PERSONNEL PSYCHOLOGY

2016, 69, 67–121

LEADER–MEMBER EXCHANGE (LMX) AND

PERFORMANCE: A META-ANALYTIC REVIEW
ROBIN MARTIN
University of Manchester

YVES GUILLAUME
Aston University

GEOFF THOMAS
University of Surrey

ALLAN LEE
University of Manchester

OLGA EPITROPAKI
ALBA Graduate Business School and Aston University

This paper reports a meta-analysis that examines the relationship be-

tween leader–member exchange (LMX) relationship quality and a mul-
tidimensional model of work performance (task, citizenship, and coun-
terproductive performance). The results show a positive relationship
between LMX and task performance (146 samples, ρ = .30) as well
as citizenship performance (97 samples, ρ = .34), and negatively with
counterproductive performance (19 samples, ρ = –.24). Of note, there
was a positive relationship between LMX and objective task perfor-
mance (20 samples, ρ = .24). Trust, motivation, empowerment, and job
satisfaction mediated the relationship between LMX and task and cit-
izenship performance with trust in the leader having the largest effect.
There was no difference due to LMX measurement instrument (e.g.,
LMX7, LMX-MDM). Overall, the relationship between LMX and per-
formance was weaker when (a) measures were obtained from a different
source or method and (b) LMX was measured by the follower than the
leader (with common source- and method-biased effects stronger for
leader-rated LMX quality). Finally, there was evidence for LMX lead-
ing to task performance but not for reverse or reciprocal directions of
effects.
Introduction

Within the field of leadership, an approach that examines the qual-

ity of the relationship between a leader and a follower (leader–member

We thank the action editor (Fred Morgeson) and the reviewers for helpful comments on
previous drafts of this paper.
Correspondence and requests for reprints should be addressed to Robin Martin, Manch-
ester Business School, University of Manchester, Booth Street West, Manchester, M15
6PB, United Kingdom; [Link]@[Link].

C 2015 Wiley Periodicals, Inc. doi: 10.1111/peps.12100

67
68 PERSONNEL PSYCHOLOGY

exchange theory, LMX) has been popular (Yammarino, Dionne, Chun, &
Dansereau, 2005). LMX theory was introduced by Dansereau, Graen, and
colleagues during the 1970s and was originally referred to as the vertical
dyad linkage (VDL) approach (Dansereau, Graen, & Haga, 1975; Graen
& Cashman, 1975). The main tenant of LMX theory is that, through dif-
ferent types of exchanges, leaders differentiate in the way they treat their
followers (Dansereau et al., 1975), leading to different quality relation-
ships between the leader and each follower. Research shows that high
LMX quality relates to a range of positive follower outcomes (for reviews
see Anand, Hu, Liden, & Vidyarthi, 2011; Martin, Epitropaki, Thomas,
& Topakas, 2010; Schriesheim, Neider, & Scandura, 1998; van Breuke-
len, Schyns, & Le Blanc, 2006). Given the above reviews, one might
conclude that we have already gained a comprehensive understanding of
how LMX affects various outcomes, in accordance with LMX theory.
However, we believe there are some important theoretical issues that re-
main unanswered with respect to the relationship between LMX and work
performance that could be addressed through a meta-analytic review. We
briefly describe three main research issues that we aim to address that
significantly contribute to the LMX literature.
First, although the relationship between LMX and task and contex-
tual performance has been established (e.g., Dulebohn, Bommer, Liden,
Brouer, & Ferris, 2012; Gerstner & Day, 1997; Ilies, Nahrgang, & Morge-
son, 2007), no prior meta-analysis has focused on the relationship between
LMX and counterproductive performance (i.e., negative behaviors that
harm others in the organization, such as property misuse, theft), despite
an increasing number of studies examining this aspect of performance.
There are many theoretically important reasons to examine counterproduc-
tive performance, which are elucidated in more detail later, including the
fact that it is highly predictive of overall performance (Rotundo & Sack-
ett, 2002) and that, unlike task and citizenship performance, it is based
on negative, rather than positive, follower behaviors. In terms of LMX
theory, it is important to determine whether the benefits of positive LMX
relationships generalize to this important aspect of work performance.
Second, little is known of the potential mediators between LMX and
performance. Although there are strong theoretical underpinnings to LMX
theory (e.g., role theory, Graen, 1976; Graen & Scandura, 1987; social
exchange theory, Blau, 1964; Sparrowe & Liden, 1997; Thibaut & Kelley,
1959; self-determination theory, Deci & Ryan, 1985; Liden, Wayne, &
Sparrowe, 2000), it is not clear what the specific mediating mechanisms
between LMX and performance are. Although these theories provide
different accounts of how LMX leads to performance, each proposes a
different set of mediators (e.g., role clarity, role theory; job satisfaction,
 ROBIN MARTIN ET AL. 69

social exchange theory; motivation, self-determination theory). By

examining a range of theoretically proposed mediators, the current study
provides a much-needed opportunity to test some of the underlying
mechanisms explaining how LMX affects performance.
The third issue concerns the direction of effect between LMX and
performance. LMX theory assumes, but rarely tests, the assumption that
relationship quality has a direct effect on performance. Although there
might be strong reasons to propose such a link, it seems plausible to
assume that performance affects LMX or that the relationship between the
two is reciprocal (Danserau et al., 1975; see also Nahrgang, Morgeson,
& Ilies, 2009). It seems important, therefore, to examine the direction of
the effect between LMX and performance as this will allow a test of the
assumption in LMX theory that relationship quality determines outcomes
or whether the other direction of effects exists.
In summary, the meta-analysis makes a number of contributions to
examining LMX theory: The use of wider sample selection criteria to
obtain a larger sample size of studies permits the examination of some im-
portant theoretical relationship that have not yet been examined in detail
(e.g., the relation between LMX and objective performance); examina-
tion of a multidimensional model of performance with the inclusion, for
the first time in a meta-analysis, of counterproductive performance; ex-
amination of alternative theoretically-derived mediational models based
on role, social exchange, and self-determination theories; examination of
important moderators (e.g., performance type, LMX measurement, LMX
rater, and same vs. nonsource effects); and, finally, the first attempt to
meta-analytically examine the causal direction of effects in the LMX–
performance relationship.
In the following section, we first briefly outline a multidimensional
model of work performance that guides our meta-analysis and then de-
velop specific hypotheses concerning the main theoretical issues in this
meta-analysis (concerning main effects, mediators, moderators, and di-
rection of effect between LMX and performance).

LMX and Work Performance: Unresolved Theoretical Issues

Previous meta-analyses of LMX have taken a narrow view of the con-

cept of performance. In some cases, the conceptualization of performance
has been “performance ratings” and “objective performance” (Gerstner &
Day, 1997) or measures combined into one category of “job performance”
(Dulebohn et al., 2012). However, performance is a multidimensional
concept (e.g., Kaplan, Bradley, Luchman, & Haynes, 2009; Rotundo &
Sackett, 2002; Viswesvaran & Ones, 2000), with each dimension relating
70 PERSONNEL PSYCHOLOGY

to a different aspect of performance, and therefore it is important to deter-

mine if predictions from LMX theory apply across different performance
dimensions.
Performance has been conceptualized in numerous ways (e.g., Camp-
bell, 1993; Murphy, 1989), but most of these can be captured within
Rotundo and Sackett’s (2002) three-component model of performance:
task, citizenship, and counterproductive performance (see also Judge &
Kammeyer-Mueller, 2012). Task performance (or in-role performance)
refers to “a group of behaviors involved in the completion of tasks . . .
[and] includes behaviors that contribute to the production of a good or
the provision of a service” (p. 67). This concept covers issues related
to the quantity and quality of work output and the accomplishment of
work duties and responsibilities associated with the job. Citizenship per-
formance (or extra-role performance) concerns a “group of activities that
are not necessarily task-related but that contribute to the organization in
a positive way” (p. 67). Examples of activities that fall within this cate-
gory are altruism, helping and supporting peers, making good suggestions,
spreading goodwill, and defending and supporting organizational objec-
tives. Counterproductive performance is defined as “a group of behaviors
that detract from the goals of the organization . . . [and] as voluntary
behavior that harms the well-being of the organization” (p. 69). There
are a range of activities in this category including property, production
and political deviance, personal aggression, theft, and drug misuse. It
also covers negative behaviors that harm others in the organization and
not following rules and procedures. Counterproductive performance has
some similarities to citizenship performance but tends to focus more on
negative, rather than positive, behaviors. Given the utility of the three-
component view of performance, we shall use this framework to guide the
meta-analysis.
We now turn to examine the relationship between LMX and these three
dimensions of performance (task, citizenship, and counterproductive) in
terms of the main theoretical contributions stated earlier (main effects,
mediators, moderators, and direction of effects).

Main Effects of the LMX and Performance Relationship

Research in LMX has traditionally relied on role and social exchange

theories to explain how different types of LMX relationships develop.
Low LMX relationships are based primarily on the employment contract
and involve mainly economic exchanges (Blau, 1964) that focus on the
completion of work. By contrast, high LMX relationships extend beyond
the formal job contract where the aim is to increase follower’s ability
and motivation to perform at a high level. In high LMX relationships,
 ROBIN MARTIN ET AL. 71

the exchanges are more social in nature involving mutual respect, affect,
support and loyalty, and felt obligation (Uhl-Bien & Maslyn, 2003).
Based on role and social exchange theories, research in LMX (Blau,
1964; Graen & Scandura, 1987; Sparrowe & Liden, 1997; Thibaut &
Kelley, 1959) suggests that a variety of rules and norms govern the pattern
of exchanges between people. For example, a common rule is that of
reciprocity, where the actions of one person lead to the expectation that
the other person will reciprocate with an equally valued exchange (Blau,
1964; Sparrowe & Liden, 1997). The favorable treatment the follower
receives from the leader leads to feelings of obligation to “pay back”
the leader by working hard as a means of reciprocation. In addition, the
positive exchanges between the leader and follower increase feelings of
affect and liking for the leader, and this also motivates followers to want
to meet the leader’s work demands. This should in turn enhance task and
contextual performance.
These arguments are well supported by the empirical evidence as far as
task performance is concerned, irrespective of whether task performance
is measured with leader, peer, or follower ratings (Dulebohn et al., 2012;
Gerstner & Day, 1997; Rockstuhl, Dulebohn, Ang, & Shore, 2012). In
addition, the relationship with objective task performance measures was
found to be much weaker, yet still positive (Gerstner & Day, 1997). There
is also meta-analytic evidence showing that LMX is positively related to
citizenship performance (Dulebohn et al., 2012; Ilies et al., 2007; Scott,
Craven, & Green, 2006). Therefore, based on these meta-analyses and
consistent with LMX theory, we propose the following two hypotheses.
Hypothesis 1: There is a positive relationship between LMX and task
performance.

Hypothesis 2: There is a positive relationship between LMX and citi-

zenship performance.
There are important theoretical and practical reasons to examine the
relationship between LMX and counterproductive performance. First, be-
ing to ensure that the impact of LMX is assessed against all aspects
of performance, not only to fully assess LMX theory but also from a
practical perspective with organizations becoming ever more concerned
with ethical conduct. The importance of this is shown by Rotundo and
Sackett’s (2002) finding that counterproductive performance (together
with task performance) contributed more to judgments of overall work
performance than did citizenship performance. Furthermore, they found
that for some managers, counterproductive performance had the greatest
weight, out of the three performance dimensions, in predicting overall per-
formance judgments. Second, counterproductive performance is the one
72 PERSONNEL PSYCHOLOGY

performance dimension that is based on negative rather than positive fol-

lower behaviors. Because positive and negative social exchanges can have
different effects on relationships (Sparrowe, Liden, Wayne, & Kraimer,
2001), it is therefore important to determine whether high LMX not only
leads to positive work behaviors (e.g., task and citizenship performance)
but also to less engagement in negative behaviors (i.e., counterproductive
performance).
In terms of the relationship between LMX and counterproductive per-
formance, we make the following prediction. In high LMX relationships,
followers feel an obligation to pay back the leader by meeting work de-
mands, which should make it less likely that the follower engages in
behaviors that harm the leader or the organization (as this could impact on
their performance levels). By contrast, in low LMX relationships, follow-
ers might deal with perceived inequity or unfair treatment by their leader
by engaging in counterproductive behaviors so as to harm the leader and
organization. Therefore, based on this, we expect LMX should be nega-
tively related to counterproductive behaviors.

Hypothesis 3: There is a negative relationship between LMX and

counterproductive performance.

Mediators of the LMX and Performance Relationship

The second theoretical issue concerns the mediators between LMX

and performance. LMX theory points to a number of possible medi-
ators explaining why high LMX leads to performance. Therefore, we
test the most common theoretical approaches (role theory, social ex-
change theory, and self-determination theory) that seek to explain how
LMX leads to enhanced performance. The findings will help clarify not
only what mediates LMX effects but also which theory accounts best for
this effect. We describe each of these potential mediators in more detail
below.
Based on role theory (see Graen & Scandura, 1987), good relationships
develop when there is role clarity associated with each person. The labels
“leader” and “follower” (and indeed, “leader–follower” relationship) bring
with them a set pattern of expected behaviors (in a similar way to how
followers have implicit theories of leaders; Epitropaki & Martin, 2004).
For example, the leader role is one in which the person is expected to take
responsibility, make decisions, coordinate resources, and so forth. The
role expectations of the leader and follower will significantly affect the
pattern of social exchanges and the resources that can be exchanged. Given
this, one might expect that when the follower has a good relationship with
the leader, then the nature of the exchanges should reduce uncertainty in
 ROBIN MARTIN ET AL. 73

the work environment and create clear paths to good performance. On

this basis we argue that role clarity is likely to mediate the relationship
between LMX and performance.
Social exchange theory (Cropanzano & Mitchell, 2005) leads to the
expectation that trust in the leader is a potential mediator between LMX
and performance. Trust is at the heart of the LMX construct as LMX has
been defined as a trust-building process (Bauer & Green, 1996; Liden,
Wayne, & Stilwell, 1993; Scandura & Pellegrini, 2008). Through a series
of social exchanges, the leader and follower develop trust with each other
so that there is an expectation that the positive exchanges will continue
(Sue-Chan, Au, & Hackett, 2012). In the leadership literature, more gen-
erally, the relationship between trust and behavioral outcomes such as per-
formance and OCB has been well established (e.g., Burke, Sims, Lazzara,
& Salas, 2007; Colquitt, LePine, Piccolo, Zapata, & Rich, 2012; Dirks &
Ferrin, 2002; Pillai, Schriesheim, & Williams, 1999; Yang & Mossholder,
2010). Research has also shown that trustworthiness leads to trust, which
in turn leads to performance (trustworthiness-trust-performance; Colquitt,
Scott, & LePine, 2007). Based on prior research and LMX theory, we ex-
pect trust to mediate the relationship between LMX and performance.
In addition, job satisfaction and organizational commitment are work
reactions followers exchange with their leaders in return for rewards and
valued outcomes. Prior meta-analyses have examined work attitudes only
as consequences of LMX rather than as an explanatory mechanism of
the relationship between LMX and performance (e.g., Gerstner & Day,
1997). LMX theory proposes that high LMX is an interpersonal rela-
tionship characterized by high levels of affect and liking, and this leads
to increased satisfaction and commitment to both the leader and the or-
ganization (Dulebohn et al., 2012). More generally, the relationship be-
tween work attitudes and performance has received considerable attention
(e.g., Harrison, Newman, & Roth, 2006; Judge, Thoresen, Bono, & Pat-
ton, 2001; Riketta, 2005). The premise that attitudes lead to behavior is
grounded in the social psychological literature (Fishbein & Ajzen, 1975).
Based on this, we suggest there is reliable evidence, and strong theoreti-
cal grounds, to propose that work attitudes (in this case, job satisfaction
and organizational commitment) will be an important mechanism through
which LMX affects performance outcomes.
Self-determination theory (Deci & Ryan, 1985, 2000; for similar ar-
guments, see theorizing on empowerment, Spreitzer, 1995) is a relevant
framework for understanding how high LMX can lead to enhanced per-
formance. Self-determination theory represents a broad framework for
understanding human motivation that focuses on intrinsic and extrinsic
sources of motivation. People are motivated by both external (e.g., re-
ward systems, evaluations) and internal (e.g., interests, curiosity, values)
74 PERSONNEL PSYCHOLOGY

factors. Conditions that support an individual’s experience of autonomy,

competence, and relatedness encourage motivation and engagement in
work-related activities, including enhanced performance and creativity. It
is clear that high LMX relationships tap into all three components of the
theory; autonomy from greater job discretion provided by the leader, com-
petence from increased leader feedback and support on performance, and
relatedness from an enhanced interpersonal relationship with the leader.
Therefore LMX should be positively related to followers’ motivation and
sense of empowerment (see also Liden et al., 2000). We therefore suggest
that motivation and empowerment mediate the relationship between LMX
and performance.

Hypothesis 4: The relationship between LMX and task and citizen-

ship performance will be mediated by role clarity, trust,
job satisfaction, organizational commitment, motiva-
tion, and empowerment.

Moderators of the LMX and Performance Relationship

The third theoretical issue examines moderators of the LMX and per-
formance relationship. Previous reviews show that there is much unex-
plained variation in the relationship between LMX and performance, and
have examined a number of moderators (e.g., Dulebohn et al., 2012;
Gerstner & Day, 1997). This is important not only to provide boundary
conditions for when LMX might lead to performance but also to address
key theoretical issues. In this paper, we examine some important moder-
ators that have not been examined (or not comprehensively). We do not
make specific hypotheses concerning the moderators because, in some
cases, they are not theoretically predicted and in others they are examined
as possible boundary conditions. We examine three potential moderators.
The first concerns common source and common method bias, which
refers to potential problems of measuring LMX and performance from
the same source or method (e.g., leader-rated LMX quality and leader-
assessed performance) and from different sources or methods (e.g., leader-
rated LMX quality and objective performance). It is well known that ef-
fect sizes become inflated when they suffer from common method or
common source bias or when employees rate their own performance
(Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). As a case in point,
Gerstner and Day (1997) found that LMX was more strongly related to
leader-rated performance when LMX was measured by the leader (com-
mon source, ρ = .55) than by the follower (noncommon source, ρ =
.30). Gerstner and Day (1997) noted that the leader-rated LMX and per-
formance correlation may be confounded with same-source bias. It is
 ROBIN MARTIN ET AL. 75

therefore surprising that recent meta-analyses did not distinguish between

performance that was follower-rated, leader-rated, or obtained from an
objective source, and suffered from common source and common method
bias and not on whether the effect sizes were obtained from a separate
source or with a different method (Dulebohn et al., 2012; Ilies et al., 2007;
Rockstuhl et al., 2012). Moreover, we would expect objective performance
measures to be less positively related to LMX. Objective performance
measures (e.g., sales, productivity, accidents) are less prone to rater bias
but may also capture performance aspects that are less under the control
of either the follower or leader. Indeed, Gerstner and Day (1997) in a
meta-analysis reported a corrected r with LMX of .11 (eight samples).
However, the removal of just one study (a field experiment by Graen,
Novak, & Sommerkamp, 1982) resulted in the corrected correlation be-
coming .07.
The second moderator is the type of LMX measure. The LMX literature
is dominated by two measures: First, the LMX-7 scale described by Graen
and Uhl-Bien (1995; see also Dansereau et al., 1975; Scandura & Graen,
1984) consists of seven items reflecting a unidimension of LMX based
on the observation that the LMX dimensions are so highly correlated that
they tap into a single measure and, second, the multidimensional measure
(LMX-MDM) developed by Liden and Maslyn (1998), which consists
of 12 items reflecting four dimensions (contribution, loyalty, affect, and
professional respect). Although there is broad consensus that LMX is a
higher order construct and the correlation between the two main measures
is extremely high (corrected r = .90; Joseph, Newman, & Sin, 2011),
it would be prudent to examine this as a potential moderator, as each
measure tends to be employed by different research teams.
The third moderator is the type of rater. In most cases, LMX is evalu-
ated by the follower as, typically, this is related to follower-level outcomes
(e.g., follower well-being and performance). Meta-analyses have found
higher correlations between leader-rated LMX with performance than
with follower-rated LMX (e.g., Gerstner & Day, 1997). These differences
might be conceptual or methodological in nature (see Schyns & Day,
2010). It would seem, therefore, important to test whether there are dif-
ferences between leader- and follower-rated effects on task, citizenship,
and counterproductive performance, and whether these effects hold even
when common source or method bias is controlled for.

Direction of Effects in LMX and Performance Relationship

The fourth theoretical issue concerns the direction of effect in the

LMX and performance relationship. It is an assumption in LMX theory
that LMX quality directly effects outcomes, that is, the higher the LMX
76 PERSONNEL PSYCHOLOGY

quality, the better will be a range of outcomes (including performance;

e.g., Cogliser, Schriesheim, Scandura, & Gardner, 2009; Maslyn & Uhl-
Bien, 2001; Uhl-Bien, 2006). For example, Dulebohn et al. (2012) state
“it is the nature or quality of leader-follower relationships (i.e., the way
in which the leader and follower characteristics and perceptions com-
bine) that determines critical outcomes,” and also Anand et al. (2011)
state “LMX literature maintains that dyadic relationship quality exerts
significant influence on a wide variety of organizational outcomes.” This
is reflected in research in which LMX is treated as the “independent”
variable predicting other dependent variables (Liden et al., 1997). The
assumption that LMX relationship quality causes outcomes is central in
a number of LMX theories (e.g., Scandura & Lankau, 1996; model of
impact of diverse leaders) and also in research designs in which LMX is
often conceptualized as the mediating variable between antecedents and
outcomes (e.g., LMX is tested as the mediator between transformational
leadership and performance; Howell & Hall-Merenda, 1999).
Although there are strong theoretical reasons to suggest that LMX
affects outcomes (like performance), one might also argue that the re-
verse can occur (i.e., outcomes affect LMX relationship quality). Indeed,
the general attitude-to-behavior link, which underlies much theorizing in
management science, has been questioned and alternative models of effect
direction have been proposed. For example, theories such as expectancy-
based models of motivation (Lawler & Porter, 1967; Vroom, 1964) explic-
itly state that the manipulation of follower rewards leads to performance,
which in turn affects job satisfaction. Indeed, reverse causality has been
examined in a number of meta-analyses between performance and work
reactions (e.g., job satisfaction, Judge et al., 2001; organizational com-
mitment, Riketta, 2008; and attitudes, Harrison et al., 2006) or it has been
advocated for future research (conflict, De Drue & Weingart, 2003; team
efficacy, Gully, Beaubien, Incalcaterra, & Joshi, 2002; business-level sat-
isfactions, Harter, Schmidt, & Hayes, 2002). Finally, one might propose
that the relationship between LMX and performance is reciprocal. Be-
cause social exchanges between the leader and follower occur over time
and follower performance is an important exchange resource, it is possible
that LMX and performance operate as a reciprocal process. Some theo-
rists have expanded this analysis to include concepts from social network
analysis, which emphasize the reciprocity inherent in leader–follower in-
teractions (Sparrowe & Liden, 1997).
The issue of direction of effect between LMX and performance has not
been examined in previous meta-analyses, possibly because most studies
have been cross-sectional in design. However, more recently, there have
been a sufficient number of studies that measure LMX and performance
at different time points, allowing for issues of direction of effects to
 ROBIN MARTIN ET AL. 77

be addressed. Although there are strong theoretical reasons to propose

that LMX determines performance, it would be fruitful to also examine
the possibility of different directions of effects such as reverse causality
(i.e., good performance leads to enhanced LMX relationship quality) or
reciprocal causality.

Hypothesis 5: There is a positive relationship between LMX and per-

formance and this is stronger than the relationship be-
tween performance and LMX.

Method
Literature Search

To locate suitable studies investigating the relationship between LMX

with task, citizenship, and counterproductive performance, we searched
Proquest, PsychInfo, EBSCO, and ISI Web of Science until the year 2012
using keywords such as “Leader–Member Exchange,” “LMX,” “Vertical
Dyad,” “Team Member Exchange,” “TMX,” “Co-Worker Exchange,”
“CWX,” “Leader–Leader Exchange,” and “LLX.” This search included
journal articles, dissertations, book chapters, and conference proceedings.
We also searched the reference lists from relevant review articles (Avolio,
Walumbwa, & Weber, 2009; Graen & Uhl-Bien, 1995; Martin et al., 2010)
and previous meta-analyses (Dulebohn et al., 2012; Gerstner & Day, 1997;
Ilies et al., 2007, Rockstuhl et al., 2012). Furthermore, we contacted
academics that publish regularly in the area of LMX asking if they had
or knew of any unpublished papers or papers that were currently under
review. This initial search resulted in 622 journal articles, dissertations,
book chapters, and papers published in conference proceedings. These
publications were all retrieved and scrutinized using the study inclusion
criteria discussed next.

Study Inclusion

A study had to meet a number of criteria to be included. First, it

had to provide a zero-order correlation between any measure of LMX
and any of the three performance outcomes (i.e., task, citizenship, or
counterproductive) or provide sufficient information to calculate the zero-
order correlation. Second, LMX and the performance outcome had to
be measured at the individual level of analysis. Accordingly, all studies
that measured LMX or the performance outcome at the group level were
excluded. Third, to calculate the sampling error, the study had to report
sample size. Finally, the sample had to be independent and not overlap
with another sample; if a sample appeared in more than one publication,
78 PERSONNEL PSYCHOLOGY

it was only included once. One hundred and ninety-five publications and
207 independent samples (several publications reported multiple samples)
met these criteria. We encountered one redundancy of data (i.e., in which
the same dataset has been published twice).

Dataset

Applying the specified inclusion criteria resulted in an initial set of 146

correlations for the relationship between LMX with task performance, 97
for the relationship between LMX and citizenship performance, and 19
for the relationship between LMX and counterproductive performance.
Independent datasets were constructed for each of the specific categorical
moderator analyses. Dependent correlations in the dataset were repre-
sented by unit-weighted composite correlations.

Coding

The initial coding scheme, along with instructions, was jointly de-
veloped by all authors on the basis of the extant LMX literature. Using
this initial coding scheme, all authors coded 10 randomly selected studies.
The coding was discussed between the authors; any ensuing discrepancies
and problems were resolved, resulting in a refined coding scheme. On the
basis of this refined coding scheme, one of the authors coded all studies; a
nonauthor (who is conducting research in leadership) double checked 20%
of the coding. No discrepancies were encountered. Data requiring sub-
jective judgments (see later for details) were rated by two of the authors.
The overall interrater reliability for the subjective judgments was 98.7%
(performance: 96%, and mediators: 100%). Any discrepancies between
the two raters were resolved by re-examining the original articles; if the
discrepancies could not be resolved, the other authors were consulted.
The type of LMX measure (i.e., LMX-7, LMX-MDM, and LMX
Other) was coded along with the specified performance outcome (i.e.,
task, citizenship, and counterproductive), sample size, reliabilities of
either variable, and moderators (i.e., whether the leader or follower
rated LMX; whether the performance outcome was objective or leader-,
follower-, peer-, or customer-rated; whether LMX was measured before
or after the performance outcome). We coded the LMX measure as
LMX-7 when it was measured with one of the three available LMX-7
measures (Graen et al., 1982; Graen & Uhl-Bien, 1995; Scandura &
Graen, 1984); LMX-MDM was coded when the LMX-MDM scale
developed by Liden and Maslyn (1998) was used. Measures of LMX
Other included modified versions of the LMX scales (Dunegan, Duchon,
& Uhl-Bien, 1992; Stark & Poppler, 2009; Yi-feng & Tjosvold, 2008) as
 ROBIN MARTIN ET AL. 79

well as dyad linkage (VDL) scales (Cashman, 1975; Graen & Cashman,
1975; Snyder & Bruning, 1985; Wakabayashi, Graen, & Graen, 1988) and
the leader–member social exchange (LMXS) scale (Bernerth, Armenakis,
Field, Giles, & Walker, 2007).
We coded two main types of task performance: In-role performance
was assessed with objective measures, such as average sales per hour
(e.g., Klein & Kim, 1998), frequency and magnitude of errors (Vecchio,
1987), and piece-rate pay systems (Lam, Huang, & Snape, 2007); leader-,
peer-, customer-, and self-ratings of commonly used in-role performance
scales, such as the ones developed by Williams and Anderson (1991) and
Podsakoff and Mackenzie (1989), or performance appraisal data based on
supervisor or peer reports, was retrieved from organizational files. Citi-
zenship performance was coded when the study employed self-, leader-,
or peer-rated measures of organizational citizenship behaviors (OCB),
contextual performance, or extra-role behaviors, such as those developed
by Podsakoff, MacKenzie, Moorman, and Fetter (1990) and Williams and
Anderson (1991). Counterproductive performance coding included ob-
jective measures of absenteeism (e.g., van Dierendonck, Le Blanc, & van
Breukelen, 2002), withdrawal behaviors (e.g., Erdogan & Bauer, 2010),
and reported accidents (e.g., Hofmann & Morgeson, 1999); self-rated
measures of psychological withdrawal (e.g., Aryee & Chen, 2006), re-
sistance to change (e.g., van Dam, et al., 2008), and counterproductive
behavior (e.g., Lindsay, 2009); leader-rated scales of retaliation behav-
ior (e.g., Townsend, Phillips, & Elkins, 2000) and social loafing (e.g.,
Murphy, 1998; Murphy, Wayne, Liden, & Erdogan, 2003).
When a study included potential mediators, we also coded the relation-
ship between LMX with the mediator, and the relationship between the
mediator with any of the three performance outcomes (i.e., task, citizen-
ship, or counterproductive). The most common mediators were role clarity,
trust, job satisfaction, organizational commitment, motivation, and em-
powerment. Mediator variables in the primary studies were all self-rated
by the follower. Role clarity included a range of variables such as reverse-
coded role ambiguity and role conflict. Role clarity, role conflict, and role
ambiguity were almost exclusively measured with scales developed by
Rizzo, House, and Lirtzman (1970). Trust included measures of followers
trust with their supervisors or, in one case, their management in general
(van Dam, Oreg, & Schyns, 2008). Trust was most often measured with the
Podsakoff et al. (1990) instrument. Measures of job satisfaction included
one-dimensional scales with items focusing only on the job (e.g., general
job satisfaction items from the revised job descriptive survey; Hackman
& Oldham, 1980) to multidimensional instruments designed to assess
various aspects related to job satisfaction (e.g., the satisfaction with the
work itself scale of the job descriptive index; Smith et al., 1987) and
80 PERSONNEL PSYCHOLOGY

the Minnesota Satisfaction Questionnaire (Weiss, Dawis, England, &

Lofquist, 1967). Commitment generally referred to commitment to the
organization and, most commonly, affective organizational commitment
measured using the Meyer, Allen, and Smith (1993) scale. Motivation
included a number of different variables, the most common referred to
the employee’s intrinsic motivation for their job (e.g., Amabile, 1985).
Finally, empowerment was measured using Spreitzer’s (1995) scale or
facets thereof (e.g., Basu & Green, 1997; Ozer, 2008).

Meta-Analytic Techniques

The meta-analysis relied on the widely-used Hunter and Schmidt

(1990, 2004) approach: A random effects model that accounts for sam-
pling bias and measurement error. Accordingly, we calculated a sample-
weighted mean correlation (r), and a sample-weighted mean correlation
corrected individually for unreliability in both criterion and predictor
variables, hereafter referred to as the corrected population correlation (ρ).
Missing artifact values (i.e., reliability of either predictor or criterion) were
estimated by inserting the mean value across the studies in which informa-
tion was given, as recommended by Hunter and Schmidt (2004). Objective
performance data were not corrected for unreliability, because researchers
frequently argue that measures based on objective performance data are
unbiased (Riketta, 2005) and because no procedure is currently available
to correct for unreliability of such measures.
In addition, we report the 90% confidence intervals (90% CI) of
the sample-weighted mean correlation and the 80% credibility intervals
(80% CV) of the corrected population correlation. Confidence intervals
estimate variability in the sample-weighted mean correlation that is due to
sampling error; credibility intervals estimate variability in the individual
correlations across studies that is due to moderator variables (Whitener,
1990). If the 90% confidence interval around a sample-weighted mean
correlation does not include zero, we can be 95% confident that the
sample-weighted mean correlation is different from zero. Moreover,
confidence intervals can be used to test whether two estimates differ from
each other; two estimates are considered different when their confidence
intervals are nonoverlapping. As some authors question the use of
significance testing in meta-analyses, we also interpret the effect size of
the corrected population correlation using the rule of thumb for small,
medium, and large effect sizes (.10, .30, and .50) as suggested by Cohen
(1992). If the 80% credibility interval of the corrected population corre-
lation is large and includes zero, this indicates that there is considerable
variation across studies, and moderators are likely operating.
 ROBIN MARTIN ET AL. 81

To further corroborate that moderators were present, we assessed

whether sampling error and error of measurement accounted for more
than 75% of the variance between studies in the primary estimates (Hunter
& Schmidt, 1990); accordingly, we report the percentage of variance ac-
counted for in the corrected population correlation by sampling and mea-
surement error (% VE). Moderators are assumed to be operating when
sampling and measurement error account for less than 75% of the vari-
ance. Categorical moderators were computed using Hunter and Schmidt’s
(1990, 2004) subgroup analyses techniques by conducting separate meta-
analyses at each of the specified moderator level. To examine whether
there are significant differences between the mean-corrected correlations
of sublevels of the hypothesized moderator variable, we compared their
confidence intervals as discussed.
For the mediation and causal analyses, we applied the respective mod-
els discussed in the hypothesis section to the matrix of corrected mean
correlations. To minimize common source variance and common method
bias in the mediation analysis, the correlations between LMX and the
performance outcomes, and between the mediators and the performance
outcomes, were based on noncommon source estimates (Podsakoff et al.,
2003). Following recommendations by Hom, Caranikas-Walker, Prussia,
and Griffeth (1992; cf. Viswesvaran & Ones, 1995), we tested the me-
diation and causal models using structural equation modeling and the
maximum likelihood estimate method in Mplus (Muthén & Muthén,
1998–2006). Given that sample sizes varied across the various cells of
the inputted correlation matrices, we used the harmonic mean of each
subsample to calculate model estimates and standard errors (Viswesvaran
& Ones, 1995). Using the harmonic mean results in more conservative
estimates, as less weight is given to large samples.

Results

Main Effects of the LMX and Performance Relationship

There is a positive relationship between LMX with task and citi-

zenship performance and a negative relationship with counterproduc-
tive performance (supporting Hypotheses 1 to 3). As can be seen in
Tables 1–3, LMX (overall) had a moderately strong positive effect on
task performance (ρ = .30, 90% CI [.25, .28]), a moderately strong pos-
itive effect on citizenship performance (ρ = .34, 90% CI [.27, .32]), and
a moderately strong negative effect on counterproductive performance
(ρ = –.24, 90% CI [–28, –.16]). Because the relationship with objec-
tive performance has only been reported in one previous meta-analysis
(Gerstner & Day, 1997), we specifically report this relationship. In our
82 PERSONNEL PSYCHOLOGY

TABLE 1
Meta-Analytic Results for the Relationship Between LMX and Task Performance

90% CI 80% CV

Variable k N r Lower Upper ρ SDρ %VE Lower Upper

LMX follower- or leader-rated
LMX overall 146 32,670 .27 .25 .28 .30 .13 22.38 .13 .47
LMX-7 86 20,766 .27 .25 .29 .31 .13 21.21 .14 .48
LMX-MDM 27 6,065 .24 .21 .28 .28 .12 27.29 .13 .43
LMX Other 37 7,168 .26 .23 .30 .30 .14 24.64 .13 .48
Noncommon source
LMX overall 121 26,574 .25 .23 .26 .28 .10 32.36 .15 .41
LMX-7 75 17,838 .25 .23 .27 .28 .10 32.07 .15 .41
LMX-MDM 25 5,671 .24 .21 .28 .27 .12 27.57 .13 .42
LMX Other 27 4,491 .22 .19 .25 .24 .09 45.28 .13 .36
Objective performance
LMX overall 20 4,398 .22 .18 .26 .24 .11 29.73 .10 .38
LMX-7 14 3,742 .24 .19 .28 .26 .10 29.06 .13 .39
LMX-MDM 0
LMX Other 6 656 .11 .03 .20 .12 .08 61.92 .02 .23
External performance ratings
LMX overall 109 23,877 .25 .23 .27 .28 .10 34.65 .16 109
LMX-7 67 15,721 .25 .23 .27 .28 .10 33.26 .16 67
LMX-MDM 25 5,671 .24 .21 .28 .27 .12 27.57 .13 25
LMX Other 24 4,006 .24 .21 .27 .27 .07 57.74 .18 24
Common source
LMX overall 43 9,016 .35 .31 .40 .42 .20 10.94 .16 .68
LMX-7 26 5,433 .38 .32 .44 .44 .22 9.02 .16 .73
LMX-MDM 3 816 .27 .15 .39 .33 .15 18.58 .14 .52
LMX Other 15 3,189 .34 .27 .41 .41 .17 15.96 .19 .63
LMX follower-rated
LMX overall 134 31,140 .25 .24 .27 .29 .11 28.70 .15 .43
LMX-7 80 19,977 .25 .23 .27 .28 .11 28.14 .14 .42
LMX-MDM 27 6,012 .24 .20 .28 .27 .12 27.81 .12 .42
LMX Other 35 6,925 .26 .23 .29 .29 .10 35.05 .16 .43
Noncommon source
LMX overalla 118 26,294 .25 .23 .27 .28 .10 33.03 .15 .41
LMX-7 72 17,173 .25 .23 .27 .28 .10 3.88 .15 .41
LMX-MDM 25 5,671 .24 .21 .28 .27 .11 28.50 .13 .42
LMX Other 22 3,855 .24 .21 .27 .27 .08 52.13 .17 .37
Objective performance
LMX overall 17 4,004 .23 .18 .27 .25 .11 28.81 .11 .39
LMX-7 13 3,617 .24 .19 .29 .26 .11 26.20 .13 .40
LMX-MDM 0
LMX Other 4 387 .11 .03 .19 .13 .00 100.00 .13 .13
External performance ratings
LMX overall 108 23,672 .26 .24 .27 .29 .09 37.12 .17 .41
LMX-7 66 15,507 .25 .23 .27 .29 .09 38.09 .18 .40
LMX-MDM 25 5,671 .24 .21 .28 .27 .11 28.50 .13 .42
LMX Other 21 3,820 .24 .21 .27 .27 .08 51.34 .17 .37
continued
 ROBIN MARTIN ET AL. 83

TABLE 1 (continued)

90% CI 80% CV

Variable k N r Lower Upper ρ SDρ %VE Lower Upper

Leader-rated performance
LMX overall 107 23,998 .25 .24 .27 .29 .09 37.70 .17 .41
LMX-7 66 15,507 .25 .23 .27 .29 .09 38.16 .18 .40
LMX-MDM 25 5,671 .24 .20 .28 .27 .11 28.67 .13 .42
LMX Other 21 3,820 .24 .21 .27 .27 .08 51.34 .17 .37
Peer-rated performance
LMX overall 2 313 .36 .32 .39 .38 .00 100.00 .38 .38
LMX-7 1 163 .33 .36
LMX-MDM 1 150 .39 .40
LMX Other 0
Common source
LMX overall 22 5,763 .26 .21 .31 .31 .16 16.12 .11 .51
LMX-7 8 2,436 .23 .13 .32 .26 .18 11.25 .04 .49
LMX-MDM 2 394 .18 .04 .32 .21 .10 40.36 .09 .34
LMX Other 12 2,933 .30 .25 .36 .37 .11 27.51 .22 .51
LMX leader-rated
LMX overall 27 4,118 .45 .39 .50 .52 .20 12.34 .26 .78
LMX-7 20 3,343 .46 .40 .52 .54 .16 16.85 .34 .75
LMX-MDM 1 422 .36 .48
LMX Other 7 775 .41 .24 .58 .45 .30 7.79 .06 .83
Noncommon source
LMX overall 6 722 .13 .07 .19 .14 .04 84.07 .08 .20
LMX-7 3 387 .14 .09 .19 .16 .00 100.00 .16 .16
LMX-MDM 3 335 .11 .01 .22 .12 .08 62.37 .01 .22
LMX Other 0
Objective performance
LMX overall 4 477 .08 .03 .14 .09 .00 100.00 .09 .09
LMX-7 2 224 .10 .06 .13 .11 .00 100.00 .11 .11
LMX-MDM 0
LMX Other 2 253 .07 -.02 .17 .07 .00 100.00 .07 .07
External performance ratings
LMX overall 3 300 .22 .17 .26 .26 .00 100.00 .26 .26
LMX-7 1 163 .20 .24
LMX-MDM 0
LMX Other 2 137 .24 .16 .32 .27 .00 100.00 .27 .27
Follower-rated performance
LMX overall 3 300 .22 .17 .26 .26 .00 100.00 .26 .26
LMX-7 1 163 .20 .24
LMX-MDM 0
LMX Other 2 137 .24 .16 .32 .27 .00 100.00 .27 .27
Peer-rated performance
LMX overall 1 163 .20 .24
LMX-7 1 163 .20 .24
LMX-MDM 0
LMX Other 0
continued
84 PERSONNEL PSYCHOLOGY

TABLE 1 (continued)

90% CI 80% CV

Variable k N r Lower Upper ρ SDρ %VE Lower Upper

Common source
LMX overall 27 3,971 .49 .44 .54 .58 .15 19.02 .38 .77
LMX-7 21 3,394 .49 .44 .54 .57 .15 19.21 .39 .76
LMX-MDM 1 422 .36 .46
LMX Other 6 577 .56 .45 .67 .63 .16 19.18 .42 .84

Note. Results are corrected for criterion and predictor unreliability. k = number of correla-
tions; N = number of respondents; r = sample-weighted mean correlation; ρ = corrected
population correlation; SDρ = standard deviation of the corrected population correlation;
% VE = percentage of variance attributed to sampling error in corrected population corre-
lation; 90% CI = 90% confidence interval around the sample-weighted mean correlation;
80% CV = 80% credibility interval around the corrected population correlation.
a
Corrected population correlation served as input for mediation analyses.

meta-analysis there was a positive relationship between LMX and ob-

jective task performance (20 samples, ρ = .24, 90% CI [.18, .26]) and
a negative relationship between LMX and objective counterproductive
performance (6 samples, ρ = –.11, 90% CI [–13, –.07]), though the
small number of samples for the last finding should be noted. Due to the
nature of citizenship performance, there were no studies with objective
measures.

Mediators of the LMX and Performance Relationship

To test for mediation, we first derived the meta-analytic correlations for

the relationship between LMX (follower-rated, noncommon source, LMX
overall) and the mediating variables; role clarity, trust, job satisfaction,
organizational commitment, motivation, and empowerment (all follower-
rated). The results are displayed in Table 4; all the effects were significant
(i.e., none of the 90% CIs included 0), positive, and strong (ρ = .48
for role clarity; ρ = .65 for trust; ρ = .61 for job satisfaction; ρ =
.49 for organizational commitment; ρ = .31 for motivation; ρ = .34 for
empowerment).
Next, we meta-analyzed the effects of these mediating variables (all
follower-rated) on all measures of performance (all from noncommon
sources). The results are displayed in Table 5. All correlations were
significant and positive. For task performance, the effects ranged from
medium to small (ρ = .12 for role clarity; ρ = .24 for trust; ρ = .20
for job satisfaction; ρ = .15 for organizational commitment; ρ = .21 for
motivation; ρ = .23 for empowerment). For citizenship performance, the
effects were stronger and ranged from large to medium (ρ = .19 for role
 ROBIN MARTIN ET AL. 85

TABLE 2
Meta-Analytic Results for the Relationship Between LMX and Citizenship
Performance

90% CI 80% CV

Variable k N r Lower Upper ρ SDρ %VE Lower Upper

LMX follower- or leader-rated
LMX overall 97 23,039 .29 .27 .32 .34 .15 17.94 .15 .53
LMX-7 62 14,800 .28 .25 .30 .32 .15 18.55 .13 .51
LMX-MDM 25 5,332 .28 .24 .32 .32 .11 30.94 .18 .45
LMX Other 14 3,913 .36 .29 .43 .42 .16 12.62 .21 .62
Noncommon source
LMX overall 74 16,186 .27 .25 .30 .31 .12 25.63 .16 .47
LMX-7 49 10,902 .27 .24 .30 .31 .13 24.19 .15 .47
LMX-MDM 21 4,568 .27 .23 .32 .30 .12 25.33 .15 .46
LMX Other 9 1,853 .25 .20 .29 .29 .06 62.08 .21 .36
Common source
LMX overall 32 8,977 .33 .28 .39 .39 .19 9.56 .15 .64
LMX-7 21 5,764 .31 .24 .38 .36 .20 9.02 .10 .62
LMX-MDM 6 1,417 .30 .24 .37 .35 .09 34.76 .23 .47
LMX Other 5 2,060 .44 .33 .56 .52 .14 9.22 .33 .70
LMX follower-rated
LMX overall 94 22,362 .29 .26 .31 .33 .14 18.41 .14 .51
LMX-7 59 14,123 .27 .24 .29 .31 .14 19.63 .12 .49
LMX-MDM 25 5,332 .27 .23 .31 .31 .11 30.36 .17 .45
LMX Other 14 3,913 .35 .28 .42 .41 .16 12.31 .20 .62
Noncommon source
LMX overalla 72 15,365 .27 .25 .30 .31 .13 24.91 .15 .48
LMX-7 46 9,950 .27 .24 .30 .31 .13 23.43 .14 .48
LMX-MDM 21 4,568 .27 .23 .32 .30 .13 24.10 .14 .47
LMX Other 9 1,853 .25 .20 .29 .29 .06 62.08 .21 .36
Common source
LMX overall 25 7,611 .30 .24 .35 .35 .18 10.71 .13 .58
LMX-7 15 4,556 .24 .18 .31 .29 .16 14.19 .09 .49
LMX-MDM 5 995 .25 .20 .30 .29 .02 95.37 .27 .31
LMX Other 5 2,060 .44 .33 .56 .52 .14 9.22 .33 .70
LMX leader-rated
LMX overall 10 2,318 .42 .34 .50 .50 .16 14.57 .30 .70
LMX-7 9 2,160 .43 .34 .52 .51 .16 13.00 .31 .72
LMX-MDM 1 422 .44 .50
LMX Other 1 158 .56 .74
Noncommon source
LMX overall 4 1,374 .30 .21 .40 .37 .11 23.15 .23 .50
LMX-7 3 952 .24 .17 .32 .30 .07 44.74 .21 .39
LMX MDM 1 422 .44 .50
LMX Other 0 continued
86 PERSONNEL PSYCHOLOGY

TABLE 2 (continued)

90% CI 80% CV

Variable k N r Lower Upper ρ SDρ %VE Lower Upper

Common source
LMX overall 8 1,497 .52 .46 .57 .60 .09 31.46 .48 .72
LMX-7 7 1,339 .51 .44 .58 .59 .09 32.30 .48 .71
LMX-MDM 0
LMX Other 1 158 .56 .74

Note. Results are corrected for criterion and predictor unreliability. k = number of correla-
tions; N = number of respondents; r = sample-weighted mean correlation; ρ = corrected
population correlation; SDρ = standard deviation of the corrected population correlation;
% VE = percentage of variance attributed to sampling error in corrected population corre-
lation; 90% CI = 90% confidence interval around the sample-weighted mean correlation;
80% CV = 80% credibility interval around the corrected population correlation.
a
Corrected population correlation served as input for mediation analyses.

clarity; ρ = .46 for trust; ρ = .27 for job satisfaction; ρ = .24 for organi-
zational commitment; ρ = .29 for motivation; ρ = .18 for empowerment).
Only five studies were available for counterproductive performance; the
effects were weak (ρ = .09 for job satisfaction; ρ = .04 for organizational
commitment; ρ = –.07 for motivation) and, due to the small number of
available studies, were inconclusive.
To enhance the validity of our results, we only included those media-
tors in our analyses for which we obtained at least three studies for each
link of the mediation sequence (cf. Harrison et al., 2006). Due to the small
number of available primary studies, this left us with role clarity, trust, job
satisfaction, organizational commitment, motivation, and empowerment
as mediators of the relationships between LMX and task and citizen-
ship performance. The results of our mediation analyses are displayed in
Table 6.
As can be seen in Table 6, and in support of Hypothesis 4, the re-
sults suggest that trust, job satisfaction, motivation, and empowerment
mediate the effects between LMX (follower-rated) and task performance
(externally-rated or based on objective measures); organizational commit-
ment and role clarity did not mediate this relationship. The mediator that
explained most of the variance in task performance was trust (25.0%), fol-
lowed by empowerment (17.9%), motivation (14.3%), and job satisfaction
(10.7%).
The effects of LMX (follower-rated) on citizenship performance
(externally-rated) are accounted for by trust, job satisfaction, organiza-
tional commitment, motivation and empowerment (see Table 6), which
also supports Hypothesis 4. Role clarity did not mediate these effects.
Trust appears as the mediator with the highest predictive validity; trust
 ROBIN MARTIN ET AL. 87

TABLE 3
Meta-Analytic Results for the Relationship Between LMX and
Counterproductive Performance

90% CI 80% CV

Variable k N r Lower Upper ρ SDρ %VE Lower Upper

LMX follower- or leader-rated
LMX overall 19 6,342 −.22 −.28 −.16 −.24 .18 9.53 −.48 −.01
LMX-7 14 5,207 −.21 −.29 −.14 −.24 .19 8.01 −.48 .00
LMX-MDM 3 848 −.32 −.33 −.30 −.37 .00 100.00 −.37 −.37
LMX Other 2 287 −.04 −.14 .06 −.04 .03 89.16 −.08 .00
Noncommon source
LMX overall 13 4,308 −.13 −.18 −.08 −.14 .10 25.49 −.27 −.01
LMX-7 10 3,899 −.13 −.18 −.08 −.15 .10 24.31 −.27 −.02
LMX-MDM 2 246 −.34 −.35 −.33 −.37 .00 100.00 −.37 −.37
LMX Other 2 287 −.04 −.14 .06 −.04 .03 89.16 −.08 .00
Objective performance
LMX overall 6 3,122 −.10 −.13 −.07 −.11 .02 88.67 −.13 −.08
LMX-7 4 2,835 −.10 −.13 −.08 −.11 .00 100.00 −.11 −.11
LMX-MDM 0
LMX Other 2 287 −.04 −.14 .06 −.04 .03 89.16 −.08 .00
External performance ratings
LMX overall 7 1,186 −.23 −.34 −.12 −.26 .17 18.23 −.48 −.04
LMX-7 5 940 −.20 −.34 −.06 −.23 .19 14.78 −.47 .01
LMX-MDM 2 246 −.34 −.35 −.33 −.37 .00 100.00 −.37 −.37
LMX Other 0
Common source
LMX overall 9 3,188 −.33 −.42 −.24 −.38 .19 7.97 −.62 −.14
LMX-7 8 2,586 −.33 −.44 −.22 −.38 .21 7.06 −.65 −.12
LMX-MDM 1 602 −.31 −.36
LMX Other 0
LMX follower-rated
LMX overall 18 6,230 −.22 −.28 −.16 −.25 .18 9.50 −.47 −.02
LMX-7 13 5,095 −.22 −.29 −.14 −.24 .18 7.88 −.47 .00
LMX-MDM 3 848 −.32 −.33 −.30 −.37 .00 100.00 −.37 −.37
LMX Other 2 287 −.04 −.14 .06 −.04 .03 89.16 −.08 .00
Noncommon source
LMX overall 11 3,778 −.14 −.19 −.08 −.15 .11 22.91 −.29 −.02
LMX-7 7 3,245 −.13 −.19 −.07 −.15 .09 21.83 −.27 −.02
LMX-MDM 2 246 −.34 −.35 −.33 −.37 .00 100.00 −.37 −.37
LMX Other 2 287 −.04 −.14 .06 −.04 .03 89.16 −.08 .00
Objective performance
LMX overall 6 3,122 −.10 −.13 −.07 −.11 .02 88.67 −.13 −.08
LMX-7 4 2,835 −.10 −.13 −.08 −.11 .00 100.00 −.11 −.11
LMX-MDM 0
LMX Other 2 287 −.04 −.14 .06 −.04 .03 89.16 −.08 .00
External performance ratings
LMX overall 5 656 −.34 −.42 −.25 −.38 .08 54.23 −.48 −.28
LMX-7 3 410 −.34 −.47 −.20 −.38 .12 32.76 −.53 −.22
continued
88 PERSONNEL PSYCHOLOGY

TABLE 3 (continued)

90% CI 80% CV

Variable k N r Lower Upper ρ SDρ %VE Lower Upper

LMX-MDM 2 246 −.34 −.35 −.33 −.37 .00 100.00 −.37 −.37
LMX Other 0
Common source
LMX overall 8 3,038 −.31 −.41 −.22 −.36 .18 8.18 −.59 −.13
LMX-7 7 2,436 −.32 −.43 −.20 −.36 .20 7.15 −.62 −.11
LMX-MDM 1 602 −.31 −.36
LMX Other 0
LMX leader-rated
LMX overall 3 680 −.19 −.40 .02 −.22 .24 8.42 −.52 .09
LMX-7 3 680 −.19 −.40 .02 −.22 .24 8.42 −.52 .09
LMX-MDM 0
LMX Other 0
Noncommon source
LMX overall 2 530 −.08 −.13 −.02 −.08 .00 100.00 −.08 −.08
LMX-7 2 530 −.08 −.13 −.02 −.08 .00 100.00 −.08 −.08
LMX-MDM 0
LMX Other 0
Objective performance
LMX overall 0
LMX-7 0
LMX-MDM 0
LMX Other 0
External performance ratings
LMX overall 2 530 −.08 −.13 −.02 −.08 .00 100.00 −.08 −.08
LMX-7 2 530 −.08 −.13 −.02 −.08 .00 100.00 −.08 −.08
LMX-MDM 0
LMX Other 0
Common source
LMX overall 1 150 −.60 −.67
LMX-7 1 150 −.60 −.67
LMX-MDM 0
LMX Other 0

Note. Results are corrected for criterion and predictor unreliability. k = number of correla-
tions; N = number of respondents; r = sample-weighted mean correlation; ρ = corrected
population correlation; SDρ = standard deviation of the corrected population correlation;
% VE = percentage of variance attributed to sampling error in corrected population corre-
lation; 90% CI = 90% confidence interval around the sample-weighted mean correlation;
80% CV = 80% credibility interval around the corrected population correlation.

accounted for 93.6% of the variance in the direct effect of LMX on citi-
zenship performance suggesting full mediation. Job satisfaction explained
25.8% of the variance; motivation explained 22.6%; organizational com-
mitment explained 19.4% of the variance in the direct effect; empower-
ment accounted for 9.7% of the variance. Due to the small numbers on
 ROBIN MARTIN ET AL. 89

TABLE 4
Meta-Analytic Results for the Relationship Between LMX (Follower-Rated) and
Mediators (Follower-Rated)

90% CI 80% CV

Variable k N r Lower Upper ρ SDρ %VE Lower Upper

Role clarity 8 1,636 .39 .31 .46 .48 .11 30.37 .34 .62
Trust 8 1,217 .55 .45 .66 .65 .19 10.88 .41 .89
Job satisfaction 48 13,493 .53 .49 .56 .61 .15 9.47 .41 .81
Commitment 42 10,332 .43 .40 .45 .49 .11 24.18 .35 .63
Motivation 8 3,447 .26 .23 .28 .31 .04 67.16 .26 .36
Empowerment 15 3,110 .29 .23 .35 .34 .13 24.04 .17 .51

Note. Results are corrected for criterion and predictor unreliability. k = number of correla-
tions; N = number of respondents; r = sample-weighted mean correlation; ρ = corrected
population correlation; SDρ = standard deviation of the corrected population correlation;
% VE = percentage of variance attributed to sampling error in corrected population corre-
lation; 90% CI = 90% confidence interval around the sample-weighted mean correlation;
80% CV = 80% credibility interval around the corrected population correlation.

the second stage of the mediation model (i.e., fewer than three available
studies), the findings for motivation should be interpreted with caution.
For the sake of completeness, we also tested whether job satisfaction,
organizational commitment, and motivation accounted for the relationship
between LMX and counterproductive performance; they did not. However,
in light of the small number of available primary studies, these findings
call for more corroborating evidence in the future.

Moderators of the LMX and Performance Relationship

The low amount of explained variation in, and the large credibil-
ity intervals around, the effects of LMX (overall) on task performance
(22.38%, 80% CV [.13, .47]), citizenship performance (17.94%, 80% CV
[.15, .53]), and counterproductive performance (9.53%, 80% CV [–.48,
–.01]) in Table 1–3 suggest that moderators are operating.
Common source and common method bias concerns whether the LMX
and performance measure were obtained from the same or different source
or method. Tables 1–3 suggest that the effects of LMX (overall) on the
performance outcomes tend to be lower when LMX and outcome measures
were obtained from a different source or were assessed with a different
method. When there was no bias, LMX had a weaker effect on task
performance (ρ = .28 vs. ρ = .42), citizenship performance (ρ = .31 vs.
ρ = .39), and counterproductive performance (ρ = –.14 vs. ρ = –.38).
The 90% CIs were nonoverlapping for task performance ([.23, .26]; [.31,
.40]) and counterproductive performance ([–.18, –.08]; [–.42, –.24]) but
90 PERSONNEL PSYCHOLOGY

TABLE 5
Meta-Analytic Results for the Relationship Between Mediators (Follower-Rated)
and Performance

90% CI 80% CV

Variable k N r Lower Upper ρ SDρ %VE Lower Upper

Task performance (externally-rated or based on objective measures)
Role clarity 6 816 .11 .06 .15 .12 .00 100.00 .12 .12
Trust 5 599 .20 .16 .24 .24 .00 100.00 .24 .24
Job satisfaction 27 5,825 .18 .14 .21 .20 .09 40.13 .08 .33
Commitment 30 7,029 .13 .10 .17 .15 .11 30.38 .01 .30
Motivation 5 835 .19 .12 .25 .21 .08 55.64 .11 .31
Empowerment 12 2,452 .21 .14 .27 .23 .13 26.05 .07 .40
Citizenship performance (externally)
Role clarity 3 589 .16 .01 .31 .19 .16 19.44 −.02 .40
Trust 3 563 .42 .32 .52 .46 .11 27.88 .33 .60
Job satisfaction 16 3,238 .23 .20 .27 .27 .06 59.94 .19 .35
Commitment 20 4,785 .21 .17 .25 .24 .09 37.34 .12 .35
Motivation 2 395 .25 .13 .37 .29 .11 33.33 .14 .43
Empowerment 5 1,010 .15 .03 .26 .18 .15 22.33 −.01 .38
Counterproductive performance (externally-rated or based on objective measures)
Role clarity 0
Trust 0
Job satisfaction 3 907 .08 .05 .11 .09 .00 100.00 .09 .09
Commitment 1 276 .04 .04
Motivation 1 1,924 −.06 −.07
Empowerment 0

Note. Results are corrected for criterion and predictor unreliability. k = number of correla-
tions; N = number of respondents; r = sample-weighted mean correlation; ρ = corrected
population correlation; SDρ = standard deviation of the corrected population correlation;
% VE = percentage of variance attributed to sampling error in corrected population corre-
lation; 90% CI = 90% confidence interval around the sample-weighted mean correlation;
80% CV = 80% credibility interval around the corrected population correlation.

not for citizenship performance ([.25, .30]; [.28, .39]). However, when
we increased the CI for citizenship performance to 80%, the two effects
appeared to be different ([.16, .47]; [.15, .64]). This suggests that the
effects of LMX (overall) on performance are indeed weaker for all three
performance outcomes under conditions in which measures were obtained
from a different source or assessed with a different method. For task
performance (see Table 1), it seems not to matter whether performance
is assessed with objective measures or with external ratings, the two
effects of LMX (overall) on these outcomes are similar and their 90% CIs
are overlapping (ρ = .24, [.18, .26] vs. ρ = .28, [.23, .27]). Similarly,
the effect of LMX on counterproductive performance (see Table 3) for
objective measures and external ratings is the same (ρ = –.11, [–.13,
–.07] vs. ρ = –.26, [–.34, –.12]).
 ROBIN MARTIN ET AL. 91

TABLE 6
Mediator Analyses for LMX (Follower-Rated) on Task and Citizenship
Performance

Mediator N a b ab c’ c %VE
Task performance (externally-rated or based on objective measures)
Role clarity 1,600 .48∗∗∗ −.02 −.01 .29∗∗∗ .28∗∗∗ 0
Trust 1,186 .65∗∗∗ .10∗∗ .07∗∗ .22∗∗∗ .28∗∗∗ 25.00
Job satisfaction 10,570 .61∗∗∗ .05∗∗∗ .03∗∗∗ .25∗∗∗ .28∗∗∗ 10.71
Commitment 10,827 .49∗∗∗ .02 .01 .27∗∗∗ .28∗∗∗ 3.57
Motivation 1,966 .31∗∗∗ .14∗∗∗ .04∗∗∗ .24∗∗∗ .28∗∗∗ 14.29
Empowerment 3,909 .34∗∗∗ .15∗∗∗ .05∗∗∗ .23∗∗∗ .28∗∗∗ 17.86
Citizenship performance (externally-rated)
Role clarity 1,264 .48∗∗∗ .05 .03 .28∗∗∗ .31∗∗∗ 9.68
Trust 1,127 .65∗∗∗ .45∗∗∗ .29∗∗∗ .02 .31∗∗∗ 93.55
Job satisfaction 6,696 .61∗∗∗ .13∗∗∗ .08∗∗∗ .23∗∗∗ .31∗∗∗ 25.81
Commitment 8,089 .49∗∗∗ .12∗∗∗ .06∗∗∗ .25∗∗∗ .31∗∗∗ 19.35
Motivationa 1,039 .31∗∗∗ .21∗∗∗ .07∗∗∗ .24 .31∗∗∗ 22.58
Empowerment 2,179 .34∗∗∗ .08∗∗∗ .03∗∗∗ .28 .31∗∗∗ 9.68

Note. N is harmonic mean. Standardized coefficients are presented. a = first stage mediation
effect; b = second stage mediation effect; ab = indirect effect; c’ = direct effect; c = overall
effect; %VE = variance explained in overall effect by indirect effect. Each mediator was
entered separately into the mediation analysis.
a
Second stage mediation effect based on only two studies.
∗
p < .05. ∗∗ p < .01. ∗∗∗ p < .001.

Type of measurement referred to the use of the LMX-7, LMX-MDM,

or LMX Other scales and, as can be seen in Tables 1 and 2, type of
measurement did not moderate the relationships between LMX and task
performance, or between LMX and citizenship performance; the respec-
tive 90% CIs were overlapping for LMX-7, LMX-MDM, and LMX Other
with task performance ([.25, .29]; [.21, .28]; [.23, .30]) and with citizenship
performance ([.25, .30]; [.24, .32]; [.29, .43]). The results for counterpro-
ductive performance (see Table 3) are inconclusive. Although 14 studies
looked at LMX-7, there are only three studies that looked at LMX-MDM
and two that looked at LMX Other. The effects for LMX-MDM seem to be
the most negative ([–.33, –.30]), followed by LMX-7 ([–.29, –.14]), and
there are no effects for LMX Other ([–.14, .06]). Overall, this suggests that
type of measurement does not moderate the LMX–performance relation-
ship, at least for task and citizenship performance. To further corroborate
these findings, we also meta-analyzed the intercorrelations between the
different types of measures; as can be seen in Table 7, the three measures
correlated very highly with each other (average ρ = .87). Although the
number of studies is too low to draw any firm conclusion, this provides
further support for the idea that the different measures are tapping into the
same overarching construct.
92 PERSONNEL PSYCHOLOGY

TABLE 7
Intercorrelations Between LMX Measures

90% CI 80% CV

Variable K N r Lower Upper ρ SDρ %VE Lower Upper

LMX-7 with 3 811 .71 .63 .80 .80 .08 13.76 .69 .90
LMX-MDM
LMX-7 with 1 195 .86 .95
LMX Other
LMX-MDM with 1 195 .79 .87
LMX Other
Note. Results are corrected for criterion and predictor unreliability. k = number of correla-
tions; N = number of respondents; r = sample-weighted mean correlation; ρ = corrected
population correlation; SDρ = standard deviation of the corrected population correlation;
% VE = percentage of variance attributed to sampling error in corrected population corre-
lation; 90% CI = 90% confidence interval around the sample-weighted mean correlation;
80% CV = 80% credibility interval around the corrected population correlation.

Type of rater refers to whether LMX was assessed by the leader or

follower. Whether the follower or leader rates, LMX has an effect on the
relationship between LMX (overall) with all performance measures. As
can be seen in Tables 1–3, the effects tend to be weaker when the follower
rates LMX as opposed to the leader, as indicated by their respective effect
sizes and 90% CIs (task performance: ρ = .29, [.24, .27] vs. ρ = .52,
[.39, .50]; citizenship performance: ρ = .33, [.26, .31] vs. ρ = .50, [.34,
.50]; counterproductive performance: ρ = –.25, [–.28, –.16] vs. ρ = –.22,
[–.40, .02]). However, due to the relatively smaller number of leader-rated
LMX studies versus follower-rated LMX studies and the overrepresenta-
tion of leader-rated LMX studies that are prone to common source and
common method bias, we caution not to read too much into these results.
Even so, when we take common method and common method bias into
account, the effects of follower-rated LMX and leader-rated LMX on the
three performance outcomes are different. For task performance, the ef-
fects of LMX (overall) rated by the follower has a stronger effect on task
performance than leader-rated LMX (overall) when there is no bias (non-
common source; ρ = .28, [.23, .27] vs. ρ = .14, [.07, .19]); in contrast
when there is bias (common source), the effects of follower-rated LMX
(overall) on task performance are much weaker than those for leader-
rated LMX (ρ = .31, [.21, .31] vs. ρ = .58, [.44, .54]). For citizenship
performance, the effects are similar when there is no bias (ρ = .31, [.25,
.30] vs. ρ = .37, [.21, .40]), however, when there is bias, follower-rated
LMX (overall) effects are again weaker (ρ = .35, [.24, .35] vs. ρ =
.60, [.46, .57]). For counterproductive performance, there are no differ-
ences between unbiased follower- and leader-rated LMX overall effects
 ROBIN MARTIN ET AL. 93

(ρ = –.15, [–.19, –.08] vs. ρ = –.08, [–.13, –.02]). Only one study looked at
biased leader-rated LMX effects on counterproductive performance. With
this caveat, there seems to be some indication that biased leader-rated ef-
fects on counterproductive performance also tend to be stronger (i.e., more
negative) than follower-rated LMX overall effects on counterproductive
performance (ρ = –.36, [–.41, –.22] vs. ρ = –.67).

Direction of Effects in the LMX and Performance Relationship

Whether LMX at Time 1 has a stronger effect on performance at Time 2

than performance at Time 1 on LMX at Time 2 could only be tested for task
performance due the availability of primary studies. The meta-analytic
results for the studies with a time gap between LMX (follower-rated)
and task performance, as well as the respective cross-sectional studies
are displayed in Table 8. LMX at Time 1 had a significant, positive, and
strong effect on LMX at Time 2 (ρ = .63), and a significant, positive,
and moderate effect on task performance at Time 2 (ρ = .31). Similarly,
task performance at Time 1 had a significant, positive, and strong effect
on task performance at Time 2 (ρ = .54); and a significant, positive, and
moderately strong effect on LMX at Time 2 (ρ = .21). The cross-sectional
correlations between LMX and task performance were also significant,
positive, and of medium size (ρ = .39).
Next, we subjected these meta-analytic correlations to a structural
equation model as displayed in Figure 1. As can be seen, LMX at Time
1 had a small significant positive effect on task performance at Time 2
(γ = .12, p < .001), whereas task performance at Time 1 did not have any
effect on LMX at Time 2 (γ = –.04, ns). This was further corroborated by
a Wald test that showed that both parameter estimates were significantly
different from each other (χ 2 (1) = 12.63, p < .001). Thus, LMX does
affect task performance, supporting Hypothesis 5, but not the other way
round.
Discussion

This paper reports a meta-analysis of the relationship between LMX

relationship quality and performance. In doing this, we report an up-to-
date review to reflect the rapid increase in research on LMX and perfor-
mance. For example, in terms of the number of samples examining LMX
and performance: Gerstner and Day (1997) reported 50 samples (42 per-
formance ratings, 8 objective), Ilies et al. (2007) reported 50 samples (all
OCB), Dulebohn et al. (2012) reported 135 samples (108 job performance,
27 OCB), and Rockstuhl et al. (2012) reported 200 samples (116 task
performance, 84 OCB). By contrast, this meta-analysis reports 262 sam-
ples (146 task, 97 citizenship, 19 counterproductive performance).
94 PERSONNEL PSYCHOLOGY

TABLE 8
Cross-Lagged Correlations Between LMX (Follower-Rated) and Task
Performance

90% CI 80% CV

Variable k N r Lower Upper ρ SDρ %VE Lower Upper

LMX (T1) – 13 3,469 .28 .23 .32 .31 .12 23.26 .16 .45
Perfor-
mance
(T2)
Performance 5 1,021 .20 .12 .27 .21 .08 48.13 .12 .31
(T1) –
LMX (T2)
Performance 4 610 .36 .25 .46 .39 .09 47.88 .27 .51
(T1/T2) –
LMX
(T1/T2)
LMX (T1) – 4 756 .58 .51 .65 .63 .07 37.26 .55 .72
LMX (T2)
Performance 5 897 .51 .43 .60 .54 .12 18.05 .38 .70
(T1) – Per-
formance
(T2)
Note. Results are corrected for criterion and predictor unreliability. k = number of correla-
tions; N = number of respondents; r = sample-weighted mean correlation; ρ = corrected
population correlation; SDρ = standard deviation of the corrected population correlation;
% VE = percentage of variance attributed to sampling error in corrected population corre-
lation; 90% CI = 90% confidence interval around the sample-weighted mean correlation;
80% CV = 80% credibility interval around the corrected population correlation; T =
measurement point.

Summary of Findings and Implications for Theory and Research

We identified four theoretical issues in the introduction (main effects,

mediating variables, moderating variables, and direction of effects), and
we summarize the findings in each of these areas with reference to the
implications of these findings for LMX theory and research.
Main effects. Guided by the integration of LMX and social exchange
theories and the multidimensional model of work performance (Rotundo
& Sackett, 2002), the first theoretical issue was to examine the main ef-
fects of LMX on a broader range of performance dimensions than had
been previously conducted. The meta-analyses supported Hypotheses 1
to 3. There was a significant positive relationship between LMX and task
performance with a corrected correlation of .30. This result compares
to the two most recent meta-analyses: Namely, Dulebohn et al. (2012)
and Rockstuhl et al. (2012) who reported corrected correlations between
 ROBIN MARTIN ET AL. 95

.65***
LMX (T1) LMX (T2)

-.04

.21***
.39***

.12***

Task Task
Performance (T1) .49***
Performance (T2)

Figure 1: Cross-Lagged Panel Correlation Analysis of LMX (Follower

Rated) and Task Performance.
Note. N = 859 (harmonic mean). Standardized maximum likelihood parameter estimates. T =
measurement point. *p < .05. **p < .01. ***p < .001.

LMX and “job performance” of .34, and .30 and .29 for individualistic
and collectivist countries, respectively. However, in both the Dulebohn
et al. (2012) and Rockstuhl et al. (2012) meta-analyses, only a global
finding is reported and the results are not given for different raters of
LMX or information on whether the correlations were from the same or
different sources or methods. This is important if one wants to examine
the relationship between LMX and objective measures of performance be-
cause these provide the most unbiased measure. Gerstner and Day (1997)
reported a corrected correlation of .11 based on eight samples, and led
them to conclude that “its practical meaningfulness is questionable” (p.
835). Because the recent meta-analyses did not differentiate between the
source of different performance measures (i.e., subjective vs. objective),
the relationship between LMX and objective performance has not yet
been clearly established. The present meta-analysis can answer this ques-
tion as it included 20 samples with objective performance (more than
twice Gerstner & Day, 1997) and found a corrected correlation of .24
(as opposed to the .11 effect size reported by Gerstner & Day, 1997).
Overall, the current meta-analysis confirms that LMX is positively asso-
ciated with task performance, even if it involves objective measures of
performance.
For citizenship performance, the corrected correlation was .34 and
is similar to the corrected correlations of .37, .39, and .35/.28 (individ-
ualistic/collectivistic cultures) found by Ilies et al. (2007), Dulebohn et
al. (2012), and Rockstuhl et al. (2012), respectively. Collectively, these
results show a positive relationship between LMX and citizenship per-
formance. Finally, the present meta-analysis, for the first time, examined
96 PERSONNEL PSYCHOLOGY

counterproductive performance. As explained in the Introduction, it is

important to include this dimension of performance as it is significantly
related to judgments of overall performance and an aspect of perfor-
mance about which organizations are becoming ever more concerned. As
expected, the corrected correlation was negative between LMX and coun-
terproductive performance (–.24), which was also evident with objective
measures of counterproductive behaviors (–.11), although we should be
cautious with the latter finding due to a small number of samples (6).
Our meta-analytic results give greater confidence to the veracity of the
LMX–performance relationship for three reasons. First, we triangulate
across different kinds of evidence to show that LMX is robustly asso-
ciated with performance, regardless of rating perspective, rating source,
and type of measure. Moreover, we find larger effects for objective perfor-
mance than did Gerstner and Day (1997: .24 as opposed to .11). Second,
we extended LMX theory by incorporating the multidimensional approach
to work performance (Rotundo & Sackett, 2002) and demonstrated that
the predictive power of LMX extends across all three dimensions of work
performance (task, citizenship, counterproductive performance). As such,
we provided the first meta-analytic test of the LMX–counterproductive
relationship and showed that high LMX reduces the incidence of neg-
ative work behaviors (i.e., counterproductive performance). Given that
primary studies have revealed mixed results including both negative (e.g.,
Townsend et al., 2000) and null effects (e.g., Chullen , Dunford, Anger-
meier, Boss, & Boss, 2010) of LMX on counterproductive performance,
our meta-analysis clarified the size and nature of this relationship. The
finding concerning counterproductive performance is particularly impor-
tant in extending LMX theory by showing that high LMX relationships
not only lead to more positive work behaviors (i.e., task and citizenship
performance) but also to fewer negative work behaviors (i.e., counter-
productive performance). Third, this is the first meta-analysis to show
that initial levels of LMX predict later task performance (and not vice
versa). It is important to note that our cross-lagged analyses (see Figure 1)
represent a particularly robust test of the LMX–performance relationship
because it disentangled the effects of LMX and task performance, con-
trolled for baseline levels of performance and LMX, and helped to rule
out the alternative explanations of reverse and reciprocal causality. Thus,
taken together, this meta-analysis triangulated across different sources,
methods, performance dimensions, and time lags to provide both novel
theoretical and empirical insights and the most compelling evidence to
date for the LMX–performance relationship.
Mediators. The second theoretical issue was to examine potential me-
diators between LMX and performance. Based on role, social exchange,
 ROBIN MARTIN ET AL. 97

and self-determination theories, we identified a number of potential me-

diating variables between LMX and performance.
The results showed that trust, motivation, empowerment, and job sat-
isfaction mediated the relationship between LMX and task performance
and between LMX and citizenship performance (supporting Hypothe-
sis 4). In addition, organizational commitment mediated the relationship
between LMX and citizenship performance. These findings are fully in
line with self-determination theory. One would expect on the basis of
self-determination theory that because high-quality relationships fulfill
people’s need for competence, autonomy, and relatedness, they should be
both motivating and satisfying. At first glance it appears that these findings
are more difficult to align with social exchange theory, as one would expect
that followers to reciprocate high-quality relationships with higher levels
of organizational commitment, mediating both the relationship between
LMX and task performance and with contextual performance. However,
another explanation for these findings may have to do with the elusive
attitude–performance relationship (cf. Harrison et al., 2006), in that fol-
lowers may pay back their obligation when it comes to task performance
with higher levels of motivation rather than with higher levels of organi-
zational commitment.
In contrast, our findings are not supportive of role theory. The results
showed that role clarity did not mediate the relationship between LMX
and either task or citizenship performance. According to role theory’s
account of LMX, one would expect better LMX relationships to be ones
in which the leader clearly designs and clarifies the role for the follower
(see Graen & Scandura, 1987). However, high LMX relationships might
be ones in which the leader gives the follower considerable discretion
over their work and opens up new work opportunities, and therefore the
follower’s job role remains unclear. Notwithstanding the plausibility of
these arguments, further research is needed to provide more evidence on
the role of different types of exchanges on the relationship between role
clarity and LMX.
Overall, trust in the leader accounted for the most variance in the
mediation models for both task performance and citizenship behavior.
This finding shows the importance of trust in the leader as an important
mechanism between LMX and performance. This is to be expected given
that LMX is conceptualized as a trust-building process (Bauer & Green,
1996; Graen & Cashman, 1975; Liden et al., 1993; Scandura & Pellegrini,
2008). However, some caution is warranted when interpreting this effect as
some LMX measures include items that are highly related to the concept
of trust. For example, in the development of the LMX-MDM, Liden
and Maslyn (1998) noted a strong overlap between the items in their
loyalty scale and trust. As noted earlier, future research would benefit by
98 PERSONNEL PSYCHOLOGY

examining the different dimensions of LMX on performance to determine

whether differential effects occur.
It is important to note, however, that our meta-analysis is not just
a summary (i.e., the average effect size) of previous empirical studies
that have tested mediation. In fact, there are surprisingly few empirical
studies that have directly tested mediational models of LMX, despite
the frequent calls in the literature (see Erdogan & Liden, 2002; Graen
& Uhl-Bien, 1995; Martin et al., 2010). As such, we have advanced
extant knowledge by examining the underlying process by which LMX
affects task and citizenship performance, including some mediators that
have not been tested before (e.g., trust, job satisfaction). Our findings
speak little to what accounts for the LMX–counterproductive performance
relationship because there were too few studies that would have allowed
us to test for mediators of this effect. This is clearly an area for future
research.
Moderators. The third theoretical issue of the meta-analysis concerns
potential moderators of the LMX and performance relationship. First,
confirming previous meta-analyses (e.g., Gerstner & Day, 1997), the cor-
relation between LMX and performance was stronger when both measures
were obtained from the same source or method than from different ones for
all three performance measures. However, although same source/method
data can potentially inflate correlations, this cannot explain the main
effects observed earlier. When measures were obtained from different
sources, the main effects were still statistically significant (task, 121 sam-
ples, ρ = .28; citizenship, 74 samples, ρ = .31; counterproductive, 13
samples, ρ = –.14).
Second, there was not a moderating effect of LMX measurement in-
strument. The effects of LMX-7, LMX-MDM, and LMX Other on all
three performance outcomes were of equal size. The LMX-7 and LMX-
MDM dominate the LMX literature despite some authors suggesting that
neither sufficiently captures the quality of the exchanges between leader
and follower (see Bernerth et al., 2007). Because LMX is seen as one
higher order factor (Graen & Uhl-Bien, 1995), studies tend to collapse
across the four dimensions of the LMX-MDM to give a single score
of relationship quality. Unfortunately, there were insufficient samples in
which the correlations are provided between the individual dimensions of
the LMX-MDM and performance. However, this is something that should
be encouraged in future studies to explore whether different dimensions
of LMX differentially predict performance indexes.
Third, the type of rater (leader vs. follower) has also been identified
as a potential moderator. This meta-analysis found the correlation be-
tween LMX and all performance measures was weaker when LMX was
measured by the follower than the leader. However, a different and more
 ROBIN MARTIN ET AL. 99

complex pattern emerged when we controlled for common source and

method bias. Across all outcomes, common source- and method-biased
effects were stronger for leader-rated LMX than for follower-rated LMX
(task: .58 vs. .31, citizenship: .60 vs. .35, counterproductive: –.67 vs. –
.36), whereas the differences were reversed (task: .14 vs. .28) or nullified
(citizenship: .37 vs. .31) when the effects were unbiased. One reason for
this might be due to leader ratings suffering from response inflation be-
cause items in LMX measures focus heavily on the leader and are thus
perceived by leaders as a self-rating of their own performance (see Sin,
Nahrgang, & Morgeson, 2009). When leaders rate LMX and follower
performance ratings in the same questionnaire, this might prime leaders
to also inflate follower performance ratings because they might perceive
the follower performance rating items as an assessment of their effective-
ness as a leader rather than follower performance. One implication of this
would be to vary the order of measurement (LMX vs. performance rating)
to determine whether this priming effect still occurs when the order is
reversed. Furthermore, the same-source rating suggests that for leaders,
high performance is almost conceptually equivalent with high levels of
LMX, whereas for followers, LMX and performance are conceptually
much more distinct. In contrast, the lower correlations for the unbiased
ratings suggest that leader ratings of their LMX relationship are much less
predictive of task and counterproductive performance than for citizenship
performance.
Direction of effects. The fourth theoretical issue was to examine di-
rectionality in the LMX–performance relationship. Due to a low number
of studies, we could only examine this issue in relation to task perfor-
mance. However, we found that LMX predicts task performance but not
vice versa (which supports Hypothesis 5). Although this direction of ef-
fect has been assumed in models of LMX (e.g., Cogliser et al., 2009;
Maslyn & Uhl-Bien, 2001; Uhl-Bien, 2006), we provide the first evidence
from a meta-analysis to support this crucial theoretical assumption. Our
results, however, found no evidence for reverse causality or reciprocal
causal effects. The lack of a temporal effect of performance on LMX may
be due to the small number of studies that have measured initial levels
of performance and later measures of LMX quality. Further research is
needed to establish the temporal characteristics of the LMX–performance
relationship, and in particular cross-lagged panel designs that help detect
changes in both LMX quality and performance over time. In one of the
few studies to measure the effects of performance on LMX over time,
Nahrgang et al. (2009) showed that performance/competence was an im-
portant determinant in the embryonic stages (the first few weeks) of the
LMX relationship. However, it is possible that, once initial impressions
100 PERSONNEL PSYCHOLOGY

are formed, subsequent changes in performance may have less influence

on LMX development.
Although, our meta-analysis is the first to go beyond concurrent effects
and test for temporal direction, we should exercise caution in reaching
causal conclusions based upon correlational data. Although we provide
a more rigorous test of causality, we cannot rule out alternative causal
explanations (e.g., third variables that may covary with both LMX and
performance). In addition, it is not clear whether the longitudinal samples
included in our meta-analysis incorporated the optimal time lag between
measurement points for detecting causal effects. Indeed, little is known
about how long it takes for the effects of LMX manifest in changes
in performance (and vice versa). Furthermore, panel designs are often
designed according to logistical constraints rather than based upon theo-
retical considerations of the optimal time lag for measurement (Riketta,
2008; Williams & Podsakoff, 1989). Despite these limitations, we found
initial evidence for the temporal effect of LMX on performance.
Due to a limited number of samples, we were only able to examine this
issue in relation to task performance. However, given the consistent pattern
of results across all three measures of performance, we would also predict
similar findings for citizenship and counterproductive performance.
Tests of causal direction have important implications for theory de-
velopment. This meta-analysis is a valuable starting point for teasing
apart causal effects and extending LMX theory by showing evidence for
prospective effects of LMX on performance but not reverse or reciprocal
causality (albeit based upon a relatively small sample of studies that tested
the prospective effect of performance on later LMX). Further research is
needed to more comprehensively examine the temporal characteristics
of the LMX relationship (e.g., how long it takes for LMX to influence
performance and vice versa, and how long these effects last), to examine
different types of models (e.g., moderated mediation; Tse, Ashkanasy,
& Dasborough, 2012), and more generally increase our understanding of
the process of LMX development. Regardless of the kind of performance
dimension measured, there is a need for more studies employing cross-
lagged panel and experimental designs, and in particular, multiple waves
that permit the examination of within-dyad change (i.e., trajectories) over
time (e.g., Nahrgang et al., 2009).

Implications for Practice

The link between LMX and performance also holds important impli-
cations for practice. At the individual level, research suggests that leaders
should try to develop high LMX relationships with all of their followers
(Graen & Uhl-Bien, 1995; Scandura, 1999) not only to enhance work
 ROBIN MARTIN ET AL. 101

performance (as shown in this meta-analysis) but to increase a wide range

of positive follower outcomes including; job satisfaction, health, and well-
being (for reviews see Anand et al., 2011; Martin et al., 2010). The pursuit
of high LMX with all followers is clearly desirable, but is it practical in
organizational settings? Developing high LMX takes time and requires
regular social exchanges and there may be several practical constraints
that might limit this occurring such as a large span of control, time con-
straints on the part of the leader, and the potential scarcity of required
material resources (Van Breukelen et al., 2006).
The range of organizational constraints (coupled with personal biases)
often mitigates against leaders developing high LMX with all followers but
to the development of different quality relationships (Liden et al., 2006),
and this can lead to poor work outcomes. For example, higher levels of
LMX differentiation (i.e., variability in LMX quality within work teams)
is associated with greater work group conflict (Hooper & Martin, 2008)
and lower individual task performance and OCB (Hu & Liden, 2013). In
practice, the ability of a leader to develop high LMX relationships with all
their followers requires not only good relationship-building skills but the
ability to manage several followers and to ensure that the leader is seen
as procedurally fair and unbiased in the treatment of all members of their
team (Hooper & Martin, 2008).
In some respects, LMX research has focused on the benefits of high
LMX without due consideration to the damaging effects of low LMX. Our
meta-analytic results suggest that the costs of low LMX may be greater
than is often recognized, in that neglected followers are likely to engage in
counterproductive and deviant behavior that undermines both the leader
and the organization (Jones, 2009). Moreover, organizations need to be
aware that requiring leaders to use punitive strategies to deal with such
counterproductive performance (see Atwater & Elkins, 2009) is likely to
lead to a downward spiral in LMX quality. Therefore, organizations should
look to use other corrective strategies to deal with counterproductive
performance. One way to address the underlying cause, however, would
be for organizations to remove the structural barriers to LMX development
such as reducing group sizes and increasing leader’s time and resources
to reduce the likelihood of LMX differentiation in work groups.
Findings from the meta-analysis also have practical implications
for human resource/personnel systems in organizations. First, the
LMX–performance effect has ramifications for employee and career de-
velopment systems in organizations. LMX quality is positively related to
perceived organizational career opportunities, career and development or-
ganizational support, career mentoring (Kraimer, Seibert, Wayne, Liden,
& Bravo, 2011), career satisfaction (Joo & Ready, 2012), speed of promo-
tion (e.g., Wakabayashi & Graen, 1984), and salary growth (Scandura &
102 PERSONNEL PSYCHOLOGY

Schriesheim, 1994). Moreover, leaders in organizations routinely operate

under a noncompensatory model of career development, in which follow-
ers must achieve both high LMX and high performance in order to progress
in their careers (e.g., Scandura, Graen, & Novak, 1986). Thus, LMX qual-
ity plays an important role in determining followers’ career progression.
The longitudinal results of this study, however, raise questions about the
fairness of this approach to employee and career development. For exam-
ple, our finding that performance failed to predict later LMX (after con-
trolling for initial LMX) suggests that once LMX is established it remains
unaffected by follower’s level of task performance (either better or worse).
Therefore, in the interests of procedural justice, organizations should con-
sider adopting a compensatory model of career and employee development
in which high performance can compensate for low LMX. In addition,
organizations need to provide other kinds of informal (e.g., career men-
toring) and formal (e.g., training workshops; career planning workshops)
employee development experiences, especially for low LMX followers to
compensate for the lack of developmental support and growth opportuni-
ties provided by their immediate leaders (Kraimer et al., 2011; Scandura &
Schriesheim, 1994).
Second, our results hold implications for performance management
systems in organizations. Prior research has sometimes called into ques-
tion the validity of leader performance evaluations of high LMX followers
because they can be unduly influenced by prior reputation or the closeness
of the LMX relationship (e.g., Duarte, Goodson, & Klich, 1993; Steiner,
1997). This is a major concern for organizations because supervisor-rated
performance is typically the primary source of data used by performance
management systems (Murphy & Cleveland, 1995). The results of our
meta-analysis are informative with respect to this important issue. On the
one hand, the fact that LMX quality predicted both leader-rated perfor-
mance and objective performance suggests that leader-rated performance
in organizations is to a certain extent accurate (see Funder, 1995), irrespec-
tive of LMX quality, and this provides some support for organizational
reliance on supervisor-rated performance data. On the other hand, the re-
lationship between LMX and objective performance is not as strong as it is
for supervisor-rated performance. There are many potential explanations
for this result including the possibility that leaders’ ratings encompass a
wider range of job performance criteria than objective measures (Smither
& London, 2009). However, it is also likely that rater errors and biases
might affect leaders’ ratings, at least in part due to LMX quality (Erdo-
gan, 2002; Martin et al., 2010; Steiner, 1997). Thus, organizations need
to ensure that leaders and HRM specialists are aware of the natural in-
clination to be more lenient toward high LMX followers. Other ways for
organizations to militate against appraisal bias and calibrate ratings is to
 ROBIN MARTIN ET AL. 103

require leaders to justify their ratings with others such as the leader’s man-
ager (Smither & London, 2009), provide leaders with frame-of-reference
training (Woehr & Huffcutt, 1994), and, where appropriate, integrate sub-
jective and objective measures of performance (Bommer, Johnson, Rich,
Podsakoff, & Mackenzie, 1995).
Finally, the meta-analysis has ramifications for leadership training
and development systems in organizations. LMX research has empha-
sized that leadership training that focuses on improving the quality of the
relationship between leader and follower is likely to have benefits for fol-
lower performance (Graen et al., 1982). The results of the meta-analysis
go beyond this and start to identify some of the mechanisms that might
account for why high LMX is beneficial and therefore what should be
addressed in leadership training programs. Leadership training that fo-
cuses on techniques to improve LMX through enhancing follower’s job
satisfaction, trust, work motivation, and empowerment is likely to result
in improvements in performance (see Korsgaard, Sapienza, & Schweiger,
2002). The meta-analysis gives a more differentiated picture of the LMX–
performance relationship than was hitherto known. For example, the re-
lationship between LMX and performance is not the same for followers
and leaders, for example, leader ratings of their LMX relationship are less
predictive of task and counterproductive performance than for citizen-
ship performance. Therefore, leadership training needs to acknowledge
that followers and leaders have different “lenses” in viewing what fac-
tors enhance performance. Leadership training could benefit from helping
leaders understand the multiple “lenses” (for the leader and followers)
that might operate in viewing work performance in helping to identify
potential biases in judgments, to understand the causal relationship be-
tween LMX and different dimensions of performance, and to better direct
leadership behaviors to enhance follower performance.

Conclusions

This meta-analysis was designed to address four main theoretical is-

sues, derived from LMX theory, with respect to the relationship between
LMX quality and performance. The main findings confirm that the effects
of LMX on various indicies of performance (positive with task and citizen-
ship performance and negative with counterproductive performance) are
of moderate to large size and also establish a moderate positive effect size
on objective performance. Also, a number of factors were found to mediate
the LMX–performance relationship, with trust in leader having the largest
effect. The findings concerning mediation supports social exchange and
self-determination (but not role) theories as well as theoretical models
of LMX that emphasize that LMX is a trust-building process. Finally,
104 PERSONNEL PSYCHOLOGY

evidence is found for a relationship between LMX and performance and

not for reverse or reciprocal causality. Based on these results, we encour-
age scholars to extend LMX theory by examining theory-guided mecha-
nisms that explain the link between LMX and the various dimensions of
performance (task, citizenship, and counterproductive performance) and
how this process develops over time.

REFERENCES

References marked with an asterisk* indicate studies included in the meta-analysis.

Amabile TM. (1985). Motivation and creativity: Effects of motivational orientation on

creative writers. Journal of Personality and Social Psychology, 48, 393–399.
DOI:org/10.1037//0022-3514.48.2.393
Anand S, Hu J, Liden RC, Vidyarthi PR. (2011). Leader-member exchange: Recent research
findings and prospects for the future. In Bryman A, Collinson D, Grint K, Jackson
B, Uhl-Bien M (Eds.). The Sage handbook of leadership (pp. 311–325). Thousand
Oakes, CA: Sage.
*Anand S, Vidyarthi PR, Liden RC, Rousseau DM. (2010). Good citizens in poor-quality
relationships: Idiosyncratic deals as a substitute for relationship quality. Academy
of Management Journal, 53, 970–988. DOI:org/10.5465/AMJ.2010.54533176
*Anderson SE, Williams LJ. (1996). Interpersonal, job, and individual factors related
to helping processes at work. Journal of Applied Psychology, 81, 282–296.
DOI:org/10.1037//0021-9010.81.3.282
*Aryee S, Chen ZX. (2006). Leader-member exchange in a Chinese context antecedents, the
mediating role of psychological empowerment and outcomes. Journal of Business
Research, 59, 793–801. DOI:org/10.1016/[Link].2005.03.003
*Asgari A, Silong AD, Ahmad A, Abu Samah B. (2008). The relationship between trans-
formational leadership behaviors, organizational justice, leader-member exchange,
perceived organizational support, trust in management and organizational citizen-
ship behaviors. European Journal of Scientific Research, 23, 227–242.
*Atwater L, Carmeli A. (2009). Leader-member exchange, feelings of energy, and in-
volvement in creative work. Leadership Quarterly, 20, 264–275. DOI:org/10.1016/
[Link].2007.07.009
Atwater L, Elkins T. (2009). Diagnosing, understanding and dealing with counterproductive
work behavior. In Smither JW, London M (Eds.). Performance management: Putting
research into practice. (pp. 359–410) San Fransisco, CA: Jossey-Bass.
Avolio BJ, Walumbwa FO, Weber TJ. (2009). Leadership: Current theories, re-
search, and future directions. Annual Review of Psychology, 60, 421–449.
DOI:org/10.1146/[Link].60.110707.163621
*Basu R. (1991). An empirical examination of LMX and transformational leadership as
predictors of innovative behavior. Ann Arbor, MI: University Microfilms Interna-
tional.
*Basu R, Green SG. (1997). Leader-member exchange and transformational lead-
ership: An empirical examination of innovative behaviors in leader-member
dyads. Journal of Applied Social Psychology, 27, 477–499. DOI:org/10.1111/
j.15591816.1997.tb00643.x
*Bauer TN, Erdogan B, Liden RC, Wayne SJ. (2006). A longitudinal study of the mod-
erating role of extraversion: Leader-member exchange, performance, and turnover
 ROBIN MARTIN ET AL. 105

during new executive development. Journal of Applied Psychology, 91, 298–310.

DOI:org/10.1037/0021-9010.91.2.298
*Bauer TN, Green SG. (1996). Development of leader-member exchange: A longitudinal
test. Academy of Management Journal, 39, 1538–1567. DOI:org/10.2307/257068
*Bernerth JB. (2006). Putting exchange back into leader-member exchange (LMX): An
empirical assessment of a social exchange (LMSX) scale and an investigation of
personality as an antecedent. Doctoral Dissertation, Auburn University Ann Arbor,
MI: ProQuest Information & Learning.
*Bernerth JB, Armenakis AA, Feild HS, Giles WF, Walker HJ. (2007). Is personality asso-
ciated with perceptions of LMX? An empirical study. Leadership & Organization
Development Journal, 28, 613–631. DOI:org/10.1108/01437730710823879
*Bettencourt LA. (2004). Change-oriented organizational citizenship behaviors: The direct
and moderating influence of goal orientation. Journal of Retailing, 80, 165–180.
DOI:org/10.1016/[Link].2003.12.001
*Bhal KT. (2006). LMX-citizenship behavior relationship: Justice as a mediator. Lead-
ership & Organization Development Journal, 27, 106–117. DOI:org/10.1108/
01437730610646615
*Blau G. (1988). An investigation of the apprenticeship organizational socialization
strategy. Journal of Vocational Behavior, 32, 176–195. DOI:org/10.1016/0001-
8791(88)90013-9
*Blau G, Moideenkutty U, Ingham K. (2010). Leader-member exchange as a sig-
nificant correlate of organizational and occupational sportsmanship behaviors
in a health services setting. Leadership in Health Services, 23, 219–232
DOI:org/10.1108/17511871011061046
Blau P. (1964). Exchange and power in social life. New York, NY: Wiley.
Bommer WH, Johnson JL, Rich GA, Podsakoff PM, MacKenzie SB. (1995). On the
interchangeability of objective and subjective measures of employee performance: A
meta-analysis. P ERSONNEL P SYCHOLOGY, 48, 587–605. DOI:org/10.1111/j.1744-
6570.1995.tb01772.x
*Brandes P, Dharwadkar R, Wheatley K. (2004). Social exchanges within orga-
nizations and work outcomes The importance of local and global rela-
tionships. Group & Organization Management, 29, 276–301. HYPER-
LINK “[Link] DOI:org/10.1177/
1059601103257405
*Brouer RL. (2007). The role of political skill in the leadership process-work outcomes
relationships. Doctoral Dissertation, Florida State University.
Burke CS, Sims DE, Lazzara EH, Salas E. (2007). Trust in leadership: A
multi-level review and integration. Leadership Quarterly, 18, 606–632.
DOI:org/10.1016/[Link].2007.09.006
*Burton J, Sablynski C, Sekiguchi T. (2008). Linking justice, performance, and citizen-
ship via leader-member exchange. Journal of Business and Psychology, 23, 51–61.
DOI:org/10.1007/s10869-008-9075-z
Campbell JP. (1990). Modeling the performance prediction problem in industrial and orga-
nizational psychology. In Dunnette MD, Hough LM (Eds.). Handbook of industrial
and organizational psychology (Vol. 1, 687–732). Palo Alto, CA: Consulting Psy-
chologists Press.
*Carter G. (2010). The relationships among social exchange, organizational citizenship,
and employee behavior. Unpublished Thesis. The George L. Graziadio School of
Business and Management Pepperdine University.
Cashman J. (1975). The nature of vertical dyadic linkages within managerial units. Doctoral
Dissertation, University of Illinois.
106 PERSONNEL PSYCHOLOGY

*Chang CHD, Johnson RE. (2010). Not all leader–member exchanges are created equal:
Importance of leader relational identity. Leadership Quarterly, 21, 796–808.
DOI:org/10.1016/[Link].2010.07.008
*Chau SL. (2007). Examining the emotional labor process: A moderated model of emo-
tional labor and its effects on job performance and turnover. Doctoral Dissertation,
University of Akron.
*Chen C, Chang W, Hu C. (2007). The relationship between leader-member exchange,
trust, supervisor support, and organizational citizenship behavior: A case study of
nurses. Paper presented at the Proceedings of the 13th Asia Pacific Management
Conference, Melbourne, Australia.
*Chen G, Kirkman BL, Kanfer R, Allen D, Rosen B. (2007). A multilevel study of leader-
ship, empowerment, and performance in teams. Journal of Applied Psychology, 92,
331–346. DOI:org/10.1037/0021-9010.92.2.331
*Chen G, Klimoski RJ. (2003). The impact of expectations on newcomer performance in
teams as mediated by work characteristics, social exchanges, and empowerment.
Academy of Management Journal, 46, 591–607. DOI:org/10.2307/30040651
*Chen Z, Lam W, Zhong JA. (2007). Leader-member exchange and member perfor-
mance: A new look at individual-level negative feedback-seeking behavior and
team-level empowerment climate. Journal of Applied Psychology, 92, 202–212.
DOI:org/10.1037/0021-9010.92.1.202
*Chen ZX, Tsui AS, Zhong L. (2008). Reactions to psychological contract
breach: A dual perspective. Journal of Organizational Behavior, 29, 527–548.
DOI:org/10.1002/job.481
*Chullen C, Dunford BB, Angermeier I, Boss R, Boss AD. (2010). Minimizing deviant
behavior in healthcare organizations: The effects of supportive leadership and job
design. Journal of Healthcare Management, 55, 381–398.
*Cogliser C, Schriesheim C, Scandura T, Gardner W. (2009). Balance in leader and
follower perceptions of leader-member exchange: Relationships with perfor-
mance and work attitudes. Leadership Quarterly, 20, 452–465. DOI:org/10.1016/
[Link].2009.03.010
Cohen J. (1992). A power primer. Psychological Bulletin, 112, 155–159.
DOI:org/10.1037//0033-2909.112.1.155
Colquitt JA, LePine JA, Piccolo RF, Zapata CP, Rich BL. (2012). Explaining the justice–
performance relationship: Trust as exchange deepener or trust as uncertainty re-
ducer? Journal of Applied Psychology, 97, 1–15. DOI:org/10.1037/a0025208
Colquitt JA, Scott BA, LePine JA. (2007). Trust, trustworthiness, and trust propensity: A
meta-analytic test of their unique relationships with risk taking and job performance.
Journal of Applied Psychology, 92, 909–927. DOI:org/10.1037/0021-9010.92.4.909
*Connell P. (2005). Transformational leadership, leader-member exchange (LMX), and
OCB: The role of motives. Graduate School Theses and Dissertations. University of
South Florida.
*Credo KR, Armenakis AA, Feild HS, Young RL. (2010) Organizational ethics,
leader-member exchange, and organizational support: Relationships with work-
place safety. Journal of Leadership & Organizational Studies, 17, 325–334.
DOI:org/10.1177/1548051810366712
Cropanzano R, Mitchell, MS. (2005). Social exchange theory: An interdisciplinary review.
Journal of Management, 31, 874–900. DOI:org/10.1177/0149206305279602
Dansereau F, Graen GB, Haga W. (1975). A vertical dyad linkage approach to leadership
in formal organizations. Organizational Behavior and Human Performance, 13,
46–78. DOI:org/10.1016/0030-5073(75)90005-7
 ROBIN MARTIN ET AL. 107

*DeConinck JB. (2010). The effect of organizational justice, perceived organizational sup-
port, and perceived supervisor support on marketing employees’ level of trust.
Journal of Business Research, 63, 1349–1355. DOI:org/10.1016/[Link].2010.
01.003
Deci EL, Ryan RM. (1985). Intrinsic motivation and self-determination in human behavior.
New York, NY: Plenum.
Deci EL, Ryan RM. (2000). The “what” and “why” of goal pursuits: Human needs
and the self-determination of behavior. Psychological Inquiry, 11, 227–268.
DOI:org/10.1207/S15327965PLI1104_01
De Dreu CK, Weingart LR. (2003). Task versus relationship conflict, team performance,
and team member satisfaction: A meta-analysis. Journal of Applied Psychology, 88,
741–749. DOI:org/10.1037/0021-9010.88.4.741
*Deluga RJ, Perry JT. (1994). The role of subordinate performance and ingratiation
in leader-member exchanges. Group & Organization Management, 19, 67–86.
DOI:org/10.1177/1059601194191004
*Dienesch RM. (1987). An empirical investigation of the relationship between quality of
leader-member exchange and subordinate performance and satisfaction. Doctoral
Dissertation, Georgia Institute of Technology.
Dirks KT, Ferrin DL. (2002). Trust in leadership: Meta-analytic findings and implica-
tions for organizational research. Journal of Applied Psychology, 87, 611–628.
DOI:org/10.1037/0021-9010.87.4.611
*Dobbins GH, Cardy RL, Platz-Vieno SJ. (1990). A contingency approach to ap-
praisal satisfaction: An initial investigation of the joint effects of organizational
variables and appraisal characteristics. Journal of Management, 16, 619–632.
DOI:org/10.1177/014920639001600307
*Doyle S. (2006). The role of organizational citizenship behavior in volunteer organiza-
tions. Doctoral Dissertation, St. Ambrose University.
Duarte NT, Goodson JR, Klich NR. (1993). How do I like thee? Let me
appraise the ways. Journal of Organizational Behavior, 14, 239–249.
DOI:org/10.1002/job.4030140304
*Duarte NT, Goodson JR, Klich NR. (1994). Effects of dyadic quality and dura-
tion on performance appraisal. Academy of Management Journal, 37, 499–521.
DOI:org/10.2307/256698
Dulebohn JH, Bommer WH, Liden RC, Brouer RL, Ferris GR. (2012). A meta-analysis of
antecedents and consequences of leader-member exchange. Journal of Management,
38, 1715–1759. DOI:10.1177/0149206311415280
*Dunegan KJ, Duchon D, Uhl-Bien M. (1992). Examining the link between
leader-member exchange and subordinate performance: The role of task an-
alyzability and variety as moderators. Journal of Management, 18, 59–76.
DOI:org/10.1177/014920639201800105
*Dunegan KJ, Uhl-Bien M, Duchon D. (2002). LMX and subordinate performance: The
moderating effects of task characteristics. Journal of Business and Psychology, 17,
275–285. DOI:10.1023/A:1019641700724
*Eisenberger R, Karagonlar G, Stinglhamber F, Neves P, Becker TE, Gonzalez-
Morales MG, Steiger-Mueller M. (2010). Leader–member exchange and affective
organizational commitment: The contribution of supervisor’s organizational embod-
iment. Journal of Applied Psychology, 95, 1085–1103. DOI:org/10.1037/a0020858
*El Akremi A, Vandenberghe C, Camerman J. (2010). The role of justice and social
exchange relationships in workplace deviance: Test of a mediated model. Human
Relations, 63, 1687–1717. DOI:10.1177/0018726710364163
108 PERSONNEL PSYCHOLOGY

*Elicker JD, Levy PE, Hall RJ. (2006). The role of leader-member exchange in
the performance appraisal process. Journal of Management, 32, 531–551.
DOI:org/10.1177/0149206306286622
Epitropaki O, Martin R. (2004). Implicit leadership theories in applied settings: Factor
structure, generalizability, and stability over time. Journal of Applied Psychology,
89, 293–310. DOI:org/10.1037/0021-9010.89.2.293
*Erdogan B. (2002). Leader-member exchange differentiation fairness: Evidence for a new
construct. Doctoral Dissertation, University of Illinois at Chicago.
*Erdogan B, Bauer TN. (2010). Differentiated leader–member exchanges: The buffer-
ing role of justice climate. Journal of Applied Psychology, 95, 1104–1120.
DOI:org/10.1037/a0020578
*Erdogan B, Enders J. (2007). Support from the top: Supervisors’ perceived organizational
support as a moderator of leader-member exchange to satisfaction and performance
relationships. Journal of Applied Psychology, 92, 321–330. DOI:org/10.1037/0021-
9010.92.2.321
Erdogan B, Liden RC. (2002). Social exchanges in the workplace: A review of recent
developments and future research directions in leader-member exchange theory.
In Neider LL, Schriescheim CA (Eds.). Leadership (pp. 65–114). Greenwich, CT:
Information Age Publishing.
*Fernandez CF, Vecchio RP. (1997). Situational leadership theory revisited: A test of an
across-jobs perspective. Leadership Quarterly, 8, 67–84. DOI:org/10.1016/S1048-
9843(97)90031-X
*Findley HM. (1996). Fairness in social exchange: The relationships of leader member ex-
change and performance appraisal fairness to organizational citizenship behavior.
Doctoral Dissertation, Auburn University.
Fishbein M, Ajzen I. (1975). Belief, attitude, intention and behavior: An introduction to
theory and research. Reading, MA: Addison-Wesley.
Funder DC. (1995). On the accuracy of personality judgment: A realistic approach. Psy-
chological Review, 102, 652–670. DOI:org/10.1037//0033-295X.102.4.652
*Furst SA, Cable DM. (2008). Employee resistance to organizational change: Managerial
influence tactics and leader-member exchange. Journal of Applied Psychology, 93,
453–462. DOI:org/10.1037/0021-9010.93.2.453
*Gaa SM. (2010). An exploratory analysis of the relationships between leadership, safety
climate, and organizational citizenship behavior within high-containment biosafety
laboratories. Unpublished Thesis. Capella University. UMI Number: 3412440.
*Gandolfo CJ. (2006). The role of Leader-member exchange theory and multi-rater feed-
back on evaluating the performance appraisal process. Doctoral Dissertation, South-
ern Illinois University Carbondale.
Gerstner CR, Day DV. (1997). Meta-analytic review of leader-member exchange the-
ory: Correlates and construct issues. Journal of Applied Psychology, 82, 827–843.
DOI:org/10.1037//0021-9010.82.6.827
*Gillis LR. (2009). A network perspective of multiple social exchange relationships. Doc-
toral Dissertation, University of Texas at Austin.
*Glibkowski BC. (2009). Negotiating Social Exchanges: The Mediating Role of Integra-
tive Negotiation in Social Exchanges. Unpublished Thesis. University of Illinois at
Chicago. UMI Number: 3394217.
*Golden T, Veiga J. (2008). The impact of superior-subordinate relationships on the com-
mitment, job satisfaction, and performance of virtual workers. Leadership Quarterly,
19, 77–88. DOI:org/10.1016/[Link].2007.12.009
 ROBIN MARTIN ET AL. 109

*Gooty J, Yammarino FJ. (2011). Dyads in organizational research: Conceptual is-

sues and multilevel analyses. Organizational Research Methods, 14, 456–483.
DOI:org/10.1177/1094428109358271
Graen GB. (1976). Role-making processes within complex organizations. In MD Dunnette
(Ed.): Handbook of industrial and organizational psychology (pp. 1201–1245).
Chicago, IL: Rand McNally.
Graen GB, Cashman JF. (1975). A role-making model of leadership in formal organisations:
A developmental approach. In Hunt JG, Larson LL (Eds.). Leadership frontiers
(pp. 143–165). Kent, OH: Kent State University Press.
Graen GB, Novak MA, Sommerkamp P. (1982). The effects of leader-member exchange and
job design on productivity and satisfaction: Testing a dual attachment model. Orga-
nizational Behavior and Human Performance, 30, 109–131. DOI:org/10.1016/0030-
5073(82)90236-7
Graen GB, Scandura T. (1987). Toward a psychology of dyadic organizing. In Staw B, Cum-
mings LL (Eds.). Research in organizational behavior (pp. 175–208). Greenwich,
CT: JAI Press.
Graen GB, Uhl-Bien M. (1995). Relationship-based approach to leadership: Develop-
ment of leader-member exchange (LMX) theory of leadership over 25 years: Ap-
plying a multi-level multi-domain perspective. Leadership Quarterly, 6, 219–247.
DOI:org/10.1016/1048-9843(95)90036-5
*Green SG, Blank W, Liden RC. (1983). Market and organizational influences on bank
employees’ work attitudes and behaviors. Journal of Applied Psychology, 68, 298–
306. DOI:org/10.1037//0021-9010.68.2.298
*Greenwood DM. (2000). Is anybody listening? Individual differences in reactions to
performance feedback. Doctoral Dissertation, George Mason University.
*Greguras GJ, Ford JM. (2006). An examination of the multidimensionality of supervisor
and subordinate perceptions of leader-member exchange. Journal of Occupational
and Organizational Psychology, 79, 433–465. DOI:org/10.1348/096317905×53859
Gully SM, Beaubien JM, Incalcaterra KA, Joshi A. (2002). A meta-analytic investigation
of the relationship between team efficacy, potency, and performance. Journal of
Applied Psychology, 87, 819–832. DOI:org/10.1037//0021-9010.87.5.819
Hackman JR, Oldham GR. (1980). Work redesign. Reading, MA: Addison-Wesley.
*Haines VY, Merrheim G, Roy M. (2001). Understanding reactions to safety incentives.
Journal of Safety Research, 32, 17–30. DOI:org/10.1016/S0022-4375(00)00051-7
*Harris KJ, Wheeler AR, Kacmar KM. (2009). Leader-member exchange and em-
powerment: Direct and interactive effects on job satisfaction, turnover inten-
tions, and performance. Leadership Quarterly, 20, 371–382. DOI:org/10.1016/
[Link].2009.03.006
Harrison DA, Newman DA, Roth PL. (2006). How important are job at-
titudes? Meta-analytic comparisons of integrative behavioral outcomes
and time sequences. Academy of Management Journal, 49, 305–325.
DOI:org/10.5465/AMJ.2006.20786077
Harter JK, Schmidt FL, Hayes TL. (2002). Business-unit-level relationship between
employee satisfaction, employee engagement, and business outcomes: A meta-
analysis. Journal of Applied Psychology, 87, 268–279. DOI:org/10.1037//0021-
9010.87.2.268
*Henderson DJ, Wayne SJ, Bommer WH, Shore LM, Tetrick LE. (2008). Leader-member
exchange, differentiation, and psychological contract fulfillment: A multilevel exam-
ination. Journal of Applied Psychology, 93, 1208–1219. DOI:org/10.1037/a0012678
110 PERSONNEL PSYCHOLOGY

*Hepperlen TM. (2003). Leader-member exchange (LMX) or fulfillment? The role of basic
psychological needs in LMX relationships. Doctoral Dissertation, University of
Nebraska.
*Hofmann DA, Morgeson FP. (1999). Safety-related behavior as a social exchange: The
role of perceived organizational support and leader-member Exchange. Journal of
Applied Psychology, 84, 286–296. DOI:org/10.1037//0021-9010.84.2.286
*Hofmann DA, Morgeson FP, Gerras SJ. (2003). Climate as a moderator of the relationship
between leader-member exchange and content specific citizenship: Safety climate as
an exemplar. Journal of Applied Psychology, 88, 170–178. DOI:org/10.1037/0021-
9010.88.1.170
Hom PW, Caranikas-Walker F, Prussia GE, Griffeth RW. (1992). A meta-analytical struc-
tural equations analysis of a model of employee turnover. Journal of Applied Psy-
chology, 77, 890–909. DOI:org/10.1037/0021-9010.77.6.890
Hooper DT, Martin R. (2008). Beyond personal leader–member exchange (LMX) qual-
ity: The effects of perceived LMX variability on employee reactions. Leadership
Quarterly, 19, 20–30. DOI:org/10.1016/[Link].2007.12.002
*Hoover ER. (2009). How personality and self-identity impact the effects of leader member
exchange on role stressors and organizational outcomes. Graduate School Theses
and Dissertations. University of South Florida.
*Howell JM, Hall-Merenda KE. (1999). The ties that bind: The impact of leader-
member exchange, transformational and transactional leadership, and distance on
predicting follower performance. Journal of Applied Psychology, 84, 680–694.
DOI:org/10.1037/0021-9010.84.5.680
*Hsiung H-H, Tsai W-C. (2009). Job definition discrepancy between supervisors and
subordinates: The antecedent role of LMX and outcomes. Journal of Occupational
and Organizational Psychology, 82, 89–112. DOI: 10.1348/096317908×292374
Hu J, Liden RC. (2013). Relative leader-member exchange within team contexts: How and
when social comparison impacts individual effectiveness. P ERSONNEL P SYCHOL -
OGY , 66, 127–172. DOI:10.1111/peps.12008
*Huang X, Chan SCH, Lam W, Nan X. (2010). The joint effect of leader member exchange
and emotional intelligence on burnout and work performance in call centers in
China. International Journal of Human Resource Management, 21, 1124–1144.
DOI:10.1080/09585191003783553
*Hui C, Law KS, Chen ZX. (1999). A structural equation model of the effects of neg-
ative affectivity, leader–member exchange, and perceived job mobility on in-role
and extra-role performance: A Chinese case. Organizational Behavior and Human
Decision Processes, 77, 3–21. DOI:10.1006/obhd.1998.2812
*Hui C, Law KS, Chen NYF, Tjosvold D. (2008). The role of co-operation and competi-
tion on leader-member exchange and extra-role performance in China. Asia Pacific
Journal of Human Resources, 46, 133–152. DOI: 10.1177/1038411108091753
*Hui C, Lee C, Rousseau DM. (2004). Employment relationships in China: do work-
ers relate to the organization or to people? Organization Science, 15, 232–240.
DOI:org/10.1287/orsc.1030.0050
Hunter J, Schmidt F. (1990). Methods of meta-analysis: Correcting for error and bias in
research findings (1st ed.). Newbury Park, CA: Sage.
Hunter J, Schmidt F. (2004). Methods of meta-analysis: Correcting for error and bias in
research findings (2nd ed.). Thousand Oaks, CA: Sage.
Ilies R, Nahrgang JD, Morgeson FP. (2007). Leader-member exchange and citizen-
ship behaviors: A meta-analysis. Journal of Applied Psychology, 92, 269–277.
DOI:org/10.1037/0021-9010.92.1.269
 ROBIN MARTIN ET AL. 111

*Ishak NA, Alam SS. (2009). The effects of leader-member exchange on organizational
justice and organizational citizenship behavior: Empirical study. European Journal
of Social Sciences, 8, 324–334. DOI:10.5539/ijbm.v4n3p52
*Janssen O, Van Yperen NW. (2004). Employees’ goal orientations, the quality of leader-
member exchange, and the outcomes of job performance and job satisfaction.
Academy of Management Journal, 47, 368–384. DOI:10.2307/20159587
*Jawahar IM, Carr D. (2007). Conscientiousness and contextual performance:
The compensatory effects of perceived organizational support and leader-
member exchange. Journal of Managerial Psychology, 22, 330–349.
DOI:org/10.1108/02683940710745923
*Johnson RE, Saboe KN. (2011). Measuring implicit traits in organizational research: De-
velopment of an indirect measure of employee implicit self-concept. Organizational
Research Methods, 14, 530–547. DOI:10.1177/1094428110363617
*Johnson J, Truxillo DM, Erdogan B, Bauer TN, Hammer L. (2009). Perceptions of overall
fairness: Are effects on job performance moderated by leader member exchange?
Human Performance, 22, 432–449. DOI:10.1080/08959280903248427
Jones DA. (2009). Getting even with one’s supervisor and one’s organization: Relationships
among types of injustice, desires for revenge, and counterproductive work behaviors.
Journal of Organizational Behavior, 30, 525–542. DOI:10.1002/job.563
Joo B, Ready K. (2012). Career satisfaction: The effects of performance goal
orientation, proactive personality, organizational learning culture, and leader-
member exchange quality. Career Development International, 17, 276–295.
DOI:org/10.1108/13620431211241090
Joseph DL, Newman DA, Sin HP. (2011). Leader-member exchange (LMX) measurement:
Evidence for consensus, construct breadth and discriminant validity. In Bergh DD,
Ketchen DJ (Eds.). Building methodological bridges: Research methodology in
strategy and management (Vol. 6, pp. 89–135). Bingley, UK: Emerald.
*Judge TA, Ferris GR. (1993). Social context of performance evaluation decisions. Academy
of Management Journal, 36, 80–105. DOI:[Link]/10.2307/256513
Judge TA, Kammeyer-Mueller JD. (2012). Job attitudes. Annual Review of Psychology, 63,
341–367. DOI:10.1146/annurev-psych-120710-100511
Judge TA, Thoresen CJ, Bono JE, Patton GK. (2001). The job satisfaction–job performance
relationship: A qualitative and quantitative review. Psychological Bulletin, 127, 376–
407. DOI:org/10.1037/0033-2909.127.3.376
*Kacmar KM, Witt LA, Zivnuska S, Gully SM. (2003). The interactive effect of leader-
member exchange and communication frequency on performance ratings. Journal
of Applied Psychology, 88, 764–772. DOI:org/10.1037/0021-9010.88.4.764
*Kacmar KM, Zivnuska S, White CD. (2007). Control and exchange: The im-
pact of work environment on the work effort of low relationship quality
employees. Leadership Quarterly, 18, 69–84. HYPERLINK “[Link]
10.1016/[Link].2006.11.002” \t “doilink” DOI:10.1016/[Link].2006.11.002
*Kamdar D, Van Dyne L. (2007). The joint effects of personality and workplace so-
cial exchange relationships in predicting task performance and citizenship perfor-
mance. Journal of Applied Psychology, 92, 1286–1298. DOI:org/10.1037/0021-
9010.92.5.1286
Kaplan S, Bradley JC, Luchman JN, Haynes D. (2009). On the role of positive and negative
affectivity in job performance: A meta-analytic investigation. Journal of Applied
Psychology, 94, 162–176. DOI:org/10.1037/a0013115
*Karriker JH, Williams ML. (2009). Organizational justice and organizational citizen-
ship behavior: A mediated multifoci model. Journal of Management, 35, 112–135.
DOI:[Link]/10.1177/0149206307309265
112 PERSONNEL PSYCHOLOGY

*Kath LM, Marks KM, Ranney J. (2010). Safety climate dimensions, leader mem-
ber exchange, and organizational support as predictors of upward safety com-
munication in a sample of rail industry workers. Safety Science, 48, 643–650.
DOI:10.1016/[Link].2010.01.016
*Katrinli A, Atabay G, Gunay G, Cangarli BG. (2011). The moderating role of leader-
member exchange in the relationship between psychological contract violation and
organizational citizenship behavior. African Journal of Business Management, 5,
1–6.
*Kim S, O’Neill JW, Cho H-M. (2010). When does an employee not help coworkers?
The effect of leader–member exchange on employee envy and organizational citi-
zenship behavior. International Journal of Hospitality Management, 29, 530–537.
DOI:10.1016/[Link].2009.08.003
*Klein HJ, Kim JS. (1998). A field study of the influence of situational constraints, leader–
member exchange, and goal commitment on performance. Academy of Management
Journal, 41, 88–95. DOI:10.2307/256900
Korsgaard MA, Sapienza HJ, Schweiger DM. (2002). Beaten before begun: The role
of procedural justice in planning change. Journal of Management, 28, 497–516.
DOI:10.1177/014920630202800402
*Kraimer ML, Seibert SE, Wayne SJ, Liden RC, Bravo J. (2011). Antecedents and
outcomes of organizational support for development: The critical role of career
opportunities. Journal of Applied Psychology, 96, 485–500. DOI:org/10.1037/
a0021452
*Kraimer ML, Wayne SJ, Jaworski RA. (2001). Sources of support and expatriate perfor-
mance: The mediating role of expatriate adjustment. P ERSONNEL P SYCHOLOGY,
54, 71–99. DOI:10.1111/j.1744-6570.2001.tb00086.x
*Kraus E. (2003). Personality and job performance: The mediating roles of leader member
exchange quality and action control. Doctoral Dissertation, Florida International
University.
*Krishnan VR. (2004). Impact of transformational leadership on followers’ influ-
ence strategies. Leadership & Organization Development Journal, 25, 58–72.
DOI:org/10.1108/01437730410512778
*Lagace RR. (1987). An investigation into the impact of interpersonal trust on the quality
of the relationship and outcome variables in the sales manager/salesperson dyad.
Doctoral Dissertation, University of Cincinnati.
*Lam W, Huang X, Snape E. (2007). Feedback-seeking behavior and leader member ex-
change: Do supervisor-attributed motives matter? Academy of Management Journal,
50, 348–363. DOI:10.5465/AMJ.2007.24634440
*Law KS, Wong C-S, Wang D, Wang L. (2000). Effect of supervisor-subordinate guanxi on
supervisory decisions in China: An empirical investigation. International Journal of
Human Resource Management, 11, 751–765. DOI:10.1080/09585190050075105
Lawler EE, Porter LW. (1967). Antecedent attitudes of effective managerial performance.
Organizational Behavior and Human Performance, 2, 122–142. DOI:10.1016/0030-
5073(67)90026-8
*Leana CR. (1986). Predictors and consequences of delegation. Academy of Management
Journal, 29, 754–774. DOI: 10.2307/255943
*Lee J. (2008). Effects of leadership and leader-member exchange on innovativeness. Jour-
nal of Managerial Psychology, 23, 670–687. DOI:org/10.1108/02683940810894747
*Li C K, Hung CH. (2009). The influence of transformational leadership on workplace re-
lationships and job performance. Social Behavior and Personality: An International
Journal, 37, 1129–1142. DOI:org/10.2224/sbp.2009.37.8.1129
 ROBIN MARTIN ET AL. 113

*Li N, Liang J, Crant JM. (2010). The role of proactive personality in job satisfaction and
organizational citizenship behavior: A relational perspective. Journal of Applied
Psychology, 95, 395–404. DOI:org/10.1037/a0018079
*Liao HUI, Liu D, Loi R. (2010). Looking at both sides of the social exchange coin:
A social cognitive perspective on the joint effects of relationship quality and
differentiation on creativity. Academy of Management Journal, 53, 1090–1109.
DOI:10.5465/AMJ.2010.54533207
*Liden RC, Erdogan B, Wayne SJ, Sparrowe RT. (2006). Leader-member exchange, dif-
ferentiation, and task interdependence: Implications for individual and group per-
formance. Journal of Organizational Behavior, 27, 723–746. DOI: 10.1002/job.409
*Liden RC, Maslyn JM. (1998). Multidimensionality of leader-member exchange: An
empirical assessment through scale development. Journal of Management, 24, 43–
72. DOI:10.1177/014920639802400105
Liden RC, Sparrowe RT, Wayne SJ. (1997). Leader-member exchange theory: The past and
potential for the future. Research in Personnel and Human Resources Management,
15, 47–119.
*Liden RC, Wayne SJ, Sparrowe RT. (2000). An examination of the mediating role of
psychological emprovement on the relations between the job, interpersonal re-
lationships, and work outcomes. Journal of Applied Psychology, 85, 407–416.
DOI:org/10.1037/0021-9010.85.3.407
*Liden RC, Wayne SJ, Stilwell D. (1993). A longitudinal study on the early develop-
ment of leader-member exchanges. Journal of Applied Psychology, 78, 662–674.
DOI:org/10.1037/0021-9010.78.4.662
*Liden RC, Wayne SJ, Zhao H, Henderson D. (2008). Servant leadership: Development of
a multidimensional measure and multi-level assessment. Leadership Quarterly, 19,
161–177. DOI:10.1016/[Link].2008.01.006
*Lindsay DR. (2009). Polychronicity and its impact on leader-member exchange and
outcome behaviors. Doctoral Dissertation, Pennsylvania State University.
*Lo MC, Ramayah T, Hui JKS. (2006). An investigation of leader member exchange
effects on organizational citizenship behavior in Malaysia. Journal of Business and
Management, 12, 5–23.
*Loi R, Ngo H-Y. (2009). Work outcomes of relational demography in Chinese vertical
dyads. International Journal of Human Resource Management, 20, 1704–1719.
DOI:10.1080/09585190903087057
*Manogran P, Stauffer J, Conlon EJ. (1994, August). Leader-member exchange as a key
mediating variable between employees’ perceptions of fairness and organizational
citizenship behavior. Paper presented at the annual meeting of the Academy of
Management, Dallas.
*Markham SE, Yammarino FJ, Murry WD, Palanski ME. (2010). Leader member exchange,
shared values, and performance: Agreement and levels of analysis do matter. Lead-
ership Quarterly, 21, 469–480. DOI:10.1016/[Link].2010.03.010
Martin R, Epitropaki O, Thomas G, Topakas A. (2010). A critical review of leader-member
relationship (LMX) research: Future prospects and directions. International Review
of Industrial and Organizational Psychology, 25, 61–91.
*Martinaityte I, Sacramento CA. (2013). When creativity enhances sales effectiveness: The
moderating role of leader–member exchange. Journal of Organizational Behavior,
34, 974–994. DOI: 10.1002/job.1835
*Maslyn JM, Fedor DB. (1998). Perceptions of politics: Does measuring different
foci matter. Journal of Applied Psychology, 83, 645–653. DOI:org/10.1037/0021-
9010.83.4.645
114 PERSONNEL PSYCHOLOGY

Maslyn JM, Uhl-Bien M. (2001). Leader-member exchange and its dimensions: Effects of
self-effort and other’s effort on relationship quality. Journal of Applied Psychology,
86, 697–708. DOI:org/10.1037/0021-9010.86.4.697
*Masterson SS, Lewis K, Goldman BM, Taylor MS. (2000). Integrating justice and social
exchange: The differing effects of fair procedures and treatment on work relation-
ships. Academy of Management Journal, 43, 738–748. DOI:10.2307/1556364
Meyer JP, Allen NJ, Smith CA. (1993). Commitment to organizations and occupations:
Extension and test of a three-component conceptualization. Journal of Applied
Psychology, 78, 538–551. DOI:org/10.1037/0021-9010.78.4.538
*Michael DF. (2004). Gender differences in information-exchange and outcomes at work.
Doctoral Dissertation, Auburn University.
*Michael JH, Guo ZG, Wiedenbeck JK, Ray CD. (2006). Production supervisor im-
pacts on subordinates’ safety outcomes: An investigation of leader-member ex-
change and safety communication. Journal of Safety Research, 37, 469–477.
DOI:10.1016/[Link].2006.06.004
*Moss SE, Sanchez JI, Brumbaugh AM, Borkowski N. (2009). The mediating role of
feedback avoidance behavior in the LMX-performance relationship. Group & Or-
ganization Management, 34, 645–664. DOI:10.1177/1059601109350986
*Murphy KR. (1989). Dimensions of job performance. In Dillion R, Pellingrino J (Eds.).
Testing: Applied and theoretical perspectives (pp. 218–247). New York, NY:
Praeger.
Murphy KR, Cleveland J. (1995). Understanding performance appraisal: Social, organi-
zational and goal-oriented perspectives. Newbury Park, CA: Sage.
*Murphy SE, Ensher EA. (1999). The effects of leader and subordinate characteristics
in the development of leader-member exchange quality. Journal of Applied Social
Psychology, 29, 1371–1394. DOI:10.1111/j.1559-1816.1999.tb00144.x
*Murphy SM. (1998). Organizational justice: An examination of antecedents and conse-
quences. Doctoral Dissertation, University of Illinois.
*Murphy SM, Wayne SJ, Liden RC, Erdogan B. (2003). Understanding social loafing: The
role of justice perceptions and exchange relationships. Human Relations, 56, 61–84.
DOI:10.1177/0018726703056001450
Muthén L, Muthén B. (1998–2006). Mplus (Version 4.0). Los Angeles, CA: Muthén &
Muthén.
*Nahrgang J, Morgeson F, Ilies R. (2009). The development of leader-member exchanges:
Exploring how personality and performance influence leader and member relation-
ships over time. Organizational Behavior and Human Decision Processes, 108,
256–266. DOI:10.1016/[Link].2008.09.002
*Ouyang Y. (2011). An exploration of LMX and personal guanxi on job performance: The
mediating effects of CSR. Journal of Global Business Issues, 5, 1–10.
*Ozer M. (2008). Personal and task-related moderators of leader-member exchange
among software developers. Journal of Applied Psychology, 93, 1174–1182.
DOI:org/10.1037/0021-9010.93.5.1174
*Paparoidamis NG, Guenzi P. (2009). An empirical investigation into the impact of relation-
ship selling and LMX on salespeople’s behaviours and sales effectiveness. European
Journal of Marketing, 43, 1053–1075. DOI:org/10.1108/03090560910961515
*Pelled LH, Xin KR. (2000). Relational demography and relationship quality in
two cultures. Organization Studies, 21, 1077–1094. DOI:[Link]/10.1177/
0170840600216003
*Piccolo RF, Bardes M, Mayer DM, Judge TA. (2008). Does high quality leader-member
exchange accentuate the effects of organizational justice? European Journal of Work
and Organizational Psychology, 17, 273–298. DOI:10.1080/13594320701743517
 ROBIN MARTIN ET AL. 115

*Piccolo RF, Colquitt JA. (2006). Transformational leadership and job behaviors: The
mediating role of core job characteristics. Academy of Management Journal, 49,
327–340. DOI:10.5465/AMJ.2006.20786079
Pillai R, Schriesheim CA, Williams ES. (1999). Fairness perceptions and trust for transfor-
mational and transactional leadership: A two-sample study. Journal of Management,
25, 897–933. DOI:10.1177/014920639902500606
Podsakoff PM, MacKenzie SB. (1998). Appendix: Measures and Assessments for the
Substitutes for Leadership Approach. Monographs in Organisational Behavior and
Industrial Relations, 24, 259–260.
Podsakoff PM, MacKenzie SB, Lee JY, Podsakoff NP. (2003). Common method biases in
behavioral research: A critical review of the literature and recommended remedies.
Journal of Applied Psychology, 88, 879–903. DOI:org/10.1037/0021-9010.88.5.879
Podsakoff PM, MacKenzie SB, Moorman RH, Fetter R. (1990). Transformational leader
behaviors and their effects on followers’ trust in leader, satisfaction, and organiza-
tional citizenship behaviors. Leadership Quarterly, 1, 107–142. DOI:10.1016/1048-
9843(90)90009-7
*Polly LM. (2002). Social exchange and customer service: The relationship between per-
ceived organizational support, leader-member exchange, and customer service be-
havior. Doctoral Dissertation, Louisiana State University.
*Restubog SLD, Bordia P, Tang RL, Krebs SA. (2010). Investigating the moderating
effects of leader–member exchange in the psychological contract breach employee
performance relationship: A test of two competing perspectives. British Journal of
Management, 21, 422–437. DOI:10.1111/j.1467-8551.2009.00673.x
*Reynolds D. (2002). The moderating effect of leader-member exchange in the relationship
between self-efficacy and performance. Journal of Human Resources in Hospitality
& Tourism, 1, 77–90. DOI:10.1300/J171v01n03_06
Riketta M. (2005). Organizational identification: A meta-analysis. Journal of Vocational
Behavior, 66, 358–384. DOI:10.1016/[Link].2004.05.005
Riketta M. (2008). The causal relation between job attitudes and performance: A
meta-analysis of panel studies. Journal of Applied Psychology, 93, 472–481.
DOI:org/10.1037/0021-9010.93.2.472
Rizzo JR, House RJ, Lirtzman, SI. (1970). Role conflict and ambiguity in complex organiza-
tions. Administrative Science Quarterly, 15, 150–163. DOI:[Link]/stable/2391486
Rockstuhl T, Dulebohn JH, Ang S, Shore LM. (2012). Leader–member exchange (LMX)
and culture: A meta-analysis of correlates of LMX across 23 countries. Journal of
Applied Psychology, 97, 1097–1130. DOI:org/10.1037/a0029978
*Rosen CC, Harris KJ, Kacmar KM. (2011). LMX, context perceptions, and performance:
An uncertainty management perspective. Journal of Management, 37, 819–838.
DOI:10.1177/0149206310365727
Rotundo M, Sackett PR. (2002). The relative importance of task, citizenship and counterpro-
ductive performance to global ratings of performance: A policy-capturing approach.
Journal of Applied Psychology, 87, 66–80. DOI:org/10.1037/0021-9010.87.1.66
*Sanders K, Moorkamp M, Torka N, Groeneveld S, Groeneveld C. (2010). How to sup-
port innovative behaviour? The role of LMX and satisfaction with HR practices.
Technology and Investment, 1, 59–68. DOI:[Link]/10.4236/ti.2010.11007
*Scaduto A, Lindsay D, Chiaburu DS. (2008). Leader influences on training effectiveness:
Motivation and outcome expectation processes. International Journal of Training
and Development, 12, 158–170. DOI:10.1111/j.1468-2419.2008.00303.x
Scandura TA. (1999). Rethinking leader-member exchange: An organizational justice per-
spective. Leadership Quarterly, 10, 25–40. DOI:10.1016/S1048-9843(99)80007-1
116 PERSONNEL PSYCHOLOGY

Scandura TA, Graen GB. (1984). Moderating effects of initial leader–member exchange
status on the effects of a leadership intervention. Journal of Applied Psychology, 69,
428–436. DOI:org/10.1037/0021-9010.69.3.428
*Scandura TA, Graen GB, Novak MA. (1986). When managers decide not to decide
autocratically: An investigation of leader-member exchange and decision influence.
Journal of Applied Psychology, 71, 579–584. DOI:org/10.1037/0021-9010.71.4.579
Scandura T.A, Lankau MJ. (1996). Developing diverse leaders: A leader-member ex-
change approach. Leadership Quarterly, 7, 243–263. DOI:10.1016/S1048-9843(96)
90043-0
Scandura TA, Pellegrini EK. (2008). Trust and leader-member exchange: A closer look
at relational vulnerability. Journal of Leadership & Organizational Studies, 15,
101–110. DOI:10.1177/1548051808320986
*Scandura TA, Schriesheim CA. (1994). Leader-member exchange and supervisor career
mentoring as complementary constructs in leadership research. Academy of Man-
agement Journal, 37, 1588–1602. DOI:10.2307/256800
*Schaninger WS. (2002). The workplace social exchange network: An empirical examina-
tion. Doctoral Dissertation, Auburn University.
Schriesheim CA, Castro SL, Cogliser CC. (1999). Leader-member exchange (LMX) re-
search: A comprehensive review of theory, measurement, and data-analytic practices.
Leadership Quarterly, 10, 63–113. DOI:10.1016/S1048-9843(99)80009-5
*Schriesheim CA, Castro SL, Yammarino FJ. (2000). Investigating contingencies: An ex-
amination of the impact of span of supervision and upward controllingness on leader-
member exchange using traditional and multivariate and between-entities analysis.
Journal of Applied Psychology, 85, 659–677. DOI:org/10.1037/0021-9010.85.5.659
*Schriesheim CA, Neider LL, Scandura TA. (1998). Delegation and leader member ex-
change: Main effects, moderators, and measurement issues. Academy of Manage-
ment Journal, 41, 298–318. DOI:10.2307/256909
Schyns B, Day D. (2010). Critique and review of leader-member exchange theory: Issues
of agreement, consensus, and excellence. European Journal of Work and Organiza-
tional Psychology, 19, 1–29. DOI:10.1080/13594320903024922
Scott J, Craven AE, Green C. (2006). A meta-analysis of organizational citizenship behavior
and leader-member exchange. Journal of Business and Economics Research, 4, 19–
36.
*Scott KD. (2009). The development and test of an exchange-based model of interpersonal
workplace exclusion. Doctoral Dissertation, University of Kentucky.
*Scott SG, Bruce RA. (1994). Determinants of innovative behavior: A path model of
individual innovation in the workplace. Academy of Management Journal, 37, 580–
607. DOI:10.2307/256701
*Sears G. (2005). The dispositional antecedents of leader-member exchange and organi-
zational citizenship behaviour: A process perspective. Doctoral Dissertation, Mc-
Master University, Ontario.
*Seers A. (1989). Team-member exchange quality: A new construct for role-making re-
search. Organizational Behavior and Human Decision Processes, 43, 118–135.
DOI:10.1016/0749-5978(89)90060-5
*Sekiguchi T, Burton JP, Sablynski CJ. (2008). The role of job embeddedness on
employee performance: The interactive effects with leader-member exchange
and organization-based self-esteem. P ERSONNEL P SYCHOLOGY, 61, 761–792.
DOI:10.1111/j.1744-6570.2008.00130.x
*Settoon RP, Bennett N, Liden RC. (1996). Social exchange in organizations: Perceived
organizational support, leader-member exchange, and employee reciprocity. Journal
of Applied Psychology, 81, 219–227. DOI:org/10.1037/0021-9010.81.3.219
 ROBIN MARTIN ET AL. 117

*Shalhoop JH. (2004). Social-exchange as a mediator of the relationship between organi-

zational justice and workplace outcomes. Doctoral Dissertation, The University of
Akron.
*Shull CK. (1995). The effects of leader-member exchange relations on organizational
citizenship behaviors. Doctoral Dissertation, Kansas State University.
Sin HP, Nahrgang JD, Morgeson FP. (2009). Understanding why they don’t see eye to eye:
An examination of leader-member exchange (LMX) agreement. Journal of Applied
Psychology, 94, 1048–1057. DOI:org/10.1037/a0014827
Smith CA, Organ DW, Near JP. (1983). Organizational citizenship behavior: Its nature and
antecedents. Journal of Applied Psychology, 68, 653–663. DOI:org/10.1037/0021-
9010.68.4.653
*Smith ML Jr. (2003). Reciprocity and social exchange relationships in organizations:
Examining why and how individuals contribute to organizational social capital.
Doctoral Dissertation, Purdue University.
Smith PC, Balzer W, Brannick M, Eggleston S, Gibson W, Ironson G, Josephson H, Paul
K, Reilly C, Whalen M. (1987). Manual for the revised JD1 and job in general
scales. Bowling Green, OH: Bowling Green State University.
Smither JW, London M. (2009). Best practices in performance management. In Smither
JW, London M (Eds.). Performance management: Putting research into action (pp.
585–625). San Francisco, CA: Jossey-Bass.
*Snyder RA, Bruning NS. (1985). Quality of vertical dyad linkages: Congruence of super-
visor and subordinate competence and role stress as explanatory variables. Group
& Organization Management, 10, 81–94. DOI:10.1177/105960118501000105
*Somech A, Wenderow M. (2006). The impact of participative and directive leadership on
teachers’ performance: The intervening effects of job structuring, decision domain,
and leader-member exchange. Educational Administration Quarterly, 42, 746–772.
DOI:10.1177/0013161×06290648
Sparrowe RT, Liden RC. (1997). Process and structure in leader-member exchange.
Academy of Management Review, 22, 522–552. DOI:10.2307/3069458
Sparrowe RT, Liden RC, Wayne SJ, Kraimer ML. (2001). Social networks and the perfor-
mance of individuals and teams. Academy of Management Journal, 44, 316–325.
DOI:10.2307/3069458
*Sparrowe RT, Soetjipto BW, Kraimer ML. (2006). Do leaders’ influence tactics relate to
members’ helping behavior? It depends on the quality of the relationship. Academy
of Management Journal, 49, 1194–1208. DOI:10.5465/AMJ.2006.23478645
Spreitzer GM. (1995). Psychological empowerment in the workplace: Dimensions, mea-
surement, and validation. Academy of Management Journal, 38, 1442–1465.
DOI:10.2307/256865
*Stark E, Poppler P. (2009). Leadership, performance evaluations, and all the usual suspects.
Personnel Review, 38, 320–338. DOI:org/10.1108/00483480910943368
Steiner DD. (1997). Attributions in leader member exchanges: Implications for prac-
tice. European Journal of Work and Organizational Psychology, 6, 59–71.
DOI:10.1080/135943297399295
*Suazo MM. (2003). An examination of antecedents and consequences of psychological
contract breach. Doctoral Dissertation, University of Kansas.
*Suazo MM. (2011). The impact of affect and social exchange on outcomes of psychological
contract breach. Journal of Managerial Issues, 23, 190–205.
Sue-Chan C, Au AKC, Hackett RD. (2012). Trust as a mediator of the relationship between
leader/member behavior and leader-member-exchange quality. Journal of World
Business, 47, 459–468. DOI:10.1016/[Link].2011.05.012
118 PERSONNEL PSYCHOLOGY

*Sue-Chan C, Chen Z, Lam W. (2011). LMX, coaching attributions, and em-

ployee performance. Group & Organization Management, 36, 466–498. DOI:
10.1177/1059601111408896
*Tansky JW. (1993). Justice and organizational citizenship behavior: What is
the relationship? Employee Responsibilities and Rights Journal, 6, 195–207.
DOI:10.1007/BF01419444
*Tejeda MJ. (2006). Nondiscrimination policies and sexual identity disclosure: Do they
make a difference in employee outcomes? Employee Responsibilities and Right
Journal, 18, 45–59. DOI:10.1007/s10672-005-9004-5
*Tekleab AG, Taylor MS. (2003). Aren’t there two parties in an employment relation-
ship? Antecedents and consequences of organization-employee agreement on con-
tract obligations and violations. Journal of Organizational Behavior, 24, 585–608.
DOI:10.1002/job.204
*Tepper BJ, Uhl-Bien M, Kohut GF, Rogelberg SG, Lockhart DE, Ensley MD. (2006).
Subordinates’ resistance and managers’ evaluations of subordinates’ performance.
Journal of Management, 32, 185–209. DOI:10.1177/0149206305277801
Thibaut JW, Kelley HH. (1959). The social psychology of groups. Oxford, England: Wiley.
*Tierney P, Bauer TN. (1996, August). A longitudinal assessment of lmx on extra-role
behavior. Paper presented at the Academy of Management Best Papers Proceedings.
*Tierney P, Farmer SM, Graen GB. (1999). An examination of leadership and employee
creativity: The relevance of traits and relationships. P ERSONNEL P SYCHOLOGY,
52, 591–620. DOI:10.1111/j.1744-6570.1999.tb00173.x
*Townsend J, Phillips JS, Elkins TJ. (2000). Employee retaliation: The neglected conse-
quence of poor leader-member exchange relations. Journal of Occupational Health
Psychology, 5, 457–463. DOI:org/10.1037/1076-8998.5.4.457
*Townsend JC, Da Silva N, Mueller L, Curtin P, Tetrick LE. (2002). Attributional com-
plexity: A link between training, job complexity, decision latitude, leader member
exchange, and performance. Journal of Applied Social Psychology, 32, 207–221.
DOI:10.1111/j.1559-1816.2002.tb01427.x
Tse HHM, Ashkanasy NM, Dasborough MT. (2012). Relative leader–member exchange,
negative affectivity and social identification: A moderated-mediation examination.
Leadership Quarterly, 23, 354–366. DOI:10.1016/[Link].2011.08.009
Uhl-Bien M. (2006). Relational leadership theory: Exploring the social pro-
cesses of leadership and organizing. Leadership Quarterly, 17, 654–676.
DOI:10.1016/[Link].2006.10.007
*Uhl-Bien M, Maslyn JM. (2003). Reciprocity in manager-subordinate relationships:
Components, configurations, and outcomes. Journal of Management, 29, 511–532.
DOI:10.1016/S0149-2063_03_00023-0
*Uhl-Bien M, Tierney PS, Graen GB, Wakabayashi M. (1990). Company paternalism and
the hidden-investment process identification of the “ Right Type” for line managers
in leading Japanese organizations. Group & Organization Management, 15, 414–
430. DOI:10.1177/105960119001500406
Van Breukelen W, Schyns B, Le Blanc P. (2006). Leader-member exchange theory
and research: Accomplishments and future challenges. Leadership, 2, 295–316.
DOI:10.1177/1742715006066023
*van Dam K, Oreg S, Schyns B. (2008). Daily work contexts and resistance to organisational
change: The role of leader-member exchange, development climate, and change
process characteristics. Applied Psychology, 57, 313–334. DOI:10.1111/j.1464-
0597.2007.00311.x
 ROBIN MARTIN ET AL. 119

*van Dierendonck D, Le Blanc PM, van Breukelen W. (2002). Supervisory behavior,

reciprocity and subordinate absenteeism. Leadership & Organization Development
Journal, 23, 84–92. DOI:org/10.1108/01437730210419215
*Van Dyne L, Jehn KA, Cummings A. (2002). Differential effects of strain on two forms
of work performance: Individual employee sales and creativity. Journal of Organi-
zational Behavior, 23, 57–74. DOI:10.1002/job.127
*Van Dyne L, Kamdar D, Joireman J. (2008). In-role perceptions buffer the negative
impact of low LMX on helping and enhance the positive impact of high LMX
on voice. Journal of Applied Psychology, 93, 1195–1207. DOI:org/10.1037/0021-
9010.93.6.1195
*Vecchio RP. (1987). Situational leadership theory: An examination of a prescriptive theory.
Journal of Applied Psychology, 72, 444–451. DOI:org/10.1037/0021-9010.72.3.444
*Vecchio RP. (1998). Leader-member exchange, objective performance, employment du-
ration, and supervisor ratings: Testing for moderation and mediation. Journal of
Business and Psychology, 12, 327–341. DOI:10.1023/A:1025027514081
*Vecchio RP, Brazil DM. (2007). Leadership and sex-similarity: A comparison in a
military setting. P ERSONNEL P SYCHOLOGY, 60, 303–335. DOI:10.1111/j.1744-
6570.2007.00075.x
*Vecchio RP, Bullis RC, Brazil DM. (2006). The utility of situational leadership the-
ory: A replication in a military setting. Small Group Research, 37, 407–424. doi:
10.1177/1046496406291560
*Vecchio RP, Gobdel BC. (1984). The vertical dyad linkage model of leadership: Prob-
lems and prospects. Organizational Behavior and Human Performance, 34, 5–20.
DOI:10.1016/0030-5073(84)90035-7
*Vidyarthi PR, Liden RC, Anand S, Erdogan B, Ghosh S. (2010). Where do I stand? Exam-
ining the effects of leader–member exchange social comparison on employee work
behaviors. Journal of Applied Psychology, 95, 849–861. DOI:org/10.1037/a0020033
Viswesvaran C, Ones DS. (1995). Theory testing: Combining psychometric meta-analysis
and structural equations modelling. P ERSONNEL P SYCHOLOGY, 48, 865–885.
DOI:10.1111/j.1744-6570.1995.tb01784.x
Viswesvaran C, Ones DS. (2000). Perspectives on models of job performance. International
Journal of Selection and Assessment, 8, 216–226. DOI:10.1111/1468-2389.00151
Vroom VH. (1964). Work and motivation. Oxford, UK: Wiley.
*Waismel-Manor R, Tziner A, Berger E, Dikstein E. (2010). Two of a kind? leader-
member exchange and organizational citizenship behaviors: The moderating role
of leader-member similarity. Journal of Applied Social Psychology, 40, 167–181.
DOI:10.1111/j.1559-1816.2009.00568.x
Wakabayashi M, Graen G. (1984). The Japanese career progress study: A seven-year
follow up. Journal of Applied Psychology, 69, 603–614. DOI:org/10.1037/0021-
9010.69.4.603
*Wakabayashi M, Graen G, Graen M. (1988). Japanese management progress: Mo-
bility into middle management. Journal of Applied Psychology, 73, 217–227.
DOI:org/10.1037/0021-9010.73.2.217
*Walumbwa, FO, Cropanzano R, Goldman BM. (2011). How leader–member exchange
influences effective work behaviors: Social exchange and internal–external effi-
cacy perspectives. P ERSONNEL P SYCHOLOGY, 64, 739–770. DOI:10.1111/j.1744-
6570.2011.01224.x
*Walumbwa FO, Cropanzano R, Hartnell CA. (2009). Organizational justice, voluntary
learning behavior, and job performance: A test of the mediating effects of iden-
tification and leader-member exchange. Journal of Organizational Behavior, 30,
1103–1126. DOI:10.1002/job.611
120 PERSONNEL PSYCHOLOGY

*Wang H, Law KS, Chen ZX. (2008). Leader-member exchange, employee per-
formance, and work outcomes: An empirical study in the Chinese con-
text. International Journal of Human Resource Management, 19, 1809–1824.
DOI:10.1080/09585190802323926
*Wang H, Law KS, Hackett RD, Wang D, Chen ZX. (2005). Leader member exchange
as a mediator of the relationship between transformational leadership and follow-
ers’ performance and organizational citizenship behavior. Academy of Management
Journal, 48, 420–432. DOI:10.5465/AMJ.2005.17407908
*Wang L, Chu X, Ni J. (2010). Leader-member exchange and organizational citizenship
behavior: A new perspective from perceived insider status and Chinese traditionality.
Frontiers of Business Research in China, 4, 148–169. DOI:org/10.1007/s11782-010-
0007-1
*Wang X, Liao J, Xia D, Chang T. (2010). The impact of organizational justice
on work performance: Mediating effects of organizational commitment and
leader-member exchange. International Journal of Manpower, 31, 660–677.
DOI:org/10.1108/01437721011073364
*Wat D, Shaffer MA. (2005). Equity and relationship quality influences on organizational
citizenship behaviors: The mediating role of trust in the supervisor and empower-
ment. Personnel Review, 34, 406–422. DOI:org/10.1108/00483480510599752
*Wayne SJ, Ferris GR. (1990). Influence tactics, affect, and exchange quality in supervisor-
subordinate interactions: A laboratory experiment and field study. Journal of Applied
Psychology, 75, 487–499. DOI:org/10.1037/0021-9010.75.5.487
*Wayne SJ, Shore LM, Bommer WH, Tetrick LE. (2002). The role of fair treatment
and rewards in perceptions of organizational support and leader-member exchange.
Journal of Applied Psychology, 87, 590–598. DOI:org/10.1037/0021-9010.87.3.590
*Wayne SJ, Shore LM, Liden RC. (1997). Perceived organizational support and leader-
member exchange: A social exchange perspective. Academy of Management Jour-
nal, 40, 82–111. DOI:10.2307/257021
*Wech BA. (2002). Team-member exchange and trust contexts: Effects on individual level
outcome variables beyond the influence of leader-member exchange. Doctoral Dis-
sertation, Louisiana State University.
*Wee GY, Ahmad KZ, Fen YS. (2011). Promoting organizational citizenship behavior
through high involvement human resource practices: An attempt to reduce turnover
intention. Paper Presented at Annual International Conference on Human Resource
Management and Professional Development for the Digital Age (HRM&PD)
Weiss DJ, Dawis RV, England GW, Lofquist LH. (1967). Manual for the Minnesota
satisfaction questionnaire. Work Adjustment Project, Industrial Relations Center.
Minneapolis, MN: University of Minnesota.
*Weng LC, Su CW, Lai YC. (2011). Superior service performance through transformational
leadership: A cross-level study of a large Taiwanese commercial bank. Asia Pacific
Management Review, 16, 181–195.
Whitener EM. (1990). Confusion of confidence intervals and credibility intervals in meta-
analysis. Journal of Applied Psychology, 75, 315–321. DOI:org/10.1037/0021-
9010.75.3.315
*Whittington JL. (1998). An integrative model of transformational leadership and follower
behavior. Doctoral Dissertation, The University of Texas at Arlington.
*Wilhelm, CC, Herd AM, Steiner DD. (1993). Attributional conflict betweenmanagers
and subordinates: An investigation of leader-member exchange effects. Journal of
Organizational Behavior, 14, 531–544. DOI:10.1002/job.4030140603
 ROBIN MARTIN ET AL. 121

Williams LJ, Anderson SE. (1991). Job satisfaction and organizational commitment as pre-
dictors of organizational citizenship and in-role behaviors. Journal of Management,
17, 601–617. DOI:10.1177/014920639101700305
Williams LJ, Podsakoff PM. (1989). Longitudinal field methods for studying reciprocal
relationships in organizational behavior research: Toward improved causal analysis.
Research in Organizational Behavior, 11, 247–292.
Woehr DJ, Huffcutt AI. (1994). Rater training in performance appraisal: A quantitative
review. Journal of Organizational and Occupational Psychology, 67, 189–205.
DOI:10.1111/j.2044-8325.1994.tb00562.x
Yammarino FJ, Dionne SD, Chun JU, & Dansereau F. (2005). Leadership and levels
of analysis: A state-of-the-science review. Leadership Quarterly, 16, 879–919.
DOI:10.1016/[Link].2005.09.002
Yang J, Mossholder KW. (2010). Examining the effects of trust in leaders: A bases-and-foci
approach. Leadership Quarterly, 21, 50–63. DOI:10.1016/[Link].2009.10.004
*Yi-feng NC, Tjosvold D. (2008). Goal interdependence and leader-member relationship for
cross-cultural leadership in foreign ventures in China. Leadership & Organization
Development Journal, 29, 144–166. DOI:org/10.1108/01437730810852498
*Yuan L. (2007). A motivational model of organizational citizenship behavior. Doctoral
Dissertation, The University of Illinois at Chicago.
Copyright of Personnel Psychology is the property of Wiley-Blackwell and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.