ArticlePDF Available

Abstract and Figures

Although transformational leadership has been studied extensively, the magnitude of the relationship between transformational leadership and follower performance across criterion types and levels of analysis remains unclear. Based on 117 independent samples over 113 primary studies, the current meta-analytic study showed that transformational leadership was positively related to individual-level follower performance across criterion types, with a stronger relationship for contextual performance than for task performance across most study settings. In addition, transformational leadership was positively related to performance at the team and organization levels. Moreover, both meta-analytic regression and relative importance analyses consistently showed that transformational leadership had an augmentation effect over transactional leadership (contingent reward) in predicting individual-level contextual performance and team-level performance. Contrary to our expectation, however, no augmentation effect of transformational leadership over contingent reward was found in predicting individual-level task performance. Instead, contingent reward explained incremental variance in individual-level task performance beyond that explained by transformational leadership. [ABSTRACT FROM AUTHOR] Copyright of Group & Organization Management is the property of Sage Publications Inc. 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. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Content may be subject to copyright.
http://gom.sagepub.com/
Management
Group & Organization
http://gom.sagepub.com/content/36/2/223
The online version of this article can be found at:
DOI: 10.1177/1059601111401017
2011 36: 223Group & Organization Management
Gang Wang, In-Sue Oh, Stephen H. Courtright and Amy E. Colbert
Levels: A Meta-Analytic Review of 25 Years of Research
Transformational Leadership and Performance Across Criteria and
Published by:
http://www.sagepublications.com
can be found at:Group & Organization ManagementAdditional services and information for
http://gom.sagepub.com/cgi/alertsEmail Alerts:
http://gom.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://gom.sagepub.com/content/36/2/223.refs.htmlCitations:
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Group & Organization Management
36(2) 223 –270
© The Author(s) 2011
Reprints and permission: http://www.
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1059601111401017
http://gom.sagepub.com
Transformational
Leadership and
Performance Across
Criteria and Levels:
A Meta-Analytic Review
of 25 Years of Research
Gang Wang
1
, In-Sue Oh
2
,
Stephen H. Courtright
1
,
and Amy E. Colbert
1
Abstract
Although transformational leadership has been studied extensively, the
magnitude of the relationship between transformational leadership and follower
performance across criterion types and levels of analysis remains unclear. Based
on 117 independent samples over 113 primary studies, the current meta-
analytic study showed that transformational leadership was positively related
to individual-level follower performance across criterion types, with a stronger
relationship for contextual performance than for task performance across
most study settings. In addition, transformational leadership was positively
related to performance at the team and organization levels. Moreover, both
meta-analytic regression and relative importance analyses consistently showed
that transformational leadership had an augmentation effect over transactional
leadership (contingent reward) in predicting individual-level contextual
performance and team-level performance. Contrary to our expectation,
however, no augmentation effect of transformational leadership over contingent
reward was found in predicting individual-level task performance. Instead,
1
University of Iowa, Iowa City
2
Virginia Commonwealth University, Richmond
Corresponding Author:
Gang Wang, Department of Management and Organizations, University of Iowa,
John Pappajohn Business Building, W252, PBB, Iowa City, IA 52242-1994
Email: gang-wang@uiowa.edu
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
224 Group & Organization Management 36(2)
contingent reward explained incremental variance in individual-level task
performance beyond that explained by transformational leadership.
Keywords
transformational leadership, meta-analysis, performance
The title of Bass’s (1985) seminal book on transformational leadership,
Leadership and Performance Beyond Expectations, has garnered much atten-
tion over the past quarter century. In his book, Bass contrasts transactional or
exchange-based forms of leadership, in which leaders clarify expectations and
reward followers for fulfilling them, with transformational leadership, in
which leaders motivate their followers to move beyond self-interest and
work for the collective good (Avolio & Yammarino, 2002; Bass, 1985; Burns,
1978). Using examples like Mahatma Gandhi and John F. Kennedy, Bass pro-
posed that transformational leaders increase followers’ confidence and the
intrinsic value of performance, resulting in higher levels of motivation (Seibert,
Wang, & Courtright, in press). Thus, while transactional leadership may lead
to expected performance, transformational leadership has the potential to
result in performance beyond expectations. As our understanding of different
types and levels of performance has become more precise (e.g., Borman &
Motowidlo, 1993; Klein, Dansereau, & Hall, 1994; Organ, 1988; Yammarino,
Dionne, Chun, & Dansereau, 2005), a growing body of research has investi-
gated the range of potential performance implications of transformational
leadership.
However, despite the abundance of primary studies linking transformational
leadership and performance, the current transformational leadership literature
does not provide a clear understanding of the generalizability of the “beyond
expectation” role of transformational leadership in performance across crite-
rion types and levels of analysis. Meta-analysis can be used to estimate the
true magnitude of the role of transformational leadership in performance and
its generalizability across studies in several ways. First, at the most basic
level, meta-analysis allows us to estimate the more precise magnitude of the
relationship between transformational leadership and follower individual
performance than any of the primary studies included in the meta-analysis.
While theory suggests that transformational leadership is associated with
higher levels of performance from followers, prior meta-analyses have pro-
vided limited information about the size of this relationship. Indeed, among
the five meta-analyses conducted on transformational leadership (summarized
in Table 1), only DeGroot, Kiker, and Cross (2000) attempted to estimate the
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
225
Table 1. Summary of Previous and Present Meta-Analytic Findings on Outcomes of Transformational Leadership
Variable k N r
_
ρ
ˆ
Source
Comparison between previous
meta-analyses and the present study
Follower attitudinal and motivational outcomes
Follower job satisfaction 14 3,832 .70 .77 DeGroot, Kiker, and Cross
(2000)
NA
6 2,175 .27 .30 Dumdum, Low, and Avolio
(2002)
NA
18 5,279 .58 Judge and Piccolo (2004) NA
Follower satisfaction with
leader
9 2,457 .49 .57 Dumdum et al. (2002) NA
23 4,349 .71 Judge and Piccolo (2004) NA
12 2,680 .71 .80 Fuller, Patterson, Hester,
and Stringer (1996)
NA
Follower motivation 16 4,773 .53 Judge and Piccolo (2004) NA
Follower organizational
commitment
3 2,040 .39 .43 DeGroot et al. (2000) NA
Follower effort 12 3,807 .65 .73 DeGroot et al. (2000) NA
Follower behavioral outcome
Follower performance 4 715 .19 .21 DeGroot et al. (2000) This estimate was based on a limited
number of studies. The authors did
not estimate the relationship between
transformational leadership and any
specific types of subordinate individual
performance. Furthermore, the sources
of performance measures were unclear.
(continued)
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
226
Table 1. (continued)
Variable k N r
_
ρ
ˆ
Source
Comparison between previous
meta-analyses and the present study
Follower behavioral outcome
Follower individual-level
performance
62 16,809 .22 .25 Present study Present estimate of the meta-analytic
correlation between transformational
leadership and follower individual-level
performance is based on a much larger
number of primary studies in which
performance measures were non-self-
reported.
Follower task performance 31 7,016 .19 .21 Present study NA
Follower contextual
performance
28 7,970 .26 .30 Present study NA
Follower creative
performance
14 3,728 .19 .21 Present study NA
Leader outcomes
Leader effectiveness 23 5,577 .68 .74 DeGroot et al. (2000) NA
18 7,262 .43 .50 Dumdum et al. (2002) NA
27 5,415 .64 Judge and Piccolo (2004) NA
10 1,524 .68 .78 Fuller et al. (1996) NA
47 6,485 .62 .71 Lowe, Kroeck, and
Sivasubramaniam (1996)
NA
Leader job performance 13 2,126 .27 Judge and Piccolo (2004) NA
Supraindividual-level outcomes
Overall performance 27 4,611 0.39 0.45 Fuller et al. (1996) No distinction between follower, team,
and organization performance ratings;
No distinction among specific job
performance dimension at each level.
(continued)
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
227
Table 1. (continued)
Variable k N r
_
ρ
ˆ
Source
Comparison between previous
meta-analyses and the present study
Group or organizational
performance
41 6,197 .26 Judge and Piccolo (2004) Group and organization performance
were lumped together as the same
criterion.
Team performance 7 432 .42 .49 DeGroot et al. (2000) This estimate was based on a limited
number of primary studies, and the
sources of performance measure were
unclear.
Team-level performance 34 2,830 .24 .33 Present study Present study distinguishes group
performance from organizational
performance and is based on sufficient
number of primary studies in which all
performance measures are either non-
self-reported or objective.
Organizational-level
performance
27 2,408 .19 .27 Present study
Note: k = number of correlations; N = combined sample size; r
_
= sample-size weighted mean uncorrected correlation; ρ
ˆ
= estimated corrected mean
correlation; NA = not applicable.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
228 Group & Organization Management 36(2)
relationship of the charismatic component of transformational leadership
with follower individual performance. However, the small number of stud-
ies that were used to estimate this relationship (k = 4) suggests the poten-
tial for second-order sampling error (Hunter & Schmidt, 2004). Thus, the
first purpose of our article is to provide a more precise estimate of the
relationship between transformational leadership and follower individual
performance and to examine the generalizability of this relationship across
settings.
Second, when Bass (1985) initially suggested that transformational leader-
ship motivates followers to perform “beyond expectations,” researchers were
only beginning to differentiate between various types of performance criteria
(Austin & Villanova, 1992). Thus, the exact meaning of “performance
beyond expectations” was not clearly specified. On one hand, transforma-
tional leadership may motivate followers to work harder, exerting more
effort than would be expected from transactional leadership and resulting in
higher levels of task performance. On the other hand, Podsakoff, MacKenzie,
and Bommer (1996) have proposed that transformational leadership moti-
vates followers to go beyond the minimum requirements of their job descrip-
tions, resulting in higher levels of contextual performance. Finally, the focus
of transformational leaders on challenging the status quo suggests that per-
formance beyond expectations may result in higher levels of creativity and
innovation among followers. Despite the fact that primary studies have
examined the relationship of transformational leadership with task, contex-
tual, and creative performance, none of the prior meta-analyses on transfor-
mational leadership have estimated the magnitude of these relationships.
Thus, the second purpose of our meta-analysis is to investigate the relative
impact of transformational leadership on follower task, contextual, and cre-
ative performance.
Third, transformational leadership theory suggests that transformational
leadership is related not only to individual follower performance but also to
performance at the group and organization levels (Bass, 1985; Conger &
Kanungo, 1998; Shamir, House, & Arthur, 1993). However, no previous
meta-analysis has provided a comparison of the relationship of transforma-
tional leadership with performance at all three levels. DeGroot et al.
(2000) provided initial evidence that transformational leadership is posi-
tively related to team performance, but this analysis was based on a lim-
ited number of primary studies on team performance (k = 7). Judge and
Piccolo (2004) identified a larger number of primary studies examining the
relationship between transformational leadership and performance at the
group and organizational levels (k = 41), but they combined these studies in
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 229
their meta-analysis, reporting the relationship between transformational
leadership and group/organization performance. However, individual,
group, and organization performance is likely influenced by different fac-
tors and through different mechanisms (Dansereau, Cho, & Yammarino,
2006). As a result, the magnitude of the relationships of transformational
leadership with performance at the three levels of analysis may differ
(Yammarino et al., 2005). Thus, a third purpose of our study is to estimate
and compare the relationship of transformational leadership with individual,
group, and organization performance.
Finally, one of the most interesting theoretical claims of Bass (1997) is
that transformational leadership has one-way augmentation effects over
transactional leadership. That is, transformational leadership is hypothesized
to predict follower performance beyond the effects of transactional leader-
ship. Yet this proposition has not been systematically examined in predict-
ing follower performance across performance criteria and levels of analysis.
A small number of primary studies by Bass and his colleagues (Bass, Avolio,
Jung, & Berson, 2003; Howell & Avolio, 1993) examined and found sup-
port for the augmentation effects at the group and organizational levels.
Furthermore, Judge and Piccolo (2004) showed that transformational lead-
ership had an augmentation effect on employee attitudes over contingent
reward but no effect on leader job performance, suggesting the existence of
possible boundary conditions of the augmentation hypothesis. Judge and
Piccolo did not test the augmentation hypothesis for follower performance.
Accordingly, the generalizability of the augmentation effect remains unclear
across levels of analysis and across various performance criteria (task and
contextual performance). Thus, the fourth purpose of our research, testing
the generalizability of the augmentation effect, will allow us to not only test
the overall validity of transformational leadership but also potentially make
critical refinements to the theory.
In sum, after decades of research on transformational leadership, the num-
ber of primary studies that link transformational leadership and performance
is sufficient to allow us to better understand this relationship across crite-
rion type and levels of analysis. This research has the potential to clarify
the precise ways in which transformational leadership impacts performance
and may increase the practical utility of transformational leadership theory
(Corley & Gioia, 2011). Moreover, by comparing the relative effects of
transformational and transactional leadership on different types and levels
of performance, we can learn more about how these two types of leadership
may work together to facilitate both effective performance across types and
levels.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
230 Group & Organization Management 36(2)
Hypotheses
Transformational Leadership
and Follower Performance at the Individual Level
According to Bass (1985), transformational leaders exhibit four primary
behaviors. First, through the behavior of inspirational motivation, transforma-
tional leaders develop and articulate a shared vision and high expectations that
are motivating, inspiring, and challenging. Second, transformational leaders
exhibit the behavior of idealized influence, serving as a role model by acting
in ways that are consistent with the articulated vision. Third, transformational
leaders intellectually stimulate their followers to challenge existing assump-
tions and solicit followers’ suggestions and ideas. Finally, through the behavior
of individualized consideration, transformational leaders attend to the needs
of their followers and treat each follower as a unique individual, thereby foster-
ing feelings of trust in and satisfaction with the leader (Podsakoff, MacKenzie,
Moorman, & Fetter, 1990). Taken together, these transformational leadership
behaviors are expected to motivate followers to perform at higher levels.
Although theory suggests that transformational leaders motivate individuals
to achieve higher levels of performance (Bass, 1985), this relationship may
differ across specific performance criteria. Transformational leadership has
been theoretically and empirically linked with task performance (i.e., in-role
performance, focal performance), contextual performance (i.e., extrarole per-
formance, organizational citizenship behavior), and creative performance
(e.g., Gong, Huang, & Farh, 2008; Shin & Zhou, 2007). Therefore, the pres-
ent study investigates transformational leadership as it relates to these three
performance criteria.
Task performance refers to work behaviors that are stipulated by a formal
job description (Borman & Motowidlo, 1993; Harrison, Newman, & Roth,
2006). Transformational leaders enable and motivate their followers to fulfill
their assigned job duties in a number of ways. First, transformational leaders
link followers’ work roles to a compelling vision of the future of the organiza-
tion, causing followers of transformational leaders to view their work as more
meaningful and significant and thus increasing its intrinsic motivating poten-
tial (Bono & Judge, 2003; Zhu, Avolio, & Walumbwa, 2009). In addition,
transformational leaders instill in their followers a belief that they can achieve
the goals that are set for them (Shamir et al., 1993), and these increased levels
of self-efficacy positively affect performance (Bandura, 1986). Finally, trans-
formational leaders serve as effective coaches and mentors to their followers,
providing them with the support and tools that they need to accomplish their
jobs (e.g., Howell & Hall-Merenda, 1999). For these reasons, research studies
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 231
have consistently found positive relationships between transformational leader-
ship and task performance (e.g., Liao & Chuang, 2007; MacKenzie, Podsakoff, &
Rich, 2001).
Contextual performance, however, refers to voluntarily motivated work
behaviors that go beyond prescribed job roles but contribute to the psycho-
logical and social contexts around the job (Borman & Motowidlo, 1993;
Harrison et al., 2006). As Bass (1985) noted, transformational leaders moti-
vate followers to work for the good of the group by increasing social identifi-
cation, thereby inspiring followers to engage in altruistic behaviors (e.g.,
helping coworkers in need of help) and to dedicate themselves to their jobs and
organizations (e.g., working extra hours and promoting organizational public
images; Bass & Avolio, 1993; Kouzes & Posner, 2002; Pillai, Schreisheim, &
Williams, 1999; Sosik, 2005). As followers of transformational leaders inter-
nalize the goals of the collective, they likely view actions that support the
psychological and social context of their work as meaningful and consistent
with their own self-concept. Furthermore, transformational leaders influence
their followers to engage in contextual performance by serving as role models
who are willing to sacrifice their own interests for the collective good and by
bolstering a sense of group belongingness and cohesion (Podsakoff et al., 1990;
van Knippenberg & van Knippenberg, 2005). Thus, we expect that transfor-
mational leadership is also positively related to contextual performance.
In addition, transformational leaders encourage and intellectually stimu-
late followers to challenge the status quo, question assumptions, take risks,
suggest innovative ideas, and engage in divergent thinking (Bass, 1985).
Transformational leaders empower their followers to be problem solvers,
such that employees grow in their creativity by learning from their failures
and experimenting with various options without fear of failure (e.g., Jung,
2001; Jung, Chow, & Wu, 2003; Shin & Zhou, 2003). Thus, we expect that
transformational leadership is positively related to creative performance. In
sum, we posit the following hypotheses:
Hypothesis 1: Transformational leadership is positively related to indi-
vidual follower (a) task, (b) contextual, and (c) creative performance.
Although transformational leadership theory suggests a link between trans-
formational leadership and multiple types of individual follower performance,
we know little about the relative magnitude of these relationships. Drawing
from research on the determinants of task and contextual performance, we
expect that transformational leadership has a stronger relationship with contex-
tual performance as compared to task performance. As noted above, the
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
232 Group & Organization Management 36(2)
exercise of contextual performance is voluntary and volitional in nature
whereas the exercise of task performance is heavily constrained by the learning
and mastery of task-related ability, knowledge, and skills. In other words, task
performance is mainly determined by “can-do” factors (e.g., ability, knowledge,
skills), whereas contextual performance is mainly determined by “will-do”
factors (e.g., motivation; Borman & Motowidlo, 1993). As such, we expect
that followers’ responses to transformational leadership are more strongly
reflected in their contextual performance than their task performance because
transformational leaders affect their followers more by raising their levels of
positive emotion and motivation than their levels of task-related skills and
ability. In fact, Podsakoff et al. (1990; see also Pillai et al., 1999) argued that
the true essence of transformational leadership is in enabling followers to
engage in extrarole behaviors. Thus, it is expected that transformational lead-
ership, operating through its motivational effects, has a stronger relationship
with contextual performance than with task performance.
Hypothesis 2: Transformational leadership has a stronger positive rela-
tionship with individual follower contextual performance than with
individual follower task performance.
Transformational Leadership and
Performance at the Team and Organizational Levels
Transformational leadership theory emphasizes the critical role of transfor-
mational leaders in improving performance across all levels of organizations
(House & Aditya, 1997; Yammarino et al., 2005). Although more empirical
work has focused on testing the relationship between transformational leader-
ship and follower performance at the individual level (Lim & Ployhart, 2004),
transformational leadership theory suggests a number of ways in which trans-
formational leaders may impact team and organizational performance as well.
At the team level, transformational leaders communicate a vision for the group
and motivate team members to work toward the collective vision (Bass, 1985).
Motivation to achieve team-level goals is enhanced by the increased levels
of social identification that are characteristic of followers of transformational
leaders (Kark, Shamir, & Chen, 2003). Transformational leaders also express
their confidence that teams will achieve their goals, leading to higher levels
of team potency (Bass et al., 2003; Schaubroeck, Lam, & Cha, 2007). Further-
more, transformational leaders encourage higher levels of team cohesion (Bass
et al., 2003), which facilitates coordination and cooperation among group
members. Thus, we propose the following hypotheses:
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 233
Hypothesis 3: Transformational leadership is positively related to team-
level performance.
At the organizational level, transformational leaders once again positively
influence performance through the communication of a vision that serves to
motivate employees and align their efforts (Bass, 1985). Transformational
leaders may also positively affect organizational performance through their
direct leadership of the top management team. By increasing team cohesion,
motivation, and goal congruence within the top management team, transfor-
mational leaders facilitate higher levels of organizational performance
(Colbert, Kristof-Brown, Bradley, & Barrick, 2008; Waldman & Yammarino,
1999). Furthermore, transformational leaders at the top of organizations may
serve as role models for leaders at lower levels, encouraging (cascading down)
transformational leadership throughout the organization (Waldman & Yammarino,
1999). Finally, transformational leaders may influence organizational perfor-
mance through their impact on organizational climates, systems, and strategies,
resulting in work environments more conducive to transformational leader-
ship (Jung et al., 2003; Liao & Chuang, 2007).
Hypothesis 4: Transformational leadership is positively related to
organizational-level performance.
The Augmentation Effect of Transformational
and Transactional Leadership on Performance
In addition to proposing links between transformational leadership and per-
formance across levels, transformational leadership theory also suggests that
transformational leadership influences performance beyond the effect of
transactional leadership (Bass, 1985; Bass & Avolio, 1993). This augmenta-
tion effect refers to the extent to which “transformational leadership styles
build on the transactional base in contributing to the extra effort and performance
of followers(Bass, 1998, p. 5). According to Bass (1985), true transformational
leaders display both transformational and transactional leadership behaviors,
and transformational leadership should explain variance in performance
beyond the effects of transactional leadership. Results of early studies of the
augmentation effect of transformational leadership over transactional leader-
ship were generally supportive. For example, Hater and Bass (1988) and
Waldman, Bass, and Yammarino (1990) found that transformational leader-
ship was significantly related to individual follower performance, controlling
for transactional leadership.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
234 Group & Organization Management 36(2)
The augmentation hypothesis has been interpreted by some (e.g., Bycio,
Hackett, & Allen, 1995) as suggesting not only that transformational leader-
ship explains unique variance in outcomes beyond transactional leadership
but also that the reverse is not true. If transactional leadership does not pre-
dict performance beyond the effects of transformational leadership, this sug-
gests that the predictive power of transactional leadership is solely due to its
overlap with transformational leadership. However, two recent studies ques-
tion this interpretation of the augmentation hypothesis. Both Schriesheim,
Castro, Zhou, and DeChurch (2006) and Vecchio, Justin, and Pearce (2008)
found that contingent reward, the most important indicator of transactional
leadership, was significantly related to individual follower performance after
controlling for transformational leadership. Furthermore, when contingent
reward and a transformational leadership dimension (e.g., vision, modeling,
high expectations) were entered together as predictors of individual perfor-
mance, the transformational leadership dimensions were not significantly
related to performance. Thus, these studies question the augmentation effect
of transformational leadership and instead provide support for an augmenta-
tion effect of the most important dimension of transactional leadership: con-
tingent reward. The conflicting results in the literature may be due to sampling
error that cannot be corrected for in a single study. Hence, it is beneficial to
examine the augmentation effect using meta-analytic methods.
Specifically, to account for these conflicting findings, we suggest that
both the type and level of performance need to be considered to fully under-
stand the augmentation effects of transformational and transactional leadership.
For example, with regard to individual follower task performance, we pro-
pose that transformational leadership augments transactional leadership and
that transactional leadership augments transformational leadership primarily
because these two types of leadership work through different motivational
mechanisms. Transactional leaders (especially those using contingent reward
behaviors) clearly specify performance expectations and provide rewards for
the achievement of these expectations (Bass, 1985). Thus, the task perfor-
mance of followers is expected to be higher when they work with transac-
tional leaders because of the motivational effects of performance goals
(Locke & Latham, 1990) and because of the clear link between performance
and rewards (Vroom, 1964). Transformational leadership may further enhance
individual task performance through such mechanisms as increased effort-
accomplishment expectancies and enhanced meaningfulness of goal accom-
plishment (Shamir et al., 1993). Thus, we expect that both transactional leadership
and transformational leadership explain unique variance in individual follower
task performance.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 235
However, in predicting individual follower contextual performance, we
expect that the augmentation effect of transformational leadership over trans-
actional leadership is stronger than the augmentation effect of transactional
leadership over transformational leadership. As noted earlier, followers of
transformational leaders are likely to exhibit higher levels of contextual per-
formance due to their identification with the organization, their internaliza-
tion of group goals, and the presence of transformational leaders as role
models. In contrast, transactional leaders motivate employees by clarifying
goals and rewarding goal achievement. Given that contextual performance is
defined as voluntarily motivated behaviors that go beyond work roles, it is less
likely to be a part of formal goal setting and reward systems. Thus, it is
likely that the intrinsic motivation engendered by transformational leadership
will predict contextual performance above and beyond the more extrinsic
motivation of transactional leadership.
Hypothesis 5a: Transformational leadership explains unique variance
in individual follower task performance beyond the effects of trans-
actional leadership.
Hypothesis 5b: Transactional leadership explains unique variance in
individual follower task performance beyond the effects of transfor-
mational leadership.
Hypothesis 6: Transformational leadership explains unique variance in
individual follower contextual performance beyond the effects of
transactional leadership.
Although most research has examined the unique relationships of transformational
and transactional leadership with individual-level outcomes, augmentation effects
can be proposed for unit-level performance as well. However, the small
number of studies that examine the relationship between transactional leadership
and organizational performance prevented us from meta-analytically examining
augmentation effects at the organizational level. Therefore, we focus our discussion
on augmentation effects on team performance.
As discussed above, prior research has established a number of mecha-
nisms by which transformational leadership may influence team perfor-
mance. Transformational leaders provide direction for the team through
their vision, enhance motivation to work toward team objectives, and
encourage team potency and cohesion (Bass et al., 2003; Schaubroeck et al.,
2007). At the same time, it has been argued that transactional leadership (in
particular, contingent reward) may also be associated with higher levels of
team performance (Howell & Avolio, 1993). For example, transactional
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
236 Group & Organization Management 36(2)
leaders may motivate higher levels of team performance by clarifying
team goals and rewarding the team for the achievement of those goals.
Thus, because the effects of transformational and transactional leader-
ship on team performance occur through different mechanisms, it is
possible that each type of leadership may explain unique variance in team
performance.
However, while results of empirical research have generally supported
the augmentation effect of transformational leadership over and above
transactional leadership at the team level, they have not supported the aug-
mentation effect of transactional leadership over and above transforma-
tional leadership. For example, Bass and colleagues (2003) found that
transformational leadership affected platoon unit performance beyond the
effects of contingent reward behaviors (although only after some items
were removed). In a study of business-unit performance in a financial
institution, Howell and Avolio (1993) found that transformational leader-
ship predicted business unit performance 1 year later, controlling for trans-
actional leadership. Similarly, Rowold and Heinitz (2007) found that
transformational leadership augmented the effects of transactional leader-
ship on branch profit in a study of a large public transport company. Inter-
estingly, in two of these studies (Howell & Avolio, 1993; Rowold &
Heinitz, 2007), transactional leadership was negatively related to unit
performance after controlling for transformational leadership. In sum,
these results seem to suggest that the relationship of transactional leader-
ship and team performance is primarily due to the overlap between trans-
actional and transformational leadership and that transactional leadership
explains little unique variance in team performance. Thus, we hypothesize
as follows:
Hypothesis 7: Transformational leadership explains unique variance in
team performance beyond transactional leadership.
Method
Literature Search
We conducted an extensive electronic and manual search for both published
and unpublished transformational leadership studies to minimize publica-
tion bias (Cooper, 2003). For the electronic search, we searched electronic
databases such as EBSCO, PsycINFO, Web of Science, ABI/Inform, Dis-
sertation Abstracts, and Google Scholar using combinations of the following
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 237
keywords: transformational leadership, charismatic leadership,
1
charisma,
transactional leadership, contingent reward, management by exception,
performance, citizenship behavior, productivity, profit, creativity, and inno-
vation. For the manual search, we consulted the reference lists of existing
narrative and meta-analytic reviews,
2
all issues of major journals published
as of August 2010 to account for articles that were not yet included in the
electronic databases, and the Society for Industrial and Organizational Psy-
chology and the Academy of Management conference programs for 2007,
2008, 2009, and 2010. Finally, we also searched for possible unpublished
and in-press studies by sending email solicitations to members of the Acad-
emy of Management listservs.
Inclusion Criteria
Primary studies had to meet the following criteria to be included in the meta-
analysis. First, we focused on samples of adults working in organizational
settings. Second, we only included studies in which study participants were
direct leader–follower dyads and leadership was naturally occurring under
field conditions (i.e., not experimentally manipulated). Third, we included
studies measuring transformational or transactional leadership and one or
more of the following non-self-report performance criteria: task, contextual,
and creative performance at individual, team, or organizational levels. Stud-
ies based on self-report performance measures were not included because of
concerns about common source bias (Podsakoff, MacKenzie, Lee, & Podsa-
koff, 2003) as well as potential self-enhancement biases in self-report
performance measures. Finally, studies without the statistical information
needed to calculate the correlations among the variables were excluded. It is
noteworthy that we did not focus on transformational leadership that was
measured by any particular scale, such as The Multifactor Leadership
Questionnaire (MLQ; Bass & Avolio, 1995). We think it is methodologi-
cally and theoretically important to include transformational leadership
measured with different scales to triangulate the generalizability of findings
in this study. We did not discriminate against either survey or experimental
design. However, as few experimental studies were conducted using field
samples (working adults), our second inclusion criterion may have unin-
tendedly excluded studies based on experimental designs using undergrads.
As a result, a total of 113 primary studies with 117 independent samples,
including 78 published articles and 35 unpublished doctoral dissertations and
working papers, met the inclusion criteria and were included in the
meta-analysis.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
238 Group & Organization Management 36(2)
Coding of Information
Information needed to test our hypotheses and potential moderators was
coded from each study. Effect sizes (e.g., Pearson correlations), sample
sizes, reliability coefficients, and moderator information (e.g., study setting,
leader level) were coded. It should be noted that for individual-level out-
comes, the sample size referred to the number of followers; for team or
organizational level outcomes, the sample size referred to the number of
teams or organizations.
For moderator analyses, study setting was coded as private if studies were
conducted in for-profit organizations (e.g., Fortune 500 companies) or as
public if studies were conducted in not-for-profit organizations (e.g., government
agencies, colleges, military). Leader level (position) was coded as supervisory
level (e.g., first-line managers) and mid- or upper level (e.g., CEOs, execu-
tives) according to the information provided in the method section. Geographic
regions or countries where studies were conducted were recorded. Studies
were coded as cross-sectional if variables of interest were measured at the
same period. Studies were coded as longitudinal when criteria were mea-
sured after leadership variables. Published and in press studies were coded
as published, whereas unpublished dissertations, conference papers, and
working papers were coded as unpublished. Finally, survey instruments that
were used to measure leadership variables were also coded into two catego-
ries: Bass and his associates’ (e.g., MLQ-Form 5X short; Bass & Avolio,
1995) and all the others’ (e.g., Transformational Leadership Inventory
[TLI]; Podsakoff et al., 1990).
The coding process was very straightforward and needed a few subjective
judgments. Nevertheless, to maximize coding accuracy, the first two authors
independently coded included studies and compared their coding results.
The agreement rate was 93%. Most of the discrepancies revolved around sub-
jective judgments and were resolved through discussion. The remaining dis-
crepancies were clerical errors, which were corrected with reference to
original studies.
Measurements of Variable
Independent variables. The majority of leadership variables (i.e., transfor-
mational leadership, charismatic leadership, contingent reward, management-
by-exception [MBE]-active, MBE-passive) were measured with various
forms of MLQ by Bass and his associates (e.g., Bass & Avolio, 1995) across
levels of analysis. Specifically, for individual-level relationships, 48 (or 77%)
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 239
studies used Bass and his associates’ scales to measure transformational or
charismatic leadership, 11 (or 18%) studies used the TLI (Podsakoff et al.,
1990), and the remaining 3 studies (or 5%) used scales developed by Conger
and Kanungo (1994), Pearce and Sims (2002), and Wang and Howell (2010),
respectively. For team-level relationships, 29 (or 85%) studies used Bass and
his associates’ scales to measure transformational or charismatic leadership,
3 (or 9%) used Podsakoff et al.’s (1990) TLI, and the remaining 2 studies (or
6%) used scales developed by Pearce and Sims (2002) and Wang and Howell
(2010), respectively. For organizational-level relationships, 20 (or 74%) stud-
ies used Bass and his associates’ scales to measure transformational or charis-
matic leadership, 3 (or 11%) studies used Podsakoff and colleagues’ TLI, the
remaining 4 studies (or 15%) used scales developed by other researchers (e.g.,
Cox, 1994; Pearce & Sims, 2002). At the individual level of analysis, all but
one study asked direct reports to assess their supervisors’ transformational or
charismatic leadership. At the team level of analysis, all but one study aggre-
gated team members’ reports of team leaders’ transformational or charismatic
leadership behavior. At the organizational level of analysis, the majority of
studies (23 or 85%) aggregated top management team members’ or employ-
ees’ reports of their organizational leaders’ (e.g., CEO) transformational or
charismatic leadership. Most of these studies provided acceptable justifica-
tions for aggregation. The remaining four studies asked either a HR manger
or a senior executive to report the CEO’s transformational leadership or a
CEO to assess his top management team’s transformational leadership.
Dependent variables. Reliable scales and objective measures were used to
assess the performance criteria of interest in most studies. At the individual
level of analysis, task performance includes role-based performance (Bor-
man & Motowidlo, 1993), such as in-role performance, sales performance,
and service performance. Among the 30 studies, objective measures (mainly
sales performance) of follower task performance were used in 7 studies.
3
Nonself (mainly supervisor) ratings were used in the remaining 23 studies.
William and colleaguesscales (e.g., William, 1989; Williams & Anderson,
1991) were most frequently used (nine studies), followed by scales measur-
ing followers quality and quantity of work (six studies), and Liao and
Chuang’s (2004) scale measuring service performance (two studies). Con-
textual performance consists of explicit OCB and extrarole behavior. No
studies used implicit measures of contextual performance such as prosocial
behaviors. Followers’ contextual performance was assessed by their super-
visors in all studies. Podsakoff and associates’ (e.g., Podsakoff et al., 1990)
scales were most frequently used (nine studies), followed by Organ and
colleagues’ (e.g., Organ, 1988; Smith, Organ, & Near, 1983) scales (five
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
240 Group & Organization Management 36(2)
studies), Williams and Anderson’s (1991) scale (four studies), and Lee and
Allen’s (2002) scale (two studies). Supervisors assessed their direct reports
creative behavior in all but one study. Scales by George and Zhou (2001)
and Welbourne, Johnson, and Erez (1998) were most frequently used (three
studies each), followed by Oldham and Cummings’ (1996) scale (two stud-
ies) and Tierney, Farmer, and Graen’s (1999) scale (two studies). Team
performance was objectively measured in 15 out of the 34 studies.
4
External
leaders, team leaders, or external experts (Bass et al., 2003) assessed team
performance in the remaining 19 studies. No any particular scale was used
more than twice. Organizational performance was objectively evaluated in
all studies. Performance indices such as productivity, return of asset, and
outputs of patents were used as indicators of organizational performance.
Meta-Analytic Techniques
We estimated the true-score correlations of transformational leadership and
follower performance using Hunter and Schmidt’s (2004) psychometric
random-effects meta-analysis methods. We corrected each effect size (corre-
lation) for measurement error in both variables using reliability coefficients
reported in the original study to be comparable with prior meta-analyses. For
meta-analytic calculations, we used the Hunter-Schmidt meta-analysis pack-
age program (VG6 Module–individual correction methods for correlations;
Schmidt & Le, 2004).
Psychometric corrections for measurement error are very important
because this study attempt to estimate “true-score (construct-level)
relationships across levels of analysis. Internal consistency reliability
(alpha) coefficients for both variables were used for individual-level
relationships and in few cases where a selected score model was used for
team- and organizational-level relationships (e.g., CEO perceptions of
transformational leadership at the organizational level; Garcia-Morales,
Matias-Reche, & Hurtado-Torres, 2008; Rogers, 2001). Since alpha coef-
ficients do not capture transient error and scale-specific (rater-specific)
error, they overestimate reliability and thus cause underestimates of true
score relationships (Le, Schmidt, & Putka, 2009). Nonetheless, we used
alpha coefficients to be directly comparable with previous meta-analyses
(e.g., Judge & Piccolo, 2004). ICC(2) as a reliability estimate for the aggre-
gate group or organizational mean scores (when a direct-consensus compo-
sition model [Chan, 1998] used for either transformational leadership or
performance for team and organizational level relationships) were used.
Relatedly, Whitman, Van Rooy, and Viswesvaran (2010) wrote that “ICC(2)
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 241
indicates the reliability of the unit-level means . . . Therefore, correcting for
ICC(2) should provide a meta-analytic estimate based on ‘true’ mean
scores(p. 56). The reliability of objective performance measures was not
reported in all but one primary study. Reliability estimates of objective
performance measures reported in previous meta-analyses were used. Spe-
cifically, .92 (Barrick & Mount, 1991) was used to correct for measurement
error in individual-level objective performance measures and .82 (Combs,
Liu, Hall, & Ketchen, 2006) was used to correct for measurement error in
team and organizational levels of performance measures; even when the
reliability of objective performance measures was set to one to be conser-
vative, we found that results changed by less than or equal to .02 in all
cases. For those studies that did not report estimated reliabilities, we used
an imputation procedure based on the average reliability estimated from the
other original studies that examined the same relationship at the same level
(e.g., Judge & Piccolo, 2004, p. 758).
Subgroup comparison method was used for moderator analyses (Hunter &
Schmidt, 2004). With reference to the moderator information coded, we
grouped studies into two categories and performed separate meta-analysis for
each of the subgroup. Prior research shows that this method has higher statis-
tical power to detect the presence of moderator than other methods (Hunter
& Schmidt, 2004, pp. 423-424; Schmidt, 2008).
To ensure that effect sizes included in our meta-analysis were statisti-
cally independent, we computed a composite correlation when original
studies reported multiple estimates of the correlations within a single sam-
ple (e.g., correlations between several facets of transformational leadership
and performance); otherwise, the average of the correlations was used to
be compared with prior meta-analyses (e.g., Judge & Piccolo, 2004).
This ensured that a sample contributed only once in each meta-analytic
estimate.
To test the augmentation hypotheses, usefulness analysis (Darlington,
1968) was conducted. This technique employs hierarchical regression analy-
sis to examine the incremental validity of one predictor over the other
predictor(s) with respect to a particular criterion. We also used general domi-
nance analysis (Budescu, 1993) to show the relative importance (weight) of
transformational and transactional leadership in the prediction of perfor-
mance criteria. Dominance analysis is an alternative method to test relative
importance of predictors when the predictors of interest are highly corre-
lated; in the presence of multicollinearity, regression coefficients as an index
of relative importance cause interpretation problems (see Budescu, 1993 for
more details).
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
242 Group & Organization Management 36(2)
Results
As shown in Table 2, transformational leadership is positively related to indi-
vidual follower performance across job performance criterion types (ρ
ˆ
= .25).
The 95% confidence interval (CI) and 80% credibility interval (CV) do not
include zero, which suggests that the overall validity estimate of transforma-
tional leadership for follower individual-level performance is distinguishable
from zero and is generalizable in most situations, respectively. More specifi-
cally, it was found that transformational leadership has positive relationships
with follower task (
ρ
ˆ
= .21), contextual (ρ
ˆ
= .30), and creative performance
(ρ
ˆ
= .21). Similarly, the 95% CIs and the 80% CVs for these performance
criteria do not include zero. These results are consistent with Hypothesis 1.
Apart from the three specific performance criteria tested as part of Hypothe-
sis 1, some studies did not distinguish individual-level performance criteria
from each other but focused on the relationship between transformational
leadership and follower overall or general job performance. Not surprisingly,
Table 2 shows that transformational leadership was also positively related
to follower general job performance (ρ
ˆ
= .18), with the 95% CI and 80% CV
excluding zero. Finally, as shown in Table 3, transformational leadership was
positively related to individual follower job performance and its two major
dimensions—that is, task and contextual performance—across moderators,
including study setting, leader level, geographical region, research design,
publication status, and leadership scale. None of the 95% CIs for these mod-
erators includes zero. Furthermore, given that the 95% CIs of the mean effect
sizes under each prescribed moderator overlap substantially with each other,
the relationships of transformational leadership with individual follower
performance were not moderated by these situational and methodological
variables.
Findings in Table 2 also support Hypothesis 2, in which we posited that
transformational leadership is more strongly related to follower contextual
performance than to follower task performance. The corrected mean correla-
tion for follower contextual performance (ρ
ˆ
= .30; 95% CI = .26, .34; 80%
CV = .18, .42) is 34% greater than that for follower task performance (ρ
ˆ
=
.21; 95% CI = .16, .26; 80% CV = .03, .38). We note that there is a very slight
overlap between the 95% CIs for these two performance criteria, indicating
that the two corrected mean correlations are considerably, though not fully,
different. However, because the two corrected mean correlations for task and
contextual performance were not based on completely independent samples,
we conducted the Hotelling-Williams Test, which compares two dependent
correlations (Judge & Piccolo, 2004, p. 759; Steiger, 1980). To use this test,
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 243
Table 2. Relationships of Transformational Leadership and Transactional Leadership
With Individual-Level Follower Performance
Criterion k N r
_
ρ
ˆ
SD
ρ
80% CV
ρ
95% CI
ρ
ˆ
Individual-level performance
Transformational leadership 62 16,809 .22 .25 .11 .10 .39 .21 .28
Contingent reward
a
50 9,108 .20 .22 .09 .10 .34 .19 .25
Management by
exception-active
5 525 –.11 –.13 .29 –.50 .25 –.40 .14
Management by
exception-passive
6 555 –.05 –.06 .13 –.23 .11 –.20 .07
Task performance
Transformational leadership 31 7,016 .19 .21 .14 .03 .38 .16 .26
Contingent reward
a
17 6,180 .26 .28 .06 .20 .36 .25 .31
Management by
exception-active
5 525 –.11 –.12 .30 –.51 .27 –.40 .16
Management by
exception-passive
6 555 –.07 –.08 .10 –.22 .05 –.20 .03
Contextual performance
Transformational leadership 28 7,970 .26 .30 .09 .18 .42 .26 .34
Contingent reward
b
5,568 .23
Management by
exception-active
2 356 –.22 –.29 .16 –.49 –.09 –.53 –.05
Management by
exception-passive
2 356 –.01 –.03 .19 –.27 .22 –.31 .26
Creative performance
Transformational leadership 14 3,728 .19 .21 .10 .08 .34 .15 .27
General performance
Transformational leadership 13 4,017 .16 .18 .07 .09 .27 .13 .23
Note: k = number of correlations; N = combined sample size; r
_
= sample-size weighted mean
uncorrected correlation;
ρ
ˆ
= estimated corrected mean correlation; SD
ρ
= estimated standard
deviation of true-score correlations; CV = credibility interval; CI = confidence interval.
a. Taken from Podsakoff, Bommer, Podsakoff, and MacKenzie (2006, table 4).
b. True-score correlation of contingent reward with the composite of the five facets of
citizenship behavior: altruism, courtesy, conscientiousness, sportsmanship, and civic virtue in
Podsakoff et al. (2006, table 4).
we first calculated the corrected mean correlation between task and contex-
tual performance among the dependent samples, which was found to be .69
( r
_
= .60; k = 19). The results of the Hotelling-Williams Test revealed that the
corrected mean correlation of transformational leadership with contextual
performance, when controlled for statistical dependence, would be higher than
that of transformational leadership with task performance as hypothesized in a
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
244
Table 3. Moderator Analyses for the Relationship of Transformational Leadership and Individual-Level Follower Performance
Study moderator k N r
_
ρ
ˆ
SD
ρ
80% CV
ρ
95% CI
ρ
ˆ
Study setting
Private Total 53 15,133 .21 .24 .12 .09 .40 .21 .28
Task 28 6,444 .19 .21 .13 .04 .38 .15 .26
Contextual 21 6,761 .26 .30 .09 .18 .42 .25 .34
Public Total 7 1,315 .25 .28 .00 .28 .28 .23 .33
Task 2 424 .28 .30 .05 .24 .36 .19 .41
Contextual 4 704 .26 .29 .00 .29 .29 .22 .35
Leader level
Supervisory Total 45 10,470 .22 .26 .13 .10 .42 .22 .30
Task 23 4,644 .17 .19 .16 –.01 .39 .12 .26
Contextual 19 4,442 .28 .32 .12 .17 .47 .26 .38
Mid- or upper-level Total 10 2,585 .17 .19 .07 .09 .28 .13 .25
Task 4 610 .21 .22 .10 .09 .35 .10 .35
Contextual 2 292 .25 .28 .02 .25 .31 .17 .39
Geographic region
North America Total 37 10,257 .22 .24 .12 .09 .40 .20 .29
Task 19 4,730 .17 .19 .12 .03 .34 .13 .25
Contextual 18 5,768 .25 .29 .06 .21 .37 .25 .33
East Asia Total 16 4,605 .20 .23 .04 .18 .28 .19 .26
Task 6 1,312 .15 .16 .06 .08 .24 .09 .23
Contextual 5 921 .25 .29 .00 .29 .29 .23 .35
Research design
Cross-sectional Total 21 4,327 .21 .24 .08 .14 .35 .20 .29
Task 10 1,429 .19 .21 .14 .03 .39 .11 .31
Contextual 6 993 .29 .34 .00 .34 .34 .29 .40
(continued)
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
245
Table 3. (continued)
Study moderator k N r
_
ρ
ˆ
SD
ρ
80% CV
ρ
95% CI
ρ
ˆ
Longitudinal Total 37 11,434 .22 .25 .13 .08 .41 .20 .29
Task 21 5,587 .19 .21 .13 .04 .38 .14 .27
Contextual 21 6,786 .26 .30 .10 .17 .43 .25 .35
Publication status
Published Total 41 10,764 .21 .24 .10 .12 .36 .20 .27
Task 24 5,587 .19 .21 .11 .07 .34 .16 .26
Contextual 18 5,893 .25 .28 .05 .21 .35 .25 .32
Unpublished Total 21 6,045 .22 .26 .14 .08 .44 .20 .32
Task 7 1,429 .19 .21 .21 –.06 .48 .04 .37
Contextual 10 2,077 .31 .35 .14 .16 .53 .25 .44
Leadership scale
Bass and his associates’ Total 48 11,509 .20 .23 .11 .09 .37 .19 .26
Task 24 4,531 .18 .20 .13 .03 .36 .14 .26
Contextual 18 3,712 .25 .28 .07 .19 .36 .23 .32
Others’ Total 14 5,300 .25 .29 .11 .15 .43 .22 .35
Task 7 2,485 .21 .24 .14 .06 .42 .13 .35
Contextual 10 4,258 .33 .39 .14 .21 .58 .30 .49
Note: k = number of correlations; N = combined sample size; r
_
= sample-size weighted mean uncorrected correlation; ρ
ˆ
= estimated corrected mean
correlation; SD
ρ
= estimated standard deviation of true-score correlations; CV = credibility interval; CI = confidence interval.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
246 Group & Organization Management 36(2)
“typical primary” study setting (t = 2.00, p < .05). Finally, as recommended
by Hunter and Schmidt (2004), we tested this moderating effect of type of
criterion across various conditions in a hierarchical manner to ensure that the
observed difference between the relationship of transformational leadership
with follower task and contextual performance was not confounded by situ-
ational and methodological variables. Table 3 shows that transformational
leadership was more strongly related to contextual than task performance in
all conditions, except public sector (possibly due to second-order sampling
error, as the number of studies ranges from 2 to 4). Taken together, these
analyses provide support for Hypothesis 2.
The relationships of transformational leadership with team- and organizational-
level performance are reported in Table 4. Transformational leadership was
found to be positively related to overall team performance (
ρ
ˆ
= .33), with the
95% CI and 80% CV excluding zero. Thus, Hypothesis 3 was supported.
Furthermore, while the overall relationship of transformational leadership
with organizational-level performance was somewhat smaller (ρ
ˆ
= .27) than
for team-level performance but greater than for individual-level performance,
the 95% CI excluded zero, signifying that transformational leadership has a
positive relationship with organizational performance. However, the 80%
CV included zero, which indicates that there is a large variance in the cor-
rected correlations included in the analysis at the organizational level. Never-
theless, these results provide support for Hypothesis 4, which predicted that
transformational leadership is significantly associated with organizational
performance. We should note that we did not examine the effects of transfor-
mational leadership on different types of team- or organizational-level per-
formance because most studies focused on team task performance and
organizational financial performance and only a very limited number of stud-
ies examined team- and organizational-level contextual and creative perfor-
mance (k = 3, 4 for team-level contextual and creative performance,
respectively; k = 1, 4 for organizational contextual and creative performance,
respectively). This may be due to the fact that theory and research on the
multidimensionality of performance at the unit level is not as abundant or
robust as it is at the individual-level of analysis (Mathieu, Maynard, Rapp, &
Gilson, 2008).
Interestingly, results in Tables 2 and 4 demonstrate that transformational
leadership generally shows the highest relationship with team performance (ρ
ˆ
=
.33 for team performance vs. ρ
ˆ
= .25 and .27 for individual and organizational
performance, respectively). However, we were not able to directly compare
the observed differences across levels of analysis, given that data points are not
comparable across levels of analysis; specifically, for individual-, team-, and
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 247
organizational-level studies, one data point is one person, one team (work-unit,
group), and one organization, respectively.
Finally, to meta-analytically test the augmentation effect of transforma-
tional leadership (Viswesvaran & Ones, 1995), the meta-analytic relation-
ships of transactional leadership behaviors, including contingent reward,
active management-by-exception, and passive management-by-exception,
with individual, team and organizational performance were needed. Meta-
analytic estimates of the relationships of leader contingent reward behavior
with follower individual- and team-level performance criteria were taken from
Podsakoff, Bommer, Podsakoff, and MacKenzie (2006). However, no previous
meta-analyses have examined the relationships of active management-by-
exception and passive management-by-exception leadership behaviors with
follower individual-, team-, and organizational-level performance. In fact, as
shown in Tables 2 and 4, only a small number of primary studies have exam-
ined these relationships. Therefore, to avoid second-order sampling error
(Hunter & Schmidt, 2004) and to be more conservative (as noted, the effects
of management by exception on performance are negative), we excluded
these active and passive management-by-exception from the subsequent anal-
yses. Thus, we tested the augmentation effect of transformational leadership
over contingent reward transactional leadership behavior.
Table 4. Relationships of Transformational Leadership and Transactional Leadership
With Team- and Organizational-Level Performance
Criterion k N r
_
ρ
ˆ
SD
ρ
80% CV
ρ
95% CI
ρ
ˆ
Team-level performance
Transformational leadership 34 2,830 .24 .33 .07 .24 .42 .29 .37
Contingent reward
a
19 1,361 .21 .24 .24 –.07 .55 .15 .33
Management by
exception-active
5 254 –.16 –.18 .15 –.37 .00 –.36 –.01
Management by
exception-passive
5 254 .10 .13 .17 –.09 .34 –.07 .32
Organizational-level performance
Transformational leadership 27 2,408 .19 .27 .24 –.04 .59 .17 .37
Contingent reward
a
3 243 .13 .15 .05 .09 .21 .01 .28
Management by exception-
passive
2 205 –.19 –.25 .00 –.25 –.25 –.38 –.12
Note: k = number of correlations; N = combined sample size; r
_
= sample-size weighted mean
uncorrected correlation;
ρ
ˆ
= estimated corrected mean correlation; SD
ρ
= estimated standard
deviation of true-score correlations; CV = credibility interval; CI = confidence interval.
a. Taken from Podsakoff, Bommer, Podsakoff, and MacKenzie (2006, table 4).
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
248 Group & Organization Management 36(2)
Table 5. Relative Importance of Transformational and Transactional Leadership in
the Prediction of Follower Performance
Leadership
dimension
Individual-level task
performance
(ρ
ˆ
/b/dominance)
Individual-level
contextual
performance
(ρ
ˆ
/b/dominance)
Team-level
performance
(ρ
ˆ
/b/dominance)
Transformational
leadership
.21/–.04/28% .30/.32/71% .33/.38/72%
Contingent reward .28
a
/.31/72% .23
b
/–.03/29% .24
a
/–.07/28%
Overall R .28 .30 .33
ΔR over
Contingent
reward
.00 .07 .09
ΔR over
Transformational
leadership
.07 .00 .00
a. Taken from Podsakoff, Bommer, Podsakoff, and MacKenzie (2006, table 4).
b. True-score correlation of contingent reward with the composite of the five facets of
citizenship behavior: altruism, courtesy, conscientiousness, sportsmanship, and civic virtue in
Podsakoff et al. (2006, table 4).
Results of usefulness analysis in Table 5 showed that transformational
leadership did not explain unique variance beyond contingent reward in pre-
dicting follower individual-level task performance R = .00), whereas contin-
gent reward explained unique variance beyond transformational leadership in
predicting follower individual-level task performance R = .07). These results
failed to support Hypothesis 5a but are consistent with Hypothesis 5b. In con-
trast, consistent with Hypothesis 6, for individual-level contextual performance,
transformational leadership augmented the effect of contingent reward (ΔR =
.07), but contingent reward did not explain incremental variance beyond that
explained by transformational leadership R = .00). The overall Rs for
both analyses were similar: .28 for task performance and .30 for contextual
performance.
For team- and organizational- level performance, we were only able to test
the augmentation effect for team-level performance given the availability of
primary studies. As shown in Table 5, transformational leadership augmented
the effect of contingent reward in predicting team-level performance R =
.09). However, when transformational leadership was entered first, contingent
reward did not predict incremental variance in team-level performance (ΔR =
.00). The overall R was .33. Thus, Hypothesis 7 was supported.
Given the high meta-analytic correlation between transformational leader-
ship and contingent reward (.80; Judge & Piccolo, 2004), we also conducted
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 249
dominance analyses to show the relative importance of transformational lead-
ership and contingent reward in the prediction of performance outcomes
(see Table 5). Consistent with the usefulness analysis, contingent reward
accounted for more explained variance in individual-level task performance
than did transformational leadership (72% vs. 28%), while transformational
leadership accounted for more explained variance in individual-level contex-
tual performance (71% vs. 29%) and team performance (72% vs. 28%) than
did contingent reward.
Discussion
As implied by the title of his seminal book on transformational leadership,
Bass (1985) contended that transformational leaders motivate their followers
to “perform beyond expectations.” However, until now this hypothesized
link between transformational leadership and performance across individual-
level criterion types and levels of analysis had not been fully examined from
a meta-analytic perspective. Based on a large number of empirical studies
from the past quarter century, we thus sought to critically review the transfor-
mational leadership literature in relation to performance and meta-analytically
estimate the magnitude of influence of transformational leadership on perfor-
mance in organizations. In general, our results show that transformational
leadership exhibits a positive relationship with performance across several
individual performance criteria, including task, contextual, and creative per-
formance; the influence of transformational leadership is stronger for
contextual performance than for task performance across most study settings
examined. The positive relationship between transformational leadership
and individual performance holds across organizational type, leader level,
and geographic region. Moreover, transformational leadership has posi-
tive effects on performance across levels of analysis (i.e., individual, team,
and organizational levels) with the relationship being higher at the team
level and augments the effect of transactional leadership on individual-level
contextual performance and team-level performance (but not individual-level
task performance).
At this point, it seems informative to briefly compare the current findings
with prior meta-analytic findings as shown in Table 1. The effect sizes on per-
formance reported in this study are generally smaller than those on follower
attitudinal and motivational outcomes reported in previous meta-analyses (e.g.,
DeGroot et al., 2000; Dumdum, Low, & Avolio, 2002; Fuller, Patterson,
Hester, & Stringer, 1996; Judge & Piccolo, 2004). That is, transformational
leadership has a stronger effect on employee attitudes and motivation than on
employee performance; theoretically, this makes sense. The correlation
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
250 Group & Organization Management 36(2)
between transformational leadership and follower individual-level perfor-
mance that we estimated in this article (
ρ
ˆ
= .25, k = 62, N = 16, 809) was
slightly higher than the correlation between charisma and follower individual-
level performance (ρ
ˆ
= .21, k = 4, N = 715) reported by DeGroot et al. (2000).
This was likely due to the stronger relationship between transformational
leadership and contextual performance (ρ
ˆ
= .30). However, transformational
leadership had a smaller relationship with team performance (ρ
ˆ
= .33, k = 34,
N = 2,830) than its charismatic component in DeGroot et al. (2000; ρ
ˆ
= .49,
k = 7, N = 432). Transformational leadership had a larger relationship with
team and organizational performance (ρ
ˆ
= .27, k = 27, N = 2,408) than with
Judge and Piccolo’s (2004) combined team and organizational performance
(ρ
ˆ
= .26, k = 41, N = 6,197). These complex differences notwithstanding, we
believe that our estimates are more accurate given that our results clearly
distinguished team and organizational performance.
The results of our meta-analysis help clear the aforementioned
ambiguities (represented as research hypotheses) in transformational
leadership theory and thus contribute to the increased precision of trans-
formational leadership theory and practice. First, our results indicate that
transformational leadership is positively related to individual follower
performance, but perhaps more interestingly, that transformational lead-
ership is more strongly related to contextual performance than task per-
formance across most conditions. Given the increasingly interdependent
nature of contemporary work systems and processes, contextual perfor-
mance is becoming more critical than ever before for achieving collective
goals (Podsakoff, Whiting, Podsakoff, & Blume, 2009). As our results
indicate, transformational leadership serves as a crucial component in
encouraging contextual performance. However, our findings also suggest
potential moderators of the outcomes of transformational leadership that
should be addressed in future research given the nonnegligible true varia-
tion across studies (SD
ρ
) found. For example, given that transformational
leadership has stronger effects on contextual performance than task
performance, transformational leadership may be less effective in envi-
ronments with low degrees of interdependence where interpersonal coop-
eration among employees is less critical. Similarly, for contexts in which
task performance is the predominant outcome of interest, transforma-
tional leadership may be less effective than transactional leadership
(Keller, 2006). These and other research questions stemming from our
meta-analysis provide an interesting and useful guide for future research
endeavors.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 251
Second, our results show that transformational leadership has positive
relationships with performance not only at the individual level of analysis but
also at team and organizational levels (Dansereau et al., 2006), supporting
Klein and House’s (1995) argument that transformational leadership theory
“is truly a meso theory, a theory that cuts across organizational levels” (p.
197). Interestingly, the strongest relationship between transformational lead-
ership and performance occurred at the team level. Although the lack of com-
parability of data units across levels of analysis prevents us from conducting
direct comparisons of these relationships, the result that the relationship
between transformational leadership and performance is strongest at the team
level is consistent with several aspects of transformational leadership theory.
To begin, transformational leaders motivate followers by emphasizing the
followers’ ties to the collective group, fostering team identity and team
potency and efficacy (Gully, Incalcaterra, Joshi, & Beaubien, 2002; Kirk-
man, Chen, Farh, Chen, & Lowe, 2009; Shamir et al., 1993). Transforma-
tional leaders also function as role models that their followers desire to
emulate, thereby eliciting higher team commitment, cooperation, and perfor-
mance. Furthermore, transformational leaders care about their followers and
appeal to them on an emotional level. For example, when followers work
together in a team led by a transformational leader, “they have many oppor-
tunities to reinforce (vs. douse) each other’s commitment to their common
cause” through a process of social influence and emotional contagion (Klein
& House, 1995, p. 192; see also Barsade, 2002). Taken together, it seems
possible that transformational leaders boost up team performance by generat-
ing synergy among team members. Relatedly, Chun, Yammarino, Dionne,
Sosik, and Moon (2009) found an empirical support for Kark and Shamir’s
(2002) “dual effects of transformational leadership not only utilizing ideal-
ized influence and inspirational motivation for collective purpose but also
demonstrating intellectual stimulation and individually considerate behav-
iors for individual team members” (Chun et al., 2009, p. 704). That is, the
stronger effect of transformational leadership at the team level may piggy-
back on its individual-level effect.
Our examination of performance across levels suggests two important
directions for future research. The first is to further investigate the mediators
of transformational leadership at all levels of analysis. We argue that trans-
formational leadership may affect performance at different levels through dif-
ferent mechanisms. For example, transformational leadership may increase
individual-level performance through its effects on follower motivation and
attitudes (Bono & Judge, 2003; Piccolo & Colquitt, 2006). However, at the
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
252 Group & Organization Management 36(2)
team level, transformational leaders may affect team processes and emergent
states (e.g., team cohesion and potency), leading to higher team performance
(Bass et al., 2003). Finally, transformational leaders may influence organiza-
tional performance not only through affecting individual-level and team-level
processes and performance but also through affecting organizational cultures,
systems, and strategies (Jung et al., 2003; Liao & Chuang, 2007). Second, we
suggest that future research should examine the effect of transformational
leadership on performance across levels of analysis using hierarchically
nested data to examine trickle-down effects from the organizational level to
team and individual levels. Although our meta-analysis provides a prelimi-
nary step in this endeavor by reporting and discussing the differential effects
of transformational leadership on performance across levels of analysis, mul-
tilevel research incorporating nested data will truly enable researchers and
organizations to understand the multilevel effects of transformational leader-
ship in organizations. Third, the greater magnitudes of relationships (.33 and
.27 for the team and organizational levels vs. .25 at the individual level)
between transformational leadership and performance at supraindividual level
are once again noteworthy. As noted by Harter, Schmidt, and Hayes (2002),
“unit-level research also provides opportunities to establish linkages to out-
comes that are directly relevant to most businesses . . . item-level measurement
error is less of a concern” and has direct implications for “issues on which
managers can develop specific action plans” (pp. 268-269). That is, the supra-
individual relationships better represent unit-level validity, which is a more
direct and accurate estimate of business performance than individual-level
validity; furthermore, they provide more direct implications for organiza-
tional interventions.
Finally, our results suggest boundary conditions of the augmentation
hypothesis that transformational leadership explains variance in performance
outcomes beyond the effects of transactional leadership. Although Bass (1985,
1998) argued that transformational leadership augments the effect of transac-
tional leadership, he and other theorists did not specify boundary conditions
for the augmentation effect, such as the criteria on which the augmentation
effect exists. The usefulness analyses (Table 5) suggest that the augmentation
effect of transformational leadership may be contingent on the type of perfor-
mance criteria and the level of analysis. Transformational leadership does add
to the effects of contingent reward when predicting individual-level contex-
tual performance and team-level performance. These results are consistent
with the idea that transactional leadership is a foundation for transforma-
tional leadership (Bass, 1998). However, transformational leadership did not
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 253
incrementally predict follower task performance beyond the effects of con-
tingent reward leadership. Instead, contingent reward predicted follower-
individual-level task performance beyond transformational leadership. This
unexpected reverse augmentation effect indicates that contingent reward
leadership is needed to clarify goals and provide rewards to motivate followers
to perform the tasks that are expected of them. Thus, while contingent reward
alone is sufficient for motivating task performance, it is not enough to solicit
extrarole behaviors and team-level performance. Transformational leader-
ship, with its emphasis on achieving significant goals that challenge the status
quo as a team, is needed for these purposes.
Managerial Implications
In addition to providing guidance for future theory and research, our find-
ings also have several implications for practitioners. First, the moderately
positive meta-analytic relationships of transformational leadership with
various performance criteria across follower individual-, team-, and organi-
zational- levels and various situations (e.g., study setting, leader level,
geographic region) indicate that transformational leadership tends to be a
robust predictor of desirable performance outcomes across situations (Bass,
1997). Thus, organizations need to pay particular attention to intervention
programs that enhance their managers leadership style. Fortunately, prior
research shows that transformational leadership style is trainable, with the
results of such training being quite substantial (Barling, Weber, & Kelloway,
1996). However, even with the benefits of training, transformational leader-
ship can also be predicted from some individual differences such as
extraversion and emotional stability (Bono & Judge, 2004). Hence, we sug-
gest that organizations should focus on selecting and promoting individuals
on such traits for upper-level managerial positions as such individuals are
more likely to become transformational leaders.
Furthermore, although it is important for organizations to promote transfor-
mational leadership through selection practices and training interventions,
organizations should also bear in mind that they are likely to derive the most
value from transformational leadership in settings involving teamwork and
collaboration among employees (Podsakoff, Ahearne, & MacKenzie, 1997).
Indeed, given that transactional leadership by itself is effective at raising task
performance, there may be some settings in which transactional styles of lead-
ership are most desired, particularly when jobs involve little interdependence
or collaboration.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
254 Group & Organization Management 36(2)
Contributions Beyond Previous Meta-Analyses
Compared with prior meta-analyses on transformational leadership, this
meta-analysis makes several unique contributions to transformational leader-
ship research. To begin, our meta-analysis is the first to estimate the
relationships between transformational leadership and three individual-level
follower performance criteria—task, contextual, and creative performance.
As noted earlier, prior meta-analyses (e.g., Judge & Piccolo, 2004) shed light
on the motivational and attitudinal outcomes of transformational leadership,
but they did not examine behavioral outcomes of transformational leadership
in terms of follower performance. In addition, an early meta-analysis by
Lowe, Kroeck, and Sivasubramaniam (1996) on the MLQ showed a moder-
ate relationship between transformational leadership and self-reported extra
effort. In contrast, we only included non-self-report performance measures in
examining the relationship between transformational leadership and perfor-
mance, thereby yielding a less biased (by removing common method bias)
estimate of the magnitude of this relationship.
Second, our estimates of the overall transformational leadership–individual
performance relationship are based on a larger number of primary studies than a
previous meta-analytic study on charismatic leadership that examined this rela-
tionship (DeGroot et al., 2000). Thus, our results are less subject to second-order
sampling error and thus are more credible. Third, our study is the first to sepa-
rately examine relationships of transformational leadership with team and
organizational levels of performance. Prior meta-analyses investigating these
relationships combined performance criteria that differed in level of analysis and
content (Fuller et al., 1996; Judge & Piccolo, 2004). Finally, our study is the first
to meta-analytically examine the augmentation effect of transformational-trans-
actional leadership on various follower performance outcomes, including indi-
vidual-level task and contextual performance, and team-level performance.
Limitations and Directions for Future Research
This study also has some limitations. First, although our meta-analysis includes
both cross-sectional and longitudinal studies, the relationships are correla-
tional in nature. Thus, conclusions cannot be drawn regarding the causal
direction of the relationship between transformational leadership and perfor-
mance. Second, although we examined the relationship between transformational
leadership with three individual performance criteria, we were unable to inves-
tigate the relationship of transformational leadership and counterproductive
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 255
performance because of a limited number of primary studies (see Brown &
Treviño, 2000, for an exception). Given the important role of counterproduc-
tive work behaviors in influencing managers’ overall job performance ratings
(Rotundo & Sackett, 2002) and the ethical/moral implications of transforma-
tional leadership theory (e.g., Howell & Avolio, 1992), we suggest that more
studies should examine the effect of transformational leadership on various
follower counterproductive behaviors so that we can have a better understand-
ing of the relationship between transformational leadership and the whole
spectrum of individual performance.
Finally, the issue of differential effects of transformational leadership on
performance across levels of analysis should ideally be examined using the
same sample following multilevel analysis principles (Kozlowski & Klein,
2000) to ensure that differences in effect sizes across levels of analysis are
attributable to differences in levels of analysis alone.
Conclusion
In summary, the current study shows that transformational leadership is posi-
tively related to performance across criterion types and levels of analysis.
Moreover, transformational leadership has a stronger relationship with individual-
level contextual performance than with individual-level task performance.
Transformational leadership is also positively related to team- and organization-
level performance. Finally, transformational leadership has an augmentation
effect over contingent reward leadership in predicting follower individual-level
contextual and team-level performance. Overall, our results support that trans-
formational leaders lead not only their individual followers but also their
teams and organizations to achieve higher levels of performance.
Acknowledgments
The authors thank Greg L. Stewart and Terry L. Boles for constructive comments and
advice on an earlier version of the article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the authorship
and/or publication of this article.
Funding
The author(s) received no financial support for the research and/or authorship of this
article.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
256 Group & Organization Management 36(2)
Notes
1. Although transformational and charismatic leadership developed from distinct
traditions, many scholars have noted the overlap among the behaviors included in
these two constructs (Conger & Kanungo, 1998). In addition, Judge and Piccolo
(2004) found that the validities for studies of transformational and charismatic
leadership were similar. In the current study, there are only four, five, and eight
primary studies in which charismatic leadership was measured at the individual,
team, and organizational levels of analysis, respectively. Given these small num-
bers, we have reported the results of meta-analyses that combine transformational
and charismatic leadership studies; results with and without primary studies us-
ing charismatic leadership were found to be virtually the same at the individual
and team levels. We also conducted separate meta-analyses for primary studies
assessing transformational and charismatic leadership at the organizational level
of analysis where there were relatively enough number of studies for charismatic
leadership, and the 95% confidence intervals for the two results were found to
substantially overlap; the difference between the two meta-analytic estimate does
not statistically differ from zero (for transformational leadership,
ρ
ˆ
= .25, k = 19,
N = 1,785, 95% CI = .12, .37, 80% CV = –.09, .58; for charismatic leadership,
ρ
ˆ
= .36, k = 8, N = 623, 95% CI = .24, .48, 80% CV = .19, .53).
2. We thank an anonymous reviewer who recommended two relevant review studies.
3. Separate meta-analyses show that the relationship of transformational leadership
with follower task performance did not significantly differ regardless of whether
follower task performance was objectively or subjectively measured.
4. Separate meta-analyses show that the relationship of transformational leadership
with team performance did not significantly differ regardless of whether team
performance was objectively or subjectively measured.
References
References marked with an asterisk indicate studies included in the meta-analysis.
*Agle, B. R., Nagarajan, N. J., Sonnenfeld, J. A., & Srinivasan, D. (2006). Does CEO
charisma matter? An empirical analysis of the relationships among organizational
performance, environmental uncertainty, and top management team perceptions of
CEO. Academy of Management Journal, 49, 161-174.
*Arena, F. L., Ayman, R., & Daniels, A. L. (2008). Moderating effects of environ-
mental uncertainty on charismatic leadership and corporate profitability. Paper
presented at the 68th annual meeting of the Academy of Management, Anaheim,
CA.
Austin, J. T., & Villanova, P. (1992). The criterion problem: 1917-1992. Journal of
Applied Psychology, 77, 836-874.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 257
*Avolio, B. J., Howell, J. M., & Sosik, J. J. (1999). A funny thing happened on the
way to the bottom line: Humor as a moderator of leadership style effects. Academy
of Management Journal, 42, 219-227.
Avolio, B. J., & Yammarino, F. (2002). Introduction to, and overview of, transforma-
tional and charismatic leadership. In B. J. Avolio & F. Yammarino (Eds.), Trans-
formational and charismatic leadership: The road ahead (Vol. 2 of Monographs
in Leadership and Management). St. Louis, MO: Elsevier.
Bandura, A. (1986). The explanatory and predictive scope of self-efficacy theory.
Journal of Social & Clinical Psychology, 4, 359-373.
Barling, J., Weber, T., & Kelloway, E. K. (1996). Effects of transformational leader-
ship training on attitudinal and financial outcomes: A field experiment. Journal of
Applied Psychology, 81, 827-832.
Barrick, M. R., & Mount, M. K. (1991). The Big Five personality dimensions and job
performance. Personnel Psychology, 44, 1-26.
Barsade, S. G. (2002). The ripple effect: Emotional contagion and its influence on
group behavior. Administrative Science Quarterly, 47, 644-675.
Bass, B. M. (1985). Leadership and performance beyond expectations. New York,
NY: Free Press.
Bass, B. M. (1997). Does the transactional–transformational leadership paradigm
transcend organizational and national boundaries? American Psychologist,
52, 130-139.
Bass, B. M. (1998). Transformational leadership: Industrial, military, and educa-
tional impact. Mahwah, NJ: Lawrence Erlbaum.
Bass, B. M., & Avolio, B. J. (1993). Transformational leadership: A response to cri-
tiques. In M. M. Chemers & R. Ayman (Eds.), Leadership theory and research:
Perspectives and directions (pp. 49-80). San Diego, CA: Academic Press.
Bass, B. M., & Avolio, B. J. (1995). MLQ Multifactor Leadership Questionnaire for
research. Redwood City, CA: Mind Garden.
*Bass, B. M., Avolio, B. J., Jung, D. I., & Berson, Y. (2003). Predicting unit per-
formance by assessing transformational and transactional leadership. Journal of
Applied Psychology, 88, 207-218.
*Basu, R., & Green, S. G. (1997). Leader–member exchange and transformational
leadership: An empirical examination of innovative behaviors in leader–member
dyads. Journal of Applied Social Psychology, 27, 477-499.
*Baum, J. R., Locke, E. A., & Kirkpatrick, S. A. (1998). A longitudinal study of the
relation of vision and vision communication to venture growth in entrepreneurial
firms. Journal of Applied Psychology, 83, 43-54.
*Boerner, S., Eisenbeiss, S. A., & Griesser, D. (2007). Follower behavior and organi-
zational performance: The impact of transformational leader. Journal of Leadership
and Organizational Studies, 13, 15-26.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
258 Group & Organization Management 36(2)
*Bono, J. E., & Judge, T. A. (2003). Self-concordance at work: Toward understand-
ing the motivational effects of transformational leaders. Academy of Management
Journal, 46, 554-571.
Bono, J. E., & Judge, T. A. (2004). Personality and transformational and transactional
leadership: A meta-analysis. Journal of Applied Psychology, 89, 901-910.
Borman, W. C., & Motowidlo, S. J. (1993). Expanding the criterion domain to include
elements of contextual performance. In N. Schmitt & W. C. Borman (Eds.), Person-
nel selection in organizations (pp. 71-98). San Francisco, CA: Jossey-Bass.
*Brossoit, K. B. (2001). Understanding employee empowerment in the workplace:
Exploring the relationships between transformational leadership, employee per-
ceptions of empowerment, and key work outcomes (Unpublished doctoral disserta-
tion). The Claremont Graduate University, Claremont, CA.
*Brown, F. W., & Dodd, N. G. (1999). Rally the troops or make the trains run on time:
The relative importance and interaction of contingent reward and transformational
leadership. Leadership and Organization Development Journal, 20, 291-299.
Brown, M. E., & Treviño, L. K. (2000). Socialized charismatic leadership, values con-
gruence, and deviance in work groups. Journal of Applied Psychology, 91, 954-962.
Budescu, D. V. (1993). Dominance analysis: A new approach to the problem of relative
importance of predictors in multiple regression. Psychological Bulletin, 114, 542-551.
Burns, J. M. (1978). Leadership. New York, NY: Harper & Row.
Bycio, P., Hackett, R. D., & Allen, J. S. (1995). Further assessments of the Bass
(1985) conceptualization of transactional and transformational leadership. Journal
of Applied Psychology, 80, 468-478.
*Casimir, G., Waldman, D. A., Bartram, T., & Yang, S. (2006). Trust and the rela-
tionship between leadership and follower performance: Opening the black box in
Australia and China. Journal of Leadership and Organizational Studies, 12, 68-84.
*Cavazotte, F., & Bernardo, J. (2008). Transformational leadership and performance
in Brazil: A replication and test of mediating processes. Paper presented at the
68th annual meeting of the Academy of Management, Anaheim, CA.
Chan, D. (1998). Functional relations among constructs in the same content domain at
different levels of analysis: a typology of composition models. Journal of Applied
Psychology, 83, 234-246.
*Chen, C. H. V., & Tang, Y. Y. (2009). A multilevel investigation on mechanisms link-
ing transformational leadership and innovative behavior. Paper presented at the
69th annual meeting of the Academy of Management, Chicago, IL.
*Chun, J. U., Yammarino, F. J., Dionne, S. D., Sosik, J. J., & Moon, H. K. (2009).
Leadership across hierarchical levels: Multiple levels of management and multiple
levels of analysis. Leadership Quarterly, 20, 689-707.
*Cho, J. W., & Dansereau, F. (2010). Are transformational leaders fair? A multi-level
study of transformational leadership, justice perceptions, and organizational citi-
zenship behaviors. Leadership Quarterly, 21, 409-421.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 259
*Cohen, A., Tura, E. B., & Vashdi, D. R. (2008). Group characteristics as moderators
of the relationship between exchange variables and OCB. Paper presented at the
68th annual meeting of the Academy of Management, Anaheim, CA.
*Colbert, A. E. (2004). Understanding the effects of transformational leadership: the
mediating role of leader-follower value congruence (Unpublished doctoral dis-
sertation). University of Iowa, Iowa City.
*Colbert, A. E., Kristof-Brown, A., Bradley, B. H., & Barrick, M. R. (2008). CEO
transformational leadership: The role of goal importance congruence in top man-
agement teams. Academy of Management Journal, 51, 81-96.
Combs, J., Liu, Y., Hall, A., & Ketchen, D. (2006). How much do high-performance
work practices matter? A meta-analysis of their effects on organizational performance.
Personnel Psychology, 59, 501-528.
*Comer, L. B., Jolson, M. A., Dubinsky, A. J., & Yammarino, F. J. (1995). When the
sales manager is a woman: An exploration into the relationship between salespeople’s
gender and their responses to leadership styles. Journal of Personal Selling and
Sales Management, 15, 17-32.
Conger, J. A., & Kanungo, R. (1994). Charismatic leadership in organizations: Perceived
behavioral attributes and their measurement. Journal of Organizational Behavior,
15, 439-452.
Conger, J. A., & Kanungo, R. N. (1998). Charismatic leadership in organizations.
Thousand Oaks, CA: SAGE.
Cooper, H. (2003). Editorial. Psychological Bulletin, 129, 3-9.
Corley, K., & Gioia, D. A. (2011). Building theory about theory building: What con-
stitutes a theoretical contribution? Academy of Management Review, 36, 12-32.
*Cowen, S. S. (1990). A study of relationships between perceived leadership behaviors
of presidents at public four-year institutions of higher education in the U. S. and
the changes in FTE enrollment, perceptions of effectiveness, subordinate satis-
faction, and other factors of the presidency (Unpublished doctoral dissertation).
Gonzaga University, Spokane, WA.
Cox, J. F. (1994). The effects of superleadership training on leader behavior, subordi-
nate self-leadership behavior, and subordinate citizenship (Unpublished doctoral
dissertation). University of Maryland, College Park.
*Curphy, G. J. (1991). An empirical evaluation of Bass’ (1985) theory of transforma-
tional and transactional leadership (Unpublished doctoral dissertation). University
of Minnesota, Twin Cities Campus.
Dansereau, F., Cho, J., & Yammarino, F. J. (2006). Avoiding the “fallacy of the wrong
level”: A within and between analysis (WABA) approach. Group and Organiza-
tion Management, 31, 536-577.
Darlington, R. B. (1968). Multiple regression in psychological research. Psychologi-
cal Bulletin, 79, 161-182.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
260 Group & Organization Management 36(2)
*DeClerk, C. C. (2007). The relationship between retail store manager leadership
styles and employee generational cohort, performance, and satisfaction (Unpub-
lished doctoral dissertation). University of Phoenix, Phoenix, AZ.
*de Hoogh, A. H. B., Hartog, D. N. D., Koopman, P. L., Thierry, H., van den Berg, P. T.,
van der Weide, J. G., & Wilderom, C. P. M. (2004). Charismatic leadership, envi-
ronmental dynamism, and performance. European Journal of Work and Organiza-
tional Psychology, 13, 447-471.
DeGroot, T., Kiker, D. S., & Cross, T. C. (2000). A meta-analysis to review organiza-
tional outcomes related to charismatic leadership. Canadian Journal of Adminis-
trative Sciences, 17, 356-371.
*Degula, R. J. (1995). The relationship between attributional charismatic leadership
and organizational citizenship behavior. Journal of Applied Social Psychology,
25, 1652-1671.
*Dubinsky, A. J., Yammarino, F. J., Jolson, M. A., & Spangler, W. D. (1995). Transfor-
mational leadership: An initial investigation in sales management. Journal of Per-
sonal Selling & Sales Management, 15, 17-31.
Dumdum, U. R., Low, K. B., & Avolio, B. J. (2002). A meta-analysis of transforma-
tional and transactional leadership correlates of effectiveness and satisfaction: An
update and extension. In B. J. Avolio & F. Yammarino (Eds.), Transformational
and charismatic leadership: The road ahead (Vol. 2 of Monographs in Leadership
and Management; pp. 35-66). St. Louis, MO: Elsevier.
*Eisenbeiss, S. A., Knippenberg, D. V., & Boerner, S. (2008). Transformational leader-
ship and team innovation: Integrating team climate principles. Journal of Applied
Psychology, 93, 1438-1446.
*Ensley, M. D., Hmieleski, K. M., & Pearce, C. L. (2006). The importance of vertical
and shared leadership within new venture top management teams: Implications for
the performance of startups. Leadership Quarterly, 17, 217-231.
*Ensley, M. D., Pearce, C. L., & Hmieleski, K. M. (2006). The moderating effect of
environmental dynamism on the relationship between entrepreneur leadership
behavior and new venture performance. Journal of Business Venturing, 21, 243-263.
*Fu, P. P., Tsui, A., Liu, J., Song, J. W., Jiang, Y., Jia, L. D., . . . Wu, W. K. (n.d.). CEO
personal values and middle manager responses: The mediating role of leadership
behavior and organizational culture. Retrieved from http://www.baf.cuhk.edu.hk/
asia-aom/PDW06/T4_Fuetal.pdf
Fuller, J. B., Patterson, C. E. P., Hester, K., & Stringer, D. Y. (1996). A quantitative
review of research on charismatic leadership. Psychological Reports, 78, 271-287.
*Garcia-Morales, V. J., Matias-Reche, F., & Hurtado-Torres, N. (2008). Influence
of transformational leadership on organizational innovation and performance
depending on the level of organizational learning in the pharmaceutical sector.
Journal of Organizational Change Management, 21, 188-212.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
Wang et al. 261
George, J. M., & Zhou, J. (2001). When openness to experience and conscientious-
ness are related to creative behavior: An interactional approach. Journal of Applied
Psychology, 86, 513-524.
*Gerstner, C. R. (1998). Leadership relationships and work group effectiveness:
A multi-level empirical examination (Unpublished doctoral dissertation). The
Pennsylvania State University, University Park.
*Geyer, A. L. J., & Steyrer, J. M. (1998). Transformational leadership and objective
performance in banks. Applied Psychology: An International Review, 47, 397-420.
*Gong, Y. P., Hung, C. J., & Farh, J. L. (2009). Employee learning orientation, trans-
formational leadership, and employee creativity: The mediating role of employee
creative self-efficacy. Academy of Management Journal, 52, 765-778.
*Goodwin, V. L., Wofford, J. C., & Whittington, J. L. (2001). A theoretical and empir-
ical extension to the transformational leadership construct. Journal of Organiza-
tional Behavior, 22, 759-774.
*Griffith, G. (2004). Relation of principal transformational leadership to school staff
job satisfaction, staff turnover, and school performance. Journal of Educational
Administration, 42, 333-356.
Gully, S. M., Incalcaterra, K. A., Joshi, A., & Beaubien, J. M. (2002). A meta-analysis of
team-efficacy, potency, and performance: Interdependence and level of analysis as
moderators of observed relationships. Journal of Applied Psychology, 87, 819-832.
*Gumusluoglu, L., & Ilsev, A. (2009). Transformational leadership, creativity, and
organizational innovation. Journal of Business Research, 62, 461-473.
*Hacket, R., Farh, J. L., Chen, Z. J., Wang, A., & Cheng, B. S. (2008). Trans-
formational leadership and OCB: A test of a moderated mediated model.
Paper presented at the 68th annual meeting of the Academy of Management,
Anaheim, CA.
*Han, J., Liao, H., Taylor, S. M., & Kim, S. (2010). The effects of HRM and organiza-
tional context on transformational leadership and team performance. Paper presented
at the annual meeting of the Academy of Management, Montreal, Quebec, Canada.
Harrison, D. A., Newman, D. A., & Roth, P. L. (2006). How important are job atti-
tudes? Meta-analytic comparisons of integrative behavioral outcomes and time
sequences. Academy of Management Journal, 49, 305-325.
Harter, J. K., Schmidt, F. L., & Hayes, T. L. (2002). Business-unit-level relationship
between employee satisfaction, employee engagement, and business outcomes:
A meta-analysis. Journal of Applied Psychology, 87, 268-279.
*Hater, J. J., & Bass, B. M. (1988). Superiors’ evaluations and subordinates’ perceptions
of transformational and transactional leadership. Journal of Applied Psychology, 73,
695-702.
*Hoffman, D. V., & Johnes, L. M. (2005). Leadership, collective personality, and
performance. Journal of Applied Psychology, 90, 509-522.
at The University of Iowa Libraries on August 26, 2011gom.sagepub.comDownloaded from
262 Group & Organization Management 36(2)
House, R. J., & Aditya, R. N. (1997). The social scientific study of leadership: Quo
vadis? Journal of Management, 23, 409-473.
Howell, J. M., & Avolio, B. J. (1992). The ethics of charismatic leadership: Submission
or liberation? Academy of Management Executive, 6, 43-54.
*Howell, J. M., & Avolio, B. J. (1993). Transformational leadership, transactional
leadership, locus of control, and support for innovation: Key predictors of con-
solidated business-unit performance. Journal of Applied Psychology, 78, 891-902.
*Howell, J. M., & Hall-Merenda, K. E. (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.
*Hult, G. T. M., Ketchen, D. J., Jr., & Chabowski, B. R. (2007). Leadership, the buy-
ing center, and supply chain performance: A study of linked users, buyers, and
suppliers. Industrial Marketing Management, 36, 393-403.
Hunter, J. E., & Schmidt, F. L. (2004). Methods of meta-analysis: Correcting error
and bias in research findings (2nd ed.). Thousand Oaks, CA: SAGE.
Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional leadership: A
meta-analytic test of their relative validity. Journal of Applied Psychology, 89, 755-768.
Jung, D. I. (2001). Transformational and transactional leadership and their effects on
creativity in groups. Creativity Research Journal, 13, 185-195.
*Jung, D. I., Chow, C., & Wu, A. (2003). The role of transformational leadership in
enhancing organizational innovation: Hypotheses and some preliminary findings.
Leadership Quarterly, 14, 525-544.
*Jung, D. I., Wu, A., & Chow, C. (2008). Towards understanding the direct and indi-
rect effects of CEOs’ transformational leadership on firm innovation. Leadership
Quarterly, 19, 582-594.
Kark, R., Shamir, B., & Chen, G. (2003). The two faces of transformational leader-
ship: Empowerment and dependency. Journal of Applied Psychology, 88, 246-255.
Kark, R., & Shamir, B. (2002). The dual effect of transformational leadership:
Priming relational and collective selves and further effects on followers. In
B. J. Avolio & F.J. Yammarino (Eds.), Transformational and charismatic leader-
ship: The road ahead (pp. 67 -91). Oxford, UK: Elsevier Science.
*Kearney, E., & Gebert, D. (2009). Managing diversity and enhancing team Out-
comes: The promise of transformational leadership. Journal of Applied Psychology,
94, 77-89.
*Keller, R. T