Journal of Business and Psychology

https://doi.org/10.1007/s10869-018-9542-0

ORIGINAL PAPER

A Social Exchange Perspective of Abusive Supervision and Knowledge

Sharing: Investigating the Moderating Effects of Psychological Contract
Fulfillment and Self-Enhancement Motive
Woohee Choi 1 & Seckyoung Loretta Kim 2 & Seokhwa Yun 3

# Springer Science+Business Media, LLC, part of Springer Nature 2018

Abstract
Given the importance of knowledge management in this competitive environment, the purpose of the present study is to fill the
gap in contemporary literatures of knowledge sharing behavior and abusive supervision by observing the main effect, mecha-
nism, and moderators. Based on social exchange perspective, we propose a theoretical model that links abusive supervision to
employee knowledge sharing as mediated by leader-member exchange (LMX) with conditional processes. Employing a sample
of 184 supervisor-subordinate dyads, we carried out a survey in large listed companies in South Korea. To test our hypoth-
eses, we conducted multiple regression analyses and used bootstrapping procedures. Our results suggest that LMX mediates
the abusive supervision and knowledge sharing relationship. Most significantly, findings show that this mediated relationship
is contingent on the level of psychological contract fulfillment and self-enhancement motive. One of the most critical
implications of our work is that negative influence of hostile behaviors of supervisors on knowledge sharing via LMX
may actually be attenuated by perceptions of employees formed both from the organization (i.e., psychological contract
fulfillment) and from oneself (i.e., self-enhancement motive). Moreover, it also provides practical insights for both the
management practitioner and the organization. Extending from earlier studies, this research enriches our understanding of
organizational behavior research by demonstrating an overall complete picture of a moderated-mediation model between
abusive supervision and knowledge sharing by uncovering a mediator explaining the mechanism and moderators buffering the
negative effect of abusive supervision.

Keywords Knowledge sharing . Abusive supervision . Leader-member exchange . Psychological contract . Self-enhancement
motive

Introduction

Organizational knowledge is recognized as an invaluable re-

* Seckyoung Loretta Kim source to a firm’s competitive advantage in this dynamic so-
[email protected] ciety (Grant, 1996; Spender & Grant, 1996). Companies ex-
pend great effort on ensuring effective knowledge manage-
Woohee Choi
[email protected] ment to encourage individuals to share their knowledge
(Quigley, Tesluk, Locke, & Bartol, 2007). However, since
Seokhwa Yun
each person shares his or her knowledge, it is important to
[email protected]
identify factors that influence knowledge sharing at the indi-
1
Fisher College of Business, The Ohio State University, 700 Fisher vidual level. Despite the growing significance of sharing
Hall, 2100 Neil Avenue, Columbus, OH 43210, USA knowledge between individuals, it is not easy to stimulate
2
College of Business Administration, Incheon National University, employees to share their knowledge in the competitive work-
119 Academy-ro, Yeonsu-gu, Incheon, South Korea place, because this behavior is not mandatory in most organi-
3
College of Business Administration, Seoul National University, 1 zations (Cabrera, & Cabrera, 2005; Kim, Han, Son, & Yun,
Gwanak-ro, Gwanak-gu, Seoul, South Korea 2017). Moreover, knowledge is a critical asset for each
 J Bus Psychol

individual in the workplace (French & Raven, 1959); there- knowledge to deepen our understanding of employees’
fore, employees may feel risks in sharing their valuable knowledge sharing behavior. Prior research noted that the im-
knowledge with others. Given its voluntary nature, employees pact of negative events is much stronger than that of positive
are likely to engage in knowledge sharing when they receive ones (Baumeister, Bratslavsky, Finkenauer, & Vohs, 2001).
benefits or feel obligated towards a supervisor or organization Thus, this research contributes to the literature by investigat-
(Kim et al., 2017). On the other hand, employees may choose ing abusive supervision as an obstacle to employee knowl-
not to exhibit knowledge sharing behavior when they receive edge sharing.
negative treatment from major actors in the workplace. From Examining a mediating mechanism is meaningful in ad-
this perspective, this research examines knowledge sharing vancing our understanding of the relationship between abu-
predictors, particularly leader behavior with its mechanism sive supervision and knowledge sharing. This study explores
and moderators based on a social exchange framework. leader-member exchange (LMX), which is rooted in social
The importance of leader behavior has been emphasized exchange theory, as a mediating mechanism in the relationship
because of its critical role in organizational efficiency and goal between abusive supervision and knowledge sharing. LMX
achievement and in employee engagement (Dierdorff, Rubin, has been widely used as a mediator in the relationship between
& Morgeson, 2009). Since the behavior of a leader could be a leader behaviors and outcomes (e.g., Newman, Schwarz,
significant factor affecting the level of support or resources, Cooper, & Sendjaya, 2015). Many researchers have investi-
much of the leadership research on knowledge sharing focuses gated individual attitudes or emotional and relational factors
on effective leader behaviors (e.g., Srivastava, Bartol, & as mediators of the link between abusive supervision and out-
Locke, 2006). However, the influence of leaders on em- comes (e.g., Tepper, 2000; Xu, Huang, Lam, & Miao, 2012).
ployees is not always positive (Tepper, 2000, 2007). Leaders Among various possible mediators, we selected a relational
are often regarded as key decision-makers and have unique factor, namely LMX, since it may best represent the relation-
responsibilities in the organization. As such, their destructive ship between a supervisor and subordinate and explains why
behaviors are likely to reduce employees’ job-related perfor- abused employees exhibit negative attitudes and behaviors
mance. Recently, scholars have been concerned about abusive within a social exchange framework. Since high-quality
supervision, a widespread phenomenon in today’s organiza- LMX relationships are likely to develop mutual trust and re-
tional context (Tepper, 2007). Abusive supervision, which spect (Graen & Uhl-Bien, 1995), these employees are likely to
refers to Bsubordinates’ perceptions of the extent to which engage beyond mandatory tasks. In contrast, employees with
supervisors engage in the sustained display of hostile verbal low-quality LMX relationships are likely to perceive low trust
and nonverbal behaviors, excluding physical contact^ and obligation (Graen & Uhl-Bien, 1995), leading to a low
(Tepper, 2000, p. 178), is considered an important factor level of voluntary tasks. Drawing on insights from social ex-
representing Bthe dark side of leadership.^ Previous studies change theory (Blau, 1964), we suggest that abused em-
significantly and negatively related abusive supervision to ployees are likely to reduce their knowledge sharing behav-
employees’ attitudes and behaviors (Tepper, 2007). For exam- iors, since they may not build high-quality LMX relationships
ple, abusive supervision tends to decrease followers’ job sat- based on negative reciprocity consequent to abusive
isfaction, psychological well-being, and performance out- supervision.
comes (Martinko, Harvey, Brees, & Mackey, 2013). Based Although leader behavior is a factor needed to alter the
on these observations, we assume that abusive supervision, a quality of exchange relationships or the behaviors of em-
common type of destructive leadership, may act as a key bar- ployees in the workplace, the influence of such behavior could
rier to employees’ knowledge sharing behavior. However, the be changed through situational or personal factors in organi-
relationship between abusive supervision and knowledge zational settings (Chiaburu, Lorinkova, & Van Dyne, 2013).
sharing has not been investigated much. Numerous previous studies adopted an interactionist perspec-
Considering that employees may perceive high risks in tive to examine situational and personal factors that may
sharing their knowledge, leaders’ abusive behavior is harmful change the effect of leader behavior on outcomes (e.g.,
to employees’ knowledge sharing behaviors. As such, abused Garcia, Wang, Lu, Kiazad, & Restubog, 2015; Harvey,
employees may not automatically share their knowledge with Stoner, Hochwarter, & Kacmar, 2007; Kacmar, Bachrach,
others. Knowledge sharing is usually regarded as voluntary Harris, & Zivnuska, 2011; Tepper, Duffy, & Shaw, 2001).
behavior despite the benefits to the team and organization The present study adopts an interactional perspective and con-
(Kim, Kim, & Yun, 2015). Therefore, employees may per- siders as moderators one’s perceptions towards the organiza-
ceive that they have more flexibility to behave (Judge, tion and oneself. Specifically, we selected psychological con-
LePine, & Rich, 2006), easily withdrawing such action than tract fulfillment for the purposes of this study, which is defined
mandatory behaviors. Given the importance of sharing knowl- as Bindividual beliefs shaped by the organization regarding the
edge in this knowledge-based society, it is necessary to deter- terms of an exchange agreement between individuals and their
mine when employees may decide not to share their organization^ (Rousseau, 1995, p. 9). Psychological contract
J Bus Psychol

theory (Rousseau, 1995; Robinson & Morrison, 2000) sug- these employees focus on enhancing their impression on
gests that employees form certain perceptions of mutual obli- others, they are less likely to build a low-quality LMX rela-
gations between themselves and their organizations. Zhao, tionship with their abusive supervisor. Thus, we argue that the
Wayne, Glibkowski, and Bravo (2007) viewed this construct self-enhancement motive could be a critical personal factor
as a critical lens through which to understand the quality of the that alleviates the negative effect of abusive supervision on
exchange relationship between an employer and employee. LMX.
As such, psychological contract fulfillment can be regarded Notably, aside from providing a broader view of the mech-
as a relevant situational factor explaining the exchange rela- anism through which abusive supervision negatively impacts
tionship between an employer and employee based on the knowledge sharing, the final goal of our study is to investigate
exchange relationship (Zhao et al., 2007). Moreover, the or- an expansive view of the organizational and individual aspects
ganizational variable is suitable in mitigating the negative ef- within which abusive supervision occurs. We demonstrate a
fect of abusive supervision (Kim et al., 2015). Despite the moderated-mediation framework of the abusive supervision–
usefulness of situational factors that buffer the negative effect knowledge sharing relationship through LMX by suggesting
of abusive supervision on outcomes, few studies examine sit- aspects of individuals’ perceptions within the organization as
uational factors as moderators. Aryee, Sun, Chen, and Debrah two moderating variables. One is the notion of the organiza-
(2008) investigated the work unit structure as a moderator in tion (i.e., psychological contract fulfillment), and the other is
the relationship between abusive supervision and contextual the notion of oneself (i.e., self-enhancement motive). We pre-
performance dimensions of interpersonal facilitation and job dict that these two moderators dynamically interact with abu-
dedication. Thus, since employees with high psychological sive supervision to predict and alter the degrees of LMX and
contract fulfillment develop an implicit obligation to recipro- knowledge sharing behavior.
cate the organization with stronger affective bonds (Blau, In summary, the current study offers four research purposes
1964), we predict that the negative effect of abusive supervi- based on social exchange theory. First, this research intends to
sion on LMX could weaken if focal employees perceive their enhance understanding of knowledge sharing by examining
psychological contract as fulfilled. abusive supervision as a predictor. Second, the research pro-
In addition, as a personal factor, we selected as an individ- poses LMX as a mediating mechanism in the relationship
ual moderator the self-enhancement motive, which represents between abusive supervision and knowledge sharing. Third,
the positive aspect of impression management, since the neg- this study suggests psychological contract fulfillment and the
ative impact of abusive supervision varies depending on indi- self-enhancement motive as moderators that could mitigate
vidual characteristics (c.f., Tepper et al., 2001). In the work- the detrimental impact of abusive supervision on LMX.
place, employees engaged in impression management attempt Last, we aim to explore a complete picture of a moderated-
to improve the image they portray to others including super- mediation model between abusive supervision and knowledge
visors (Bolino, Kacmar, Turnley, & Gilstrap, 2008). Although sharing through LMX by examining psychological contract
abusive supervision tends to ruin the quality of LMX, not all fulfillment and the self-enhancement motive.
abused employees engage in low-quality LMX relationships.
In particular, individuals with a certain motive may adjust
their feelings (Grant & Mayer, 2009), thereby reducing the Theoretical Background and Hypotheses
harmful effects of abusive supervision on LMX.
Recognizing the critical role of personal factors, previous Knowledge Sharing and Abusive Supervision
studies indicated that individuals may or may not cope with
stressful situations such as abusive supervision depending on Bartol and Srivastava (2002) defined knowledge sharing as
their abilities, motivations, or characteristics (Chi & Liang, individuals sharing relevant information, ideas, suggestions,
2013). Scholars explained that emotional contagion, negative and expertise with others across the organization. Given the
reciprocity beliefs, and emotional intelligence may serve as discretionary nature of knowledge sharing behaviors, focal
key moderators in the relationship between abusive supervi- employees may have autonomy in determining whether to
sion and outcomes (Chi & Liang, 2013; Frieder, Hochwarter, donate their knowledge depending on how others serve them.
& DeOrtentiis, 2015; Nandkeolyar, Shaffer, Li, Ekkirala, & With growing interest in and the importance of knowledge,
Bagger, 2014; Mitchell & Ambrose, 2007). For example, Hu earlier literature mostly investigated the positive relation be-
(2012) demonstrated that emotionally intelligent employees tween constructive leader behaviors and knowledge sharing
are likely to mitigate the negative effect of abusive supervision (Srivastava et al., 2006). For example, scholars examined
on the emotional labor burden. Here, it is meaningful to in- management support, transformational leadership, and
spect the self-enhancement motive as a moderator because empowering leadership as key predictors of knowledge shar-
employees are likely to respond to abusive supervision in ing behaviors (e.g., Bryant, 2003; Srivastava et al., 2006).
different ways depending on their level of this motive. Since However, recent research recognizes the detrimental impact
 J Bus Psychol

of malicious leader behaviors such as abusive supervision on employees may easily withdraw such voluntary behavior.
followers and the organization as a whole (Tepper, 2007). Hence, we predict the following:
Thus, this research examines the significant role of abusive
supervision as a determinant of employee knowledge sharing. Hypothesis 1. Abusive supervision is negatively related to
Abusive supervision is a low base-rate phenomenon in or- knowledge sharing.
ganizations; however, researchers have paid much attention to
the concept over the past 20 years. Previous studies noted that The Mediating Effect of Leader-Member Exchange
while abusive supervision does not frequently occur, its dam-
age to the organization is devastating (e.g., Tepper, Moss, LMX describes the quality of an exchange relationship that
Lockhart, & Carr, 2007; Harris, Kacmar, & Zivnuska, 2007). develops between supervisors and employees (Graen, 1976;
Workplace supervisors have unique roles and responsibilities Graen & Scandura, 1987). According to the LMX theory,
in the organization (Carroll & Gillen, 1987; Dierdorff et al., supervisors and subordinates develop their relationships
2009); therefore, abusive supervision is likely to bring about through certain expectations (Graen & Cashman, 1975).
negative attitudes and behaviors (Tepper, 2000). For example, These expectations include trust, competence (Liden &
for attitudes, abusive supervision is linked to decreased job Graen, 1980), and mutual influence (Yukl, 1981). Since em-
satisfaction (Tepper, 2000), low self-efficacy (Duffy, ployees experiencing abusive supervision perceive negative
Ganster, & Pagon, 2002), and poor organizational commit- treatment from their supervisor, they may be reluctant to pro-
ment (Duffy et al., 2002; Tepper, 2000). Furthermore, it is vide valuable support and less willing to frequently interact
negatively related to various emotional and psychological out- with their abusive supervisor. As such, the abusive supervisor
comes such as emotional exhaustion (Grandey, Kern, & is likely to build a low-quality relationship with employees
Frone, 2007), psychological distress (Tepper, 2000), and other who exhibit low levels of interaction and support (c.f.,
negative affectivities (Martinko et al., 2013; Chan & Nahrgang, Morgeson, & Ilies, 2009). Prior studies reported
McAllister, 2014). Extant research connected abusive super- that a low-quality relationship is an exchange in accordance
vision with pernicious employee behaviors such as workplace with the employment contract, whereas a high-quality LMX
deviance (Mitchell & Ambrose, 2007; Duffy et al., 2002) and relationship is likely to form when there is mutual trust and
decreased performance—both job performance (Harris et al., respect, which is beyond the employment contract (Graen &
2007) and organizational citizenship behaviors (OCB) Uhl-Bien, 1995). Therefore, employees with higher LMX re-
(Zellars, Tepper, & Duffy, 2002). Aligned to previous re- lationships tend to work more than the requirements stipulate
search, this study explains the negative relationship between and perform voluntary tasks (Graen & Uhl-Bien, 1995). In
abusive supervision and employee knowledge sharing behav- contrast, abused employees who develop low-quality LMX
ior by applying social exchange theory. relationships do not engage in voluntary tasks such as knowl-
Based on the norm of reciprocity (Blau, 1964; Cropanzano edge sharing.
& Mitchell, 2005), when one behaves in a friendly manner The principle of social exchange theory (Blau, 1964) pro-
towards others, one may expect to receive the return of favor- poses that in a binary relationship, if something is given, there
able treatment (Gouldner, 1960). However, negative reciproc- is a silent promise to return it equivalently (Gouldner, 1960;
ity may exist, wherein hostile action is repaid with adverse Perugini & Gallucci, 2001). Based on social exchange theory,
treatment (Cropanzano & Mitchell, 2005). When subordinates we expect that abused employees are likely to generate poor
are abused by their leaders, they cannot expect trust or support social exchange relationships with leaders, because they feel
from their supervisors, which is likely to result in unfriendly they receive or expect to receive a lack of trust or support from
behaviors or responses as per the norm of negative reciprocity their supervisors. Numerous previous studies related abusive
(Tepper, 2000). However, the nature of supervisor- supervision to unfavorable attitudinal and psychological out-
subordinate relationships, which may involve power distance comes such as interpersonal conflicts, emotional exhaustion,
and hierarchy, means that employees do not usually engage in and emotional distress (Duffy et al., 2002; Grandey et al.,
unfavorable behaviors towards their supervisors. Rather, they 2007; Tepper, 2000). In this way, the exchange relationship
may passively take revenge on their leaders. Given that work- between hostile supervisors and employees is non-supportive
related knowledge sharing might benefit both the organization and disrespectful (Uhl-Bien, Graen, & Scandura, 2000),
and its representatives, employees who feel exploited by their which results in a low-quality relationship.
supervisors will respond antagonistically and therefore refrain Previous studies positively related LMX to task perfor-
from sharing their knowledge with others. Moreover, although mance and OCB (Ilies, Nahrgang, & Morgeson, 2007; Wang,
engaging in knowledge sharing may be not directly targeted at Law, Hackett, Wang, & Chen, 2005). Moreover, a recent re-
supervisors, but targeted at coworkers and at the organization, view by Wang and Noe (2010) suggested LMX as a potential
abused employees may decide not to share their knowledge as factor in predicting knowledge sharing behaviors. In the con-
negative reciprocity (Kim et al., 2015). It is because text of high LMX, employees who perceive trust and support
J Bus Psychol

from their leaders are likely to reciprocate by engaging in fa- their obligations by maintaining high-quality LMX relation-
vorable behaviors such as knowledge sharing. However, in ships. Although employees under abusive supervision may
lower quality LMX relationships, employees are likely to feel not expect to receive positive treatment or support from their
a low obligation to perform beyond in-role performance (Wang supervisor (Tepper, 2000), they are likely to feel obligated to
et al., 2005). Ultimately, we expect that LMX mediates the maintain a quality relationship with their supervisors when
negative relationship between abusive supervision and employ- they perceive high psychological contract fulfillment. This
ee knowledge sharing. Abused employees are likely to develop belief includes focal employees’ perceptions of the incentives
low-quality LMX relationships, which leads them to expend (e.g., pay, recognition) agreed on with their organizations.
less effort on knowledge sharing, because of reactive behavior When a psychological contract is fulfilled, employees feel
based on the norm of reciprocity according to social exchange increased responsibilities towards the organization (Turnley
theory. Thus, we predict the following: et al., 2003), which may have been depreciated by abusive
supervision. In this context, the quality of the exchange rela-
Hypothesis 2. LMX mediates the relationship between abu- tionships of an employee and supervisor is less likely to be
sive supervision and knowledge sharing. harmed, even when supervisors exhibit abusive supervision.
Moreover, employees may decide to maintain the quality of
The Moderating Effects of Psychological Contract LMX to remain in the organization, where they perceive high
Fulfillment psychological contract fulfillment.
On the other hand, if a psychological contract is violated,
In a work situation, employees engage in social exchange with negative feelings towards supervisors, which are already de-
both individuals and the organization (Eisenberger, veloped through abusive supervision, escalate. Much research
Huntington, Hutchison, & Sowa, 1986; Wayne, Shore, & on the breach of the psychological contract has demonstrated
Liden, 1997; Masterson, Lewis, Goldman, & Taylor, 2000). its strong impact on negative cognitions or attitudes (Conway
Employees often interact and exchange their resources with & Briner, 2005; Zhao et al., 2007). As a result, employees are
their organizations (Eisenberger et al., 1986; Masterson et al., likely to decrease or withdraw their efforts to build high-
2000). Psychological contract fulfillment can be regarded as quality relationships with their supervisors, which may further
one’s perception of the organization, since it refers to the em- ruin the LMX relationship.
ployee perception that the organization has equitably fulfilled
its terms of the contract (Rousseau, 1990, 1995). Depending Hypothesis 3. Psychological contract fulfillment moderates
on how an organization supports employees through valuable the negative relationship between abusive supervision and
resources or fair treatment, they perceive either a breach or LMX quality, such that the negative relationship is likely to
fulfillment of their psychological contracts. The extant litera- be weakened when psychological contract fulfillment is high
ture relates perceptions of psychological contract fulfillment rather than when it is low.
to various favorable organizational outcomes such as in-role More important, we propose that psychological contract
or extra-role performance and organizational commitment fulfillment moderates the relationship between abusive super-
(Turnley, Bolino, Lester, & Bloodgood, 2003). In applying vision and knowledge sharing via LMX. We argue that LMX
social exchange theory to the employee–organization relation- serves as the mechanism through which abusive supervision
ship, psychological contract theory assumes that employees affects knowledge sharing. Incorporating this notion with
tend to view the organization as an exchange partner and feel Hypothesis 3, namely that psychological contract fulfillment
reciprocal obligations (Rousseau, 1995). Thus, we expect that moderates the effect of abusive supervision on LMX, we de-
the psychological contract can be an important organizational veloped a moderated mediation model, which is represented
factor that mitigates the negative effect of abusive supervision in Fig. 1. As a factor pertaining to one’s perspective towards
on LMX and knowledge sharing via LMX based on social the organization, psychological contract fulfillment may act as
exchange theory. a buffering factor that alleviates the negative effect of abusive
Workplace supervisors serve dual functions in the LMX supervision on knowledge sharing via LMX. Since employees
relationship. For example, they are interpersonal exchange who perceive high psychological contract fulfillment may feel
partners for employees and symbolic archetypes of the orga- obligated towards the organization, they may decide to engage
nization in the employee–organization relationship in knowledge sharing by building a high-quality LMX rela-
(Henderson, Wayne, Shore, Bommer, & Tetrick, 2008). tionship as reciprocity, since knowledge sharing is an effective
From this viewpoint, focal employees who are satisfied with way to promote organizational effectiveness. Therefore, we
their organization, because they perceive high psychological contend that the negative relationship between abusive super-
contract fulfillment, may feel obligated towards it. Since a vision and knowledge sharing via LMX lessens when an em-
leader often represents the organization, employees with high ployee has a high level of psychological contract fulfillment.
psychological contract fulfillment may decide to reciprocate In sum, we hypothesize the following:
 J Bus Psychol

interaction, support, and rewards (Dienesch & Liden, 1986).

We argue that individuals with a high self-enhancement mo-
tive exhibit positive attitudes towards their supervisors to en-
hance their image.
In contrast, those with a low self-enhancement motive are
less sensitive to external factors. These employees exhibit a
lower self-enhancement motive and are less concerned about
the image they portray to others including their leaders.
Fig. 1 Research framework Therefore, when they experience a leader’s destructive behav-
ior, they are likely to demonstrate negative attitudes and be-
Hypothesis 4. Psychological contract fulfillment moderates haviors, similar to how leaders behave towards them. These
the strength of the mediated relationship between abusive su- employees do not make efforts to improve relationships with
pervision and knowledge sharing via LMX such that the me- their abusive supervisors. In summary, employees with a high
diated relationship is likely to be weakened when psycholog- self-enhancement motive are likely to expend their efforts on
ical contract fulfillment is high than when it is low. maintaining high-quality LMX relationships with their super-
visors, even under abusive supervision, to maintain their pos-
The Moderating Effects of Self-Enhancement Motive itive image.

Academics and scholars have a great interest in examining the Hypothesis 5. Self-enhancement motive moderates the nega-
motives underlying human behaviors and attitudes using var- tive relationship between abusive supervision and LMX qual-
ious approaches (e.g., Bettman, 1979; Hull, 1943). Along with ity, such that the relationship is likely to be weakened when
exchange relationships, motivation has been identified as a self-enhancement motive is high rather than when it is low.
critical factor in the decision-making process, as it affects Similar to psychological contract fulfillment, we developed
the direction of behavior and attitudes (Bettman, 1979). a moderated mediation model. When employees have a high
Although abusive supervision tends to ruin the quality of self-enhancement motive, they likely engage in knowledge
LMX, not all abused employees engage in low-quality LMX sharing via LMX, even under abusive supervision, since it is
relationships. In particular, individuals with certain motives important to enhance their image as perceived by significant
may reduce the harmful effect of abusive supervision on others. Given the importance of knowledge sharing in enhanc-
LMX. For example, employees with a high self- ing organizational effectiveness, employees who have much
enhancement motive may maintain high-quality relationships concern for other people’s evaluation may choose to maintain
with their leaders, even under abusive supervision, because of high-quality LMX, which leads to high level of knowledge
their motivation to achieve a positive self-image. Thus, this sharing. Thus, we propose that the negative effect of abusive
study suggests that an individual’s self-enhancement motive, supervision on knowledge sharing via LMX lessens when an
an aspect of impression management, is a buffering factor that employee has a high self-enhancement motive. Therefore, we
reduces the negative influence of abusive supervision on predict the following:
knowledge sharing.
Employees engaged in impression management focus on Hypothesis 6. Self-enhancement motive moderates the
improving the image they portray to others (Bolino et al., strength of the mediated relationship between abusive super-
2008). Self-enhancement is the desire or observed reality of vision and knowledge sharing via LMX such that the mediat-
seeing oneself in the most positive light (Pfeffer & Fong, ed relationship is likely to be weakened when self-
2005). Employees with a high self-enhancement motive are enhancement motive is high than when it is low.
more sensitive to others’ perceptions of them, and their level
of motivation to adapt behaviors is higher to make a good
impression on others (Yun, Takeuchi, & Liu, 2007). Even Methods
when facing abusive leaders, employees with a high self-
enhancement motive are less likely to be unfriendly to leaders, Participants and Procedures
given their powerful position. Rather, they strive to build a
better relationship with hostile leaders to maintain the positive The study was conducted in large companies located in South
image they portray to significant others including leaders. As Korea. The industries of these firms were various including
such, the core premise of LMX theory implies that leaders financial services, consulting, IT service, and manufacturing
differentiate the treatment of subordinates in the same group companies. Most participants in this study held positions in
(Liden & Graen, 1980). Generally, leaders value employees the management or research sectors, in which the exchange of
who belong to the Bin-group,^ characterized by high trust, information and knowledge is needed to improve their tasks.
J Bus Psychol

Separately sealed questionnaire packets for full-time em- supervision deliver BMy immediate supervisor ridicules me^
ployees and their direct managers were prepared and distrib- and Btells my thoughts or feelings are stupid.^ Cronbach’s
uted to 216 subordinate-supervisor dyads. To select subordi- alpha for the scale was .98.
nates, we provided specific guidelines on how to select a sub-
ordinate to constitute the pair to ensure objectivity. Moreover, LMX The most widely used LMX measure—LMX-7 scale—
in the survey packets distributed to participants, we included a developed by Graen and Uhl-Bien (1995) was adapted to
cover letter emphasizing that no individual subject would be measure the exchange relationship between supervisors and
identified and that the results of this study would only be used subordinates. The sample item says BI have enough confi-
for educational and academic purposes. After completion, par- dence in my supervisor to defend and justify my decisions
ticipants were required to enclose the survey in the envelope when I am not present to do so.^ Cronbach’s alpha for the
provided and seal it on their own to ensure the confidentiality scale was .93.
of their responses.
From 216 subordinate-supervisor dyadic samples, 184 Psychological Contract Fulfillment Psychological contract ful-
pairs were returned, giving a response rate of 85%. After fillment was measured using a measure developed by
matching the results of employee surveys with those of man- Robinson and Morrison (2000). This five-item measure cap-
agerial surveys, the final sample dyads used in the current tures an employee’s psychological contract fulfillment percep-
analysis included 175 pairs. A few responses could not be tion towards the organization. The items asked participants to
involved in the research because the responses were incom- rate the extent to which they agreed or disagreed with state-
plete. We also removed some samples that only one answered ments such as BI feel that my organization has come through
the survey between the two (i.e., subordinate and supervisor). in fulfilling the promises made to me when I was hired^ and
Of the subordinates, 75.43% were male, and the average age Bso far my organization has done an excellent job of fulfilling
of employees was 36.59 (SD = 6.58). The most frequently its promises to me.^ Cronbach’s alpha for the scale was .93.
reported employee’s level of education (72%) was a bachelor
degree. Self-Enhancement Motive Six items developed by Yun et al.
(2007) were used to measure self-enhancement motive of em-
Measures ployees. The sample item reads BI try to modify my behaviors
to give good images to others.^ Cronbach’s alpha for the scale
All items used in the present study were originally developed was .91.
in English. The questionnaires were translated into Korean
using conventional method of back translation (Brislin, Control Variables As suggested by previous studies (e.g., Xu
1980). Two Korean bilingual academics individually translat- et al., 2012; Zellars et al., 2002), age, gender, and education
ed the measures into Korean and back translated them again level could be related to our independent and dependent var-
separately to ensure semantic equivalence. All variables were iables. Thus, we controlled for in this research. Age was mea-
measured with established measures. The focal employees sured in years. Gender was assessed as a dichotomous vari-
were requested to assess their supervisors’ abusive behaviors, able, coded as 0 for male and 1 for female. We measured
LMX, their own self-enhancement motives, and psychologi- education on a scale that ranged from B1 = high school^ to
cal contract fulfillment. To reduce the issues for common B4 = master’s degree or higher.^ However, similar results were
method bias, immediate supervisors evaluated their em- obtained when we conducted the analyses without controlling
ployees’ levels of knowledge sharing. All items were mea- for those variables.
sured on a seven-point Likert-type scale ranging from
Bstrongly disagree^ to Bstrongly agree.^ The specific mea- Analytical Procedures
sures are described below, along with the results of calculation
of Cronbach’s alpha coefficients. We conducted confirmatory factor analyses (CFAs) to ensure
construct validity among variables. We included all five vari-
Knowledge Sharing Knowledge sharing level of focal em- ables (i.e., abusive supervision, leader-member exchange,
ployees was measured using seven items from the Srivastava psychological contract fulfillment, self-enhancement motive,
et al. (2006) study. The sample item states BThis employee knowledge sharing) in our research model to estimate the
shares his/her special knowledge and expertise with others.^ overall model fit. Specifically, we examined the values of
Cronbach’s alpha for the scale was .95. chi-square (χ2), Comparative Fit Index (CFI), Tucker-Lewis
index (TLI), and the Root Mean Square Error of
Abusive Supervision We measured subordinates’ perceptions Approximation (RMSEA) to assess the overall model fit. We
of abusive supervisory behaviors using the 15-item scale de- then compared our hypothesized five-factor model to a series
veloped by Tepper (2000). The sample items of abusive of competing models.
 J Bus Psychol

To test our hypotheses, we conducted regression analyses Results

for the main effect and the moderation effects and used
PROCESS for testing the indirect effects. First of all, to test The hypothesized measurement model consisted of five vari-
the main effect of abusive supervision on knowledge sharing, ables: abusive supervision, leader-member exchange, psycho-
we used hierarchical multiple regression method. In step 1, we logical contract fulfillment, self-enhancement motive, and
included control variables in order to diminish the spurious knowledge sharing. As exhibited in Table 1, when the hypoth-
effects. Demographic information of employees—age, gen- esized model is compared with a series of competing models,
der, and education level—was controlled. In step 2, we our five-factor model indicated the best fit of all. The values
inserted abusive supervision to test its main effect on employ- on the fit indices showed that the five-factor CFA model pro-
ee knowledge sharing behavior (i.e., Hypothesis 1). vided a good fit for the data (χ2 = 847.538, CFI = .929,
To test the mediation hypothesis (i.e., Hypothesis 2), we TLI = .918, and RMSEA = .071). This result offered a signif-
employed SPSS PROCESS developed by Preacher and Hayes icant improvement in chi-square over a series of competing
(2004) in order to test indirect effect of abusive supervision on models.
knowledge sharing via LMX by using the bootstrapping pro- The correlations and descriptive statistics including means
cedure (across 1000 samples). This PROCESS promotes an and standard deviations are provided in Table 2. High reliabil-
estimation of indirect effect with a bootstrap approach by pre- ities of Cronbach’s alpha coefficients were observed through
senting confidence intervals. all variables ranging from .91 or higher. Independent and
Moderation effects of both psychological contract ful- moderating variables were mean-centered to prevent potential
fillment (i.e., Hypothesis 3) and self-enhancement motive multicollinearity problem (Aiken & West, 1991).
(i.e., Hypothesis 5) were evaluated with moderated regres- Hypothesis 1 proposed that abusive supervision is nega-
sion analysis. Hierarchical regression is the most common tively related to knowledge sharing. As shown in Table 3,
method to test moderation models (Leung & Zhou, 2008). the result of the regression analysis signifies that abusive su-
Before performing the regression analysis, we mean- pervision was significantly and negatively related to knowl-
centered all independent and moderating variables (i.e., edge sharing. This supported Hypothesis 1 (Model 2; b =
abusive supervision, psychological contract fulfillment, − .20, p ≤ .01).
and self-enhancement motive) in order to prevent potential
multicollinearity problems (Aiken & West, 1991). For the Test of Mediation
regression, we included covariates in step 1, abusive super-
vision in step 2, and psychological contract fulfillment and Hypothesis 2 suggested that LMX mediates the relationship
self-enhancement motive were put together in step 3. between abusive supervision and knowledge sharing. We
Lastly, we entered two interaction terms, the multiplication employed bootstrap method to test the indirect mediation ef-
terms of mean-centered variables—one is the multiplica- fect by using SPSS PROCESS template model 4. As shown in
tion term of abusive supervision and psychological con- Table 4, the indirect effect of abusive supervision on knowl-
tract fulfillment and another one is the multiplication term edge sharing via LMX was negative (−.11) and bootstrapped
of abusive supervision and self-enhancement motive—in 95% CI around the indirect effect did not include zero (−.22,
step 4. To support hypotheses, it requires statistically sig- − .05). Hence, the bootstrapping method using PROCESS
nificant increasing values in the variance explained (R2) well supported Hypothesis 2.
with the addition of interaction terms and predicted pat-
terns of consistent with our hypotheses. Test of Moderation
Finally, the moderated-mediation effects were tested by
again using SPSS PROCESS developed by Preacher, Hypothesis 3 predicted an interaction effect of psychological
Rucker, and Hayes (2007). Following the Preacher et al. contract fulfillment and abusive supervision on LMX. The
(2007) recommendation, we set each of high and low levels results shown in Table 5 demonstrated that the interaction term
of psychological contract fulfillment (i.e., Hypothesis 4) and of psychological contract fulfillment and abusive supervision
self-enhancement motive (i.e., Hypothesis 6) as one standard on LMX was significant (Model 4; b = .12, p ≤ .05),
deviation above and below the mean score of each psycholog- supporting Hypothesis 3. We plotted the results adapting
ical contract fulfillment and self-enhancement motive. Aiken and West’s (1991) method of ± 1 standard deviation.
PROCESS provides regions of significance of the bootstrap As presented in Fig. 2, the negative relationship between abu-
estimates of the conditional indirect effects, as well as boot- sive supervision and LMX was weakened when psychological
strap estimates based on bias-corrected and accelerated confi- contract fulfillment is high. The simple slope test displayed
dence intervals to test hypotheses. We have also entered our that the negative relationship between abusive supervision and
control variables (i.e., age, gender, and education) in the mod- LMX is significant both when an employee’s psychological
erated mediation analyses tested by SPSS PROCESS. contract fulfillment is high (b = − .33, t = − 3.60, p = .00)
J Bus Psychol

Table 1 Confirmatory factor analysis results

Model Description χ2 DF CFI TLI RMSEA △χ2 △DF

Hypothesized model Five-factor modela 847.538 454 .929 .918 .071

Model 1 Four-factor modelb 1722.530 458 .773 .738 .126 874.992 4.00
Model 2 Three-factor modelc 2223.323 461 .684 .638 .148 500.793 3.00
Model 3 Two-factor modeld 3081.551 463 .530 .464 .180 858.228 2.00
Model 4 One-factor modele 4206.172 464 .328 .235 .215 1124.621 1.00

CFI Comparative Fit Index, TLI Tucker-Lewis Index, RMSEA Root Mean Square Error of Approximation
a
Five factors: abusive supervision; leader-member exchange; psychological contract fulfillment self-enhancement motive; knowledge sharing
b
Four factors: abusive supervision; leader-member exchange; psychological contract fulfillment and self-enhancement motive combined; knowledge
sharing
c
Three factors: abusive supervision and leader-member exchange combined; psychological contract fulfillment and self-enhancement motive combined;
knowledge sharing
d
Two factors: abusive supervision, leader-member exchange, psychological contract fulfillment, and self-enhancement motive combined; knowledge
sharing
e
One factor: abusive supervision, leader-member exchange, psychological contract fulfillment, self-enhancement motive, and knowledge sharing
combined

and when an employee’s psychological contract fulfillment is (b = − .66, t = − 5.39, p = .00). Accordingly, Hypothesis 5
low (b = − .57, t = − 6.37, p = .00). Therefore, Hypothesis 3 was also supported.
received support.
Hypothesis 5 proposed an interaction effect of self- Test of Moderated Mediation
enhancement motive and abusive supervision on LMX. The
results shown in Table 5 demonstrated that the interaction term To examine the conditional indirect effect of abusive supervi-
of self-enhancement motive and abusive supervision on LMX sion on knowledge sharing (through LMX) with psychological
was significant (Model 4; b = .21, p ≤ .05). We plotted the contract fulfillment, which is Hypothesis 4, we used SPSS
results using Aiken and West’s (1991) procedure of ± 1 stan- PROCESS developed by Preacher and his colleagues (2007).
dard deviation. As shown in Fig. 3, the negative relationship Following Preacher and his colleagues’ (2007) recommenda-
between abusive supervision and LMX was weakened when tion, we set high and low level of psychological contract ful-
self-enhancement motive is high. The simple slope test exhib- fillment as one standard deviation above and below the mean
ited that the negative relationship between abusive supervi- score of psychological contract fulfillment. As hypothesized,
sion and LMX is significant both when an employee’s self- the indirect effect of abusive supervision on knowledge sharing
enhancement motive is high (b = − .24, t = − 2.30, p = .02) via LMX was conditional to the level of psychological contract
and when an employee’s self-enhancement motive is low fulfillment. As stated in Table 6, the indirect effect was weaker

Table 2 Means, standard deviations, and correlations

Mean SD 1 2 3 4 5 6 7 8

1. Agea 36.59 6.58

2. Gendera .25 .43 − .27***
3. Educationa 3.13 .55 .09 − .19**
4. Abusive supervisiona 1.65 .89 − .08 − .04 .03 (.98)
5. Leader-member exchangea 5.25 .96 .22** − .17* .19** − .43*** (.93)
6. Psychological contract fulfillmenta 4.64 1.20 .00 − .16* − .17* − .12 .32*** (.93)
7. Self-enhancement motivea 5.12 .91 .12 − .12 .10 .02 .14 .11 (.91)
8. Knowledge sharingb 5.18 .97 .07 .07 − .01 − .20** .26*** .09 − .05 (.95)

N = 175. Reliabilities are on the diagonal in parentheses. Age was measured in years. For gender, 0 = male, 1 = female. Education was measured on a
scale that ranged from 1 (high school) to 4 (master’s degree or higher)
*p ≤ .05; **p ≤ .01; ***p ≤ .001 (two-tailed)
a
Self-rated
b
Supervisor-rated
 J Bus Psychol

Table 3 Regression results for main effect indirect effect not containing zero (−.35, − .07). Therefore,
Knowledge sharing b Hypothesis 6 was supported that the negative indirect effect
of abusive supervision on knowledge sharing was weakened
Model 1 Model 2 at the high level of self-enhancement motive.
Step 1. Control variables
Age .02 .01
Discussion
Gender .22 .19
Education − .00 .01
Building on the social exchange perspective, the purpose of
Step 2. Main effect
this study was to expand our understanding in the contempo-
Abusive supervisiona − .20** rary literature on knowledge sharing behavior and abusive
Overall F .83 2.21 supervision by observing the main effects, mechanisms, and
R2 .01 .05 moderators thereof. Extending earlier work, the current study
△F 6.26** broadened the research focus by exploring a moderated medi-
△R2 .04 ation model. We suggest that LMX fully mediates the negative
N = 175
relationship between abusive supervision and knowledge
sharing. Moreover, our findings demonstrated how psycho-
*p ≤ .05; **p ≤ .01; ***p ≤ .001 (two-tailed)
a logical contract fulfillment and the self-enhancement motive
Self-rated
b buffer the negative effect of abusive supervision on LMX.
Supervisor-rated
Most important, we found that the mediated relationship be-
tween abusive supervision and knowledge sharing via LMX is
and not significant at the high level of psychological contract contingent on the level of one’s perception of the organization
fulfillment (−.06, p = .07) but was stronger and significant at (i.e., psychological contract fulfillment) and of the individual
the low level of psychological contract fulfillment (−.15, (i.e., self-enhancement motive). Specifically, our findings
p = .01), with a bootstrapped 95% CI around the indirect effect showed that when psychological contract fulfillment and
not containing zero (−.28, − .06). Thus, it can understood that self-enhancement motive are high rather than low, the medi-
the negative indirect impact of abusive supervision on knowl- ated negative relationship is weakened. Our results provide
edge sharing was weakened at the high level of psychological compelling implications to both theory and managerial
contract fulfillment, supporting Hypothesis 4. practices.
For Hypothesis 6, which also predicts the conditional
indirect effect of abusive supervision on knowledge sharing Theoretical Implications
(through LMX) with self-enhancement motive, we examined
using the same method—SPSS PROCESS—with Our study supplements the theoretical discussion by providing
Hypothesis 4. As hypothesized, the indirect effect of abusive three major theoretical implications. First, the current study
supervision on knowledge sharing via LMX was conditional contributes to the existing knowledge sharing literature by
to the level of psychological contract fulfillment. As stated examining a leader factor as a critical predictor of knowledge
in Table 6, the indirect effect tested by PROCESS was sharing behavior. Numerous previous studies examined the
weaker and not significant at the high level of self- antecedents that enhance knowledge sharing such as work-
enhancement motive (−.06, p = .10) but was stronger and place atmosphere, organizational justice, and trust (Wang &
significant at the low level of self-enhancement motive Noe, 2010). Moreover, previous research indicated the posi-
(−.18, p = .01), with a bootstrapped 95% CI around the tive effect of constructive leader behaviors on knowledge
sharing (e.g., Bryant, 2003; Srivastava et al., 2006).
Table 4 Results of bootstrap for indirect effect through leader-member
However, research examining leaders’ destructive behaviors
exchange as predictors of employee knowledge sharing is lacking. To
fill the gap in the literature, our results contribute to knowl-
Bias-corrected confidence intervals edge sharing by investigating abusive supervision as a barrier.
Dependent variable Indirect effect SE (boot) Lower CI Upper CI In addition, our research adopted social exchange theory to
explain the knowledge sharing process. Previous research
Knowledge sharing − .11 .04 − .22 − .05 positively related the norm of reciprocity to individuals’
knowledge sharing behaviors (Chiu, Hsu, & Wang, 2006).
N = 175. Confidence interval does not include zero. Thus, indirect effect
is indeed significantly different from zero at p < .05 (two-tailed). Control
Considering knowledge as a valuable resource for each indi-
variables: employee’s age, gender, education. Number of samples used vidual, social exchange theory is a useful framework for un-
for indirect effect confidence intervals = 1000 derstanding when individuals may or may not share their
J Bus Psychol

Table 5 Multiple regression results

Leader-member exchangea

Model 1 Model 2 Model 3 Model 4a Model 4b Model 4

Step 1. Control variables

Age .03** .02 .02** .02* .02** .02**
Gender − .22 − .28 − .13 − .12 − .13 − .09
Education .27** .28** .37*** .34** .37*** − .33**
Step 2. Main effect
Abusive supervisiona − .46*** − .43*** − .43*** − .43*** − .45***
Step 3. Moderators
Psychological contract fulfillment (PCF)a .23*** .25*** .26*** .24***
Self-enhancement motive (SEM)a .08 .08 .11 .10
Step 4. Interaction effects
Abusive supervision × PCF .15** .12*
Abusive supervision × SEM .25** .21*
Overall F 5.17** 15.49*** 15.27*** 14.34*** 14.61*** 13.47***
R2 .08 .27 .35 .38 .38 .39
△F 42.68*** 11.13*** 6.03** 7.27** 5.58**
△R2 .18 .09 .02 .03 .04

N = 175. a Self-rated. b Supervisor-rated. * p ≤ .05; ** p ≤ .01; *** p ≤ .001 (two-tailed)

Model 4a and Model 4b present the results of each moderator separately, reporting the R-squared change by comparing these models to Model 3 (the
controls + predictors only models). Model 4 presents the results of two moderators simultaneously, reporting the R-squared change by comparing this to
Model 3

valuable knowledge (Cabrera & Cabrera, 2005). Based on the would be meaningful to identify other factors that may prevent
norm of reciprocity (Gouldner, 1960), social exchange theory employees from sharing their knowledge.
(Blau, 1964) implies that employees may exhibit positive at- Second, our investigation extends and complements the
titudes and behaviors when they receive positive treatment. In abusive supervision literature by examining knowledge shar-
contrast, employees may react negatively based on the norm ing as an outcome of abusive supervision. Past research inves-
of negative reciprocity. As such, our research demonstrated tigated the negative effects of abusive supervision on employ-
that abused employees are likely to reduce the level of knowl- ee attitudes, behaviors, and job-related performance. For ex-
edge sharing as negative reciprocity after receiving negative ample, abusive supervision decreases job satisfaction, organi-
treatment from their abusive supervisors. In future research, it zational commitment, task performance, and organizational

Fig. 2 Interaction of abusive supervision and psychological contract Fig. 3 Interaction of abusive supervision and self-enhancement motive
fulfillment on leader-member exchange on leader-member exchange
 J Bus Psychol

Table 6 Moderated mediation

results for knowledge sharing Moderator Level Knowledge sharingb
across the level of psychological
contract fulfillment and self- Conditional SE z p 95% bias-corrected CI
enhancement motive indirect effect
Lower Upper

Psychological contract fulfillmenta Low − .15 .05 − 2.84 .01 − .28 − .06
High − .06 .03 − 1.83 .07 − .16 − .02
Self-enhancement motivea Low − .18 .07 − 2.66 .01 − .35 − .07
High − .06 .04 − 1.67 .10 − .17 − .01

N = 175. Control variables: age, gender, education. Bootstrap sample size = 1000
a
Self-rated
b
Supervisor-rated

citizenship behavior (Harris et al., 2007; Martinko et al., 2013; meta-analytic study on LMX found that the quality of LMX
Tepper, 2000; Zellars et al., 2002). As a significant outcome in can be determined not only by individual factors such as per-
the knowledge-based economy, our research demonstrated the sonalities (i.e., extraversion and agreeableness) or affectivity
negative influence of abusive supervision on knowledge shar- but also by leader behaviors (i.e., transformational leadership
ing. Furthermore, this research investigated the mediating role and contingent rewarding behaviors) (Dulebohn, Bommer,
of LMX in the relationship between abusive supervision and Liden, Brouer, & Ferris, 2012). In addition, numerous earlier
knowledge sharing. Previous research demonstrated how sub- studies have shown that LMX is mostly connected with pos-
ordinates’ justice perceptions mediate relationships between itive attitudinal and performance-related variables of focal em-
abusive supervision and job satisfaction, life satisfaction, or- ployees. For example, higher overall satisfaction (e.g., Graen,
ganizational commitment, and work-family conflict (Tepper, Novak, & Sommerkamp, 1982), stronger organizational com-
2000). Another study indicated that affective commitment mitment (e.g., Nystrom, 1990), and higher performance rat-
mediates the relationship between abusive supervision and ings (e.g., Liden, Wayne, & Stilwell, 1993) were examined
organization deviance (Tepper, Henle, Lambert, Giacalone, with employees who perceived high-quality LMX. LMX
& Duffy, 2008). In addition, emotional exhaustion has been scholars proposed LMX as a mediator in the relationship be-
explored as a mediator between abusive supervision and in- tween leader behavior and employee outcomes (e.g., Wang
terpersonal facilitation relationships (Aryee et al., 2008). As a et al., 2005; Chan & Mak, 2012). For example, Chan and
relational factor, our research explored LMX and examined Mak (2012) recently showed that LMX mediates benevolent
how it mediates the relationship between abusive supervision leadership and employee task performance as well as OCB.
and knowledge sharing. Future research might investigate oth- Wang et al. (2005) also found that LMX mediates the relation-
er relational factors as mediators in the relationship between ship between transformational leadership and both employee
abusive supervision and outcomes. task performance and OCB. Similarly, we extended previous
Importantly, our results demonstrated how organizational research to present LMX as a critical mediator of the relation-
and individual factors may mitigate the pernicious effects of ship between abusive supervision and employee knowledge
abusive supervision on knowledge sharing. Specifically, this sharing.
study showed that the negative impacts of abusive supervision Moreover, our research has shown that the interaction be-
on knowledge sharing via LMX are attenuated if the psycho- tween abusive supervision and organizational (i.e., psycholog-
logical contract is fulfilled or if the self-enhancement motive is ical contract fulfillment) or personal factors (i.e., self-
high. In future studies, it would be interesting to examine how enhancement motive) may change the quality of LMX. Our
psychological contract fulfillment or the self-enhancement research demonstrated that the negative influence of abusive
motive minimizes the negative effects of abusive supervision supervision on LMX is likely to weaken when both psycho-
on other outcomes such as task performance or citizenship logical contract fulfillment the self-enhancement motive are
behaviors. Moreover, identifying other buffering factors that high. Future research should examine other organizational and
reduce the negative impacts of abusive supervision on out- personal factors that may influence the quality of LMX.
comes is worthwhile.
Last, the findings of this study add to our existing knowl- Managerial Implications
edge on LMX. Drawing on the social exchange perspective,
our results confirmed LMX as an underlying mechanism This research also provides practical insights for both the man-
linking abusive supervision and knowledge sharing. A recent agement practitioner and the organization. Although many
J Bus Psychol

companies have introduced policies and procedures to inspire prosocial behaviors in this knowledge-based society (Kim
employees to share their knowledge through, for example, the et al., 2017), we believe that self-rating of knowledge sharing
development of a knowledge sharing culture or knowledge might bring an inflation issue. Nonetheless, a better approach
management practices, these policies may not work effective- could be employed to assess and to collect knowledge sharing
ly when there is abusive supervision (Bock, Kankanhalli, & data from various sources including supervisors, coworkers,
Sharma, 2006). The present study reveals that abusive leader and the self in future research. Since knowledge sharing tends
behaviors can stop employees from sharing their valuable to be implicit and subtle, it might be meaningful to compare
knowledge. Understanding the harmful consequences of abu- how multiple sources of knowledge sharing influence the re-
sive supervision, a number of organizations should be aware sults in different ways. Lastly, it is worthwhile conducting
of the importance of the damaging influence of supervisors on research in another context to enhance the generalizability of
outcomes, which can discourage employees from building our findings. Our research was conducted in South Korea, the
quality relationships with supervisors and sharing knowledge culture of which is characterized by high power distance and a
in the workplace. However, abusive supervision is not easily hierarchical structure. In future research, it is necessary to
detected or properly managed (Tepper et al., 2007). Therefore, replicate our results in different contexts.
it is necessary to identify factors that contribute to abusive Despite some limitations, this research enriches our under-
leader behaviors. Organizations should closely monitor occur- standing of knowledge sharing by examining abusive super-
rences of abusive supervision to understand the main causes. vision as a predictor and LMX as a mediating mechanism of
In addition, establishing an organizational culture that rewards the abusive supervision–knowledge sharing relationship.
good leaders and punishes destructive behaviors is required. Furthermore, the present study provides empirical evidence
Leadership training programs could be helpful in preventing that psychological contract fulfillment and the employee
such abusive behaviors. Our results have shown how psycho- self-enhancement motive may mitigate the negative impact
logical contract fulfillment may buffer the negative influence of abusive supervision on LMX and knowledge sharing via
of abusive supervision on knowledge sharing through LMX LMX. Recognizing the importance of knowledge sharing, our
relationships. As our research indicated, psychological con- research provides insights into how employees share their
tract fulfillment can be a practical solution to decrease the knowledge with others.
unfavorable effects of abusive supervision on LMX and
knowledge sharing via LMX. Acknowledgements This work was supported by Incheon National
University (International Cooperative) Research Grant in 2016.

Limitations and Conclusion

References
The present study is subject to limitations. First, we used a
cross-sectional design, meaning we cannot infer causality. Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and
Future research may implement longitudinal research or an interpreting interactions. Thousand Oaks, CA: Sage.
experimental design to strengthen our findings. Second, there Aryee, S., Sun, L. Y., Chen, Z. X. G., & Debrah, Y. A. (2008). Abusive
supervision and contextual performance: The mediating role of
might be a potential risk of common method bias. We tried to emotional exhaustion and the moderating role of work unit structure.
minimize this matter by collecting data from two different Management and Organization Review, 4, 393–411.
sample sources, namely employees and their supervisors. Bartol, K. M., & Srivastava, A. (2002). Encouraging knowledge sharing:
For example, we applied supervisors’ data on knowledge shar- The role of organizational rewards systems. Journal of Leadership
ing, while subordinates observed a supervisor’s abusive su- and Organization Studies, 9, 64–76.
Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D.
pervision to increase objectivity. Moreover, common method (2001). Bad is stronger than good. Review of General
bias is less likely to be a problem for this study because it Psychology, 5, 323–370.
exhibits significant interaction effects (Evans, 1985). Bettman, J. R. (1979). An information processing theory of consumer
However, it is possible that subordinates would have rated choice. Reading, MA: Addison-Wesley Publishing Company.
abusive supervision and LMX being aware that their supervi- Blau, P. M. (1964). Exchange and power in social life. New York, NY:
Wiley.
sor is also taking the survey. Recognizing this issue, we made
Bock, G. W., Kankanhalli, A., & Sharma, S. (2006). Are norms
every effort to maintain the anonymity of participants’ re- enough? The role of collaborative norms in promoting organi-
sponses and to keep the confidentiality. Nevertheless, future zational knowledge seeking. European Journal of Information
research should consider this issue and take various ap- Systems, 15, 357–367.
proaches to measure these variables to enhance objectivity. Bolino, M., Kacmar, K. M., Turnley, W. H., & Gilstrap, J. B. (2008). A
multi-level review of impression management motives and behav-
Third, employee knowledge sharing was rated by their imme- iors. Journal of Management, 34, 1080–1109.
diate supervisor to increase objectivity in this study. Since Brislin, R. W. (1980). Translation and content analysis of oral and written
knowledge sharing could be viewed as one of important material. In H. C. Triandis & J. W. Berry (Eds.), Handbook of cross-
 J Bus Psychol

cultural psychology, vol. 2: Methodology (pp. 389–444). Boston, Gouldner, A. W. (1960). The norm of reciprocity: A preliminary state-
MA: Allyn & Bacon. ment. American Sociological Review, 25, 161–178.
Bryant, S. E. (2003). The role of transformational and transactional lead- Graen, G., & Cashman, J. F. (1975). A role-making model of leadership
ership in creating, sharing and exploiting organizational knowledge. in formal organizations: A developmental approach. Leadership
Journal of Leadership & Organizational Studies, 9, 32–44. Frontiers, 143–165.
Cabrera, E. F., & Cabrera, A. (2005). Fostering knowledge sharing Graen, G., Novak, M. A., & Sommerkamp, P. (1982). The effects of
through people management practices. International Journal of leader–member exchange and job design on productivity and satis-
Human Resource Management, 16, 720–735. faction: Testing a dual attachment model. Organizational Behavior
Carroll, S. J., & Gillen, D. I. (1987). Are the classical management func- and Human Performance, 30, 109–131.
tions useful in describing managerial work? Academy of Graen, G. B. (1976). Role making processes within complex organiza-
Management Review, 12, 38–51. tions. In M. D. Dunnette (Ed.), Handbook of industrial and organi-
Chan, M. E., & McAllister, D. J. (2014). Abusive supervision through the zational psychology (pp. 1201–1245). Chicago, IL: Rand-McNally.
lens of employee state paranoia. Academy of Management Review, Graen, G. B., & Scandura, T. A. (1987). Toward a psychology of dyadic
39, 44–66. organizing. Research in Organizational Behavior, 9, 175–208.
Chan, S. C., & Mak, W. M. (2012). Benevolent leadership and follower Graen, G. B., & Uhl-Bien, M. (1995). Relationship-based approach to
performance: The mediating role of leader–member exchange leadership: Development of leader-member exchange (LMX) theory
(LMX). Asia Pacific Journal of Management, 29, 285–301. of leadership over 25 years: Applying a multi-level multi-domain
Chi, S. C. S., & Liang, S. G. (2013). When do subordinates' emotion- perspective. The Leadership Quarterly, 6, 219–247.
regulation strategies matter? Abusive supervision, subordinates' Grandey, A. A., Kern, J., & Frone, M. (2007). Verbal abuse from out-
emotional exhaustion, and work withdrawal. The Leadership siders versus insiders: Comparing frequency, impact on emotional
Quarterly, 24, 125–137. exhaustion, and the role of emotional labor. Journal of Occupational
Chiaburu, D. S., Lorinkova, N. M., & Van Dyne, L. (2013). Employees’ Health Psychology, 12, 63–79.
social context and change-oriented citizenship: A meta-analysis of Grant, A. M., & Mayer, D. M. (2009). Good soldiers and good actors:
leader, coworker, and organizational influences. Group & Prosocial and impression management motives as interactive pre-
Organization Management, 38, 291–333. dictors of affiliative citizenship behaviors. Journal of Applied
Chiu, C. M., Hsu, M. H., & Wang, E. T. (2006). Understanding knowl- Psychology, 94, 900–912.
edge sharing in virtual communities: An integration of social capital Grant, R. M. (1996). Toward a knowledge-based theory of the firm.
and social cognitive theories. Decision Support Systems, 42, 1872– Strategic Management Journal, 17, 109–122.
1888. Harris, K. J., Kacmar, K. M., & Zivnuska, S. (2007). An investigation of
Conway, N., & Briner, R. B. (2005). Understanding psychological con- abusive supervision as a predictor of performance and the meaning
tracts at work: A critical evaluation of theory and research. Oxford of work as a moderator of the relationship. The Leadership
University Press. Quarterly, 18, 252–263.
Cropanzano, R., & Mitchell, M. S. (2005). Social exchange theory: An Harvey, P., Stoner, J., Hochwarter, W., & Kacmar, C. (2007). Coping with
interdisciplinary review. Journal of Management, 31, 874–900. abusive supervision: The neutralizing effects of ingratiation and
Dienesch, R. M., & Liden, R. C. (1986). Leader-member exchange model positive affect on negative employee outcomes. The Leadership
of leadership: A critique and further development. Academy of Quarterly, 18, 264–280.
Management Review, 11, 618–634. Henderson, D. J., Wayne, S. J., Shore, L. M., Bommer, W. H., & Tetrick,
Dierdorff, E. C., Rubin, R. S., & Morgeson, F. P. (2009). The milieu of L. E. (2008). Leader-member exchange, differentiation, and psycho-
managerial work: An integrative framework linking work context to logical contract fulfillment: A multilevel examination. Journal of
role requirements. Journal of Applied Psychology, 94, 972–988. Applied Psychology, 93, 1208–1219.
Duffy, M. K., Ganster, D. C., & Pagon, M. (2002). Social undermining in Hu, H.-H. (2012). The influence of employee emotional intelligence on
the workplace. Academy of Management Journal, 45, 331–351. coping with supervisor abuse in a banking context. Social Behavior
Dulebohn, J. H., Bommer, W. H., Liden, R. C., Brouer, R. L., & Ferris, G. and Personality, 40, 863–874.
R. (2012). A meta-analysis of antecedents and consequences of Hull, C. L. (1943). Principles of behavior. New York: Appleton–
leader-member exchange integrating the past with an eye toward Century–Crofts.
the future. Journal of Management, 38, 1715–1759. Ilies, R., Nahrgang, J. D., & Morgeson, F. P. (2007). Leader-member
Eisenberger, R., Huntington, R., Hutchison, S., & Sowa, D. (1986). exchange and citizenship behaviors: A meta-analysis. Journal of
Perceived organizational support. Journal of Applied Psychology, Applied Psychology, 92, 269–277.
71, 500–507. Judge, T. A., LePine, J. A., & Rich, B. L. (2006). Loving yourself abun-
Evans, M. G. (1985). A Monte Carlo study of the effects of correlated dantly: Relationship of the narcissistic personality to self- and other
method variance in moderated multiple regression analysis. perceptions of workplace deviance, leadership, and task and contex-
Organizational Behavior and Human Decision Process, 36, tual performance. Journal of Applied Psychology, 91(4), 762–776.
305–323. Kacmar, K. M., Bachrach, D. G., Harris, K. J., & Zivnuska, S. (2011).
French Jr., J. R. P., & Raven, B. H. (1959). The bases of social power. In Fostering good citizenship through ethical leadership: Exploring the
D. Cartwright (Ed.), Studies in social power (pp. 150–167). Ann moderating role of gender and organizational politics. Journal of
Arbor, MI: Institute for Social Research. Applied Psychology, 96, 633–642.
Frieder, R. E., Hochwarter, W. A., & DeOrtentiis, P. S. (2015). Kim, S. L., Han, S., Son, S. Y., & Yun, S. (2017). Exchange ideology in
Attenuating the negative effects of abusive supervision: The role supervisor-subordinate dyads, LMX, and knowledge sharing: A so-
of proactive voice behavior and resource management ability. The cial exchange perspective. Asia Pacific Journal of Management,
Leadership Quarterly, 26, 821–837. 34(1), 147–172.
Garcia, P. R. J. M., Wang, L., Lu, V., Kiazad, K., & Restubog, S. L. D. Kim, S. L., Kim, M., & Yun, S. (2015). Knowledge sharing, abusive
(2015). When victims become culprits: The role of subordinates’ supervision, and support: A social exchange perspective. Group &
neuroticism in the relationship between abusive supervision and Organization Management, 40, 599–624.
workplace deviance. Personality and Individual Differences, 72, Leung, K., & Zhou, F. (2008). Methodology for cross-cultural research.
225–229. In X.-P. Chen, A. S. Tsui & J.-L. Farh (Eds.), Empirical methodsin
J Bus Psychol

organization and management research (pp. 385–408). Beijing, Spender, J. C., & Grant, R. M. (1996). Knowledge and the firm:
China: Peking University Press. Overview. Strategic Management Journal, 17, 5–9.
Liden, R. C., & Graen, G. (1980). Generalizability of the vertical dyad Srivastava, A., Bartol, K. M., & Locke, E. A. (2006). Empowering
linkage model of leadership. Academy of Management Journal, 23, leadership in management teams: Effects on knowledge sharing,
451–465. efficacy, and performance. Academy of Management Journal,
Liden, R. C., Wayne, S. J., & Stilwell, D. (1993). A longitudinal study on 49, 1239–1251.
the early development of leader-member exchanges. Journal of Tepper, B. J. (2000). Consequences of abusive supervision. Academy of
Applied Psychology, 78, 662–674. Management Journal, 43, 178–190.
Martinko, M. J., Harvey, P., Brees, J. R., & Mackey, J. (2013). A review Tepper, B. J. (2007). Abusive supervision in work organizations:
of abusive supervision research. Journal of Organizational R e v i e w, s y n t h e s i s , a n d re s e a r c h a g e n d a . J o u r n a l o f
Behavior, 34, S120–S137. Management, 33, 261–289.
Masterson, S. S., Lewis, K., Goldman, B. M., & Taylor, M. S. (2000). Tepper, B. J., Duffy, M. K., & Shaw, J. D. (2001). Personality moderators
Integrating justice and social exchange: The differing effects of fair of the relationships between abusive supervision and subordinates’
procedures and treatment on work relationships. Academy of resistance. Journal of Applied Psychology, 86, 974–983.
Management Journal, 43, 738–748. Tepper, B. J., Henle, C. A., Lambert, L. S., Giacalone, R. A., & Duffy, M.
Mitchell, M. S., & Ambrose, M. L. (2007). Abusive supervision and K. (2008). Abusive supervision and subordinates' organization de-
workplace deviance and the moderating effects of negative reciproc- viance. Journal of Applied Psychology, 93, 721–732.
ity beliefs. Journal of Applied Psychology, 92, 1159–1168. Tepper, B. J., Moss, S. E., Lockhart, D. E., & Carr, J. C. (2007).
Nahrgang, J. D., Morgeson, F. P., & Ilies, R. (2009). The development of Abusive supervision, upward maintenance communication, and
leader–member exchanges: Exploring how personality and perfor- subordinates' psychological distress. Academy of Management
mance influence leader and member relationships over time. Journal, 50, 1169–1180.
Organizational Behavior and Human Decision Processes, 108, Turnley, W. H., Bolino, M. C., Lester, S. W., & Bloodgood, J. M. (2003).
256–266. The impact of psychological contract fulfillment on the performance
Nandkeolyar, A. K., Shaffer, J. A., Li, A., Ekkirala, S., & Bagger, J. of in-role and organizational citizenship behaviors. Journal of
(2014). Surviving an abusive supervisor: The joint roles of consci- Management, 29, 187–206.
entiousness and coping strategies. Journal of Applied Psychology, Uhl-Bien, M., Graen, G. B., & Scandura, T. A. (2000). Implications of
99, 138–150. leader-member exchange (LMX) for strategic human resource man-
Newman, A., Schwarz, G., Cooper, B., & Sendjaya, S. (2015). How agement systems: Relationships as social capital for competitive
servant leadership influences organizational citizenship behavior: advantage. Research in Personnel and Human Resources
The roles of LMX, empowerment, and proactive personality. Management, 18, 137–185.
Journal of Business Ethics, 145, 1–14. Wang, H., Law, K. S., Hackett, R. D., Wang, D., & Chen, Z. X. (2005).
Nystrom, P. C. (1990). Organizational commitment. Group & Leader-member exchange as a mediator of the relationship between
Organization Studies, 5, 296–312. transformational leadership and followers' performance and organi-
Perugini, M., & Gallucci, M. (2001). Individual differences and social zational citizenship behavior. Academy of Management Journal, 48,
norms: The distinction between reciprocators and prosocials. 420–432.
European Journal of Personality, 15, S19–S35. Wang, S., & Noe, R. A. (2010). Knowledge sharing: A review and direc-
Pfeffer, J., & Fong, C. T. (2005). Building organization theory from first tions for future research. Human Resource Management Review, 20,
principles: The self-enhancement motive and understanding power 115–131.
and influence. Organization Science, 16, 372–388. Wayne, S. J., Shore, L. M., & Liden, R. C. (1997). Perceived organiza-
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for tional support and leader-member exchange: A social exchange per-
estimating indirect effects in simple mediation models. Behavior spective. Academy of Management Journal, 40, 82–111.
Research Methods, Instruments, & Computers, 36, 717–731. Xu, E., Huang, X., Lam, C. K., & Miao, Q. (2012). Abusive supervision
Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing mod- and work behaviors: The mediating role of LMX. Journal of
erated mediation hypotheses: Theory, methods, and prescriptions. Organizational Behavior, 33, 531–543.
Multivariate Behavioral Research, 42, 185–227. Yukl, G. (1981). Leadership in organisations. Engelwood Cliffs, NJ:
Quigley, N. R., Tesluk, P. E., Locke, E. A., & Bartol, K. M. (2007). A Prentice Hall.
multilevel investigation of the motivational mechanisms under- Yun, S., Takeuchi, R., & Liu, W. (2007). Employee self-management
lying knowledge sharing and performance. Organization motives and job performance behaviors: Investigating the mod-
Science, 18, 71–88. erating effects of employee role ambiguity and managerial per-
Robinson, S. L., & Morrison, E. W. (2000). The development of psycho- ceptions of employee commitment. Journal of Applied
logical contract breach and violation: A longitudinal study. Journal Psychology, 92, 745–756.
of Organizational Behavior, 21, 525–546. Zellars, K. L., Tepper, B. J., & Duffy, M. K. (2002). Abusive supervision
Rousseau, D. M. (1990). New hire perceptions of their own and their and subordinates’ organizational citizenship behavior. Journal of
employer’s obligations: A study of psychological contracts. Applied Psychology, 86, 1068–1076.
Journal of Organizational Behavior, 11, 389–400. Zhao, H. A. O., Wayne, S. J., Glibkowski, B. C., & Bravo, J. (2007). The
Rousseau, D. M. (1995). Psychological contracts in organizations: impact of psychological contract breach on work-related outcomes:
Understanding written and unwritten agreements. Sage. A meta-analysis. Personnel Psychology, 60, 647–680.