translates into service OCB, by a process of
emotional contagion and the development of a service climate (Brown & Lam, 2008). The process of emotional contagion, which is
often evoked to explain the relationship between employee affect and customer response, corresponds to a simple affect transfer from
employees to the external context of interactions. In contrast, the service climate literature is explicit regarding how perceptions of
managerial practices and organizational environment can influence an employee's favorable responses to customer (Brown & Lam,
2008). Both these processes can partly explain how in a context of cohesive work teams, positive affect and a social exchange climate
within the team extend to outside the team and therefore enhance the will to help and demonstrate altruistic behaviors toward others
outside the team. Support for our contention can be found in the OCB (e.g., Ehrhart & Naumann, 2004), service climate (Auh, Bowen,
Aysuna, & Menguc, 2016) and group literatures. Group cohesion has been generally treated as an antecedent of OCB or prosocial
behaviors (e.g., George & Bettenhausen, 1990). Group cohesion has also been shown to create a social exchange context that enhances
production of helping behaviors toward others (e.g., George & Bettenhausen, 1990; Kidwell, Mossholder, & Bennett, 1997). Moreover,
members of cohesive groups tend to experience more positive affective states and moods (Baumeister & Leary, 1995), and we know
that these affects are positively related to prosocial behaviors (e.g., Bateman & Organ, 1983).
Hence, based on the above rationale and supporting results we contend that group cohesiveness positively influences service OCB:
Hypothesis 3b. At the group level, team cohesion is positively related to service OCB.

6
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

Perceived LMX differentiation and cohesion mediate the effect of servant leadership on team task performance and OCB

The above hypotheses combine to form two mediation hypotheses in the prediction of team task performance and service OCB.
Consistent with the integration of the servant leadership model and the theory of social comparison processes, we contend that
servant leadership offers the necessary support, resources and direction to all followers, to enhance both team task performance and
service OCB through the mediating effects of perceived LMX differentiation and team cohesion.
Several of servant leaders' characteristics partly explain why they positively impact team task performance through their influ-
ence on perceived LMX differentiation and team cohesion mediators. Given that servant leadership is based on the premise that
servant leaders build trust by selflessly focusing on serving team members' needs, they are able to build strong relationships that tend
to bring out the best in all their followers (Greenleaf, 1977; Liden et al., 2008; van Dierendonck, 2011). With this deep understanding
of followers' needs, coupled with their strong conceptual skills (Liden et al., 2008), servant leaders putatively empower and guide
each member's contribution in achieving team task performance goals, and support the team as they develop technical and problem-
solving skills that are essential for effective collaboration, and hence should positively influence team task performance (Hu & Liden,
2011; Liden et al., 2008). These supportive interactions between servant leader and all team members diminish employees' per-
ceptions of differences in relationship quality with their leader, and as servant leaders value and build confidence in all team
members, it minimizes the creation of sub-groups. Thus, it induces team members to reinforce ties between each other, to engage and
cooperate cohesively to accomplish team tasks, therefore enhance team cohesion. Empirically, servant leadership has been shown to
help satisfy psychological needs at work, leading to enhanced work engagement (van Dierendonck et al., 2014). Schaubroeck et al.
(2011) report that servant leadership positively predicts affect-based trust in the leader at the group level, which positively influences
team performance. Lastly, Hu and Liden (2011) empirically demonstrate that team potency, i.e. the shared belief in collective
capabilities (Campion, Medsker, & Higgs, 1993), mediates the relationship between servant leadership and team task performance.
Therefore, based on the above rationale and supporting empirical results we postulate that:
Hypothesis 4. At the group level, perceived LMX differentiation and group cohesion mediate the relationship between servant
leadership and team task performance.
We also predict that servant leadership positively influences service OCBs through the mediated effect of perceived LMX dif-
ferentiation and team cohesion. Servant leaders go beyond providing the necessary support to enhance team task performance; they
also play a crucial role in creating a serving climate (Auh et al., 2016; Liden et al., 2008; Liden, Wayne et al., 2014). This serving
climate begins with the way servant leaders build quality relationships, place others' best interests before their own, act ethically, and
engage in emotional support and helpful benevolent behaviors, thereby implicitly conveying how much they value and respect every
team member (Liden, Wayne et al., 2014). This positive experience with all team members should lessen employees' perceptions of
differences in relationship quality with the leader and contribute to strengthening attraction, and consequently induce greater po-
sitive affect and interactions within the team (Tyler & Blader, 2000, 2003). This serving climate spreads as employees, via social
learning (Sy, Côté, & Saavedra, 2005) and the emotional contagion process, transfer positive emotional states and emulate the servant
leaders' prosocial behaviors toward others outside the team, hence produce more service OCBs (Liden et al., 2008). Empirical re-
search indicates that servant leadership establishes team-level climates that lead team members to generate altruistic and helping
behaviors (Ehrhart, 2004; Hu & Liden, 2011). In support of our contention, Ling, Lin, and Wu (2016) find that servant leaders' service
orientation trickles down to the employees into customer service-oriented citizenship behaviors. Also, Auh et al. (2016) report that
the relationship between LMX differentiation and the service climate perceptions that employees form is mediated by relational
conflict within the team. Therefore, based on the above rationale and supporting empirical results, we postulate that:
Hypothesis 5. At the group level, perceived LMX differentiation and group cohesion mediate the relationship between servant
leadership and service OCB.

Method and measures

Sample and procedures

This study was conducted as part of a more extensive research project in a large North American company that designs and
manufactures high technology products. The sample population resulting from the data collection effort is the same as in Chiniara
and Bentein (2015).1 An electronic invitation was sent to 2508 first-level employees' work e-mail address with a link to our web-based
survey, asking them to fill in the team member questionnaire, which included demographic information, their assessment of their
supervisor's servant leadership, their team's cohesion and their perception of differentiation in leader-follower relationship quality.
We received 821 first-level employee questionnaires for a response rate of 33%. Two months later, we sent supervisors an electronic
invitation to fill in a web-based questionnaire assessing their teams' task performance and service OCBs; 157 supervisors responded,
for a response rate of 55%. All participants were entered in a draw for a bookstore gift certificate. Participants work in a variety of
departments including engineering, sales, planning, accounting and production. This sample population allowed us to match a total

1
Apart from the servant leadership construct which is common to this current study and in Chiniara and Bentein (2015), hypotheses and variables of interest of this
study are distinct from this earlier article.

7
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

of 130 teams with 403 employees. We considered employees to be members of a work team when they had the same supervisor and
when reporting task interdependence (Wageman, 2001). We excluded questionnaires with incomplete data, teams with two team
members and teams managed by a new supervisor (less than three months), and obtained a final sample of 67 teams with 229 team
members, and an average of 3.4 respondents per team. Team size varied from 3 to 29 team members, excluding the leader.
The first-level employee members of these teams are mostly Caucasian (95.5%) and Asian (3.4%). They are mainly male (65.9%)
and are, on average, 40 years old (SD = 9.5); 26.0% have been reporting to their current supervisor for 2 to 12 months, 49.4% from 1
to 3 years, 14% from 3 to 5 years, 8.1% from 5 to 10 years and 2.6% for over 10 years. Regarding tenure, 9.8% of the employees have
been with the company for 2 to 12 months, 14.9% for 1 to 3 years, 13.6% from 3 to 5 years, 17.9% from 5 to 10 years and 43.8% for
over 10 years. At this technology-driven company, most first-level employees have completed higher education: 68.3% hold a uni-
versity degree and 24.4% hold a trade school or college diploma.
Team supervisors of the teams are also largely Caucasian (97%), male (81%) and are on average 45.6 years old (SD = 7.4). They
have held their current position for 2 to 12 months (6.2%), for 1 to 3 years (46.2%), for 3 to 5 years (23.1%), for 5 to 10 years
(12.3%), for 10 to 15 years (9.2%) and for over 15 years (3.1%). Overall, supervisors have been with the company longer than first-
level employees: 44.6% have over 15 years' experience with the company, 20% have 10 to 15 years' experience, 16.9% have 5 to
10 years' experience, 9.2% have 3 to 5 years' experience, 6.2% have one to three years' experience and only 3% (two individuals)
have 2 to 12 months' experience. As for their level of education, 81.5% hold a university degree and 15.4% hold a trade school or
college diploma.

Measures

Servant leadership

Employees assessed their supervisor's servant leadership using an adaptation of Liden et al.’s (2008) 28-item, 7-dimension Servant
Leadership Scale. A sample item from the scale is “My supervisor seems to care more about my success than his/her own success”
(1 = strongly disagree to 5 = strongly agree). Following Hu and Liden (2011), we performed a confirmatory factor analysis (CFA) to
confirm a seven-factor model. Results show that the dimensions are distinct and fall under a second-order servant leadership con-
struct, χ2 (343) = 628.31, comparative fit index (CFI) = 0.92, Tucker-Lewis Fit index (TLI) = 0.92, standardized root-mean-square
residual (SRMR) = 0.06 and root-mean-square error of approximation (RMSEA) = 0.06. The factor loadings of the servant leader-
ship dimensions ranged between: 0.82 ≤ γ ≤ 0.91 (p < 0.001), with the exception of the creating value for the community dimen-
sion, which loaded at γ = 0.50 (p < 0.001) onto the second-order servant leadership construct. To be consistent with previous
studies, we included all seven dimensions and we used a latent factor to represent the servant leadership construct. The Cronbach's
alpha of this sample is = 0.91.

Perceived LMX differentiation

This construct was measured using Mayer et al.’s (2008) 4-item global measure of perceived differentiation in leader-follower
relationship quality within their work group. Employees evaluated on a 5-point scale (1 = strongly disagree to 5 = strongly agree) their
level of agreement that their leader has a high quality relationship with some employees and a low-quality relationship with others.
The four items from Mayer, Erdogan et al.’s measure are: “Some team members have a positive working relationship with my
supervisor while other team members do not,” “My supervisor has high-quality working relationships with some members of my
team, but low-quality working relationships with other team members,” “My supervisor has an effective working relationship with
some team members, but ineffective working relationships with other team members,” and “My supervisor tends to develop high-
quality working relationships with only a few trusted team members.” Cronbach's alpha (α) for our sample is 0.87. We did not select
Hooper and Martin's (2008) LMX distribution measure because in a preliminary test it showed a very low response rate, possibly
because of the greater cognitive effort it demands of participants to complete.

Team cohesion

Employees assessed their work group cohesion using an adapted version of Podsakoff, Niehoff, MacKenzie, and Williams' (1993)
6-item scale of team cohesion. This scale measures the extent to which work team members are closely knit on a social and task-
related basis. At the company's request, we adapted three items slightly by adding Carron's (1982) notion of reaching a common goal,
which better represented their team culture. The original item 3 was: “The members of my work group cooperate.” Item 3 became:
“The members of my work group cooperate to get the work done.” Items 2 and 4 were modified in the same manner and became:
“Members of my group work together as a team to meet team objectives,” and “My work group members know that they can depend on
each other to get work done.” Employees rated the extent to which they agree with each item on a 5-point scale (1 = strongly disagree
to 5 = strongly agree). The Cronbach's alpha (α) for our sample is 0.87, which is consistent with prior studies showing α > 0.86 (e.g.,
Podsakoff & Mackenzie, 1994; Podsakoff et al., 1997).

Team performance

Two classes of employee behaviors were assessed by the supervisor: team task performance and team service organizational

8
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

citizenship behaviors (OCB).

Team task performance was measured using Liden et al.’s (2006) 5-item group performance measure, which was designed and
validated in three types of organizations. We added to these five items the item initiatives taken as a team, which corresponded to a
team performance criterion recognized within the company culture. Supervisors were asked to rate their team's task performance in
terms of six criteria: quality of work, quantity of work, problem-solving speed, initiatives taken as a team, capability to complete work
on time and overall team performance. The respondents used a 7-point scale (1 = unacceptable, 2 = very poor, 3 = poor, 4 = ac-
ceptable, 5 = good, 6 = very good, and 7 = outstanding). Cronbach's alpha for this sample is α = 0.84.
Service OCB was assessed by team supervisors using an adapted version of Podsakoff, MacKenzie, Moorman, and Fetter's (1990)
organizational citizenship behaviors scale. The scale was adapted to specifically refer to the assessment of OCBs performed by a team
and directed to others outside the team. Hence, the team is considered as an entity and is assessed as such. This represents a clear
distinction from studies that define and assess team OCBs as the level of OCBs performed by individuals within a team and directed
toward members within the team (e.g., Cole, Carter, & Zhang, 2013; Hu & Liden, 2011). Service OCB is also distinct from service-
oriented OCB scale designed to assess the quality of service toward customers (Bettencourt, Gwinner, & Meuter's, 2001; Ling et al.,
2016). In addition, for ease of comprehension, three negatively worded items were reversed to the positive form. Supervisors in-
dicated the extent to which they agreed with the statement about their teams' OCBs on a 7-point scale (1 = strongly disagree to
7 = strongly agree). Our scale consists of two dimensions measuring the degree to which the team helps or shows respect and courtesy
to individuals outside the team. The three items measuring altruism are: “This team helps other teams who have heavy workloads,”
“This team willingly helps other teams who have work-related problems,” “This team is always ready to lend a helping hand to people
around the team.” The three items measuring civic virtue are: “This team considers the impact of their actions on others,” “This team
takes steps to try to prevent problems with others,” and “This team respects the rights of others.” We performed a confirmatory factor
analysis (CFA) to confirm a two-factor model. Results show that the dimensions are distinct and fall under a second-order construct,
χ2 (8) = 13.24, comparative fit index (CFI) = 0.94, Tucker-Lewis Fit index (TLI) = 0.89, standardized root-mean-square residual
(SRMR) = 0.09 and root-mean-square error of approximation (RMSEA) = 0.05. As recommended by LePine, Erez, and Johnson's
(2002) meta-analysis on OCBs dimensionality, OCB items overlap theoretically and empirically to the degree that one could use an
overall score averaging all dimensions. Accordingly, we used all 6 items as indicators for one team-level latent factor, service OCB.
Cronbach's alpha (α) for our sample is 0.77.

Control variables

Previous theoretical and empirical research on group performance found that team size and task interdependence may be related
to employees' attitudes and work-related outcomes (e.g., Liden et al., 2006). We therefore chose to include them as controls in our
analysis. Participants and supervisors stated the team size directly on the survey. Task interdependence was assessed with four items
developed by Pearce and Gregersen (1991). Team members reported their agreement with items such as “I must frequently co-
ordinate my efforts with others in my work group” using a 5-point scale (1 = strongly disagree to 5 = strongly agree). Consistent with
prior research (Liden et al., 2006), we measured task interdependence at the individual level. Cronbach's alpha (α) is 0.87.

Team level of analysis and data aggregation

Given that our model is designed at the group level of analysis, we assessed the team-level properties in several ways. First, we
examined task interdependence, a defining characteristic of work teams. Using the within-group agreement test rwg(j) (James,
Demaree, & Wolf, 1984), we verified that the homogeneity of perception of task interdependence was high, as indicated by a high
estimated median rwg(j) = 0.80. Intraclass correlations (ICC's; James, 1982) were assessed: ICC (1) was 0.15 and ICC (2) was 0.41,
suggesting acceptable levels of agreement and between-group differences (Bliese, 2000; Byrne, 2012). These statistics provide suf-
ficient support to aggregate individual scores of team interdependence at the team level. The mean level of interdependence
(M = 3.62, SD = 0.56) also supports that employees truly work in teams. Together these results provide initial support for ag-
gregating variables at the team level.
Second, we examined the appropriateness of aggregating individual scores at the team level for all our independent variables.
Estimated median rwg(j) values suggest strong agreement: 0.93 for servant leadership, 0.78 for perceived LMX differentiation and 0.88
for cohesion. Intraclass correlations ICC (1) and ICC (2) are respectively 0.15 and 0.42 for servant leadership, 0.24 and 0.56 for
perceived LMX differentiation and 0.33 and 0.66 for cohesion, indicating acceptable levels of agreement and between-group dif-
ferences (Bliese, 2000; Byrne, 2012; Julian, 2001). Together, these statistics provide sufficient support to aggregate individual scores
of all independent variables at the team level (Bliese, 2000; LeBreton & Senter, 2008).

Statistical analysis strategy

The hypothesized model was tested using multi-level structural equation modeling (MSEM) in the MPlus 7.4 package
(Muthén & Muthén, 1998-2015) using the maximum likelihood estimator (ML). Items were used as indicators for each variable,
except for the servant leadership construct, for which we created seven indicators representing the seven dimensions, and averaged
the four items by dimension. To examine the fit of the measurement and structural models, we used the Comparative Fit Index (CFI),
Tucker-Lewis Fit Index (TLI), Root Mean Square Error of Approximation (RMSEA) indices and Standardized Root Mean Square
Residual (SRMR). Together, a CFI and TLI value close to 0.95 or higher, an RMSEA value of 0.06 or lower and an SRMR value of 0.08

9
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

Table 1
Means, standard deviations, Cronbach's alphas and zero-order attenuated correlations of studied variables (unilateral significant test).

M SD 1 2 3 4 5 6 7

Individual level of analysis (N = 229 members)

1. Servant leadership 3.38 0.59 0.91
2. Perceived LMX differentiation 2.72 0.94 −0.38⁎⁎⁎ 0.87
3. Team cohesion 3.82 0.68 0.35⁎⁎⁎ − 0.44⁎⁎⁎ 0.87

Group level of analysis (N = 67 teams)

1. Team size 10.10 6.94 –
2. Task interdependence 3.62 0.54 0.05 –
3. Servant leadership 3.34 0.42 −0.24⁎ 0.01 –
4. Perceived LMX differentiation 2.73 0.75 0.25⁎ − 0.10 −0.36⁎⁎ –
5. Team cohesion 3.85 0.53 −0.18 0.31⁎⁎ 0.33⁎⁎ −0.58⁎⁎⁎ –
6. Team Task Performance 5.57 0.56 −0.13 0.14 0.09 −0.26⁎ 0.20⁎ 0.84
7. Service OCB 5.86 0,52 −0.09 0.02 0.08 −0.14 0.34⁎⁎ 0.44⁎⁎⁎ 0.77

Note: Cronbach's alpha coefficients are presented on the diagonal in bold.

⁎
p < 0.05.
⁎⁎
p < 0.01.
⁎⁎⁎
p < 0.001.

or lower indicate a good model fit (Hu & Bentler, 1999). We also report R2. To assess for full or partial mediation, we tested for size
and significance of the direct and indirect effects of the independent variable (servant leadership) on the dependent variables, using
confidence intervals (Preacher & Kelley, 2011).

Results

Measurement model and descriptive statistics

In preliminary analyses we tested a measurement model at the individual level with each of the three latent variables, namely
servant leadership, perceived LMX differentiation and team cohesion, related to their respective indicators and using the maximum
likelihood (ML) estimator setting (Muthén & Muthén, 1998-2015). This three-variable measurement model fits the data very well
with χ2 (116) = 168.36, CFI = 0.98, TLI = 0.97, RMSEA = 0.04 and SRMR = 0.04 (Hu & Bentler, 1999). All indicators loaded
significantly onto their corresponding latent constructs (p < 0.001). The factor loading range for servant leadership is 0.69 to 0.89,
and 0.44 for the community dimension; 0.68 to 0.92 for Perceived LMX differentiation, and 0.64 to 0.78 for cohesion.
We further tested the measurement model at the group level by adding the two performance latent variables, yielding a five-factor
model. The measurement model shows a good fit with the data: χ2 (483) = 538.80, CFI = 0.98, TLI = 0.97, RMSEA = 0.02 and
SRMR within = 0.03 and between = 0.13 (Hu & Bentler, 1999). All indicators loaded significantly onto their corresponding latent
constructs. At the ‘within’ level, the factor loading range for servant leadership is 0.50 to 0.86, for perceived LMX differentiation the
range is 0.59 to 0.89, and for cohesion it is 0.53 to 0.73. At the ‘between’ level, the factor loading range team task performance is 0.55
to 0.88, and is 0.44 to 0.82 for service OCB.
Table 1 reports the descriptive statistics and attenuated correlations among variables at the individual level and the group level.
Scales showed satisfactory reliability values (0.77 ≤ α ≤ 0.91). At the individual level, correlations between variables are in the
expected direction: servant leadership is significantly and negatively correlated to perceived LMX differentiation, and positively
correlated to team cohesion; perceived LMX differentiation is significantly and negatively correlated to both servant leadership and
team cohesion. At the group level, as theoretically supported and anticipated, servant leadership is significantly and negatively
correlated to perceived LMX differentiation (r = − 0.36, p < 0.01) and significantly and positively correlated to team cohesion
(r = 0.33, p < 0.01). Perceived LMX differentiation is, as anticipated, significantly and negatively correlated to team cohesion
(r = −0.58, p < 0.001). Team cohesion is significantly correlated to both components of performance: team task performance
(r = 0.20, p < 0.05) and service OCB (r = 0.34, p < 0.01).
To complete our preliminary analyses, we investigated whether our control variables team size and task interdependence were
associated with servant leadership (independent variable), perceived LMX differentiation or team cohesion (mediators) and per-
formance variables (outcomes), as recommended by Becker (2005). Correlations are reported in Table 1. Team size correlated
significantly and negatively with servant leadership (r = − 0.24, p < 0.05), and significantly and positively with perceived LMX
differentiation (r = 0.25, p < 0.05). Task interdependence correlated significantly and positively with team cohesion (r = 0.31,
p < 0.01). Therefore, we ran our analyses by controlling for team size and task interdependence. In order to reduce the number of
parameters to be estimated in our model, we entered task interdependence as an observed variable at the between level, using the
team's average task interdependence value. Results show that team size was significantly related with servant leadership
(β = −0.36, p < 0.05) and task interdependence was significantly related with team cohesion (β = 0.35, p < 0.01), we thus
included it in our model for the remaining analyses.

10
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

Fig. 2. Structural equation model of servant leadership group-level mediation model on team performance, with standardized coefficients. The effect of the control
variables team size and task interdependence are not represented here.
N = 67 teams, ** = p < 0.01, * = p < 0.05.

Structural model

The five-factor structural model representing the hypothesized relationships presented in Fig. 1 shows a good fit to the data, with
χ2 (546) = 615.54, CFI = 0.97, TLI = 0.97, RMSEA = 0.02 and SRMR within = 0.03 and between = 0.13 (Hu & Bentler, 1999).
Supporting Hypothesis 1, servant leadership significantly and negatively influences perceived LMX differentiation (β = − 0.63,
p < 0.05). In support of Hypothesis 2, perceived LMX differentiation significantly and negatively predicted team cohesion
(β = − 0.68, p < 0.01). Supporting Hypothesis 3a and 3b, team cohesion significantly and positively predicted team task perfor-
mance (β = 0.43, p < 0.05) and service OCB (β = 0.47, p < 0.05). Standardized coefficient estimates associated with Hypothesis
1, 2, 3a and 3b are presented in Fig. 2. This model accounted for 40.2% of the variance of perceived LMX differentiation, 58.7% of the
variance of team cohesion, 18.1% of the variance of team task performance, and 21.6% of service OCB.
Turning to the two mediation hypotheses, Hypotheses 4 and 5, we tested for full vs partial mediating effects by adding two direct
paths: one between servant leadership and team task performance and one between servant leadership and service OCB. Our results
show that both added paths were not significant (p = 0.52 and p = 0.87 respectively), that parameter estimates and fit indices were
exactly the same as without the direct paths, and that χ2 difference between this model and the one we hypothesized was not
significant at Δχ2(2) = 0.521, ns. The size of servant leadership's indirect effect on team task performance through perceived LMX
differentiation and team cohesion is 0.28, with a confidence interval of 90% CI [0.021, 0.536], therefore indicating that the indirect
effect was significant and supporting Hypothesis 4. Also, because the direct link between servant leadership and team task perfor-
mance was not significant, the mediation is considered complete. Servant leadership's indirect effect on service OCB through per-
ceived LMX differentiation and team cohesion is 0.16, with a confidence interval 90% CI [0.001, 0.321], therefore indicating that this
effect was significant and supporting Hypothesis 5. Again, because the direct link between servant leadership and service OCB was
not significant, the mediation is considered complete.

Discussion

The main contribution of the current study is to help answer the question of how servant leadership influences team performance.
Drawing on the theory of social comparison as a theoretical framework, this group-level study shows that servant leaders engender
follower perceptions of low differentiation in leader-follower relationship quality within a team (perceived LMX differentiation),
which in turn positively affects cohesiveness between team members, creating a social context that positively influences team task
performance and service OCB. These findings make several key theoretical contributions to the servant leadership and LMX differ-
entiation literatures.
Our findings significantly contribute to servant leadership literature in three ways. First, our results confirm servant leadership's
theoretical central premise that servant leaders' distinctive focus on serving individuals, by placing the good of the employees over
the self-interest of the leader, improves collective performance. We hypothesized and found that servant leadership has a positive
influence on team-level performance. As such, this result complements the few empirical studies that supported the significance of
servant leadership to build collective performance (e.g., Hu & Liden, 2011; Liden, Wayne et al., 2014). Second, our study goes beyond
those studies because we demonstrate that servant leadership not only predicts team-task performance, but also service OCB. Indeed,
our results substantiate another servant leadership theoretical premise suggesting that the essence of servant leaders' values in caring
and serving others, extends into followers desire to collectively be helpful and be courteous to others and become servant leaders
themselves. We postulated and found that the more leaders demonstrate servant leadership behaviors, the more team members
collectively show service OCBs toward others outside their team, but still within the organization. This finding is consistent with the
existing literature on customer service, which also examines the influence of servant leadership on pro-social behaviors. However in
that literature it is applied to front-line employees' pro-social behaviors targeting customers, who are not only outside the team, but
also outside the organization. Similar to our findings, these studies show that front-line employees are positively influenced to act
more pro-socially toward their customers when servant leaders demonstrates a greater serving attitude and altruistic values (e.g.,
Chen, Zhu, & Zhou, 2015; Hsiao, Lee, & Chen, 2015). Third, our findings shed new light on the mechanisms triggered by servant

11
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

leaders to influence team performance. Previous studies have shown that team potency (Hu & Liden, 2011), serving culture (Hunter
et al., 2013; Liden, Wayne et al., 2014), affect-based trust in leaders (Schaubroeck et al., 2011) or procedural justice climate (Ehrhart,
2004) can act as mediators of the relationship between servant leadership and team performance. By examining the focal within-
group relational processes, our study brings a new perspective to better understand how servant leadership influences its team
members. We hypothesized and found that LMX differentiation and team cohesion fully mediate the relationship between servant
leadership and both team task performance and team service OCB. Thus, our results suggest that servant leaders' positive behavior
and attitude toward all followers reduce social comparison and categorization processes. Team members perceive less differentiation
between them, in terms of the quality of the leader-member relationship, inhibiting the process of categorizing members into in-
group and out-groups, which leads to a higher team cohesion. Guided by the positive influence of servant leaders and fortified by
their cohesiveness, team members are positively induced to collectively accomplish team tasks and service OCB. This finding is
particularly interesting because even if LMX differentiation and cohesion act as mediators of the relationship between servant lea-
dership and both team task performance and service OCB, because LMX differentiation and cohesion predict two different outcomes,
we assume that our mediators act in a different way to explain their influences on each outcome. In the first case, we postulate that
servant leaders' positive support of all followers minimizes team members LMX differentiation and increases team cohesion, which
influences team members' focus, efforts, enthusiasm and commitment to perform tasks and achieve goals, thus enhancing team task
performance. As for the case of the mediation between servant leadership and team service OCB, we also suggest that servant leaders
positive relationship with all followers minimizes LMX differentiation among team members and increases team cohesion. However,
in this case, we suggest that the lower differentiation and enhanced team cohesion would facilitate the expansion of the servant
leader's service climate via followers' social learning (Sy et al., 2005) and emotional contagion processes. As servant leaders establish
and promote a climate of service and caring for others, we suggest that team members acquire positive emotional states and emulate
the servant leaders' prosocial behaviors, while transferring this positive emotional state toward others outside the team, consequently
producing more service OCBs. Hence, our findings extend existing empirical research which indicate that servant leadership es-
tablishes team-level climates that lead team members to generate altruistic and helping behaviors (Ehrhart, 2004; Hu & Liden, 2011).
Further, our findings are also consistent with very recent research that reports that service climate formation can be influenced by
team relational processes (e.g., Auh et al., 2016).
Our findings also significantly contribute to LMX differentiation literature in the following three ways. First, we contribute by
examining an antecedent of LMX differentiation that has not yet been investigated. To date, empirical research on the antecedents of
LMX differentiation has been quite scarce (Henderson et al., 2009; Anand et al., 2016). As proposed by Henderson et al. (2009), we
hypothesized and found that as servant leadership engages in supportive relationships with all team members, servant leadership
negatively affects LMX differentiation. This result also highlights how even though the nature and content of each leader-follower
relationship with the servant leader necessarily varies from one follower to another, team members still perceive less differentiation,
because servant leaders build high quality relationships with all team members. Second, our results help demonstrate the value and
interest of measuring the perception of LMX differentiation at the group level, as this allows us to observe its effect on a group
process, namely team cohesion. Perceptions of within-group variability of the quality of leader-follower relationship offer a unique
perspective for explaining basic team relational processes, such as we observed with team cohesion. Our results showed that the more
team members perceived low differentiation in the quality of the leader-follower relationship between team members, the stronger
the positive influence on team cohesion. Our central argument to explain the effect of LMX differentiation on team cohesion is that
LMX differentiation introduces relational boundaries within teams, dividing team members and fragmenting the group into sub-
groups, which undermines team cohesion (Anand et al., 2011). This result is also consistent with the studies that investigated the
effect of perceived differentiation in quality the of the leader-follower relationship in teams, and found that it was negatively related
to helping and positively related to conflict (Auh et al., 2016; Hooper & Martin, 2008; Mayer et al., 2008). The positive influence of
low perceived LMX differentiation on team cohesion can also be found in Stoverink, Umphress, Gardner, and Miner's (2014) ex-
perimental study, which reports that in a situation where all team members receive the same bad treatment from the supervisor,
teams reported greater cohesiveness, thereby highlighting that it is the perception of a differentiation in relationships that plays a key
role in augmenting team cohesion and not the leadership style alone. Third, our study also contributes to LMX differentiation
literature by using the LMX differentiation measure introduced by Mayer et al. (2008), as an alternative to Hooper and Martin's
(2008) LMX distribution measure that was used in a few other studies (e.g., Epitropaki et al., 2016) and that we were not able to use
in this study. Although the LMX distribution measure of Hooper and Martin (2008) consists of only one item, it requires the parti-
cipant to evaluate the quality of the leader-member relationship for each team member individually, which may prove to be a
cognitively challenging exercise to complete. In contrast, because Mayer, Erdogan et al.’s measure is more global, it makes it possible
to capture the broader experience of the differentiation assessment, without having to specifically assess the quality of the LMX
relationship of each individual team member, which does not necessarily accurately reflect the holistic differentiation judgment felt.
Therefore, we propose that Mayer, Erdogan et al.’s measure provides a practical way of quantifying a global perception of LMX
differentiation, which is particularly useful for studies interested in investigating its effect over global or collective outcomes such as
team performance. This proposal supports Anand et al.'s (2016) contention that there is a need to further develop a group-level
perceptual measurement and studies.

Practical implications

Our study highlights a number of practical implications for organizations. First, our findings emphasize how managers would
benefit from using servant leadership-type behavior to boost team cohesion and indirectly enhance team service OCBs. Managers

12
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

should pay particular attention to the negative impact of differentiation in the quality of relationships among followers: because this
differentiation is naturally perceived by team members, it may engender relational boundaries within the team, thwart attraction
forces within the team and hinder team cohesion and cooperation. Second, our results highlight the value of securing team cohesion
in groups and organizations. Although leadership plays a key role in nurturing and promoting team cohesion, cohesion can also
become an organizational development objective. The organization will thus adopt strategies that aim to strengthen ties and forces
that bind the team together, and to measure increases in team cohesiveness. Third, given the significant impact of servant leadership
in fostering team cohesion and enhancing team performance, it would be advantageous for organizations to select and train leaders
who are capable of putting followers' individual needs before their own and who are capable of building quality relationships with all
team members. Fourth, when servant leaders embody and promote a climate of serving those outside the team, it naturally induces
followers to be also more altruistic and show helpful and courteous behaviors toward others outside the team. Finally, and most
importantly, promoting a servant leadership-type culture in organizations where human resources development is valued and en-
couraged is particularly relevant in today's complex and competitive society, where people strive to make sense of their working
experience and hope to find human-centered and strong ethical leadership.

Strengths, limitations and future research

Our study was designed to maximize its internal validity. To reduce potential same-source bias, we collected information from
two different sources at two different times. Employees' evaluation of servant leadership, perceived LMX differentiation and team
cohesion were collected two months before direct supervisors filled in employees' performance information.
Naturally, our study has limitations that represent areas of future research. First, participants in this study possessed a high level
of formal education and worked in the private sector, representing a population that may naturally be more open to developing their
own potential and consequently be more receptive and sensitive to a servant leadership-type influence. To increase the general-
izability of our results, future studies should investigate different populations (i.e. education levels and industries) and cultural
environments. Second, although our study was designed to collect data in a time-lagged fashion, and although the theories we drew
upon support the direction of the relationships we hypothesized, this study is cross-sectional by design, and cannot make any causal
claims regarding the relationship between servant leadership, perceived LMX differentiation, team cohesion and team performance. It
is therefore possible that some of the relationships may exist in the opposite direction. For instance, Mathieu et al. (2015) analyzed 17
previous studies and examined the reciprocal and dynamic nature of the team performance/team cohesion relationship, finding that
both directions exist and that they evolve differently through time. They also found that the strength of team cohesion ➔ team
performance is higher than the relationship in the opposite direction. When a team's cohesion is high, the members cooperate and
collaborate in a way that enhances the team's performance in a stronger manner than the opposite, i.e. teams with a positive view of
their own performance have higher cohesion. Additional research could attempt to replicate and advance this knowledge. Also
considering the composition of the team, it may be interesting to examine the effect on team processes, of coworkers that are
counterproductive or loafers. This may be possible by integrating individual performance variability as a moderator into the model.2
Further, studies could examine the conditions or contexts under which servant leadership may or may not be effective. Contextual
variables such as other organizational cultures or types of work may influence the type of leadership that can be effective.
To further extend the findings of this study as well as those of other recent studies that have examined mediators of servant
leadership on outcomes (e.g., Liden, Wayne et al., 2014; van Dierendonck et al., 2014), we recommend examining cross-level in-
fluences of servant leadership. For instance, we encourage investigation of how a climate of servant leadership and team cohesion
would impact followers at the individual level for outcomes such as individual performance, creativity and increased employee
servant leadership behaviors. Building on the importance of the quality of the follower-leader relationship, it would be interesting to
investigate the dynamic nature of that relationship and its evolution over time. Individual differences such as personality traits,
values, or leadership prototype preference may also constitute influential moderators of followers' attitudes and behaviors as they
relate to servant leadership.
Finally, turning to LMX differentiation, although our study's design brings forth a negative consequence of LMX differentiation,
we still acknowledge that LMX differentiation can lead to positive consequences and that it tends to do so when LMX differentiation is
related to individual-level outcomes. LMX scholars maintain that LMX differentiation operates at both the individual and group level
simultaneously and that the nature of the relationship between LMX differentiation and outcomes may differ across levels (Anand
et al., 2016; Henderson et al., 2009). LMX differentiation seems to have a positive influence when it's examined in relation to
individual level outcomes, such as individual performance, where equity-based judgments sufficiently explain outcomes variations
(Anand et al., 2016). Yet, at group level, the norm of equality is likely violated when unequal distribution of resources is observed. In
the present study, our outcome is a group-relational mechanism, cohesion, that bonds team members to one another and to the
purpose of the team, and as such, is better served by a low degree of within-team variation in the quality of the relationships between
the supervisor and the different employees, hence likely respecting equality principles. Future research might want to continue
refining our understanding of the differential mechanisms of LMX differentiation at different levels and with different outcomes,
taking justice principles into consideration.

2
We would like to thank one of the reviewers for this suggestion.

13
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

Conclusion

In our modern and complex work context where teams' effectiveness is essential for organizational effectiveness, our study
provides theoretical and empirical insight into the growing field of servant leadership by expanding and strengthening Greenleaf's
original premise that servant leadership impacts organizational effectiveness through its focus on serving others. We investigated the
effects of servant leadership on team performance at the group level, on a sample of work teams, by drawing on social comparison
theory. Our results demonstrate that servant leadership boosts team cohesion through its influence on the perception of low dif-
ferentiation in the quality of follower-leader relationships within the team. Our study also contributes to a better understanding of
team cohesion's impact on team performance; specifically, we found that team cohesion strongly enhances both team task perfor-
mance and team service OCB. Finally, we demonstrate how servant leadership can influence team task performance and service OCB,
through the mediating effect of these two team processes: perceived LMX differentiation and team cohesion. As organizations
continue to rely on teams to successfully achieve challenging goals, and as management teams begin to embrace the ideals of servant
leadership, we underscore the importance for both managers and academics of continuing to examine the inspirational and effective
influence of this type of leadership on team outcomes.

References

Anand, S., Hu, J., Liden, R. C., & Vidyarthi, P. R. (2011). Leader-member exchange: Recent research findings and prospects for the future. In A. Bryman, D. Collinson,
K. Grint, B. Jackson, & M. Uhl-Bien (Eds.), The Sage handbook of leadership (pp. 311–325). Thousand Oaks, Ca: Sage.
Anand, S., Vidyarthi, P. R., & Park, H. S. (2016). LMX differentiation: Understanding relational leadership at individual and group levels. In T. N. E. E. Bauer, & Berrin
(Vol. Eds.), The Oxford handbook of leader-member exchange. Vol. xviii. The Oxford handbook of leader-member exchange (pp. 263–291). New York, NY, US: Oxford
University Press 435 pp.
Auh, S., Bowen, D., Aysuna, C., & Menguc, B. (2016). A search for missing links: Specifying the relationship between leader-member exchange differentiation and
service climate. Journal of Service Research, 19(3), 260–275.
Bateman, T. S., & Organ, D. W. (1983). Job satisfaction and the good soldier: The relationship between affect and employee "citizenship". Academy of Management
Journal, 26(4), 587–595. http://dx.doi.org.proxy.bibliotheques.uqam.ca:2048/10.2307/255908.
Baumeister, R. F., & Leary, M. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3),
497–529.
Beal, D. J., Cohen, R. R., Burke, M. J., & McLendon, C. L. (2003). Cohesion and performance in groups: A meta-analytic clarification of construct relations. Journal of
Applied Psychology, 88(6), 989–1004.
Becker, T. E. (2005). Potential problems in the statistical control of variables in organizational research: A qualitative analysis with recommendations. Organizational
Research Methods, 8(3), 274–289.
Benefiel, M., Fry, L. W., & Geigle, D. (2014). Spirituality and religion in the workplace: History, theory, and research. Psychology of Religion and Spirituality, 6(3),
175–187.
Bettencourt, L. A., Gwinner, K. P., & Meuter, M. L. (2001). A comparison of attitude, personality, and knowledge predictors of service-oriented organizational
citizenship behaviors. Journal of Applied Psychology, 86(1), 29–41.
Bliese, P. D. (2000). Within-group agreement, non-independence, and reliability: Implications for data aggregation and analysis. In K. J. Klein, & S. W. J. Kozlowski
(Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions (pp. 349–381). San Francisco, CA, US: Jossey-Bass.
Boies, K., & Howell, J. M. (2006). Leader-member exchange in teams: An examination of the interaction between relationship differentiation and mean LMX in
explaining team-level outcomes. The Leadership Quarterly, 17(3), 246–257.
van Breukelen, W., Schyns, B., & Le Blanc, P. (2006). Leader-member exchange theory and research: Accomplishments and future challenges. Leadership, 2(3),
295–316.
Brown, M. E., Treviño, L. K., & Harrison, D. A. (2005). Ethical leadership: A social learning perspective for construct development and testing. Organizational Behavior
and Human Decision Processes, 97(2), 117–134.
Brown, S., & Lam, S. K. (2008). A meta-analysis of relationships linking employee satisfaction to customer responses. Journal of Retailing, 84(3), 243–255.
Byrne, B. M. (2012). Structural equation modeling with Mplus: Basic concepts, applications, and programming. New York, NY: Routledge.
Campion, M. A., Medsker, G. J., & Higgs, A. C. (1993). Relations between work group characteristics and effectiveness: Implications for designing effective work
groups. Personnel Psychology, 46(4), 823–850.
Carron, A. V. (1982). Cohesiveness in sport groups: Interpretations and considerations. Journal of Sport Psychology, 4(2), 123–138.
Casey-Campbell, M., & Martens, M. L. (2009). Sticking it all together: A critical assessment of the group cohesion–performance literature. International Journal of
Management Reviews, 11(2), 223–246.
Castaño, N., Watts, T., & Tekleab, A. G. (2013). A reexamination of the cohesion–performance relationship meta-analyses: A comprehensive approach. Group Dynamics:
Theory, Research, and Practice, 17(4), 207–231.
Chen, C. V., Tang, Y.-Y., & Wang, S.-J. (2009). Interdependence and organizational citizenship behavior: Exploring the mediating effect of group cohesion in multilevel
analysis. The Journal of Psychology: Interdisciplinary and Applied, 143(6), 625–640.
Chen, Z., Zhu, J., & Zhou, M. (2015). How does a servant leader fuel the service fire? A multilevel model of servant leadership, individual self identity, group
competition climate, and customer service performance. Journal of Applied Psychology, 100(2), 511–521.
Chiniara, M., & Bentein, K. (2015). Linking servant leadership to individual performance: Differentiating the mediating role of autonomy, competence and relatedness
need satisfaction. The Leadership Quarterly, 27(1), 124–141.
Chiocchio, F., & Essiembre, H. (2009). Cohesion and performance: A meta-analytic review of disparities between project teams, production teams, and service teams.
Small Group Research, 40(4), 382–420.
Cole, M. S., Carter, M. Z., & Zhang, Z. (2013). Leader–team congruence in power distance values and team effectiveness: The mediating role of procedural justice
climate. Journal of Applied Psychology, 98(6), 962–973.
Dansereau, F., Graen, G., & Haga, W. J. (1975). A vertical dyad linkage approach to leadership within formal organizations: A longitudinal investigation of the role
making process. Organizational Behavior and Human Performance, 13(1), 46–78.
van Dierendonck, D. (2011). Servant leadership: A review and synthesis. Journal of Management, 37(4), 1228–1261.
Duchon, D., Green, S. G., & Taber, T. D. (1986). Vertical dyad linkage: A longitudinal assessment of antecedents, measures, and consequences. Journal of Applied
Psychology, 71(11), 56–60.
Ehrhart, M. G. (2004). Leadership and procedural justice climate as antecedents of unit-level organizational citizenship behavior. Personnel Psychology, 57(1), 61–94.
Ehrhart, M. G., & Naumann, S. E. (2004). Organizational citizenship behavior in work groups: A group norms approach. Journal of Applied Psychology, 89(6), 960–974.
Epitropaki, O., Kapoutsis, I., Ellen, B. P., Ferris, G. R., Drivas, K., & Ntotsi, A. (2016). Navigating uneven terrain: The roles of political skill and lmx differentiation in
prediction of work relationship quality and work outcomes. Journal of Organizational Behavior in press.
Erdogan, B., & Bauer, T. N. (2010). Differentiated leader–member exchanges: The buffering role of justice climate. Journal of Applied Psychology, 95(6), 1104–1120.
Ferris, G. R., Liden, R. C., Munyon, T. P., Summers, J. K., Basik, K. J., & Buckley, M. R. (2009). Relationships at work: Toward a multidimensional conceptualization of

14
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

dyadic work relationships. Journal of Management, 35(6), 1379–1403.

Festinger, L. (1950). Informal social communication. Psychological Review, 57(5), 271–282.
Festinger, L. (1954). A theory of social comparison processes. Human Relations; Studies Towards the Integration of the Social Sciences, 7, 117–140.
Ford, L. R., & Seers, A. (2006). Relational leadership and team climates: Pitting differentiation versus agreement. The Leadership Quarterly, 17(3), 258–270.
Fry, L. W. (2003). Toward a theory of spiritual leadership. The Leadership Quarterly, 14(6), 693–727.
George, J. M., & Bettenhausen, K. (1990). Understanding prosocial behavior, sales performance, and turnover: A group-level analysis in a service context. Journal of
Applied Psychology, 75(6), 698–709.
George, J. M., & Brief, A. P. (1992). Feeling good-doing good: A conceptual analysis of the mood at work-organizational spontaneity relationship. Psychological Bulletin,
112(2), 310–329.
Graen, G. B., & Uhl-Bien, M. (1995). Relationship-based approach to leadership: Development of leader-member exchange (LMX) theory of leadership over 25 years:
Applying a multi-level multi-domain perspective. The Leadership Quarterly, 6(2), 219–247.
Graham, J. W. (1991). Servant-leadership in organizations: Inspirational and moral. The Leadership Quarterly, 2(2), 105–119.
Greenleaf, R. K. (1977). Servant leadership: A journey into the nature of legitimate power and greatness. New York: Paulist Press.
Greenleaf, R. K. (1998). The power of servant-leadership. San Francisco: Berrett-Koehler.
Greer, L. L. (2012). Group cohesion: Then and now. Small Group Research, 43(6), 655–661.
Gully, S. M., Devine, D. J., & Whitney, D. J. (1995). A meta-analysis of cohesion and performance: Effects of levels of analysis and task interdependence. Small Group
Research, 26(4), 497–520.
Hackman, J. R. (1992). Group influences on individuals in organizations. In M. D. Dunnette, & L. M. Hough (Vol. Eds.), Handbook of industrial and organizational
psychology(2nd ed.). Vol. 3. Handbook of industrial and organizational psychology (pp. 199–267). Palo Alto, CA, US: Consulting Psychologists Press.
Harris, T. B., Li, N., & Kirkman, B. L. (2014). Leader–member exchange (LMX) in context: How LMX differentiation and LMX relational separation attenuate LMX's
influence on OCB and turnover intention. The Leadership Quarterly, 25(2), 314–328.
Henderson, D. J., Liden, R. C., Glibkowski, B. C., & Chaudhry, A. (2009). LMX differentiation: A multilevel review and examination of its antecedents and outcomes.
The Leadership Quarterly, 20(4), 517–534.
Henderson, D. J., Wayne, S. J., Shore, L. M., Bommer, W. H., & Tetrick, L. E. (2008). Leader—member exchange, differentiation, and psychological contract fulfillment:
A multilevel examination. Journal of Applied Psychology, 93(6), 1208–1219.
Hoch, J. E., Bommer, W. H., Dulebohn, J. H., & Wu, D. (2016). Do ethical, authentic, and servant leadership explain variance above and beyond transformational
leadership? A meta-analysis. Journal of Management. http://dx.doi.org/10.1177/0149206316665461 Epub ahead of print.
Hogg, M. A. (2006). Social identity theory. In P. J. Burke (Ed.), Contemporary social psychological theories. Stanford: Stanford University Press.
Hogg, M. A., & Abrams, D. (1988). Social identifications: A social psychology of intergroup relations and group processes Florence. KY, US: Routledge.
Hooper, D. T., & Martin, R. (2008). Beyond personal leader—member exchange (LMX) quality: The effects of perceived LMX variability on employee reactions. The
Leadership Quarterly, 19(1), 20–30.
Hsiao, C., Lee, Y.-H., & Chen, W.-J. (2015). The effect of servant leadership on customer value co-creation: A cross-level analysis of key mediating roles. Tourism
Management, 49, 45–57.
Hu, J., & Liden, R. C. (2011). Antecedents of team potency and team effectiveness: An examination of goal and process clarity and servant leadership. Journal of Applied
Psychology, 96(4), 851–862.
Hu, J., & Liden, R. C. (2013). Relative leader–member exchange within team contexts: How and when social comparison impacts individual effectiveness. Personnel
Psychology, 66(1), 127–172.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation
Modeling, 6(1), 1–55.
Hunter, E. M., Neubert, M. J., Perry, S. J., Witt, L. A., Penney, L. M., & Weinberger, E. (2013). Servant leaders inspire servant followers: Antecedents and outcomes for
employees and the organization. The Leadership Quarterly, 24(2), 316–331.
James, L. R. (1982). Aggregation bias in estimates of perceptual agreement. Journal of Applied Psychology, 67(2), 219–229.
James, L. R., Demaree, R. G., & Wolf, G. (1984). Estimating within-group interrater reliability with and without response bias. Journal of Applied Psychology, 69(1),
85–98.
Julian, M. W. (2001). The consequences of ignoring multilevel data structures in nonhierarchical covariance modeling. Structural Equation Modeling, 8(3), 325–352.
Kidwell, R. E., Jr., Mossholder, K. W., & Bennett, N. (1997). Cohesiveness and organizational citizenship behavior: A multilevel analysis using work groups and
individuals. Journal of Management, 23(6), 775–793.
Kozlowski, S. W. J., & Bell, B. S. (2013). Work groups and teams in organizations. In N. W. Schmitt, S. Highhouse, & I. B. Weiner (Eds.), Handbook of psychology (pp.
412–469). (2nd ed.). Hoboken, NJ, US: John Wiley & Sons Inc.
Kozlowski, S. W. J., & Chao, G. T. (2012). The dynamics of emergence: Cognition and cohesion in work teams. Managerial and Decision Economics, 33(5–6), 335–354.
Kozlowski, S. W. J., & Ilgen, D. R. (2006). Enhancing the effectiveness of work groups and teams. Psychological Science in the Public Interest, 7(3), 77–123.
Le Blanc, P. M., & González-Romá, V. (2012). A team level investigation of the relationship between leader–member exchange (LMX) differentiation, and commitment
and performance. The Leadership Quarterly, 23(3), 534–544.
LeBreton, J. M., & Senter, J. L. (2008). Answers to 20 questions about interrater reliability and interrater agreement. Organizational Research Methods, 11(4), 815–852.
LePine, J. A., Erez, A., & Johnson, D. E. (2002). The nature and dimensionality of organizational citizenship behavior: A critical review and meta-analysis. Journal of
Applied Psychology, 87(1), 52–65.
Li, A. N., & Liao, H. (2014). How do leader–member exchange quality and differentiation affect performance in teams? An integrated multilevel dual process model.
Journal of Applied Psychology, 99(5), 847–866.
Liden, R. C., Erdogan, B., Wayne, S. J., & Sparrowe, R. T. (2006). Leader-member exchange, differentiation and task interdependence: Implications for individual and
group performance. Journal of Organizational Behavior, 27(6), 723–746.
Liden, R. C., Panaccio, A., Meuser, J. D., Hu, J., & Wayne, S. J. (2014a). Servant leadership: Antecedents, processes, and outcomes. In D. V. Day (Ed.), The Oxford
handbook of leadership and organizations. Oxford library of psychology (pp. 357–379). New York, NY, US: Oxford University Press.
Liden, R. C., Wayne, S. J., Liao, C., & Meuser, J. D. (2014b). Servant leadership and serving culture: Influence on individual and unit performance. Academy of
Management Journal, 57(5), 1434–1452.
Liden, R. C., Wayne, S. J., Zhao, H., & Henderson, D. (2008). Servant leadership: Development of a multidimensional measure and multi-level assessment. The
Leadership Quarterly, 19(2), 161–177.
Ling, Q., Lin, M., & Wu, X. (2016). The trickle-down effect of servant leadership on frontline employee service behaviors and performance: A multilevel study of
Chinese hotels. Tourism Management, 52, 341–368.
Mathieu, J. E., Kukenberger, M. R., D'Innocenzo, L., & Reilly, G. (2015). Modeling reciprocal team cohesion–performance relationships, as impacted by shared
leadership and members' competence. Journal of Applied Psychology, 100(3), 713–734.
Mayer, D. M. (2010). Servant leadership and follower need satisfaction: Where do we go from here? In D. V. Dierendonck, & K. Patterson (Eds.), Servant leadership.
Developments in theory and practice (pp. 147–154). Hampshire, UK: Palgrave Macmillan.
Mayer, D. M., Erdogan, B., & Piccolo, R. F. (2008, August). Does LMX differentiation help or hinder group processes and performance? Paper presented at the Academy of
management annual meeting. California: Anaheim.
Mullen, B., & Copper, C. (1994). The relation between group cohesiveness and performance: An integration. Psychological Bulletin, 115(2), 210–227.
Muthén, L. K., & Muthén, B. O. (1998-2015). Mplus user's guide (7th ed.). Los Angeles: Muthén & Muthén.
Naidoo, L. J., Scherbaum, C. A., Goldstein, H. W., & Graen, G. B. (2011). A longitudinal examination of the effects of LMX, ability, and differentiation on team
performance. Journal of Business and Psychology, 26(3), 347–357.
Neubert, M. J., Kacmar, K. M., Carlson, D. S., Chonko, L. B., & Roberts, J. A. (2008). Regulatory focus as a mediator of the influence of initiating structure and servant
leadership on employee behavior. Journal of Applied Psychology, 93(6), 1220–1233.

15
M. Chiniara, K. Bentein The Leadership Quarterly xxx (xxxx) xxx–xxx

Northouse, P. G. (2007). Leadership: Theory and practice (4th ed.). Thousand Oaks, London, New Delhi: Sage Publications.
Page, D., & Wong, P. T. P. (2000). A conceptual framework for measuring servant leadership. In S. Adjibolooso (Ed.), The human factor in shaping the course of history
and development (pp. 69–110). Washington, DC: American University Press.
Parolini, J., Patterson, K., & Winston, B. (2009). Distinguishing between transformational and servant leadership. Leadership and Organization Development Journal, 30,
274–291.
Pearce, J. L., & Gregersen, H. B. (1991). Task interdependence and extrarole behavior: A test of the mediating effects of felt responsibility. Journal of Applied Psychology,
76(6), 838–844.
Peterson, S. J., Galvin, B. M., & Lange, D. (2012). Ceo servant leadership: Exploring executive characteristics and firm performance. Personnel Psychology, 65(3),
565–596.
Podsakoff, P. M., & MacKenzie, S. B. (1994). An examination of the psychometric properties and nomological validity of some revised and reduced substitutes for
leadership scales. Journal of Applied Psychology, 79(5), 702–713.
Podsakoff, P. M., MacKenzie, S. B., & Ahearne, M. (1997). Moderating effects of goal acceptance on the relationship between group cohesiveness and productivity.
Journal of Applied Psychology, 82(6), 974–983.
Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., & Fetter, R. (1990). Transformational leader behaviors and their effects on followers' trust in leader, satisfaction,
and organizational citizenship behaviors. The Leadership Quarterly, 1(2), 107–142.
Podsakoff, P. M., Niehoff, B. P., MacKenzie, S. B., & Williams, M. L. (1993). Do substitutes for leadership really substitute for leadership? An empirical examination of
Kerr and Jermier's situational leadership model. Organizational Behavior and Human Decision Processes, 54(1), 1–44.
Preacher, K. J., & Kelley, K. (2011). Effect size measures for mediation models: Quantitative strategies for communicating indirect effects. Psychological Methods, 16(2),
93–115.
Rothausen, T. J. (Jun 23, 2016). Integrating leadership development with ignatian spirituality: A model for designing a spiritual leader development practice. Journal
of Business Ethics, (No Pagination Specified).
Schaubroeck, J., Lam, S. S. K., & Peng, A. C. (2011). Cognition-based and affective-based trust as mediators of leader behavior influence on team performance. Journal
of Applied Psychology, 96(4), 863–871.
Schriesheim, J. F. (1980). The social context of leader–subordinate relations: An investigation of the effects of group cohesiveness. Journal of Applied Psychology, 65(2),
183–194.
Schyns, B. (2006). Are group consensus in leader-member exchange (LMX) and shared work values related to organizational outcomes? Small Group Research, 37(1),
20–35.
Sedikides, C. E., & Brewer, M. B. E. (2001). Individual self, relational self, collective self. New York, NY, US: Psychology Press.
Sherony, K. M., & Green, S. G. (2002). Coworker exchange: Relationships between coworkers, leader-member exchange, and work attitudes. Journal of Applied
Psychology, 87(3), 542–548.
Sias, P. M., & Jablin, F. M. (1995). Differential superior-subordinate relations, perceptions of fairness, and coworker communication. Human Communication Research,
22(1), 5–38.
Spears, L. C. (1995). Reflections on leadership: How Robert K. Greenleaf's theory of servant-leadership influenced today's top management thinkers. New York: John Wiley.
Stoverink, A. C., Umphress, E. E., Gardner, R. G., & Miner, K. N. (2014). Misery loves company: Team dissonance and the influence of supervisor-focused interpersonal
justice climate on team cohesiveness. Journal of Applied Psychology, 99(6), 1059–1073.
Sy, T., Côté, S., & Saavedra, R. (2005). The contagious leader: Impact of the Leader's mood on the mood of group members, group affective tone, and group processes.
Journal of Applied Psychology, 90(2), 295–305.
Tajfel, H. (1978). Social categorization, social identity and social comparison. In H. Tajfel (Ed.), Differentiation between social groups (pp. 61–76). London: Academic
Press.
Tse, H. H. M., Lam, C. K., Lawrence, S. A., & Huang, X. (2013). When my supervisor dislikes you more than me: The effect of dissimilarity in leader–member exchange
on coworkers' interpersonal emotion and perceived help. Journal of Applied Psychology, 98(6), 974–988.
Tyler, T. R., & Blader, S. L. (2000). Cooperation in groups: Procedural justice, social identity, and behavioral engagement. Essays in social psychology. New York, NY, US:
Psychology Press.
Tyler, T. R., & Blader, S. L. (2003). The group engagement model: Procedural justice, social identity, and cooperative behavior. Personality and Social Psychology Review,
7(4), 349–361.
Van Dierendonck, D., Stam, D., Boersma, P., de Windt, N., & Alkema, J. (2014). Same difference? Exploring the differential mechanisms linking servant leadership and
transformational leadership to followers outcomes. The Leadership Quarterly, 25(3), 544–562.
Vidyarthi, P. R., Liden, R. C., Anand, S., Erdogan, B., & Ghosh, S. (2010). Where do I stand? Examining the effects of leader–member exchange social comparison on
employee work behaviors. Journal of Applied Psychology, 95(5), 849–861.
Wageman, R. (2001). The meaning of interdependence. In M. E. Turner (Ed.), Groups at work: Theory and research. Applied social research (pp. 197–217). Mahwah, NJ,
US: Lawrence Erlbaum Associates Publishers.
Wood, J. V. (1996). What is social comparison and how should we study it? Personality and Social Psychology Bulletin, 22(5), 520–537.

16