The current issue and full text archive of this journal is available on Emerald Insight at:

www.emeraldinsight.com/0262-1711.htm

Leadership
Leadership self-efficacy: self-efficacy
review and leader
development implications
Laura Paglis Dwyer 637
Schroeder School of Business,
Received 9 March 2019
University of Evansville, Evansville, Indiana, USA Revised 10 May 2019
Accepted 16 August 2019

Abstract
Purpose – The purpose of this paper, on self-efficacy and leadership, has two objectives. First, it
comprehensively reviews approximately 25 years of research on leadership self-efficacy (LSE), beginning
with LSE measurement and related criticisms. Findings concerning LSE’s relationships with leader
effectiveness criteria, as well as individual and contextual influences on LSE, are presented. Second, it
examines the evidence on efficacy enhancement interventions and offers some preliminary recommendations
for increasing LSE through leadership development programs.
Design/methodology/approach – The author conducted a comprehensive literature review of the existing
research on LSE, covering the main contributors to this research stream and their findings.
Findings – The review revealed substantial diversity in LSE construct development and measurement
approaches. Regarding LSE and leader effectiveness, many studies reported positive relationships with
potential, performance and behavioral ratings of leaders. Collective (team) efficacy has emerged as a
significant mediator between LSE and group performance. Influences on LSE include several of the Big Five
personality traits, while contextual antecedents are under-researched, and potentially fruitful areas for further
study. Executive coaching and mentoring, as well as cognitive modeling techniques and training in
constructive thought patterns, received support for enhancing LSE in developing leaders.
Originality/value – This paper’s review and implications should be of substantial value to current and
future LSE researchers, as it summarizes past research, synthesizes the findings to draw out common themes
and consistent, corroborated findings, and identifies opportunities for future research. For practitioners, the
reviewed research on interventions for increasing LSE through leadership development programs provides
practical guidance.
Keywords Leadership development, Leadership, Self-efficacy
Paper type Literature review

Introduction
Self-efficacy, defined as an estimate of one’s ability to successfully execute the behavior
required to produce desired outcomes, was introduced by Bandura as a key concept in his
social cognitive theory (Bandura, 1977). Self-efficacy influences individuals’ willingness to
take on challenging assignments, the intensity of effort they apply, and their tenacity and
persistence as they encounter obstacles to successful performance (Gist, 1987). Indeed,
meta-analysis supports its positive influence on work performance (Stajkovic and
Luthans, 1998). Not surprisingly, then, self-efficacy has been a popular subject for
research in organizations during the last several decades, with researchers applying the
basic concept in a wide variety of different areas. In the present paper, the specific focus is
the role self-efficacy plays in leadership in organizations. While researchers have adopted
variant definitions of leadership self-efficacy (LSE), an early one that is consistent with
most of the published studies is “one’s self-perceived capability to perform the cognitive
and behavioral functions necessary to regulate group process in relation to goal
achievement” (McCormick, 2001, p. 30). Put more simply, LSE is an individual’s confidence
Journal of Management
in his or her ability to lead others. Self-efficacy’s effect on a person’s desire to take on a Development
leadership role and its subsequent impact on leader behavior are compelling topics of Vol. 38 No. 8, 2019
pp. 637-650
research interest with inherent, practical usefulness. The first objective here, then, is to © Emerald Publishing Limited
0262-1711
review what has been learned about LSE since it was first introduced in the literature DOI 10.1108/JMD-03-2019-0073
JMD approximately 25 years ago. A second objective is to examine the research evidence on
38,8 methods for enhancing LSE, to develop recommendations for leadership development
programs. The review begins by looking at past research on methods for measuring LSE.

Measuring LSE: a variety of approaches

Researchers have taken a number of different approaches to measure a leader’s efficacy in
638 the role. Most LSE research references Bandura’s conceptualization of self-efficacy as a
relatively narrow, task-specific capability judgment as the basis for developing measures of
the construct. Indeed, one early approach simply asked study participants to rate their
confidence in leading a group toward the achievement of specific performance levels on a
given task (Kane et al., 2002). The most-frequent method for measuring LSE has been to
begin by identifying behaviors connected to essential leadership tasks. For example, Paglis
and Green (2002) created an LSE scale consisting of confidence ratings for the behaviors of
direction-setting, gaining followers’ commitment and overcoming obstacles to change.
Kane et al. (2002) also measured LSE via leader behavior self-ratings, pairing their task
performance level scale, above, with a second scale composed of items including efficacy for
building group member confidence, developing teamwork and “taking charge” when
necessary. Another LSE scale, similar to the prior two, featured six behavioral dimensions,
including starting and leading change processes, gaining consensus of group members and
motivating people (Bobbio and Manganelli, 2009). A particularly comprehensive approach,
which seems to fit best with the behavioral perspective, was undertaken by Anderson et al.
(2008). Beginning with 88 leader attributes derived from interviews and literature review,
Anderson et al. performed principal component analysis to identify 18 LSE dimensions,
a few examples being efficacy for the leader behaviors of change, solve, involve and serve.
While confidence judgments referencing leader behaviors seem to dominate LSE
measurement, other researchers have elected to measure it through self-ratings on
leadership skills, for instance, communication, planning, delegating and problem analysis
(Chemers et al., 2000; Ng et al., 2008). Finally, in contrast to the task, behavior and skill
perspectives, some studies have employed measures of LSE that are less specific in nature,
attempting to capture a leader’s sense of overall efficacy. Example items include “How easy
would it be for you to succeed in a leadership position?” (Singer, 1991) and “I feel confident
I can be an effective leader in most of the groups I work with” (Chan and Drasgow, 2001;
Feasel, 1995; Hendricks and Payne, 2007).
In sum, measurement approaches in the past LSE research can be sorted into two
categories by the level of specificity. The first category, with a narrower focus, has been
most frequently used and includes task, behavioral and skill-based scales. Reviewing the
behavioral and skill-based scales, common themes emerge, capturing LSE as a judgment of
one’s ability to identify and build consensus around group goals and work collaboratively
with followers to achieve them. The second, more general type of measurement, involves
assessing one’s confidence for performance in the leader role overall. Indeed, this division in
LSE conceptualization and measurement has been one of the primary criticisms levied
against this research stream, discussed next.

Criticisms: LSE construct development and measurement

As studies involving LSE began to appear regularly in the literature, criticisms have
emerged along three inter-related points. The first criticism is directed at the specificity of
the LSE construct and its measurement. As noted above, many LSE researchers referenced
Bandura’s description of self-efficacy as a task-specific capability judgment as their basis
for measuring LSE through self-ratings on tasks, behaviors, or skills connected to the
leadership role. Critics contend, however, that this approach is too narrow, deficient in
capturing the complexity of leadership in organizations. Those holding this view have
expressed it as a failure to adequately distinguish between “leader” and “leadership” Leadership
(Schruijer and Vansina, 2002), and likewise, between “leader self-efficacy” and “leadership self-efficacy
efficacy” (Hannah et al., 2008). The distinction is that a leader is an individual person
performing leader role behaviors, whereas leadership is a dynamic, relational phenomenon
that involves both leader and followers. Indeed, the vast majority of past LSE research fits
squarely within the former, individualistic perspective, that is, leader self-efficacy rather
than leadership efficacy. 639
A second, related point stems from a criticism that has been directed at leadership
research in general, namely, the paucity of studies that appropriately address levels-of-
analysis issues (Yammarino et al., 2005). Connecting with the previous point, these critics
call for more research attention to leadership efficacy. Specifically, a multi-level perspective
on leadership efficacy refers to individual (leader and follower), team and organizational
levels (Hannah et al., 2008). This critique does not dismiss the utility of studying LSE from
the individualistic perspective, as nearly all researchers have done to this point. Rather, it
advocates for moving forward and expanding the research domain by investigating
leadership efficacy through a multi-level lens.
Third, LSE research has been criticized as deficient in addressing contextual factors
(Hannah et al., 2008; Schruijer and Vansina, 2002). As with the levels-of-analysis issue, the
lack of attention to context has been a criticism levied against organizational behavior and
leadership research overall (Osborn et al., 2002; Rousseau and Fried, 2001). Specific to LSE, it
seems obvious that resource availability, the leader’s role discretion, followers’ abilities and
an organization’s culture, to name just a few context factors, may have an impact on leaders’
efficacy perceptions. However, LSE research that has considered potential contextual effects
is sparse and typically does so in a tangential manner. An exception is a relatively recent
study that explicitly focused on incorporating context into the LSE scale itself (Hannah et al.,
2012). Hannah et al. developed a new measure, leader self and means efficacy (LSME), with
“means efficacy” referring to people, resources and other environmental factors that may
influence a leader’s perceived efficacy. In addition to establishing the construct validity of
their LSME scale, Hannah et al. reported predictive validity for several leader effectiveness
outcomes, as noted in the next section.

LSE and leader effectiveness

Research has conclusively shown that self-efficacy perceptions positively impact work
performance (Stajkovic and Luthans, 1998). Naturally, investigators have sought to
determine whether this effect extends to leaders and their effectiveness. Research in the late
1980s, while not specifically referencing LSE, set the stage for this line of inquiry by
reporting a positive relationship between managers’ self-efficacy and their decision-making
performance (Wood and Bandura, 1989a). The first empirical research studies specifically
investigating LSE and its relationship with outcomes used a measure developed by Murphy
(1992). She summarized the findings from this initial research, noting that high LSE leaders
reported less stress and did not show the performance decline under stressful conditions
that leaders lower in LSE did. High LSE leaders were also found to respond more
productively to negative feedback and had better performing groups (Murphy, 2002).
Another very early study found that LSE was positively related to leaders’ evaluation of
LMX (leader–member exchange) quality with subordinates (Murphy and Ensher, 1999).
Subsequent research on LSE’s relationship with leader effectiveness has used a variety
of outcome measures, incorporating ratings by self, superiors, followers or peers. These
studies have been conducted both in the field and in the laboratory. In field research,
Chemers et al. (2000) found a positive relationship between ROTC cadets’ LSE scores and
instructors’ ratings of their leadership potential. Similarly, LSE has been positively linked
with superiors’ ratings of managers’ promotability (Seibert et al., 2017), as well as follower
JMD and self-ratings of frequency of leadership attempts (McCormick et al., 2002; Paglis and
38,8 Green, 2002). Several field studies have examined LSE’s influence specifically on observers’
ratings of leadership performance and have found positive relationships with superiors’
ratings (Chemers et al., 2000; Lester et al., 2011; Ng et al., 2008; Seibert et al., 2017; Semadar
et al., 2006) and peer ratings (Chemers et al., 2000). In contrast, Ali et al. (2018) found no
relationship between LSE and subordinate-rated leader effectiveness, although LSE was
640 positively correlated with self-reported effectiveness.
Adding to these findings, a field study investigated some other factors that inform the
LSE-leader effectiveness relationship (Courtright et al., 2014). This research reported
a positive correlation between LSE and direct reports’ ratings of their leaders’
transformational leadership behavior. As well, LSE affected leaders’ response to
challenging job assignments; specifically, leaders lower in LSE reacted more negatively to
job challenges by displaying more laissez-faire leadership behavior and reporting greater
emotional exhaustion.
Two field studies concerning LSE’s relationship with leader effectiveness criteria are
singled out here for their strong research designs. The first, also discussed in the
Measuring LSE section above, used extensive interviews with 44 middle and executive
level managers as the starting point for developing a comprehensive, 18-dimension LSE
scale along with a leadership effectiveness measure consisting of nine competency areas
(Anderson et al., 2008). Next, these researchers investigated the relationship between these
two taxonomies. Using multi-source data and canonical analysis, they found that
managers’ LSE perceptions were highly related to superior, peer and subordinate ratings
of their effectiveness in various leadership areas. For instance, managers who expressed
confidence in their ability to oversee others’ work and provide direction were rated by
observers as more effective in task-oriented leadership. There were eight significant
canonical variates in all.
The second field study highlighted here involves the new LSME scale mentioned earlier
(Hannah et al., 2012). As noted, the LSME measure expands previous definitions of LSE by
incorporating means efficacy, specifically, leaders’ beliefs about resource availability and
usefulness. Concerning LSME’s connection with leader effectiveness, this research study is
noteworthy because it used multiple-source, lagged outcome measures – LSME was
measured at Time 1 and the criteria two months later. Results indicated that LSME at Time
1 was positively related to both superior and peer ratings of participants’ leadership
performance at Time 2. Also, LSME had incremental validity over an alternate LSE scale in
predicting self-ratings of contingent reward transactional leadership and transformational
leadership two months later. In addition, LSME predicted superiors’ ratings of participants’
transformational leadership at Time 2.
Findings from the limited lab-based research on LSE have been more mixed. Kane et al.
(2002) found that task-specific LSE positively predicted the goal levels leaders set in a
business simulation game, but was not significantly correlated with subsequent leader
effectiveness ratings by team members. Using the same business simulation game, however,
Hendricks and Payne (2007) did find a positive relationship between LSE and members’
ratings. Results from another laboratory study, again using a business simulation game, did
not support hypotheses linking LSE with team members’ ratings of leader effectiveness or
transformational leadership (Mesterova et al., 2015). However, this study used a small sample
of leaders (32), measured self-efficacy with a generalized rather than leadership-specific scale
and administered the self-efficacy scale after feedback was provided on the leaders’ fictional
company profits.
In sum, with few exceptions, convincing evidence indicates that LSE has a positive
influence on leader effectiveness across a variety of criteria, including LMX quality, leadership
potential, leadership attempts and performance, and transformational leadership behavior.
As the importance of LSE for leadership effectiveness emerged in the literature, researchers Leadership
began to look beyond ratings of an individual leader’s performance or behavior to explore self-efficacy
LSE’s impact on the leader’s team. Of particular interest was LSE’s relationship with collective
(team) efficacy and subsequent team performance, discussed next.

LSE and collective efficacy

As noted earlier, a criticism levied against LSE research has been its individualistic focus. 641
That is, LSE’s antecedents often take the form of leader characteristics, and LSE’s outcomes
are most frequently measured through others’ ratings of leader performance or behavior.
These individualistic models do not fully capture the social influence process inherent in
leadership or the leader’s impact on group attributes important for performance. To address
this gap, LSE researchers began to examine collective efficacy as an outcome variable.
Collective efficacy refers to a shared belief in the performance capability of the group as a
whole (Bandura, 1997). It is most often operationalized by collecting members’ perceptions
of their groups’ capability for a task, assessing within-group agreement, and if this is
satisfactory, computing the average of members’ ratings to form the group-level collective
efficacy variable. Occasionally, leaders are also asked for their perceptions of their groups’
collective efficacy.
Reviewing research on LSE and collective efficacy reveals that most studies have been
conducted in a laboratory with college student subjects. One exception used basketball
teams and measured team leaders’ self-efficacy prior to the season’s start (Time 1), finding
Time 1 LSE was positively related to teams’ average collective efficacy at Time 1. In turn,
teams’ Time 1 collective efficacy predicted teams’ overall success at season end.
Interestingly, however, Time 1 LSE was unrelated to teams’ collective efficacy at season end
(Watson et al., 2001). Regarding lab studies, Hoyt et al. (2003) measured leaders’ and
followers’ collective efficacy perceptions in three-person teams, exploring connections
between LSE, both collective efficacy measures and group performance. Structural equation
modeling supported a leadership model with collective efficacy in the mediating role.
Specifically, they found that LSE predicted leaders’ collective efficacy perceptions, which
were strongly related to followers’ collective efficacy perceptions. Followers’ perceptions of
collective efficacy, in turn, predicted group performance. Two other laboratory studies
corroborated the LSE – collective efficacy – group performance model (Taggar and Seijts,
2003; Villanueva and Sánchez, 2007).
Overall, existing research appears to support LSE’s influence on group performance, at least
in part, through collective efficacy, although confidence in this conclusion is tempered
somewhat by generalizability concerns. Besides collective efficacy, the question of how LSE
operates to positively affect performance-related outcomes, and under what conditions it has
greater or lesser effect presents an opportunity for future research. While individual past
studies have included other mediators, such as leader goals and task strategies (Kane et al.,
2002) and motivation to lead (Hendricks and Payne, 2007), and moderators including job
autonomy (Ng et al., 2008) and organizational commitment (Paglis and Green, 2002), a truly
comprehensive model has yet to be tested. To conclude this section on LSE and leader
effectiveness, research on LSE’s potentially negative consequences is discussed next.

Is high LSE always a positive influence?

Not surprisingly, almost all LSE studies contain an implicit assumption that high self-efficacy
is a desirable leader quality. Indeed, LSE has been proposed as a component of “the virtue of
courage” needed for effective team leadership under risky conditions (Amos and Klimoski,
2014, p. 110) and as the defining attribute of leaders who are best able to “navigate the waves of
uncertainty” facing business organizations today (Ellis, 2011, p. 144). A few scholars, however,
have challenged the assumption that high LSE is always beneficial. For example, Machida and
JMD Schaubroeck (2011) observe that excessive confidence in one’s leadership ability may result in
38,8 complacency and low interest in leadership skill development activities. Others point out that
leader’s efficacy level needs to be calibrated with the difficulty of the context/task (Hannah
et al., 2008). Unreasonably high LSE, relative to context, may lead to performance expectations
on followers that cannot be met. As a result, a debilitating downward spiral of performance
failure and diminished collective efficacy may occur. Also, while LSE may be critical for
642 leaders’ successful handling of extreme events that imperil an organization, excessive LSE may
lead to ill-advised risk-taking (Gist, 1987; Hannah et al., 2009).
One laboratory study tested the connection between leader overconfidence and
performance. Shipman and Mumford (2011) looked at the effectiveness of leaders’ plans
and vision statements relative to the leaders’ confidence level. Specifically, they proposed and
tested two elements of leader overconfidence: first, failing to see deficiencies or problems and
second, excessively optimistic expectations of positive outcomes. The study found leaders
who exhibited the least overconfidence on the “failing to see deficiencies” element developed
more effective plans. As the study’s authors put it, seeing many problems or deficiencies
in one’s ideas, an indicator of lower confidence, is beneficial for leader planning. As
expected, leaders’ overconfidence on “expecting positive outcomes” positively related to the
inspirational appeal of their vision statements. In sum, research is beginning to indicate that
the relationship between LSE and leader effectiveness may not be linear, and too much
confidence may impair performance. There have been very few empirical tests of this
question, though, and it appears to be an intriguing avenue for future study.

What factors account for differences in LSE?

Having concluded that well-calibrated LSE is beneficial for leadership effectiveness, attention
now turns to what factors account for differences in LSE between leaders. Past research can
be divided into two categories, individual characteristics and contextual factors.

Individual characteristics
By far, the most extensively studied influences on LSE have been leaders’ personality traits.
For example, research indicates core self-evaluation and two of its components, self-esteem
and internal locus of control, are positive correlates (Huszczo and Endres, 2017; Paglis and
Green, 2002). Several studies have investigated the Big Five traits, with many finding
extraversion to be positively related to LSE (Chan and Drasgow, 2001; Courtright et al., 2014;
Hendricks and Payne, 2007; Huszczo and Endres, 2017; Ng et al., 2008; Quigley, 2013). Given
that leadership is a social influence process, it makes sense that leaders who are outgoing
and sociable will feel more confident in the role. Research indicates three other Big Five
traits are positively linked with LSE: conscientiousness, openness to experience and
emotional stability (Chan and Drasgow, 2001; Courtright et al., 2014; Hendricks and Payne,
2007; Huszczo and Endres, 2017; Ng et al., 2008). Studies have also found positive
relationships between LSE and cognitive ability (Quigley, 2013), emotional intelligence
(Harper, 2016; Semadar et al., 2006; Villanueva and Sánchez, 2007), learning goal orientation
(Hendricks and Payne, 2007) and previous leadership experience (Chan and Drasgow, 2001;
McCormick et al., 2002; Ng et al., 2008).
In addition, researchers have explored whether gender affects LSE perceptions, with
mixed results. While two studies reported no significant gender difference (Huszczo and
Endres, 2017; Singer, 1991), at least three others have found men to be significantly higher in
LSE (Bobbio and Manganelli, 2009; Kane et al., 2002; McCormick et al., 2002). One
explanation for this gender difference may be lingering societal perceptions that associate
men more strongly with the agentic qualities traditionally associated with leadership (Eagly
and Karau, 1991). To the extent women internalize this stereotype and feel pressure to
conform to gender role expectations, confidence in their leadership abilities may weaken.
Another possible explanation is a difference in attributional style. Specifically, research Leadership
indicates women are more likely to attribute success to effort rather than ability, whereas self-efficacy
men tend toward the opposite pattern. Attributing success to innate ability, the male
pattern, is thought to be more conducive to building self-confidence (Rosenthal, 1995). Also,
fewer females in leadership positions may result in women having less access to the
stronger efficacy boost associated with same-gender role models (Barclay et al., 2007). As a
final note regarding gender and LSE, research suggests the Big Five have different 643
predictive validities for men and women; specifically, extraversion was most strongly
related to LSE for men, whereas conscientiousness and openness to experience were
stronger predictors for women (Huszczo and Endres, 2017).
To summarize, in terms of forming practical implications from the research on individual
antecedents, the strongest foundation appears to be with the Big Five traits. The quantity and
consistency of the evidence concerning the Big Five and their relationship with LSE, coupled
with the availability of psychometrically sound measures, suggest organizations should
consider incorporating these personality tests as one component in leader selection (Paglis, 2010).

Contextual factors
As noted in the earlier section on criticisms of LSE research, the influence of context on LSE
has received much less attention than individual leader characteristics. When contextual
influences have been studied, it has been done in a piecemeal fashion rather than in a truly
comprehensive model. Nevertheless, there have been some significant findings. For
instance, early laboratory research on managerial decision making found that altering the
perceived controllability of the organization influenced managers’ self-efficacy perceptions,
such that greater controllability was associated with higher self-efficacy (Wood and
Bandura, 1989a). Relatedly, Paglis and Green (2002) found subordinates’ ratings of cynicism
about change, organizational support for change and resource supply were linked to
managers’ LSE in directions consistent with Wood and Bandura’s results. Leaders’ ratings
of relationship quality with subordinates and their performance abilities also have been
positively connected to LSE (Paglis and Green, 2002; Trépanier et al., 2012). Concerning the
leader’s role definition, job autonomy has been identified as a strong, positive correlate in
two studies (Ng et al., 2008; Paglis and Green, 2002), indicating that the authority to make
decisions regarding work flow, procedures and resource allocation is important for feeling
confident as a leader. As well, research suggests leaders’ perceived role ambiguity may
negatively affect LSE (Lindberg et al., 2013).
Overall, studies of contextual influences have been relatively sparse. The development
and validation of the LSME construct (Hannah et al., 2012), which explicitly recognizes
“means efficacy” as an LSE component, may draw more attention to this topic going
forward. Also, see the section on LSE and leadership development regarding some findings
on altering perceptions about the utility of contextual resources.
In reviewing the literature on LSE, a timeline emerges of early research focusing on
construct development and measurement issues, followed by studies of LSE’s impact on
leader behavior and effectiveness. Later research added complexity to LSE models by
investigating mediating and moderating effects. Influences on LSE, especially individual
leader characteristics, have been a popular topic. Finally, as LSE’s importance in leadership
was firmly established in the literature, a research extension that has attracted significant
attention is the potential for enhancing LSE through leadership development programs.
These results will be considered next.

Building LSE through leadership development programs

As previously discussed, research indicates that LSE is partially determined by relatively fixed
variables, with particularly solid results for several of the Big Five traits. Within the
JMD parameters a person’s personality places on efficacy perceptions, however, lies the potential for
38,8 interventions designed to increase it. A research review revealed several field studies that have
used LSE as an outcome variable in the assessment of leadership development programs. For
example, Holmberg et al. (2016) evaluated a program with 86 manager-participants engaging in
12 days of leadership development activities over one year. The program emphasized reflection
through writing and group discussions, paired with lectures and development exercises
644 including role plays. The research team collected pre- and post-test LSE measures for program
participants and a control group, finding significant increases in participants’ LSE relative to
the controls. The researchers’ role was limited to assessing outcome criteria, however, so they
were unable to conclusively determine which program components accounted for the LSE
growth. They noted the relatively strong emphasis on peer feedback as one possible
mechanism, which is consistent with two information sources Bandura (1997) linked to self-
efficacy development: vicarious experience and verbal persuasion. A second study evaluated a
two-month-long leadership development program for 27 academic leaders (Evans et al., 2017).
The program consisted of eight three-hour long classroom sessions featuring multimedia
presentations, roundtable discussions, self-assessment instruments and other tools. Although
there was no control group, results showed a significant increase in LSE from pre- to post-test.
As was the case in Holmberg et al., these researchers were unable to identify specific program
elements accounting for the increase.
The findings above suggest some optimism vis-à-vis the potential for increasing LSE
through leadership development programs. They only confirmed a hypothesized outcome,
however, rather than pinpointing how or why LSE growth occurred. Gaining an
understanding of these mechanisms is necessary for improving the design and delivery of
future leadership development efforts. There is some research that has sought to fill this gap
by testing certain components of leadership development interventions, with executive
coaching and mentoring garnering the most attention. These two techniques certainly share
some common elements. However, executive coaching is often distinguished by its narrower
focus on improving coachees’ performance in specific aspects of their work roles, with a set
of predetermined goals, and the coaches may be internal members of the organization or
external consultants. Mentoring is typically more informal in scope and duration,
encompassing a broader range of career advancement and psychosocial functions. Also,
mentors are commonly longer-tenured members of the mentees’ own organization (Baron
and Morin, 2009). Regarding leadership development and LSE, there has been more research
on executive coaching than mentoring, and so these studies will be discussed first.
A literature review uncovered four field studies exploring executive coaching and LSE. Two
of these had stronger research designs, with pre- and post-test LSE measures and control
groups. The first one examined a year-long coaching program that paired external consultant-
coaches with 20 executives; it found a significant increase in the executives’ LSE at the end of
the program compared with the control (Moen and Skaalvik, 2009). Further, this study found a
positive relationship between external coaching and executives’ efficacy-enhancing attributions;
participants showed an increased frequency of taking credit for success through attributions to
their ability or strategic decisions rather than outside factors. These same executives also
participated in the study as internal coaches for 52 middle managers reporting to them.
Significant increases in LSE were found in this internal coach/middle manager group, although
the effect was not as pronounced as with the external coach/executive group. The second field
study with pre- and post-measurement and controls paired 24 middle and upper level leaders
with experienced external coaches; the dyads met for eight sessions over the course of six
months (Ladegard and Gjerde, 2014). Consistent with Moen and Skaalvik’s results, these
researchers found that LSE increased only in the coached group; further, they identified
“facilitative coach behavior” as explaining about 25 percent of the change in LSE. “Facilitative
coach behavior” consisted of challenging coachees, supporting them and giving feedback.
 Two field studies on executive coaching included pre- and post-test LSE measures, but Leadership
no control group. In the first, external coaches were paired with 31 managers from a firm self-efficacy
that had recently experienced CEO turnover and significant restructuring (Grant, 2014).
Four one-on-one coaching sessions occurred over a maximum period of four months.
Analyzing pre- and post-test LSE measures revealed a significant, positive increase over the
course of the program. A second field study focused on the amount of coaching and the
quality of the coach–coachee relationship as potential drivers of managers’ self-efficacy 645
development (Baron and Morin, 2009, 2010). (Although these researchers did not label their
criterion variable “LSE,” their scale was similar to LSE.) They studied 31 dyads in an eight-
month-long leadership development program for junior and midlevel managers, with the
executive coaching component consisting of up to 14 face-to-face sessions. The coaches were
executives from the same firm who had completed a coaching certification process. Results
showed a significant LSE increase among the participant–managers between pre- and post-
test. Further, they found the number of coaching sessions was positively related to LSE
growth, fully mediated by the quality of the coach–coachee relationship (“working alliance”).
Interestingly, the coach’s self-efficacy for facilitating coachees’ learning was positively
related to working alliance. The researchers also reported the coachees’ perceptions of the
program’s usefulness and their affective organizational commitment at the program’s start
were positively associated with the LSE increase (Baron and Morin, 2010).
Although executive coaching has been the more popular subject, two mentoring studies
also are informative. Lester et al. (2011) studied military cadets and found that a six-month-
long mentorship program was superior to a classroom-based leadership education program
in increasing cadets’ LSE. In discussing the connection between mentoring and LSE
development, the researchers proposed that effective mentors are role models and provide
feedback (i.e. Bandura’s vicarious experience and verbal persuasion), while also helping
mentees interpret their leadership experiences in an efficacy-enhancing manner (i.e. mastery
experiences). Potential antecedents of LSE growth over time also were tested in this study,
with significant, positive effects reported for cadets’ trust in the mentor and cadets’
openness to receiving critical feedback. A second mentoring study, cross-sectional in design,
found a relationship between mentees’ ratings of the quality of the mentoring relationship
and LSE, controlling for the length of the relationship (Chopin et al., 2013).
Considered together, these studies offer some support for executive coaching and
mentoring as tools for strengthening LSE. This research also suggests certain features
that may be especially important. Specifically, three studies found relationship quality,
including the level of trust between the parties, to be positively related to LSE
development (Baron and Morin, 2009; Chopin et al., 2013; Lester et al., 2011). Furthermore,
the significance of facilitative coach behavior (Ladegard and Gjerde, 2014) seems
consistent with the results concerning relationship quality. These findings suggest further
study of methods for selecting and pairing coaches/mentors with developing leaders may
be useful. In addition, several coachee/mentee antecedents of LSE growth were identified,
including perceived utility of leadership development programs, attributional style,
affective commitment and negative feedback-seeking. Regarding the first of these,
providing participants with information on the program’s goals and the expected benefits
of improving LSE is a simple, inexpensive recommendation. Some caution is warranted,
however, in drawing firm conclusions from this research stream. Limitations include small
sample sizes, unequal or absent control groups, same-source data and potential
self-selection bias in the treatment groups; as well, what comprised executive coaching or
mentoring varied widely from study to study.
Earlier research on self-efficacy growth, although not specific to LSE, offers some
additional ideas for designing programs to build leaders’ confidence. In creativity training,
Gist (1989) found cognitive modeling to be superior to traditional lecture and practice for
JMD improving managers’ self-efficacy. Cognitive modeling involves guiding trainees to monitor
38,8 their thought patterns during performance, substituting self-corrections in place of self-
defeating thoughts. Similarly, constructive thought pattern strategies, such as positive
self-talk and self-problem solving, have been found to be positively related to self-efficacy in
college students (Prussia et al., 1998). Along the same lines, Stajkovic and Luthans (1998)
advocated training in effective cognitive strategies as part of self-efficacy development
646 programs, especially in complex task domains. Specifically, they recommended
interventions that encourage trainees to adopt an incremental rather than an entity
conception of ability. Individuals who view ability as incremental rather than fixed are more
likely to adopt learning goals and view mistakes as a natural, expected part of the learning
process. In contrast, an entity conception is associated with interpreting mistakes in a
manner that diminishes self-efficacy (Elliott and Dweck, 1988; Wood and Bandura, 1989b).
More recently, researchers have applied these findings on cognitive modeling and
constructive thought patterns to make recommendations specifically for LSE development.
For example, Hannah et al. (2008) noted the importance of adaptive (i.e. constructive,
positive) vs maladaptive (i.e. ruminative, self-defeating) self-reflection for leaders, and
observed that these different patterns can be cued through instruction. Likewise, Machida
and Schaubroeck (2011, p. 464) observed that positive self-talk after receiving negative
performance feedback can prevent “downward efficacy spirals” in leaders. They also
proposed that mentors may be able to coax developing leaders to reframe certain situations,
shifting from a “you’d better do this right” performance orientation to a learning orientation
that is more efficacy-sustaining.
Rethinking the basic conceptualization of what LSE is may expose some other
connections between leaders’ thought patterns and LSE growth. For example, McDaniel and
DiBella-McCarthy (2012) suggest dividing LSE in two: first, the belief that one has the
necessary skills and abilities for success as a leader, and second, the belief that the actions
one takes as a leader will have the desired effect. The latter component brings to mind the
“means efficacy” element of LSME discussed earlier (Hannah et al., 2012). Research suggests
individuals’ perceptions about the quality and utility of available means are malleable
(Eden, 2001; Eden and Sulimani, 2002), opening the door for interventions geared toward
enhancing leaders’ appraisal of their contextual resources (Hannah et al., 2012). Likewise,
interventions focusing on altering leaders’ perceptions about the controllability of their
work environments (Wood and Bandura, 1989a), as well as developing strategies for
overcoming uncooperative features, may be worthwhile avenues for future study (McDaniel
and DiBella-McCarthy, 2012).

Conclusion

“If you think you can or you think you can’t, you’re probably right.” (Henry Ford)
Although success as a leader is undoubtedly more complicated than simply believing in
one’s abilities, a substantial body of research supports Henry Ford’s observation that
self-efficacy is a key component. This paper reviewed the research on LSE that has been
published over the past 25 years or so, summarizing the findings and offering suggestions
for future research avenues as well as some preliminary, research-based guidance on the
features of efficacy-enhancing leadership development programs. The review revealed
surprising diversity in LSE construct development and measurement, which has been a
focus of criticism. Regarding LSE’s impact on leader effectiveness, many studies have
reported positive relationships with potential, performance and behavioral ratings of
leaders. Interestingly, however, some recent research has questioned the assumption that
high LSE is always a positive leader attribute. Concerning influences on LSE, several of the
Big Five traits have significant research support, while contextual antecedents appear to be Leadership
under-researched and potentially fruitful areas for future study. Finally, executive coaching self-efficacy
and mentoring, cognitive modeling techniques and training in constructive thought patterns
are suggested for potential inclusion in leadership development programs, as research
indicates they can be effective in bolstering LSE in developing leaders.

647
References
Ali, H.E., Schalk, R., Van Engen, M. and Van Assen, M. (2018), “Leadership self-efficacy and
effectiveness: the moderating influence of task complexity”, Journal of Leadership Studies,
Vol. 11 No. 4, pp. 21-40.
Amos, B. and Klimoski, R.J. (2014), “Courage: making teamwork work well”, Group and Organization
Management, Vol. 39 No. 1, pp. 110-128.
Anderson, D.W., Krajewski, H.T., Goffin, R.D. and Jackson, D.N. (2008), “A leadership self-efficacy
taxonomy and its relation to effective leadership”, Leadership Quarterly, Vol. 19 No. 5, pp. 595-608.
Bandura, A. (1977), “Self-efficacy: toward a unifying theory of behavioral change”, Psychological
Review, Vol. 84 No. 2, pp. 191-215.
Bandura, A. (1997), Self-Efficacy: The Exercise of Control, W.H. Freeman, New York, NY.
Barclay, L.A., Mellor, S., Bulger, C.A. and Kath, L.M. (2007), “Perceived steward success and leadership
efficacy: the role of gender similarity”, Journal of Collective Negotiations, Vol. 31 No. 2, pp. 141-153.
Baron, L. and Morin, L. (2009), “The coach-coachee relationship in executive coaching: a field study”,
Human Resource Development Quarterly, Vol. 20 No. 1, pp. 85-106.
Baron, L. and Morin, L. (2010), “The impact of coaching on self-efficacy related to management soft
skills”, Leadership and Organization Development Journal, Vol. 31 No. 1, pp. 18-38.
Bobbio, A. and Manganelli, A.M. (2009), “Leadership self-efficacy scale: a new multidimensional
instrument”, Testing, Psychometrics, Methodology in Applied Psychology, Vol. 16 No. 1, pp. 3-24.
Chan, K. and Drasgow, F. (2001), “Toward a theory of individual differences and leadership:
understanding the motivation to lead”, Journal of Applied Psychology, Vol. 86 No. 3, pp. 481-498.
Chemers, M.M., Watson, C.B. and May, S.T. (2000), “Dispositional affect and leadership effectiveness:
a comparison of self-esteem, optimism, and efficacy”, Personality and Social Psychology Bulletin,
Vol. 26 No. 3, pp. 267-277.
Chopin, S.M., Danish, S.J., Seers, A. and Hook, J.N. (2013), “Effects of mentoring on the development of
leadership self-efficacy and political skill”, Journal of Leadership Studies, Vol. 6 No. 3, pp. 17-31.
Courtright, S.H., Colbert, A.E. and Choi, D. (2014), “Fired up or burned out? How developmental
challenge differentially impacts leader behavior”, Journal of Applied Psychology, Vol. 99 No. 4,
pp. 681-696.
Eagly, A.H. and Karau, S.J. (1991), “Gender and the emergence of leaders: a meta-analysis”, Journal of
Personality and Social Psychology, Vol. 60 No. 5, pp. 685-710.
Eden, D. and Sulimani, R. (2002), “Pygmalion training made effective: greater mastery through
augmentation of self-efficacy and means efficacy”, in Avolio, B.J. and Yammarino, F.J. (Eds),
Transformational/Charismatic Leadership: The Road Ahead, Elsevier, New York, NY, pp. 287-308.
Elliott, E.S. and Dweck, C.S. (1988), “Goals: an approach to motivation and achievement”, Journal of
Personality and Social Psychology, Vol. 54, pp. 5-12.
Ellis, T. (2011), “The importance of organizational leadership self-efficacy in a world of uncertainty”,
Proceedings of the Northeast Business and Economics Association Conference, Philadelphia, PA.
Evans, L., Hess, C.A., Abdelhamid, S. and Stepleman, L.M. (2017), “Leadership development in the
context of a university consolidation: an initial evaluation of the Authentic Leadership Pipeline
Program”, Journal of Leadership Studies, Vol. 10 No. 3, pp. 7-21.
JMD Feasel, K.E. (1995), “Mediating the relation between goals and subjective well-being: global and
38,8 domain-specific variants of self-efficacy”, unpublished Master’s Thesis, University of Illinois,
Urbana-Champaign, IL.
Gist, M.E. (1987), “Self-efficacy: implications for organizational behavior and human resource
management”, Academy of Management Review, Vol. 12 No. 3, pp. 472-485.
Gist, M.E. (1989), “The influence of training method on self-efficacy and idea generation among
managers”, Personnel Psychology, Vol. 42 No. 4, pp. 787-805.
648
Grant, A.M. (2014), “The efficacy of executive coaching in times of organizational change”, Journal of
Change Management, Vol. 14 No. 2, pp. 258-280.
Hannah, S.T., Avolio, B.J., Luthans, F. and Harms, P.D. (2008), “Leadership efficacy: review and future
directions”, Leadership Quarterly, Vol. 19 No. 6, pp. 669-692.
Hannah, S.T., Avolio, B.J., Walumbwa, F.O. and Chan, A. (2012), “Leader self and means efficacy:
a multi-component approach”, Organizational Behavior and Human Decision Processes, Vol. 118
No. 2, pp. 143-161.
Hannah, S.T., Uhl-Bien, M., Avolio, B.J. and Cavarretta, F.L. (2009), “A framework for examining
leadership in extreme contexts”, Leadership Quarterly, Vol. 20 No. 6, pp. 897-919.
Harper, D.S. (2016), “Correctional executives’ leadership self-efficacy and their perceptions of emotional
intelligence”, American Journal of Criminal Justice, Vol. 41 No. 4, pp. 765-779.
Hendricks, J.W. and Payne, S.C. (2007), “Beyond the Big Five: leader goal orientation as a predictor of
leadership effectiveness”, Human Performance, Vol. 20 No. 4, pp. 317-343.
Holmberg, R., Larsson, M. and Bäckström, M. (2016), “Developing leadership skills and resilience in
turbulent times: a quasi-experimental evaluation study”, Journal of Management Development,
Vol. 35 No. 2, pp. 154-169.
Hoyt, C.L., Murphy, S.E., Halverson, S.K. and Watson, C.B. (2003), “Group leadership: efficacy and
effectiveness”, Group Dynamics: Theory, Research, and Practice, Vol. 7 No. 4, pp. 259-274.
Huszczo, G. and Endres, M.L. (2017), “Gender differences in the importance of personality traits in
predicting leadership self-efficacy”, International Journal of Training and Development, Vol. 21
No. 4, pp. 304-317.
Kane, T.D., Zaccaro, S.J., Tremble, T.R. and Masuda, A.D. (2002), “An examination of the leader’s
regulation of groups”, Small Group Research, Vol. 33 No. 1, pp. 65-120.
Ladegard, G. and Gjerde, S. (2014), “Leadership coaching, leader role-efficacy, and trust in
subordinates. A mixed methods study assessing leadership coaching as a leadership
development tool”, Leadership Quarterly, Vol. 25 No. 4, pp. 631-646.
Lester, P.B., Hannah, S.T., Harms, P.D., Vogelgesang, G.R. and Avolio, B.J. (2011), “Mentoring impact
on leader efficacy development: a field experiment”, Academy of Management Learning and
Education, Vol. 10 No. 3, pp. 409-429.
Lindberg, E., Wincent, J. and Ӧrtqvist, D. (2013), “Turning stressors into something productive: an
empirical study revealing nonlinear influences of role stressors on self-efficacy”, Journal of
Applied Social Psychology, Vol. 43 No. 2, pp. 263-274.
McCormick, M.J. (2001), “Self-efficacy and leadership effectiveness: applying social cognitive theory to
leadership”, Journal of Leadership Studies, Vol. 8 No. 1, pp. 22-33.
McCormick, M.J., Tanguma, J. and López-Forment, A.S. (2002), “Extending self-efficacy theory to
leadership: a review and empirical test”, Journal of Leadership Education, Vol. 1 No. 2, pp. 34-49.
McDaniel, E.A. and DiBella-McCarthy, H. (2012), “Reflective leaders become causal agents of change”,
Journal of Management Development, Vol. 31 No. 7, pp. 663-671.
Machida, M. and Schaubroeck, J. (2011), “The role of self-efficacy beliefs in leader development”,
Journal of Leadership and Organizational Studies, Vol. 18 No. 4, pp. 459-468.
Mesterova, J., Prochazka, J., Vaculik, M. and Smutny, P. (2015), “Relationship between self-efficacy,
transformational leadership, and leader effectiveness”, Journal of Advanced Management
Science, Vol. 3 No. 2, pp. 109-122.
Moen, F. and Skaalvik, E. (2009), “The effect from executive coaching on performance psychology”, Leadership
International Journal of Evidence Based Coaching and Mentoring, Vol. 7 No. 2, pp. 31-49. self-efficacy
Murphy, S.E. (1992), “The contribution of leadership experience and self-efficacy to group performance
under evaluation apprehension”, Unpublished Doctoral Dissertation, University of Washington,
Seattle, WA.
Murphy, S.E. (2002), “Leader self-regulation: the role of self-efficacy and multiple intelligences”, in
Riggio, R.E., Murphy, S.E. and Pirozzolo, F.J. (Eds), Multiple Intelligences and Leadership,
Lawrence Erlbaum, Mahwah, NJ, pp. 163-186.
649
Murphy, S.E. and Ensher, E.A. (1999), “The effects of leader and subordinate characteristics in the
development of leader-member exchange quality”, Journal of Applied Social Psychology, Vol. 29
No. 7, pp. 1371-1394.
Ng, K., Ang, S. and Chan, K. (2008), “Personality and leader effectiveness: a moderated mediation model
of leadership self-efficacy, job demands, and job autonomy”, Journal of Applied Psychology,
Vol. 93 No. 4, pp. 733-743.
Osborn, R.N., Hunt, J.G. and Jauch, L.R. (2002), “Toward a contextual theory of leadership”, Leadership
Quarterly, Vol. 13 No. 6, pp. 797-837.
Paglis, L.L. (2010), “Leadership self-efficacy: research findings and practical applications”, Journal of
Management Development, Vol. 29 No. 9, pp. 771-782.
Paglis, L.L. and Green, S.G. (2002), “Leadership self-efficacy and managers’ motivation for leading
change”, Journal of Organizational Behavior, Vol. 23 No. 2, pp. 215-235.
Prussia, G.E., Anderson, J.S. and Manz, C.C. (1998), “Self-leadership and performance outcomes: the
mediating influence of self-efficacy”, Journal of Organizational Behavior, Vol. 19 No. 5, pp. 523-538.
Quigley, N.R. (2013), “A longitudinal, multilevel study of leadership efficacy development in MBA
teams”, Academy of Management Learning and Education, Vol. 12 No. 4, pp. 579-602.
Rosenthal, P. (1995), “Gender differences in managers’ attributions for successful work performance”,
Women in Management Review, Vol. 10 No. 6, pp. 26-31.
Rousseau, D.M. and Fried, Y. (2001), “Location, location, location: contextualizing organizational
behavior research”, Journal of Organizational Behavior, Vol. 22 No. 1, pp. 1-13.
Schruijer, S.G.L. and Vansina, L.S. (2002), “Leader, leadership, and leading: from individual characteristics
to relating in context”, Journal of Organizational Behavior, Vol. 23 No. 7, pp. 869-874.
Seibert, S.E., Sargent, L.D., Kraimer, M.L. and Kiazad, K. (2017), “Linking developmental experiences to
leader effectiveness and promotability: the mediating role of leadership self-efficacy and mentor
network”, Personnel Psychology, Vol. 70 No. 2, pp. 357-397.
Semadar, A., Robins, G. and Ferris, G.R. (2006), “Comparing the validity of multiple social effectiveness
constructs in the prediction of managerial job performance”, Journal of Organizational Behavior,
Vol. 27 No. 4, pp. 443-461.
Shipman, A.S. and Mumford, M.D. (2011), “When confidence is detrimental: influence of overconfidence
on leadership effectiveness”, Leadership Quarterly, Vol. 22 No. 4, pp. 649-665.
Singer, M. (1991), “The relationship between employee sex, length of service, and leadership
aspirations: a study from valence, self-efficacy and attribution perspectives”, Applied Psychology:
An International Review, Vol. 40 No. 4, pp. 417-436.
Stajkovic, A.D. and Luthans, F. (1998), “Self-efficacy and work-related performance: a meta-analysis”,
Psychological Bulletin, Vol. 124 No. 2, pp. 240-261.
Taggar, S. and Seijts, G.H. (2003), “Leader and staff role-efficacy as antecedents of collective-efficacy
and team performance”, Human Performance, Vol. 16 No. 2, pp. 131-156.
Trépanier, S., Fernet, C. and Austin, S. (2012), “Social and motivational antecedents of perceptions of
transformational leadership: a self-determination theory perspective”, Canadian Journal of
Behavioural Science, Vol. 44 No. 4, pp. 272-277.
Villanueva, J.J. and Sánchez, J.C. (2007), “Trait emotional intelligence and leadership self-efficacy: their
relationship with collective efficacy”, Spanish Journal of Psychology, Vol. 10 No. 2, pp. 349-357.
JMD Watson, C.B., Chemers, M.M. and Preiser, N. (2001), “Collective efficacy: a multilevel analysis”,
38,8 Personality and Social Psychology Bulletin, Vol. 27 No. 8, pp. 1057-1068.
Wood, R. and Bandura, A. (1989a), “Social cognitive theory of organizational management”, Academy
of Management Review, Vol. 14 No. 3, pp. 361-384.
Wood, R. and Bandura, A. (1989b), “Impact of conceptions of ability on self-regulatory mechanisms and
complex decision making”, Journal of Personality and Social Psychology, Vol. 56 No. 3, pp. 407-415.
650 Yammarino, F.J., Dionne, S.D., Chun, J.U. and Dansereau, F. (2005), “Leadership and levels of analysis:
a state-of-the-science review”, Leadership Quarterly, Vol. 16 No. 6, pp. 879-919.

Corresponding author
Laura Paglis Dwyer can be contacted at: [email protected]

For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: [email protected]