Summary writing

The Leadership Quarterly 28 (2017) 508–529

Contents lists available at ScienceDirect

The Leadership Quarterly

j ourna l homepage: www.e lsev ie r .com/ locate / leaqua

A meta-analytic review and future research agenda of charismatic leadership

George C. Banks a,⁎, Krista N. Engemann a,1, Courtney E. Williams a,1, Janaki Gooty a, Kelly Davis McCauley b, Melissa R. Medaugh a

a University of North Carolina at Charlotte, United States b West Texas A&M University, United States

a r t i c l e i n f o

⁎ Corresponding author at: University of North Caroli E-mail address: [email protected] (G.C. Banks).

1 Equal contribution.

http://dx.doi.org/10.1016/j.leaqua.2016.12.003 1048-9843/© 2016 Elsevier Inc. All rights reserved.

a b s t r a c t

Article history: Received 30 October 2015 Received in revised form 24 December 2016 Accepted 27 December 2016 Available online 6 January 2017

Charismatic leadership is a critical construct that draws much attention from both academic and practitioner literatures. Despite the positive attention received by the charisma construct, some have criticized its conceptualization and measurement. These critiques have, in turn, cast doubt on what we know regarding the antecedents and outcomes of charismatic leader- ship. In this review, we adopt a recently developed definition of charismatic leadership and then conduct a meta-analysis of its antecedents and objective outcomes. Following an exami- nation of 76 independent studies and 36,031 individuals, results indicate that the Big Five traits and cognitive ability vary in their association with charismatic leadership. Other findings show that dimensions of charismatic leadership predict outcomes of interest, such as supervisor- rated task performance, supervisor-rated citizenship behaviors, and group or organization performance. Several shortcomings are identified, however, in testing theoretical and method- ological moderating variables. The present research ultimately provides a roadmap for new frontiers in theoretical, measurement and empirical work on charismatic leadership.

© 2016 Elsevier Inc. All rights reserved.

Keywords: Charisma Leadership Meta-analysis

Introduction

In an organizational context, it is difficult to downplay the real or perceived importance of leadership in influencing important in- dividual- and firm-level outcomes. This statement is particularly truewhen considering the almost romantic perceptions of charismatic leaders as “larger than life” characters with a “mysterious gift” (Shamir, 1992). Ever since the introduction of charismatic leadership by Robert House (1977), a voluminous amount of theoretical and empirical work has ensued in this domain (e.g., Bryman, 1992; Conger & Kanungo, 1987; House, 1999; Shamir, House, & Arthur, 1993). Indeed, it appears that the study of charismatic leadership itself has taken on a larger than life and mysterious character in the organizational sciences.

Specifically, charisma has neither been defined consistently nor robustly in nearly six decades of theory and research on inspira- tional forms of leadership in the organizational sciences (see Antonakis, Bastardoz, Jacquart, & Shamir, 2016; Van Knippenberg & Sitkin, 2013 for reviews). The most heavily cited definitions of charismatic leadership on Google Scholar (e.g., Bass, 1985: 13,758 ci- tations; Conger & Kanungo, 1987, 1998: 3872 combined citations; House, 1977: 2466 citations & Shamir et al., 1993: 2875 citations as of May 2015) reveal two overarching themes: a) The definition of charisma has typically included outcomes or antecedents and/or b)

na at Charlotte, Belk College of Business, 9201 University City Blvd, Charlotte, NC, 28223, United States.

509G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

charisma has been defined as some unknown quality or miraculous ability. As discussed elsewhere in the literature (Antonakis, Simonton and House, 2016; Van Knippenberg & Sitkin, 2013), these two themes are particularly problematic. The former confuses the consequences or antecedents of charismawith the construct itself, while the latter is simply a useless way of defining a construct from a scientific standpoint as it could neither be operationalized nor researched in a systematic manner.

Indeed, Van Knippenberg & Sitkin (2013) called for a complete return to the drawing board with regard to the conceptualiza- tion of charisma in the organizational sciences. While we echo their concerns regarding the construct space of charisma, we do not believe that a complete return to the drawing board is a fruitful next step without first conceptually and empirically reviewing existing findings in the domain with a proper definition of charisma. For such a definition to be viable, at a minimum, it has to surpass the two critiques noted above concerning a definition that is a) free from antecedents and consequences and b) clearly and specifically measurable.

Incidentally, Antonakis et al. (2016) conducted an exhaustive examination of the definitions of charismatic leadership spanning six decades of research (1954 - 2014) with the explicit goal of offering a definition that surpasses the two criteria above. More spe- cifically, these authors reviewed the definitions of charisma across these six decades and included all of themost frequently cited def- initions noted above (e.g., Bass, 1985; Conger & Kanungo, 1987, 1998: House, 1977 & Shamir et al., 1993: See Table 1 of Antonakis et al., 2016; VanKnippenberg& Sitkin, 2013). Antonakis and colleagues offered a definition that encompasses the commondenominator across all, while maintaining construct integrity by overcoming tautological issues aswell as vast overlapwith transformational lead- ership (MacKenzie, 2003).

As explained below, their definition of charisma specifically stands alone in that it does not rely on contextual conditions, conse- quences for the leader, followers or the organization, and leader antecedent traits, such as ability or personality. Further, given the recency of this definition and the lack of empirical reviews of its predictive substantive validity, it would not be efficient or prudent for us to reinvent yet another scientific definition of charisma. To be clear, we adopt this definition as it largely improves on many noted critiques of charisma in the past. The goal and contribution in this review is to determine if such a definition of charisma relates to important follower outcomes and /or is a function of leader antecedents, such as personality and ability.

Charismatic leadership is thus defined as “values-based, symbolic, and emotion-laden leader signaling” (p. 304; Antonakis et al., 2016). Further, Antonakis et al. (2016) stated: “There is a sore need for awell-donemeta-analysis using correctmeasures of charisma, and this in models that are properly and causally specified” (p. 311). Consistent with these authors, we assert that a comprehensive review of charismatic leadership based on this latest, more rigorous literature is needed.With this definition as the backdrop, our pri- mary purpose in the current paper is tomap this new definition on to existingmeasures of charismatic leadership and then conduct a meta-analytic review of the antecedents of charismatic leadership as well as a select group of objective, individual- and unit- level outcomes. The theoretical model we test is depicted in Figure 1.

Our reviewmakes the following four contributions to the scholarly conversation in charismatic leadership. First, we align existing measures from current science with the Antonakis et al. (2016) definition of charismatic leadership. Second, we offer a comprehen- sive test of antecedents of charismatic leadership across more than 30 years (1985–2016). Our review is the first of its kind that in- vestigates a range of stable individual differences (e.g., leader cognitive ability, leader gender, education, and age) as antecedents to charismatic leadership at thedimension level, as called for by several scholars (Antonakis et al., 2016; vanKnippenberg & Sitkin, 2013; Yukl, 1999). For example, we found that certain individual differences predict dimensions of charisma to a greater extent than in past reviews.

Our reviewhighlights new findings regarding charismatic leadershipwhen conceptualized free from tautological and endogeneity biases (see Antonakis et al., 2016; Antonakis, Bendahan, Jacquart, & Lalive, 2010a, b). We only examine outcomes (where plausible) that are somewhat free from endogeneity bias because they are reported from a different source (e.g., group performance, supervisor- rated task performance). Big Five traits and cognitive ability are mostly exogenous. Evidence exists of cross-situational consistency in the measurement of such traits in adulthood. For instance, test-retest reliability serves as a reasonable indicator of this consistency (Cobb-Clark & Schurer, 2012; Hampson&Goldberg, 2006; Hertzog & Schaie, 1986), as would estimates of heritability from behavioral genetics studies (Plomin, DeFries, Knopik, & Neiderhiser, 2016; Polderman et al., 2015). Still, the relationships observed here could be explained by omitted variables or simultaneity.

Third, we consider the predictive validity of charismatic leadership of common yet objective work-related outcomes, such as supervisor-rated task performance, organizational citizenship behaviors (OCB), and group- or firm-level performance. Critically, this third contribution also highlights that despite all the attention charismatic leadership has garnered over the years, it has not been systematically studied as an antecedent to such objective outcomes. Past meta-analytic reviews have not explicitly in- vestigated charismatic leadership as the focal construct as opposed to transformational leadership (e.g., Bono & Judge, 2004;

Table 1 Conceptual overlap between Antonakis et al. (2016) charisma components and dimensions in the CL-TL literature.

Antonakis et al. (2016) charisma component Conceptually similar dimension in existing CL-TL literature

Justifying the mission by appealing to values that distinguish right from wrong and engaging in emotional displays

Idealized influence (Attributions)

Communicating in symbolic ways to make the message clear and vivid, and also symbolizing and embodying the moral unity of the collective per se

Idealized influence (Behaviors)

Demonstrating conviction and passion for the mission Inspirational motivation Articulating a vision

Fig. 1. Predictors andwork outcomes of charismatic leadership. Note: Thepredictor-charismatic leadership relations are tested independently of the charismatic leadership- outcomes relations (i.e., the two-stages are not examined simultaneously).

510 G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

DeGroot, Kiker, & Cross, 2000; Lowe, Kroeck, & Sivasubramaniam, 1996). When charisma has been studied in these past reviews, it is simply conceptualized as a sub-dimension of transformational leadership, typically by collapsing two dimensions of transfor- mational leadership together (i.e., idealized influence and inspirational motivation). Indeed, our review points at the reason why we remain enamored with charisma – a strong positive association with important outcomes. Fourth, our paper sets an agenda for future research in charismatic leadership based on current findings.

Hypotheses development

In this review, we concentrate specifically on charismatic leadership as leader signaling in line with Antonakis et al. (2016). Such signaling is enacted via expressive and passionate communication, use of symbols and metaphors as well as articulation of values (Connelly, Certo, Ireland, & Reutzel, 2011). Indeed, Antonakis et al. (2016) noted from their review of the conceptual and empirical landscape in charismatic leadership that three behaviors and/or attributions (we return to the distinction between these two below) that have been examined in prior studies are useful in defining charismatic leadership this way:

1) Justifying the mission by appealing to values that distinguish right from wrong 2) Communicating in symbolic ways to make the message clear and vivid, and also symbolizing and embodying the moral unity of the collective per se and 3) Demonstrating conviction and passion for the mission via emotional displays (Antonakis et al., 2016, p. 304)

Given the recency of this definition and the lack ofmeasures explicitly derived from this definition,wemapped these three dimen- sions on to existing ways of conceptualizing and measuring charismatic leadership. We did so by consulting all the articles that Antonakis et al. (2016) considered in their review that were marked as useful in defining charisma (see Table 1 of Antonakis et al., 2016). Most of these studies were qualitative in nature, or reviews or case studies. We found a handful of articles (e.g., Antonakis, Fenley, & Liechti, 2011; Pillai, 1996) that were quantitative studies. This review indicated to us that justification of themission by ap- pealing to followers' values and distinguishing right from wrong is conceptually similar to the attributed idealized influence dimen- sion found in earlier works in charismatic and transformational leadership. For example, in the Multifactor Leadership Questionnaire 5×, one of themostwidely used assessments of “charisma,” the idealized influence (attributed) dimension is defined as an attribution made by followers regarding a leader's higher order ideals and values (Bass & Avolio, 1995).

Interestingly, the second dimension from Antonakis et al. (2016) focusing on the collective and communication via symbols and metaphors is closest to idealized influence (behaviors) from the MLQ 5×. The MLQ 5× manual explains idealized influence (behaviors) as promoting a collective sense of mission and values to then guide one's behaviors (Bass & Avolio, 1995). The third dimension from Antonakis et al. (2016) focuses on expressive, passionate communication around the vision. We believe the Inspirational Motivation sub-dimension of the MLQ 5× along with Podsakoff, Mackenzie, Moorman and Fetter's (1990) and Conger and Kanungo's (1987, 1994) articulation of a vision dimension closely map onto the definition. We depict these definition- al overlaps in Table 1.

At first glance, the conceptual similarity abovemight seem likewhat has been done in the literature up to this point; however, it is divergent in the following ways. A majority of past empirical research has simply considered a sub-component of transformational leadership (Bass, 1985; Bass & Avolio, 1995), known as idealized influence, to be charismatic leadership (van Knippenberg & Sitkin, 2013). In such studies, a distinction is not made between attributions and behaviors per se. Further, another dominant approach sim- ply collapses idealized influence and inspirational motivation together citing high correlations between the two. We will not rehash all the literature that argues that such a high empirical correlation is due to study design issues and biases (Antonakis et al., 2010a,b; Bascle, 2008; Duncan, Magnusson, & Ludwig, 2004) and not indicative of a lack of conceptual distinctiveness (van Knippenberg & Sitkin, 2013). Instead, we adopt this literature's viewpoint that they are indeed distinct and especially so when considering attribu- tions and behaviors regarding idealized influence separately. Other conceptualizations of charismatic leadership exist, such as those by Conger and Kanungo (1987, 1994) and Shamir et al. (1993).

511G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

The conceptualization of charismatic leadership advanced by Conger and Kanungo (CK: 1987; 1994) focuses primarily on vision articulation and sensitivity to stakeholders. We discuss the vision articulation dimension of CK because sensitivity to stakeholders does not quite fit with the leader signaling definition articulated here. Other conceptualizations of charisma include those by Podsakoff et al. (1990), Rafferty and Griffin (2004), and Shamir, Zakay, Breinin and Popper (1998). Of all of these conceptualizations, only the articulating a vision dimension from Podsakoff et al. and CK fully fit the current definition of charisma and we included that component here as indicated in Table 1. To add to the conversation advanced by van Knippenberg and Sitkin (2013) and Antonakis et al. (2016), we sought to conduct a systematic search of the existing charismatic leadership literature where the current conceptual- ization of charismatic leadership could be applied (see the methods section for details).

Leader personality characteristics and charismatic leadership

We now transition to present a series of hypotheses regarding the role of charismatic leadership. We begin with a discussion of a set of exogenous (i.e., fixed) individual difference variables, whichmight predict charismatic leadership. These variables include per- sonality traits, such as the Big Five, cognitive ability, gender, and education. Such variables are largely stable (i.e., a leader's cognitive ability is not a function of the trust followers have in him or her), and therefore, can be considered to be truly exogenous and thus allow consistent, causal inferences (Antonakis, Day, & Schyns, 2012).

We then transition to explain in detail how charismatic leadership may predict outcomes, such as OCB, task performance, as well as group- or firm-level performance. We exclude from our review traditional leadership outcomes that cannot be argued to be endogenous to charisma due to endogeneity bias derived from issues such as omitted variables and simultaneity (i.e., reverse causality). Endogeneity bias can be characterized as an instance “where the effect of x on y cannot be interpreted because it includes omitted causes” (Antonakis et al., 2010a,b, p. 1087). For instance, perceptions of charismatic leadership and trust in a leader are not fixed, and consequently, causal inferences are not possible except perhaps in the most controlled experimental settings. In otherwords, charismatic leadershipmay cause trust in a leader and simultaneously trust in a leadermay cause perceptions of charismatic leadership. Hence, an endogeneity biasmay lead to erroneous results and incorrect conclusions (Hamilton&Nickerson, 2003).

We propose that the Big Five personality traits predict charismatic leadership to a varying extent. Because conscientiousness char- acterizes an individual who is organized, high achieving, and goal-directed, at first glance one might expect that such an individual may not display charismatic leadership when the latter is defined in the myriad of ways from the past (See Bono & Judge, 2004). In- deed, Bono and Judge (2004) did not hypothesize an association between conscientiousness and charisma as they argued such indi- viduals are not necessarily driven to exhibit the passion, enthusiasm, and creativity associated with charismatic leaders. We disagree with these authors and instead posit that conscientious individuals are also dependable and tend to havemore integrity as they relish in keeping their promises. As such, this personality trait might also lie at the core of being able to articulate values based messages to stakeholders, a key aspect of charismatic leadership. Furthermore, the integrity and consistency that characterize conscientiousness will translate into higher ratings of charisma as they are more likely to signal values important to the collective and follow through and act on such values.

Furthermore, there is no evidence in the personality literature that conscientious individuals do or do not display passion. From an intuitive standpoint, it is plausible that the deeply held convictions of conscientious leadersmightmake themnot only better at values based signaling, but also better able to express suchmessages in a passionate and forceful manner. The positive association between extraversion and charisma is well established in the literature. At the core of extraversion is a need for social dominance, affiliation, and positive affectivity. These core motives translate into warm and nurturing relational abilities, expression of passion and positive emotions, and ease for articulating positions (Bass & Avolio, 1995; Bono & Judge, 2004; De Hoogh, Den Hartog, & Koopman, 2005). At the opposite end of this spectrum, neuroticism is characterized by internal self-censure, guilt, and lower confidence in general. It is manifested in the experience and expression of more negative emotions and affect. As such, we expect neuroticism to be negatively related to charismatic leadership.

At the core of agreeableness is the motive to avoid conflict at all costs. We have no reason to believe this characteristic will be as- sociated with charismatic leadership.We quite agree with Bono and Judge (2004) that the lack of conflict characteristic of agreeable- ness might be associated more so with consideration type of leader behaviors, but not charisma as defined here. Finally, openness to experience should be positively associated with charismatic leadership. At the core of this trait is a thirst for new exploits, unconven- tional ideas and behaviors, as well as creativity. As such, we expect such individuals to have a wider arsenal of strategies in justifying the organization's mission, getting everyone focused on the collective and finding different forms of communication (e.g., symbols, metaphors) in expressing that mission. We thus propose the following hypothesis:

Hypothesis 1. Conscientiousness (1A), Openness to Experiences (1B) and Extraversion (1C) are positively related to charismatic leadership, whereas Neuroticism (1D) is negatively related and Agreeableness (1E) is unrelated to charismatic leadership.

Leader ability and charismatic leadership

A plethora of statements and theoretical accounts that charismatic leaders have unique and stable abilities that distinguish the charismatic leader from the non-leader are evident in the leadership domain (see Kirkpatrick & Locke, 1991; Zaccaro, Kemp, & Bader, 2004). While informative, many of these studies have largely been focused on general “visionary” or inspirational forms

512 G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

of leadership. We found little empirical research in ability and charismatic leadership per se, despite Bass (1985), Burns (1978), and House's (1977) works suggesting that charismatic leaders systematically differ from non-leaders. In line with earlier works, we posit that cognitive ability (i.e., general intelligence) will display a moderate to strong positive association with the current conceptualization of charismatic leadership. Higher cognitive ability allows individuals to scan complex and fast changing environ- ments, analyze such information and formulate solutions/messages that are novel. This last aspect is due to the link between in- telligence and creativity (Silvia & Beaty, 2012).

The signaling of what is right for the collective, while it might seem simple on the surface, should typically draw upon a keen and in-depth analysis of the complex external and internal business environment. Indeed, we argue that charismatic leadership inherently calls upon higher levels of creative thinking, sense making, and the creation of metaphors that convey the right emo- tional tone and meaning in comparison to other models of leadership (e.g., leader-member exchange or transformational). Silvia and Beaty (2012) found that the ability to generate higher quality creative metaphors was strongly related to fluid intelligence even after controlling for personality. As such, we expect to find a positive association between cognitive ability and charismatic leadership.

Hypothesis 2. Cognitive ability is positively related to charismatic leadership.

Leader demographics and charismatic leadership

Finally, following earliermeta-analytic reviews that have examined gender and leadership (e.g., Eagley, Johannsen-Schmidt, & van Engen, 2003; Paustian-Underdahl, Walker, & Woehr, 2014), we posit that women might experience a slight advantage in being per- ceived as charismatic compared tomen.While Eagly et al. (2003) found only a small magnitude effect for charismatic leadership spe- cifically, we argue that the emotion-laden communication of mission and values might come easier for women in leadership positions. On the other hand, men might not be as emotionally expressive due to stereotypical gender roles.

Past meta-analytic reviews on various styles of leadership have not considered the level of education or age of a leader (e.g., Banks, DavisMcCauley, Gardner, & Guler, 2016; Dulebohn, Bommer, Liden, Brouer, & Ferris, 2012; Harms & Credé, 2010;Wang, Oh, Courtright, & Colbert, 2011). Historical examples abound of perceived charismatic leaders who had various levels of education and age (e.g., Carson, 1987; Mio, Riggio, Levin, & Reese, 2005). Thus, there is no past evidence to suppose a largemagnitude relation between the level of a leader's education and age with their level of charismatic leadership. Further, drawing upon signaling theory (Connelly et al., 2011), we see no reason that level of education or age will affect leader signals related to emotional displays, communicating in symbolic ways, or appealing to values of followers. We propose the following:

Hypothesis 3. Women will be perceived as charismatic leaders more so than men.

Hypothesis 4. Leaders' level of education (4A) and age (4B) will display no association with charismatic leadership.

Outcomes of charisma

We next consider the extent to which charismatic leadership may relate to important work outcomes. First, charismatic leaders are those who do not maintain the status quo and strive to encourage followers to achieve levels of performance that exceed normal levels and normal roles. For this reason, researchers have argued that charismatic leaders inspire increased OCBs from followers (Babcock-Roberson & Strickland, 2010; Cho & Dansereau, 2010). It is because charismatic leaders exhibit and expect unconventional behavior from followers that higher levels of OCBs are likely to occur (Conger & Kanungo, 1994). Followers are more likely to under- stand the meaningfulness of their work when leaders who exhibit charisma clearly articulate a vision and describe the impact of achieving that vision.

Hypothesis 5. Charismatic leadership is positively related to follower OCBs.

One of the most important variables in an organizational behavior/human resource context is task performance. Charismatic leadership is expected to be related to the task performance of followers (Chen, Belkin, McNamee, & Kurtzberg, 2013; Halverson, Murphy, & Riggio, 2004). Charismatic leaders particularly excel at motivating followers to exhibit greater effort and the necessary behaviors to achieve high levels of performance (Sosik, Juzbasich, & Chun, 2011). This process is facilitated via inspiring followers to these higher expectations through the use of leader signaling regarding high standards, a collective identity and a passion for success.

Along these same lines, researchers have argued for the importance of charismatic leadership in higher positions within organizations (Waldman & Yammarino, 1999). Past empirical evidence has linked charismatic leadership to group- or firm- level performance (Awamleh & Gardner, 1999; Nohe, Michaelis, Menges, Zhang, & Sonntag, 2013). It is likely that charismatic leaders who articulate clear visions for groups and firms also provide a big picture understanding of what could be accomplished (Conger & Kanungo, 1987). As charismatic leaders emphasize collective identity, their influence is likely to motivate a large base of employment groups towards achieving goals that would not otherwise be possible without coordinated action (Shamir et al., 1998). We thus propose that charismatic leadership is positively related to task performance at the individual-level as well as group- or firm-level performance.

513G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

Hypothesis 6. Charismatic leadership is positively related to follower task performance.

Hypothesis 7. Charismatic leadership is positively related to group- or firm-level performance.

Lastly,we explore the extent towhich the position of a leader (i.e., front-line supervisors andmanagers relative tomid- and upper- level managers/executives) moderates the results of the previous hypotheses. We found evidence in the literature that a charismatic leader affects performance outcomes at both micro and macro levels as referenced above (Awamleh & Gardner, 1999; Chen, et al., 2013; Halverson et al., 2004; Nohe et al., 2013;Waldman& Yammarino, 1999), yetwe know little regarding if they aremore effective at higher or lower levels within an organization, or both,. On one hand, Antonakis et al. (2016) argue that the position of the leader does not matter:

“Charisma can be exhibited across all levels of institutions, from supervisors to team players in leaderless teams, all the way up to CEO or country presidents; it can also be exhibited in social movements and canwork from close or froma distance. Thus, whatwe have described is not a phenomenon that is uniquely linked to a position; it is a process of influence.” (p. 312)

Despite Antonakis et al.'s (2016) assertion above, our own review of the literature utilizing previous conceptualizations of cha- risma as a gift, magical quality, etc. (however flawed they might be) implicitly leans towards the examination of charisma as somewhat more appealing at higher levels in an organization (see earlier studies such as House, Spangler, & Woycke, 1991; Tosi, Misangyi, Fanelli, Waldman, & Yammarino, 2004; Waldman & Yammarino, 1999). As such, we explore the following research question:

Research question: To what extent does the position of a leader moderate the relations between charismatic leadership and its antecedents and outcomes?

Methods

Literature search

A searchwas executed to find both published and unpublished samples that investigated antecedents andwork outcomes of char- ismatic leadership. Following best practices guides from Kepes, McDaniel, Brannick, and Banks (2013), as well as Reed and Baxter (2009), we searched for samples through electronic databases, including the Social Science Citation Index, Google Scholar, ProQuest Dissertations and Thesis, as well as PapersFirst. This approach allowed us to identify both published journal articles as well as unpub- lished dissertations and conference papers. Databases were searched using combinations of charisma and charismatic paired with leader, leadership, and follower. The search was conducted using key words, titles, and abstracts. Furthermore, we performed manual searches of the Academy of Management and Southern Management Association annual conference programs. A cutoff date of August 31, 2015 was set. In order to identify any additional unpublished or forthcoming papers that were in press, a call was issued through the OB and LDRNET Academy of Management's listservs. In addition, we searched through the references of recent reviews of charis- matic leadership (e.g., Antonakis et al., 2016; van Knippenberg & Sitkin, 2013). A total of 204 direct requests were made to primary study authors in order to obtain additional samples.

Based on our operational definition of charismatic leadership (Antonakis et al., 2016), we included sub-dimensions of tradi- tional transformational leadership scales, such as the MLQ's idealized influence and inspirational motivation. There have been nu- merous published and unpublished meta-analyses on transformational leadership (e.g., Barlow, 2013; Bono & Judge, 2004; Buckman, Crawford, LePine, & Zhang, 2015; Burke et al., 2006; Dirks & Ferrin, 2002; Dumdum, Lowe, & Avolio, 2002; Eagly et al., 2003; Harms & Credé, 2010; Judge & Piccolo, 2004; Kraft & Bausch, 2015; Lowe et al., 1996; Sun, 2010; Wang et al., 2011). As part of a complementary search, we retrieved all possible samples from these past meta-analytic reviews in March 2016. As the meta-analytic reviews required updating, we searched for additional leadership studies. The same sources above were searched using combinations of transformational, idealized influence, and inspirational motivation paired with leader, leadership, and follower. Several dissertations were not available yet because of embargos. In summary, while our initial search included charisma as the focal word, we also conducted a secondary search with combinations of terms as noted above as many of these terms have been used interchangeably with charismatic leadership.

Coding procedures

In order to be included in the current study, several inclusion and exclusion criteria were created. First, the participants in the pri- mary study had to include individuals who were of the age of working adults and a part of the normal population (e.g., no probation officers; Heath, 2011). Second, studies had to report sufficient information in the form of correlation coefficients or statistics that could be converted to correlation coefficients (e.g., Cohen's d). Third, we determined to exclude studies that focused on historical leaders (e.g., through the study of speeches) to improve generalizability to the average workplace adult. Fourth, studies had to report one of the antecedents or outcomes specified in our hypotheses.

Fifth, studies had to clearly specifywhichmeasure of charisma they used. This criterionwas necessary to obtain the precise param- eter estimates desired. However, it resulted in the exclusion of studies that combined idealized influence and inspirationalmotivation without reporting unique estimates. Studies that measured charisma using the MLQ but did not specify what items were used (e.g., only idealized influence or some other combination of sub-dimensions) also had to be excluded. The one exception to this was for task performance in which we kept these types of studies for comparison purposes. Sixth, given the amount of data we

514 G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

have, we did not include estimates that have very low reliabilities in our analyses (i.e., correlations with idealized influence behavior where α = .42; von Wittich & Antonakis, 2011).

Finally, studies had to report variables that are somewhat free from endogeneity bias (Antonakis et al., 2016; Antonakis et al., 2010a,b; 2014). As previously stated, endogeneity bias prevents the interpretation of causal inferences. Thus, we elected to focus on mostly exogeneous and fixed variables in the form of individual differences and their relation to charismatic leadership when reported by others. We then focused on the relation between charismatic leadership and independently measured, other ratings of outcomes, such as OCBs, task performance, and group- or firm-level performance. Currently, only a limited number of studies have reported bias free correlations. For instance, none of the ratings of leaders were self-ratings, but all ratings of personality were self-rated. A supplemental check was performed upon completion of the coding to verify the accuracy of the coding. In total, this approach resulted in 76 samples and a total of 36,031 participants. We provide the input data in the appendix for each sample (e.g., reliabilities, sample size, correlation coefficients).

Detection heuristics were used to examine studies for potential duplicates, such as a conference paper or dissertation that had also been identified to be a published journal article (see Figure 1 for a step by step example of this process in Wood, 2008). Using Cohen's kappa (Cohen, 1960), five authors independently coded a subsample of studies (e.g., reliabilities, sample size, effect size). Inter-rater reliability was calculated across 185 coding decisions. Estimates of Cohen's kappa of less than .40 are considered to be poor, .40 to .75 is acceptable, and .75 or higher is thought to be excellent (Fleiss, 1981). Cohen's kappa estimates in this study were excellent (Cohen's kappa = 1.0 between rater 1 and rater 2; Cohen's kappa = 1.0 between rater 1 and rater 3; Cohen's kappa = 1.0 between rater 1 and rater 4; Cohen's kappa = .90 between rater 1 and rater 5). Additionally, two authors completed secondary checks of the primary data and any discrepancies were resolved through discussion.

In the event of longitudinal samples where variables were measured at multiple time points, Time 1 was coded to improve the comparability between samples that were collected using a cross-sectional design and those that used a longitudinal design. Many coded primary studies did not report reliability estimates, such as coefficient alpha. In this instance, average reliability was imput- ed using other reliability estimates of that variable in the same distribution. Helping behavior and the contentiousness OCB-sub scales were coded as proxy variables for OCBs as both constructs are specifically focused on measuring extra role behaviors. In one study, Grade Point Average was coded as a proxy for task performance (Chen, Belkin, McNamee, & Kurtzberg, 2013).

Meta-analytic procedures

Psychometric meta-analysis was used to analyze the primary samples identified to allow for the correction of measurement error (Schmidt & Hunter, 2015). In order to test the potential of moderating variables, we calculated 80% credibility intervals around the corrected effect size estimates. Wider intervals or the inclusion of zero within the intervals provides evidence consis- tent with a conclusion of moderating effects. We also report 95% confidence intervals.

Sensitivity analyses

It is important to assess the robustness of meta-analytic results, and this is particularly true in distributions with a smaller number of samples as is the case in the current study. We considered the potential for influential cases to lead to instability in the effect size estimates. Traditional approaches recommend checks for outliers, such as the sample-adjusted meta-analytic deviancy (SAMD) statistic (Beal, Corey, & Dunlap, 2002; Huffcutt & Arthur, 1995). Yet, this approach assumes a fixed-effects model and is limited as it implies an assumption of what can be classified as an outlier. Ideally, results should be reported with and without such outliers. For that reason, it is recommended to instead perform a one-sample removed analysis (Borenstein, Hedges, Higgins, & Rothstein, 2009). This approach calls for removing each sample one at a time and re-computing the effect size estimate. Consequently, one is given a range of estimates that provides a better understanding of the stability of the estimate.

A second recommended approach to consider the robustness of the estimates is to conduct publication bias analyses (for a re- view see Banks, Kepes, & McDaniel, 2015). Previous research has suggested that a minimum of 15 (Kepes, Banks, & Oh, 2014) to 20 available samples are needed to provide robust estimates (Harrison, Banks, Pollack, O'Boyle, & Short, in press). Given that many of the distributions did not meet these thresholds, we elected not to conduct publication bias analyses for most of the distributions.

As an illustrative example, we conducted trim and fill analyses (using a fixed-effects model), cumulative meta-analyses, as well as moderate selection model analyses on the gender-charisma relations (for details of these techniques see Kepes, Banks, McDaniel, & Whetzel, 2012; Harrison et al., in press). When executing the trim and fill analyses, we considered the potential for bias on both sides of the funnel plot as we were uncertain how publication bias might occur for the gender-charisma relations. We found that the trim and fill method imputed one sample on the right side of the funnel plot, but only adjusted the observed correlation by b .01 for inspirational motivation. Three, three, and zero samples were imputed for idealized influence (Δ b .01), idealized influence-attributes (Δ b .01), and idealized influence-behaviors (Δ = .00), respectively. Thus, the trim and fill analyses provided evidence that there is little to no publication bias. Both the cumulative meta-analyses and the moderate selection model analyses provided similar results. However, we encourage future research to consider the potential for this bias as more samples become available for all of the relations in this study.

515G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

Results

In this section, we review the findings of our analyses. First, we highlight the support for the hypotheses pertaining to ante- cedents of charismatic leadership. A relative weights analysis was conducted to provide a clearer understanding of the relative importance of the individual differences variables when predicting charismatic leadership. We then progress to a discussion of the outcomes of charismatic leadership and a comparison of our results to past meta-analytic findings. We conclude with a review of the one-sample removed analyses which serve as a robustness check for the meta-analytic findings.

Antecedents of charismatic leadership

Table 2 illustrates the relationships between charismatic leadership and its antecedents. First, the findings illustrate that the true-score correlation (bρ) between personality traits of the leader and ratings of charismatic leadership are mostly positive, but small in magnitude. The results are organized by the antecedents of charismatic leadership broken down by (1) idealized influ- ence, (2) inspirational motivation, and (3) vision. For comparison purposes, we break the results down further by scale, such as idealized influence (attributes and behavior), the full CK measure, as well as level of the leader. Most meta-analytic reviews combine measures of the same construct into one distribution to illustrate the true-score correlation across scales. However, be- cause of our interest in the dimensions of charismatic leadership and the specific measures that have been used, we disaggregate the distributions. This means that the distributions are noticeably smaller than the typical meta-analytic distribution. We believe that the precision of our method justifies the creation of these smaller distributions. Thus, while the distributions are small, we offer these results as the best estimates of the true population effect sizes.

From Table 2, we find that as predicted openness to experience (Idealized influence:bρ = .13, k = 6, N = 1356; Inspirational motivation: bρ = .16, k = 11, N = 1993), conscientiousness (Idealized influence:bρ = .07, k = 5, N = 1180; Inspirational motiva- tion: bρ= .06, k = 9, N = 1760), and extraversion (Idealized influence:bρ= .23, k = 4, N = 958; Inspirational motivation: bρ= .25, k = 10, N = 1908) are positively related to charismatic leadership. Thus, hypotheses 1A, 1B, and 1C, were supported; however, our findings show that the results range from small to moderate magnitude relations with extraversion having the strongest re- lation (see Bosco et al., 2015 for correlation effect size benchmarks). Further, the findings for these traits indicate that both attrib- uted and behavioral idealized influence relate to these Big Five traits.

In support of hypothesis 1D, neuroticism (Idealized influence:bρ= −.13, k = 6, N = 1310; Inspirational motivation: bρ= −.14, k = 11, N = 2038) also had a small negative relation to charismatic leadership. These associations held up for the dimensions of attributes and behaviors of idealized influence. However, counter to hypothesis 1E, agreeableness also showed a moderate level relation with charismatic leadership (Idealized influence:bρ = .26, k = 5, N = 1180; Inspirational motivation: bρ = .15, k = 10, N = 1863).

In support of hypothesis 2, cognitive ability showed a small to moderate relation with charismatic leadership (e.g., Inspiration- al motivation: bρ = .14, k = 3, N = 566). We also considered the potential role of the demographic variables of the leaders. For instance, we found little support for our assertion that females would have higher levels of charisma than males (hypothesis 3). In this context, a negative correlation indicates that women had higher levels of charisma (Idealized influence:bρ = −.05, k = 23, N = 16,851; Attributes:bρ = −.07, k = 14, N = 12,565; Behavior:bρ = −.02, k = 15, N = 12,995; Inspirational motivation: bρ = −.02, k = 27, N = 22,863). The evidence that women may score higher on charisma might be the result of selection effects whereby women could have to pass a higher barrier to emerge as leaders and only the most charismatic are selected (Antonakis et al., 2010a, b). Yet, all these relations were small in magnitude. We did find at least some evidence contrary to hy- pothesis 4A and 4B that leaders' level of education and age would be unrelated to their charismatic leadership. Age, especially, was related to leaders' idealized influence-behavior (bρ = .17, k = 4, N = 1391).

Supplemental analyses

In order to evaluate the relative importance of the individual difference variables, we performed a relative weights analysis (Johnson & LeBreton, 2004). The analysis allows for the calculation of epsilon weights for each of the predictors which sum to equal R2 and allow for a comparison via ratios. As an example, a weight of 0.40 is twice as important as a weight of 0.20 and to- gether the weights sum to 0.60 which would be the total variance explained (Tonidandel & LeBreton, 2014). After examining the individual difference variables in the current study, we elected to focus on the Big Five personality traits and cognitive ability be- cause these variables had the strongest relation with charisma. We created a meta-analytic correlation matrix using the estimates from the current study as well as estimates from past meta-analytic reviews (see Ackerman & Heggestad, 1997; Schmidt, Shaffer, & Oh, 2008; Van der Linden, te Nijenhuis, & Bakker, 2010). The meta-analytic correlation matrix served as input into an SPSS matrix regression macro for relative weights analyses (Johnson, 2001).

The results of the relative weights analyses are presented in Table 3. Quite interestingly, the results varied across the different di- mensions of charismatic leadership. When considering idealized influence-attributes, cognitive ability was the most important pre- dictor (explaining 45.6%). For idealized influence-behavior, cognitive ability was again the most important predictor (explaining 50.9%). Finally, when predicting inspirational motivation, extraversion was relatively more important (explaining 48.6%). In general, it appears that some of the Big Five traits and cognitive ability play a more important role than others in predicting charismatic leadership.

Table 2 Antecedents of charismatic leadership (CL) by measure.

Variable k N r SDr bρ SDρ CVLL CVUL CILL CIUL %Var

Openness to experience Idealized influence 6 1,356 .11 .07 .13 .09 .08 .17 .07 .19 84% Level of leader: lower 2 653 .11 .05 .14 .06 .14 .14 .07 .22 99% Idealized influence-attributes 4 596 .11 .08 .14 .10 .14 .14 .06 .22 99% Idealized influence-behavior 3 543 .08 .09 .10 .12 .01 .20 -.01 .22 63% Inspirational motivation 11 1,993 .13 .06 .16 .08 .16 .16 .11 .20 99% Level of leader: lower 2 653 .14 .02 .17 .03 .17 .17 .10 .25 99% CK 3 514 .17 .16 .23 .18 .02 .43 .03 .42 26% Vision – – – – – – – – – – –

Conscientiousness Idealized influence 5 1,180 .06 .06 .07 .07 .07 .07 .01 .13 99% Level of leader: lower 3 820 .08 .05 .10 .06 .10 .10 .03 .17 99% Idealized influence-attributes 4 596 .11 .12 .13 .16 −.03 .28 −.02 .27 42% Idealized influence-behavior 3 543 .12 .11 .15 .15 −.01 .30 −.01 .31 38% Inspirational motivation 9 1,760 .05 .08 .06 .10 .00 .13 .01 .12 76% Level of leader: lower 4 704 .07 .08 .08 .10 .02 .14 −.01 .17 79% Vision – – – – – – – – – – –

Extraversion Idealized influence 4 958 .19 .08 .23 .10 .14 .32 .14 .32 53% Level of leader: lower 2 598 .23 .05 .28 .06 .28 .28 .20 .35 99% Idealized influence-attributes 5 744 .10 .10 .13 .12 .04 .22 .04 .23 68% Idealized influence-behavior 4 691 .09 .11 .11 .14 −.02 .24 −.01 .24 48% Inspirational motivation 10 1,908 .20 .10 .25 .12 .13 .36 .17 .32 46% Level of leader: lower 3 746 .28 .02 .34 .03 .34 .34 .27 .41 99% CK 2 352 .17 .00 .21 .01 .21 .21 .11 .31 99% Vision – – – – – – – – – – –

Agreeableness Idealized influence 5 1,180 .21 .07 .26 .09 .19 .32 .19 .33 72% Level of leader: lower 2 598 .20 .07 .26 .01 .26 .26 .18 .33 99% Idealized influence-attributes 4 596 .11 .11 .15 .14 .04 .27 .03 .27 57% Idealized influence-behavior 3 543 .13 .07 .16 .09 .16 .16 .08 .25 99% Inspirational motivation 10 1,863 .12 .09 .15 .11 .06 .24 .09 .21 64% Level of leader: lower 2 598 .18 .03 .23 .04 .23 .23 .15 .30 99% Vision – – – – – – – – – – –

Neuroticism Idealized influence 6 1,310 −.11 .07 −.13 .09 −.17 −.09 −.19 −.07 86% Level of leader: lower 4 950 −.13 .07 −.16 .08 −.18 −.13 −.22 −.09 94% Idealized influence-attributes 5 744 −.07 .09 −.08 .12 −.16 −.01 −.17 .01 76% Idealized influence-behavior 4 691 −.05 .10 −.06 .14 −.19 .06 −.18 .06 52% Inspirational motivation 11 2,038 −.11 .10 −.14 .12 −.25 −.03 −.20 −.07 53% Level of leader: lower 4 876 −.16 .10 −.21 .12 −.32 −.10 −.32 −.11 49% Vision – – – – – – – – – – –

Cognitive ability Idealized influence – – – – – – – – – – – Idealized influence-attributes 3 566 .15 .15 .19 .19 −.03 .40 −.02 .39 22% Idealized influence-behavior 2 513 .17 .15 .21 .19 −.01 .43 −.04 .47 16% Inspirational motivation 3 566 .12 .14 .14 .17 −.05 .33 −.05 .32 25% Vision – – – – – – – – – – –

Age Idealized influence – – – – – – – – – – – Idealized influence-attributes 4 1,229 .00 .01 .00 .01 .00 .00 –.06 .05 99%

2 797 .00 .00 .00 .00 .00 .00 –.07 .07 99% Idealized influence-behavior 4 1,391 .14 .11 .17 .12 .03 .30 .05 .28 25% Level of leader: lower 2 797 .04 .01 .05 .01 .05 .05 –.02 .12 99% Inspirational motivation 6 1,515 .05 .07 .06 .08 .02 .10 .00 .12 84% Level of leader: lower 2 286 .16 .11 .18 .11 .10 .26 .03 .33 65% Level of leader: upper 2 797 .02 .01 .03 .01 .03 .03 –.04 .10 99% Vision – – – – – – – – – – – CK 2 1,052 −.10 .04 −.11 .04 −.11 −.11 −.17 −.05 99%

Male = 1, Female = 0 Idealized influence 23 16,851 −.04 .05 −.05 .06 −.10 .00 −.07 −.02 52% Level of leader: lower 6 1,277 .03 .13 .04 .14 −.11 .18 −.07 .14 30% Level of leader: upper 3 1,639 −.01 .04 −.01 .05 −.01 −.01 −.06 .04 99%

516 G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

Table 2 (continued)

Variable k N r SDr bρ SDρ CVLL CVUL CILL CIUL %Var

Idealized influence-attributes 14 12,565 −.06 .04 −.07 .05 −.10 −.03 −.09 −.04 64% Level of leader: lower 2 172 .06 .11 .07 .13 .02 .12 −.09 .23 90% Idealized influence-behavior 15 12,995 −.01 .04 −.02 .04 −.03 .00 −.03 .00 92% Level of leader: lower 2 172 −.07 .12 −.09 .14 −.16 −.01 −.26 .08 83% Inspirational motivation 26 22,802 −.02 .06 −.02 .06 −.09 .04 −.05 .00 35% Level of leader: lower 4 1,122 −.03 .06 −.03 .07 −.05 −.01 −.09 .03 94% Vision – – – – – – – – – – –

Education Idealized influence – – – – – – – – – – – Idealized influence-attributes 2 797 .00 .08 .00 .09 −.09 .09 −.12 .12 42% Idealized influence-behavior 2 797 .01 .04 .02 .04 .02 .02 −.05 .09 99% Inspirational motivation 2 797 .05 .09 .05 .10 −.06 .16 −.09 .19 29% Vision – – – – – – – – – – –

Note. k = number of independent samples; N = total sample size; r = sample-size-weighted mean observed correlation; SDr = sample-size-weighted observed

standard deviation of correlations; bρ = mean true-score correlation (corrected for unreliability for both variables); SDρ = standard deviation of corrected correla- tions; CVLL and CVUL = lower and upper bounds, respectively, of the 80% credibility interval; CILL and CIUL = lower and upper bounds, respectively, of the 95% confidence interval around the mean true-score correlation; %Var = percentage of variance attributable to statistical artifacts; CK: Conger & Kanungo scale; *Neg- ative correlation indicates that women were higher on charismatic leadership; male = 1 and female = 0.

517G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

In addition to relative weights presented above, we examined the incremental predictive validity of personality in predicting the dimensions of charisma over and above gender and cognitive ability (see Table 4 for these findings). Cognitive ability predicted additional variance over and above gender (ΔR2 = 0.02–0.05) as did personality traits over and above cognitive ability and gender (ΔR2 = .04–0.07)

Outcomes of charismatic leadership

We also investigated the potential for charismatic leadership to relate to supervisor-rated or objective measures of various performance outcomes. We present these results in Table 5. In support of hypothesis 5, we find evidence that idealized influence (bρ= .12, k = 2, N = 141) and inspirational motivation predict followers' engagement in OCBs (bρ= .28, k = 4, N = 523). Further, while we found support for hypothesis 6 which stated that charismatic leadership would relate to supervisor-rated task performance, the magnitude of the relation varied greatly by scale. For example, inspirational motivation (bρ = .20, k = 9, N = 1,861), vision (bρ = .21, k = 5, N = 1,318), CK (bρ = .45, k = 2, N = 411), and various combinations of scales that could not be disaggregated (bρ = .33, k = 11, N = 2,546) all predicted task performance by varying degrees. Similar results were also found in support of hypothesis 7 which predicted a relation between charismatic leadership and group- or firm-level perfor- mance (e.g., Inspirational motivation: bρ = .08, k = 4, N = 413).

Finally, we considered how our results might be moderated by the level of a leader within an organization when the data allowed for such a comparison. By and large, across the distributions, most of the studies either reported samples of lower-

Table 3 Relative-importance of individual differences as predictors of charisma dimensions

Idealized influence-attributes Idealized influence-behavior

Raw relative weights Relative weights as a % of R2 Raw relative weights Relative weights as a % of R2

Openness 0.007 9.4 0.002 2.9 Conscientiousness 0.009 13.3 0.016 18.2 Extraversion 0.008 10.9 0.006 6.7 Agreeableness 0.013 19.0 0.017 19.6 Neuroticism 0.001 1.8 0.001 1.6 Cognitive ability 0.032 45.6 0.044 50.9

R2 = 0.070 R2 = 0.086

Inspirational motivation

Raw relative weights Relative weights as a % of R2

Openness .009 10.4 Conscientiousness .002 1.7 Extraversion .043 48.6 Agreeableness .013 14.5 Neuroticism .008 8.9 Cognitive ability .014 16.0

R2 = 0.088

Table 4 The incremental validity of gender, cognitive ability, and Big Five personality traits

Idealized influence-attributes Idealized influence-behaviors

β SE β SE β SE β SE β SE β SE

Gender −0.07⁎⁎ 0.02 −0.10⁎⁎ 0.02 −0.15⁎⁎ 0.02 −0.02 0.02 −0.05⁎ 0.02 −0.11⁎⁎ 0.02 Cognitive ability 0.21⁎⁎ 0.02 0.22⁎⁎ 0.02 0.22⁎⁎ 0.02 0.26⁎⁎ 0.02 Neuroticism 0.10⁎⁎ 0.03 0.12⁎⁎ 0.03 Extraversion 0.10⁎⁎ 0.03 0.09⁎⁎ 0.03 Openness −0.01 0.03 −0.07⁎ 0.03 Agreeableness 0.12⁎⁎ 0.02 0.14⁎⁎ 0.03 Conscientiousness 0.11⁎⁎ 0.03 0.15⁎⁎ 0.03

R2 = 0.005 R2 = 0.046 R2 = 0.088 R2 = 0.000 R2 = 0.046 R2 = 0.096 Δ R2 = 0.041⁎⁎ Δ R2 = 0.042⁎⁎ Δ R2 = 0.046⁎⁎ Δ R2 = 0.050⁎⁎

Inspirational motivation

β SE β SE β SE

Gender −0.02 0.02 −0.04⁎ 0.02 0.05⁎⁎ 0.02 Cognitive ability 0.15⁎⁎ 0.02 0.12⁎⁎ 0.02 Neuroticism −0.02 0.02 Extraversion 0.22⁎⁎ 0.02 Openness 0.01 0.02 Agreeableness 0.11⁎⁎ 0.02 Conscientiousness −0.05⁎⁎ 0.02

R2 = 0.000 R2 = 0.021 R2 = 0.090 Δ R2 = 0.021⁎⁎ Δ R2 = 0.069⁎⁎

Note. Gender (male = 1; female = 0); standardized regression coefficients are reported. ⁎ p b .05 ⁎⁎ p b .01

518 G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

level leaders or mixed samples, the latter of which did not allow for a comparison. In most cases, the results were not noticeably different for samples with only lower-levels of leaders relative to the full distribution of leaders.

We summarize the results of our analyses in Table 6. Here, we also include a comparison with past meta-analytic reviews that considered the relations between antecedents of charismatic leadership and its sub-dimensions. In general, we can see that the vast majority of research has been conducted examining the idealized influence and inspirational motivation dimensions col- lapsed into one dimension with much less attention being paid specifically to idealized influence attributes and behaviors as well as vision. Of the personality traits considered, across both the Bono and Judge (2004) meta-analysis and the current study, conscientiousness tends to have the weakest relation with charisma while extraversion and agreeableness tend to have the

Table 5 Outcomes of charismatic leadership (CL) by measure.

Variable k N r SDr bρ SDρ CVLL CVUL CILL CIUL %Var

Task performance (supervisor-rated) Idealized influence – – – – – – – – – – – Inspirational motivation 9 1,861 .17 .06 .20 .06 .20 .20 .16 .25 99% Level of leader: lower 6 1,058 .18 .06 .21 .06 .21 .21 .15 .27 99% Vision 5 1,318 .19 .10 .21 .13 .07 .34 .10 .31 30% CK 2 411 .39 .09 .45 .10 .37 .54 .33 .58 51% Other combinations 11 2,546 .29 .24 .33 .27 .00 .67 .18 .49 7%

OCB (supervisor-rated) Idealized influence 2 141 .10 .02 .12 .02 .12 .12 −.05 .28 99% Inspirational motivation 4 523 .25 .08 .28 .10 .26 .31 .20 .37 96% Vision – – – – – – – – – – –

Group or firm performance Idealized influence – – – – – – – – – – – Idealized influence-attributes – – – – – – – – –– – – Idealized influence-behavior – – – – – – – – – – – Inspirational motivation 4 413 .07 .08 .08 .09 .08 .08 −.02 .17 99% Level of leader: lower 2 144 .17 .01 .19 .01 .19 .19 .03 .35 99% Vision – – – – – – – – – – –

Note. k = number of independent samples; N = total sample size; r = sample-size-weighted mean observed correlation; SDr= sample-size-weighted observed

standard deviation of correlations; bρ = mean true-score correlation (corrected for unreliability for both variables); SDρ = standard deviation of corrected correla- tions; CVLL and CVUL = lower and upper bounds, respectively, of the 80% credibility interval; CILL and CIUL = lower and upper bounds, respectively, of the 95% confidence interval around the mean true-score correlation; %Var = percentage of variance attributable to statistical artifacts; CK: Conger & Kanungo scale; OCB = Organizational citizenship behaviors.

Table 6 Summary of the relations between dimensions of charismatic leadership and its antecedents and outcomes.

Construct Measure

II IIA IIB IM II & IM Vision CK Openness to experience

B: ρ = .13 B: ρ = .14 B: ρ = .10 B: ρ = .16 BJ: ρ = .22 – B: ρ = .23

Conscientiousness B: ρ = .07 B: ρ = .13 B: ρ = .15 B: ρ = .06 BJ: ρ = .05 – – Extraversion B: ρ = .23 B: ρ = .13 B: ρ = .11 B: ρ = .25 BJ: ρ = .22 – B: ρ = .21 Agreeableness B: ρ = .26 B: ρ =. 15 B: ρ = .16 B: ρ = .15 BJ: ρ = .21 – – Neuroticism B: ρ = −.13 B: ρ = −.08 B: ρ = −.06 B: ρ = −.14 BJ:

ρ = −.17 – –

Cognitive ability – B: ρ = .19 B: ρ = .21 B: ρ = .14 – – – Age – B: ρ = .00 B: ρ = .17 B: ρ = .06 – – B:

ρ = −.11 Male or female B: d = −.10; E:

d = −.09 B: d = −.14; E: d = −.12

B: d = −.04; E: d = −.02

B: d = −.04; E: d = −.05

– – –

Education – B: ρ = .00 B: ρ = .02 B: ρ = .05 – – – Task performance – – – B: ρ = .20 – B:

ρ = .21 B: ρ = .45

OCB B: ρ = .12 – – B: ρ = .28 – – – Group or firm performance

– – – B: ρ = .08 – – –

Note: II: Idealized influence; IIA: Idealized influence-attributes; IIB: Idealized influence-behavior; IM: Inspirational motivation; CK: Conger & Kanungo; B: Banks et al. BJ: Bono & Judge (2004); E: Eagly et al. (2003); ⁎Negative Cohen's d indicates that women were higher on charismatic leadership.

519G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

strongest. In terms of predicting important outcomes, more gaps exist in the current literature. The data that are available tend to show that charisma is related to supervisor-rated task performance and OCBs.

Sensitivity analysis

One-sample removed As previously described, we conducted sensitivity analyses to determine the robustness of our findings. To accomplish this, we

completed a one-sample removed analysis using comprehensive meta-analysis (Borenstein, Hedges, Higgins, & Rothstein, 2009). For instance, the neuroticism-inspirational motivation relation illustrated in a range of just .03 if any one sample were to be removed suggests that this distribution is robust. Relatively speaking, the relation of charisma (usingmixed combinations of scales) to task per- formance was slightly less robust showing a range of .10. By comparison, the inspirational motivation-task performance link only il- lustrated a range of .02. In general, instances where the range of estimates is small, one can have greater confidence in the findings. Conversely, when the removal of individual studies results in a wide range of estimates, confidence in the results is diminished and more research is needed. We encourage further robustness checks as the amount of available data increases.

Discussion

Recently, the literature of charismatic leadership has received great criticism (Antonakis et al., 2016) and some have called for re- searchers to “go back to the drawing board” (van Knippenberg & Sitkin, 2013; Yukl, 1999). However, we argue that such criticism, while not unfounded, is shortsighted in that it is difficult to criticize the conceptual and empirical literature without first conducting a systematic review in the form of a meta-analysis to identify strengths and specific developmental needs. Our study provides three primary contributions to the literature. First, in this review, we adopted a rigorous and updated definition of charisma (Antonakis et al., 2016 based their definition on an exhaustive review spanning six decades across a variety of disciplines). This conceptualization of charismatic leadership also allowed us to examine a range of antecedents and outcomes at the individual factor level (e.g., Idealized Influence – Attributes) rather than an overall collapsed charisma dimension.

Second, we then proceeded to conduct a meta-analysis of important antecedents of charismatic leadership free from endogeneity bias. Purged of such biases, the antecedents of charismatic leadership as defined here reveal many nuanced findings. For example, while not expected, agreeableness emerged as a stronger predictor of idealized influence (attributes and behaviors) than extraversion via relative weights analyses (See Table 3). The fundamental motivation in agreeableness is harmony striving and avoidance of con- flict. Perhaps charismatic leaders who are agreeable seek to articulate values that are universal, work to reduce incompatibilities and promote harmony. Regardless of what the specific pathwaysmight be, we believe this finding can open the door for future theorizing concerning agreeableness and idealized influence.

Similarly, looking across Table 3 and Table 4, it is noteworthy that cognitive ability and conscientiousnesswere stronger predictors of idealized influence (attributes and behaviors) than extraversion. This finding again speaks to the importance of future theorizing regarding the complex associations amongst the less volatile personality characteristics such as agreeableness and conscientiousness and cognitive abilitywith idealized influence attributes and behaviors. Yet, extraversion is the strongest predictor of inspirationalmo- tivation, relative to the other individual differences considered. While past research has illustrated the potential for charismatic

520 G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

leadership to be trained (Antonakis et al., 2011; Frese, Beimel, & Schoenborn, 2003; Towler, 2003), it is also useful to identify individ- ual differences, such as personality traits and cognitive ability, that can be used to identify potential charismatic leaders.

As a third contribution, we also examined the extent to which charismatic leadership predicted important work outcomes that were from other rated sources. This is a key finding (see Table 6) relative to past reviews as there is a paucity of research establishing the substantive validity of charismatic leadership. For instance, charismatic leadership can lead to increases in positive follower be- haviors related to OCBs, as well as higher-levels of supervisor-rated task performance. What's more, charismatic leadership may also result in higher levels of group- or firm-level performance.

While the information learned about the relationships among charismatic leadership and important work outcomes is impor- tant for its practical implications, it is equally important to understand the contingency factors that might come into play. Where possible, we did identify that level of the leaders within the organization mostly did not moderate our findings. Yet, the primary study literature proved to be largely insufficient in allowing other contingency factors to be tested. Subsequently, we were unable to test for several contextual factors, for instance, that might moderate the extent to which charismatic leadership drives each of these important work outcomes.

Practical implications

Several findings from the present study merit discussion of their practical implications. First, the results support a somewhat sys- tematic, dispositional component of charismatic leaders; some trait-like individual differences demonstrated strong relations in predicting charismatic leadership while the results for other traits, abilities, and demographic variables varied in magnitude. These trait-based results may serve to inform the nature of selection criteria used in practice. Second, this study provides further insight into the conditions thatwarrant charismatic leadership. Although somehave argued that the charismatic leader is incompatible in con- texts that necessitate shared leadership (e.g., Yukl, 1999), the present workmay support otherwise. Charismamay be particularly im- portant for leaders in a professional or service context, where OCBs are valuable for front line employees. Further narrowing the scope of charisma's application, the meta-analytic results reveal positive effects on task performance. Results also situate charismatic lead- ership as a predictor of group- and firm-level performance, which may implicate selection and training procedures for management.

Limitations and future directions

We now discuss the limitations of this current research and opportunities for future research. First, different scales were used for themeasurement of charisma in thismeta-analysis. This included theMLQ, CK, and other questionnaires that were formulated often- times by a researcher working in isolation for the particular purpose of their study. There certainly are advantages to using multiple settings, samples, and scales (Sackett & Larson, 1990; Schmidt & Hunter, 2015). This is because meta-analyses capitalize on triangu- lation. Triangulation is defined as the use of “multiple reference points to locate an object's exact position” (Jick, 1979, p. 602).

The predominant use of theMLQand CK scales jointlywithin our study allowed formultiple assessments of charismatic leadership in relation to our specific targeted variables. This allows for amore robust interpretation of the relationships as they are drawing upon multiple points of reference (as compared to very large primary studies). Given the arguments put forth by Antonakis et al. (2016) and van Knippenberg and Sitkin (2013), we advocate greater exploration of conceptualizations and measures of charisma. We point to the behavioral approach by Shamir et al. (1993, 1998) as one means to accomplish this. Along this same line of thinking, fu- ture research should pursue objective measures of charisma (e.g., Fanelli, Misangyi, & Tosi, 2009; Jacquart & Antonakis, 2015; Mio, Riggio, Levin, & Reese, 2005) as well as the manipulation of charisma in experimental settings (e.g., Antonakis, Fenley, & Liechti, 2011; Awamleh & Gardner, 1999; Frese et al., 2003).

Second, if onewishes to consider the relations between charismatic leadership and important outcome variables, methodological stepsmust be taken to ensure that such relations are free from endogeneity bias (Antonakis et al., 2010a,b). This can be accomplished, for example, by correlating objectively coded measures of charisma with objectively measured outcomes. We encourage readers to exercise cautionwhen interpreting the relations between charisma and the outcomes reported in this study, because it is not possible to decisively say that such relations are free from this bias. Even if themeasurement of charismatic leadership and the outcomes come from two separate sources, it is possible that (a) there is a performance-cue effect (Lord, Binning, Rush, & Thomas, 1978) or (b) there is a simultaneity. For example, effective firmsmay select effective leaders ormay excel at training them. Further, (c) omitted variables like instrumental or transactional leadershipmay correlate with both charisma aswell as the outcome. In the current study, wewere not able to investigate the partial correlation coefficient which could have provided a more unbiased estimate of the relations of in- terest (for extensive discussions see Antonakis et al., 2010a,b; Antonakis & House, 2014).

Third, the majority of the studies we identified in our systematic search presented correlations between charisma and anteced- ents/outcomes of interest that were affected by endogeneity bias. Because we wish to present true score correlations free from such bias, these datawere excluded from the currentmeta-analytic review. Similar to past researchers,we lament themethodological limitations of these studies and recommend that future researchers conduct studies free from endogeneity bias (Antonakis et al., 2016; 2010a,b; 2014). Further, although it would have been interesting to compare charismatic leadership to similar styles, we could only do this if endogeneity bias was not a factor. We could not consider these relations because we did not have such bias free estimates. Future research should seek to address these limitations in the current literature.

Fourth, in distributions with smaller ks, the estimates will naturally be less robust. In order to combat this issue, more research is needed to examine the stability of the true-score correlations. More studies are needed to increase our confidence of the rela- tions presented in this study. One-sample removed analyses were also used to assess the strength of the true-score correlations.

521G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

As a result, some variables were highlighted as more robust than others. Further research is necessary to allow for larger ks and more significant interpretation of relationships with charismatic leadership.

Fifth, given the small number of samples, this limited our ability to test potential theoretical and methodological moderating variables. For example, the level of leader was not always discernible from the primary study. Without clear delineation, it is dif- ficult to compare between first-line supervisors, second-line supervisors, and CEOs as the composition of a sample. Future re- search should identify relationships with charismatic leadership and identify the level of leader as a possible moderating variable. It would also be valuable to look at how other individual differences in leaders, such as dark triad personality traits (O'Boyle, Forsyth, Banks, & McDaniel, 2012), in leaders predict charismatic leadership (Howell & Shamir, 2005; Walter & Bruch, 2009).

Contextual moderating variables, such as situational triggers like crises, attributional ambiguity, or information uncertainty should be identified in future research. One anonymous reviewer suggested the importance of considering the role of charismatic leadership through a cultural lens, such as by considering the differences between collectivist and individualist contexts as well as other cultural dimensions (e.g., power distance, uncertainty avoidance, gender egalitarianism). At this point, charismatic leader- ship is primarily studied in individualistic countries. The use of a collectivist perspective would be helpful for a deeper under- standing of charismatic leadership and its role within collectivist cultures (Walter & Bruch, 2009).

Context can also be measured through the use of different industries. Past research has already suggested that the importance of charismatic leadership might vary by industry (Shamir et al., 1993). More research is needed to test this assertion meta-ana- lytically. Further, researchers have also proposed that organizational size may play an important role in the predictive validity of charismatic leadership (Berson, Shamir, Avolio, & Popper, 2001). Such moderating variables should be considered as more sam- ples emerge that can be meta-analyzed. Finally, another limitation to keep in mind is the inability to prove causality. The identi- fication of more antecedents and outcomes in the study of charismatic leadership will not allow for causal statements with current settings. For example, does charisma in leaders actually cause trust or does trust lead to charismatic leadership? More ex- perimental research will be needed to address these concerns.

Overarching recommendations for theory development:

1. A systematic theoretical development of the antecedents and outcomes of the charismatic leadership construct is as yet un- available in the literature. For example, when one considers charismatic leadership as leader signaling, our relative weights analysis indicates that the somewhat less “glamorous” Big Five traits such as agreeableness and conscientiousness could matter more in some cases than the routinely studied extraversion which is associated with status striving and implicitly linked to leadership. Likewise, we know little concerning why leaders with higher cognitive ability are better at values based signaling. It could be that such leaders are better able to discern what values are shared by followers and thus are better at signaling such values. A study of other traits (e.g., core self-evaluations) could be fruitful as well. For example, as we worked on this review, we uncovered studies on emotional intelligence and charisma, but the bulk of such studies conceptualized emotional intelli- gence as a skill rather than an ability. Theories on the interaction between cognitive ability and emotional (or other forms of) intelligence are sorely needed for predicting charismatic leadership. Further, and related to traits and stable abilities, non-linear effects of such traits on charisma should be theorized and empirically examined (Simonton, 1985; Ames & Flynn, 2007). For example, at what point is an agreeable leader no longer considered charismatic? Or when does a leader's intelli- gence (perhaps at very high levels or low levels) adversely affect their charismatic behaviors and attributes? If intelligence and/or personality traits indeed have non-linear associations with charisma, the estimates reported in linear tests such as ours are attenuated.

2. Theoretical accounts of how emotions factor into passionate displays and communication are still under developed despite the voluminous research on emotion and leadership in general. Indeed, while the early works of Bass (1985) suggest inspirational leaders display more passion, we need theories that articulate the mechanisms through which displays of hope, compassion, righteous anger or sadness (i.e., all discrete emotions) are a part of the charismatic leader's signaling repertoire. As an aside, how charismatic leaders might use dark emotions and divisive values to signal crises to followers is largely unexplored and fertile domain.

3. Is charismatic leadership a static construct? We were hard pressed to locate accounts of how a leader's signaling ability might change over time and with age and/or life experiences. Our findings point at the possibility that leaders are better able to ar- ticulate visions as they get older but only at higher levels in the organizations. It is not our intent here to mimic the authentic leadership literature, but one cannot help but wonder if the leader's vision articulation is tied in some fashion to their own evolving authenticity, life experiences and moral growth.

4. Our findings indicated that lower-level leader's charismatic leadership seemed to impact higher-level outcomes. What are the theoretical mechanisms that could make such effects plausible? Is it because lower-level leaders play a key role in focusing at- tention on the collective, as they are the immediate face of the organization to followers?

5. Finally, the gender findings seem to be pretty consistent from Eagley et al. (2003) as well as our own study. Intuitively, it comes as a surprise that women are not perceived as more charismatic given all the literature that suggests that women tend to be more values oriented and more passionate in general (Eagly et al., 2003). Yet, on second glance we wonder if people hold implicit leadership prototypes (Lord, Brown, Harvey, & Hall, 2001) when it comes to values based and emotion laden sig- naling. This is a rich area for future theoretical development so we could better understand how and why women are per- ceived as more passionate in general but not so in leadership roles. Could gender role theory be at play wherein female passion is valued in certain stereotypical roles (e.g., parenting) but not as a CEO?

522 G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

Overarching recommendations for measurement and design:

1. The most immediate and urgent methodological agenda item is the development and validation of new measures that better reflect the three dimensions of charismatic leadership highlighted here. Future research should also pursue objective measures of charisma as well as manipulated charisma in experiments.

2. Despite all the attention charismatic leadership has received, we were struck by how few studies exist that rigorously test an- tecedents and outcomes of charismatic leadership free from endogeneity bias. Such studies along with enhanced measurement are urgently needed.

3. Improve reporting of study context to allow future meta-analytic reviews to test for a greater number of theoretically impor- tant moderators.

4. Tests of incremental validity from the new measurement of charismatic leadership with existing models of leadership. 5. We were limited by the smaller samples in this domain but levels of analyses issues both in theory, measurement and design

in charismatic leadership seem generally lagging behind some other leadership models. 6. Finally, future research if undertaken with existing measures should examine charisma at the dimensional level.

Conclusion

Charismatic leadership is of great importance to both academic and practitioner interests alike. The currentmeta-analysis provides estimates that reflect our best understanding of the population parameter estimate using themost precise measures (Antonakis et al., 2016). We considered the conceptualization and measurement of charisma as well as antecedents and outcomes free from endogeneity issues. Some personality traits were identified as important predictors of charismatic leadership while others require fu- ture investigation (e.g., dark triad traits). Also, charisma in leaders proved to be an important predictor of supervisor-rated task per- formance as well as group- or firm-level performance. Due to the relatively small amount of samples explicitly studied in the charismatic leadership literature, more research should focus on the antecedents and outcomes of charisma to increase the robustness of our findings. Also, close attention should be paid to the contingency factors of these relationships. Moderating variables such as the level of leader and context should be identified in future studies. Overall, charismatic leadership is a significant construct in the lead- ership literature and research should continue to focus on its conceptualization, measurement, and role within organizations.

Appendix A. Appendix

Main codes and input values of each primary study/sample included in the meta-analysis.

Author Year n r ryy rxx u Leader V. Other V.

Antonakis & House 2004 30 −0.15 0.69 0.8 1.0 IIB Agreeableness Antonakis & House 2004 30 −0.05 0.69 0.65 1.0 IIA Agreeableness Antonakis & House 2004 30 −0.13 0.69 0.78 1.0 IM Agreeableness Antonakis & House 2004 30 0.02 0.66 0.8 1.0 IIB Conscientiousness Antonakis & House 2004 30 0.01 0.66 0.65 1.0 IIA Conscientiousness Antonakis & House 2004 30 0.04 0.66 0.78 1.0 IM Conscientiousness Antonakis & House 2004 30 0.4 0.76 0.8 1.0 IIB Extraversion Antonakis & House 2004 30 0.43 0.76 0.65 1.0 IIA Extraversion Antonakis & House 2004 30 0.41 0.76 0.78 1.0 IM Extraversion Antonakis & House 2004 30 0.12 0.89 0.8 1.0 IIB Neuroticism Antonakis & House 2004 30 −0.02 0.89 0.65 1.0 IIA Neuroticism Antonakis & House 2004 30 0.16 0.89 0.78 1.0 IM Neuroticism Antonakis & House 2004 30 0.43 0.78 0.8 1.0 IIB Openness Antonakis & House 2004 30 0.43 0.78 0.65 1.0 IIA Openness Antonakis & House 2004 30 0.32 0.78 0.78 1.0 IM Openness Antonakis et al. 2016 379 −0.02 1.0 0.78 1.0 IIA Age Antonakis et al. 2016 379 0.26 1.0 0.75 1.0 IIB Age Antonakis et al. 2016 379 0.04 1.0 0.85 1.0 IM Age Antonakis et al. 2016 379 0.18 0.86 0.78 1.0 IIA Agreeableness Antonakis et al. 2016 379 0.16 0.86 0.75 1.0 IIB Agreeableness Antonakis et al. 2016 379 0.14 0.86 0.85 1.0 IM Agreeableness Antonakis et al. 2016 379 0.04 0.9 0.78 1.0 IIA Conscientiousness Antonakis et al. 2016 379 0.06 0.9 0.75 1.0 IIB Conscientiousness Antonakis et al. 2016 379 0.04 0.9 0.85 1.0 IM Conscientiousness Antonakis et al. 2016 379 0.12 0.89 0.78 1.0 IIA Extraversion Antonakis et al. 2016 379 0.06 0.89 0.75 1.0 IIB Extraversion Antonakis et al. 2016 379 0.21 0.89 0.85 1.0 IM Extraversion Antonakis et al. 2016 379 −0.05 1.0 0.75 1.0 IIB Gender Antonakis et al. 2016 379 −0.03 1.0 0.85 1.0 IM Gender Antonakis et al. 2016 379 −0.14 1.0 0.78 1.0 IIA Gender Antonakis et al. 2016 379 0.06 0.85 0.78 1.0 IIA Cognitive ability

(continued)

Author Year n r ryy rxx u Leader V. Other V.

Antonakis et al. 2016 379 0.08 0.85 0.75 1.0 IIB Cognitive ability Antonakis et al. 2016 379 0.05 0.85 0.85 1.0 IM Cognitive ability Antonakis et al. 2016 379 −0.01 0.92 0.78 1.0 IIA Neuroticism Antonakis et al. 2016 379 −0.03 0.92 0.75 1.0 IIB Neuroticism Antonakis et al. 2016 379 −0.09 0.92 0.85 1.0 IM Neuroticism Antonakis et al. 2016 379 0.09 0.87 0.78 1.0 IIA Openness Antonakis et al. 2016 379 0.04 0.87 0.75 1.0 IIB Openness Antonakis et al. 2016 379 0.08 0.87 0.85 1.0 IM Openness Alimo-Metcalfe & Alban-Metcalfe 2001 1111 −0.05 1.0 0.85 1.0 II Gender Alimo-Metcalfe & Alban-Metcalfe 2001 1034 −0.07 1.0 0.85 1.0 II Gender Ayman et al. 2000 109 −0.16 1.0 0.85 1.0 II Gender Bass 1985 256 0.67 0.86 0.94 1.0 Charisma Task performance Bass 1985 23 −0.13 1.0 0.85 1.0 II Gender Bass 1985 45 −0.11 1.0 0.85 1.0 II Gender Bass & Yammarino 1991 115 0.37 0.86 0.94 1.0 Charisma Task performance Bass et al. 1996 877 −0.12 1.0 0.85 1.0 II Gender Bass et al. 1996 271 −0.15 1.0 0.85 1.0 II Gender Bass et al. 1996 913 −0.06 1.0 0.85 1.0 II Gender Bass et al. 1996 877 −0.09 1.0 0.83 1.0 IM Gender Bass et al. 1996 271 −0.04 1.0 0.83 1.0 IM Gender Bass et al. 1996 913 −0.05 1.0 0.83 1.0 IM Gender Bommer 1995 723 0.17 0.93 0.88 1.0 Vision Task performance Bono & Judge 2000 189 0.03 1.0 0.73 1.0 IIB Gender Bono & Judge 2000 189 0.01 1.0 0.83 1.0 IM Gender Boomer 1994 61 −0.18 1.0 0.85 1.0 II Gender Boomer 1994 61 −0.09 1.0 0.75 1.0 IIA Gender Boomer 1994 61 −0.23 1.0 0.73 1.0 IIB Gender Boomer 1994 61 −0.27 1.0 0.83 1.0 IM Gender Burbach 2004 649 0.01 1.0 0.74 1.0 IIA Age Burbach 2004 649 0.04 1.0 0.70 1.0 IIB Age Burbach 2004 649 0.02 1.0 0.83 1.0 IM Age Burbach 2004 649 0.04 1.0 0.74 1.0 IIA Education Burbach 2004 649 0.03 1.0 0.70 1.0 IIB Education Burbach 2004 649 0.09 1.0 0.83 1.0 IM Education Burbach 2004 649 −0.05 1.0 0.74 1.0 IIA Gender Burbach 2004 649 −0.07 1.0 0.70 1.0 IIB Gender Burbach 2004 649 0.02 1.0 0.83 1.0 IM Gender Cavazotte et al. 2012 134 0.08 0.70 0.83 1.0 IM Agreeableness Cavazotte et al. 2012 134 0.09 0.70 0.75 1.0 IIA Agreeableness Cavazotte et al. 2012 134 0.09 0.70 0.73 1.0 IIB Agreeableness Cavazotte et al. 2012 134 0.29 0.70 0.83 1.0 IM Conscientiousness Cavazotte et al. 2012 134 0.33 0.70 0.75 1.0 IIA Conscientiousness Cavazotte et al. 2012 134 0.32 0.70 0.73 1.0 IIB Conscientiousness Cavazotte et al. 2012 134 −0.04 0.75 0.83 1.0 IM Extraversion Cavazotte et al. 2012 134 −0.04 0.75 0.75 1.0 IIA Extraversion Cavazotte et al. 2012 134 −0.04 0.75 0.73 1.0 IIB Extraversion Cavazotte et al. 2012 134 0.13 1.0 0.83 1.0 IM Gender Cavazotte et al. 2012 134 0.14 1.0 0.75 1.0 IIA Gender Cavazotte et al. 2012 134 0.14 1.0 0.73 1.0 IIB Gender Cavazotte et al. 2012 134 0.37 .84 0.83 1.0 IM Cognitive ability Cavazotte et al. 2012 134 0.42 .84 0.75 1.0 IIA Cognitive ability Cavazotte et al. 2012 134 0.42 .84 0.73 1.0 IIB Cognitive ability Cavazotte et al. 2012 134 −0.22 0.65 0.83 1.0 IM Neuroticism Cavazotte et al. 2012 134 −0.25 0.65 0.75 1.0 IIA Neuroticism Cavazotte et al. 2012 134 −0.25 0.65 0.73 1.0 IIB Neuroticism Cavazotte et al. 2012 134 0.11 0.64 0.83 1.0 IM Openness Cavazotte et al. 2012 134 0.13 0.64 0.75 1.0 IIA Openness Cavazotte et al. 2012 134 0.13 0.64 0.73 1.0 IIB Openness Center for Leadership Studies 2000 8954 −0.05 1.0 0.85 1.0 II Gender Center for Leadership Studies 2000 8954 −0.07 1.0 0.75 1.0 IIA Gender Center for Leadership Studies 2000 8954 −0.02 1.0 0.73 1.0 IIB Gender Center for Leadership Studies 2000 8954 −0.02 1.0 0.83 1.0 IM Gender Chen & Chung 2014 981 −0.11 1.0 0.83 1.0 CK Age Chen et al. 2013 279 0.03 0.86 0.81 1.0 Charisma Task performance Chun et al. 2009 218 0.23 0.86 0.89 1.0 Charisma Task performance Colbert 2004 268 0.11 0.76 0.82 1.0 II Openness Colbert 2004 268 0.10 0.91 0.83 1.0 IM Task performance Comer et al. 1995 30 0.30 0.98 0.94 1.0 Charisma Task performance Comer et al. 1995 30 0.35 0.98 0.82 1.0 IM Task performance

(continued on next page)

523G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

(continued)

Author Year n r ryy rxx u Leader V. Other V.

Courtright 2012 148 0.01 1.0 0.83 1.0 IIA Age Courtright 2012 148 0.06 1.0 0.67 1.0 IIB Age Courtright 2012 148 0.04 1.0 0.84 1.0 IM Age Courtright 2012 148 −0.16 1.0 0.83 1.0 IIA Education Courtright 2012 148 −0.06 1.0 0.67 1.0 IIB Education Courtright 2012 148 −0.15 1.0 0.84 1.0 IM Education Courtright 2012 148 0.17 0.76 0.83 1.0 IIA Extraversion Courtright 2012 148 0.22 0.76 0.67 1.0 IIB Extraversion Courtright 2012 148 0.26 0.76 0.84 1.0 IM Extraversion Courtright 2012 148 −0.05 0.59 0.83 1.0 IIA Neuroticism Courtright 2012 148 0.04 0.59 0.67 1.0 IIB Neuroticism Courtright 2012 148 −0.03 0.59 0.84 1.0 IM Neuroticism Courtright 2012 148 −0.03 1.0 0.83 1.0 IIA Gender Courtright 2012 148 −0.02 1.0 0.67 1.0 IIB Gender Courtright 2012 148 0.01 1.0 0.84 1.0 IM Gender Crant & Bateman 2000 151 0.17 0.61 0.91 1.0 CK Extraversion Crant & Bateman 2000 151 0.04 0.52 0.91 1.0 CK Openness Crant & Bateman 2000 151 0.51 0.87 0.91 1.0 CK Task performance Cuadrado 2002 118 −0.12 1.0 0.85 1.0 II Gender Davidson 1996 51 −0.15 1.0 0.85 1.0 II Gender Deluga 1995 68 −0.15 0.89 0.89 1.0 Vision Task performance Ernst 1998 1520 0.00 1.0 0.85 1.0 II Gender Ernst 1998 1520 −0.01 1.0 0.75 1.0 IIA Gender Ernst 1998 1520 0.01 1.0 0.73 1.0 IIB Gender Gerstner 1998 55 0.16 0.89 0.92 1.0 IM Group or firm performance Gillespie & Mann 2000 111 0.07 1.0 0.85 1.0 II Gender Gillespie & Mann 2000 111 0.14 1.0 0.75 1.0 IIA Gender Gillespie & Mann 2000 111 0.02 1.0 0.73 1.0 IIB Gender Gillespie & Mann 2000 111 0.03 1.0 0.83 1.0 IM Gender Golden 1999 154 −0.10 1.0 0.83 1.0 IM Gender Goodwin et al. 2001 208 0.22 0.95 0.87 1.0 IM OCB Goodwin et al. 2001 208 0.83 0.83 0.88 1.0 Charisma Task performance Goodwin et al. 2001 208 0.24 0.83 0.87 1.0 IM Task performance Graham 2013 55 0.07 0.95 0.73 1.0 IM Group or firm performance Groves 2005 162 0.40 0.87 0.82 1.0 CK Openness Hill 2000 40 −0.10 1.0 0.85 1.0 II Gender Hill 2000 40 0.09 1.0 0.75 1.0 IIA Gender Hill 2000 40 −0.26 1.0 0.73 1.0 IIB Gender Hill 2000 40 −0.25 1.0 0.83 1.0 IM Gender Hobman et al. 2011 179 0.17 0.96 0.95 1.0 IM Task performance Hobman et al. 2011 179 0.17 0.96 0.87 1.0 Vision Task performance Jantzi & Leithwood 1996 423 −0.16 1.0 0.83 1.0 IM Gender Judge & Bono 2000 169 0.28 0.75 0.85 1.0 II Agreeableness Judge & Bono 2000 169 0.21 0.75 0.83 1.0 IM Agreeableness Judge & Bono 2000 169 −0.04 0.83 0.85 1.0 II Conscientiousness Judge & Bono 2000 169 −.06 0.83 0.83 1.0 IM Conscientiousness Judge & Bono 2000 169 0.22 0.81 0.85 1.0 II Extraversion Judge & Bono 2000 169 0.24 0.81 0.83 1.0 IM Extraversion Judge & Bono 2000 169 0.01 0.74 0.85 1.0 II Neuroticism Judge & Bono 2000 169 −0.03 0.74 0.83 1.0 IM Neuroticism Judge & Bono 2000 169 0.18 0.76 0.85 1.0 II Openness Judge & Bono 2000 169 0.22 0.76 0.85 1.0 IM Openness Judge & Bono 2000 294 −0.06 1.0 0.73 1.0 IIB Gender Judge & Bono 2000 294 0.05 1.0 0.83 1.0 IM Gender Komives 1991a, 1991b 679 0.11 1.0 0.85 1.0 II Gender Komives 1991a, 1991b 679 −0.01 1.0 0.83 1.0 IM Gender Kouzes & Posner 1995 5838 0.03 1.0 0.83 1.0 IM Gender Kuchinke 1999 41 −0.10 1.0 0.85 1.0 II Gender Kuchinke 1999 104 0.06 1.0 0.85 1.0 II Gender Kuchinke 1999 41 −0.16 1.0 0.75 1.0 IIA Gender Kuchinke 1999 104 −0.02 1.0 0.75 1.0 IIA Gender Kuchinke 1999 41 −0.03 1.0 0.73 1.0 IIB Gender Kuchinke 1999 104 0.13 1.0 0.73 1.0 IIB Gender Kuchinke 1999 41 −0.15 1.0 0.83 1.0 IM Gender Kuchinke 1999 104 0.05 1.0 0.83 1.0 IM Gender Leithwood & Jantzi 1997 1253 −0.15 1.0 0.83 1.0 IM Gender Manning 2000 103 −0.01 1.0 0.83 1.0 IM Gender Mawn 2012 349 0.17 0.86 0.83 1.0 IM Task performance Meibaum 1993 214 0.0001 0.91 0.91 1.0 IM Group or firm performance Monroe 1997 206 −0.05 0.75 0.83 1.0 IM Agreeableness

524 G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

(continued)

Author Year n r ryy rxx u Leader V. Other V.

Monroe 1997 206 0.02 0.83 0.83 1.0 IM Conscientiousness Monroe 1997 206 0.20 0.81 0.83 1.0 IM Extraversion Monroe 1997 206 −0.06 0.74 0.83 1.0 IM Neuroticism Monroe 1997 206 0.13 0.77 0.85 1.0 IM Openness Moss et al. 2007 263 0.15 0.86 0.83 1.0 IM Task performance Nahum-Shani & Somech 2011 71 −0.03 1.0 0.7 1.0 IM Age Nahum-Shani & Somech 2011 70 0.12 0.80 0.92 1.0 II OCB Nahum-Shani & Somech 2011 71 0.08 0.80 0.92 1.0 II OCB Nahum-Shani & Somech 2011 70 0.40 0.80 0.7 1.0 IM OCB Nahum-Shani & Somech 2011 71 0.10 0.80 0.7 1.0 IM OCB Oreg & Berson 2015 71 0.03 1.0 0.83 1.0 CK Age Oreg & Berson 2015 201 0.17 0.89 0.83 1.0 CK Extraversion Oreg & Berson 2015 201 0.09 0.84 0.83 1.0 CK Openness Pollock 1998 218 −0.06 1.0 0.85 1.0 II Gender Pollock 1998 218 −0.01 1.0 0.83 1.0 IM Gender Rosen 1993 96 −0.17 1.0 0.85 1.0 II Gender Rozier & Hersh-Cochran 1996 545 −0.16 1.0 0.83 1.0 IM Gender Schriesheim et al. 2006 169 0.28 0.32 0.86 1.0 Vision Task performance Shao & Webber 2006 191 0.08 0.86 0.87 1.0 II Agreeableness Shao & Webber 2006 191 0.05 0.86 0.74 1.0 IM Agreeableness Shao & Webber 2006 191 0.04 0.90 0.87 1.0 II Conscientiousness Shao & Webber 2006 191 0.03 0.90 0.74 1.0 IM Conscientiousness Shao & Webber 2006 191 0.05 0.89 0.87 1.0 II Extraversion Shao & Webber 2006 191 0.02 0.89 0.74 1.0 IM Extraversion Shao & Webber 2006 191 −0.10 0.92 0.87 1.0 II Neuroticism Shao & Webber 2006 191 −0.07 0.92 0.74 1.0 IM Neuroticism Shao & Webber 2006 191 0.09 0.87 0.87 1.0 II Openness Shao & Webber 2006 191 0.13 0.87 0.74 1.0 IM Openness Shen et al. 2012 523 0.20 0.75 0.85 1.0 II Agreeableness Shen et al. 2012 523 0.19 0.75 0.83 1.0 IM Agreeableness Shen et al. 2012 523 0.05 0.83 0.85 1.0 II Conscientiousness Shen et al. 2012 523 0.04 0.83 0.83 1.0 IM Conscientiousness Shen et al. 2012 523 0.25 0.81 0.85 1.0 II Extraversion Shen et al. 2012 523 0.29 0.81 0.83 1.0 IM Extraversion Shen et al. 2012 523 −0.12 0.74 0.85 1.0 II Neuroticism Shen et al. 2012 523 −0.16 0.74 0.83 1.0 IM Neuroticism Shen et al. 2012 523 0.14 0.77 0.85 1.0 II Openness Shen et al. 2012 523 0.15 0.77 0.85 1.0 IM Openness Shen et al. 2012 144 0.21 0.71 0.89 1.0 Charisma Task performance Smith 2011 215 0.28 1.0 0.87 1.0 IIB Age Smith 2011 215 0.22 1.0 0.85 1.0 IM Age Soh 2000 718 0.11 0.86 0.91 1.0 Charisma Task performance Sosik 2005 218 0.26 0.86 0.9 1.0 Charisma Task performance Stepp 2000 328 0.02 1.0 0.85 1.0 II Gender Stepp 2000 328 −0.02 1.0 0.75 1.0 IIA Gender Stepp 2000 328 0.05 1.0 0.73 1.0 IIB Gender Stepp 2000 328 0.06 1.0 0.83 1.0 IM Gender Taylor 2012 130 0.01 0.74 0.89 1.0 II Neuroticism Taylor 2012 130 −0.37 0.74 0.89 1.0 IM Neuroticism Taylor 2012 130 0.02 0.77 0.89 1.0 II Openness Taylor 2012 130 0.09 0.77 0.86 1.0 IM Openness van Engen 2000 104 0.19 1.0 0.85 1.0 II Gender Vecchio et al. 2008 179 0.32 0.93 0.93 1.0 Vision Task performance von Wittch & Antonakis 2011 53 0.03 1.0 0.75 1.0 IIA Age von Wittch & Antonakis 2011 53 0.06 1.0 0.81 1.0 IM Age von Wittch & Antonakis 2011 53 −0.19 0.74 0.75 1.0 IIA Agreeableness von Wittch & Antonakis 2011 53 −0.05 0.74 0.81 1.0 IM Agreeableness von Wittch & Antonakis 2011 53 0.08 0.83 0.75 1.0 IIA Conscientiousness von Wittch & Antonakis 2011 53 0.19 0.83 0.81 1.0 IM Conscientiousness von Wittch & Antonakis 2011 53 −0.02 0.70 0.75 1.0 IIA Extraversion von Wittch & Antonakis 2011 53 0.10 0.70 0.81 1.0 IM Extraversion von Wittch & Antonakis 2011 53 −0.23 1.0 0.75 1.0 IIA gender von Wittch & Antonakis 2011 53 −0.04 1.0 0.81 1.0 IM gender von Wittch & Antonakis 2011 53 0.14 0.80 0.75 1.0 IIA Cognitive ability von Wittch & Antonakis 2011 53 −0.05 0.80 0.81 1.0 IM Cognitive ability von Wittch & Antonakis 2011 53 −0.13 0.83 0.75 1.0 IIA Neuroticism von Wittch & Antonakis 2011 53 0.13 0.83 0.81 1.0 IM Neuroticism von Wittch & Antonakis 2011 53 0.06 0.68 0.75 1.0 IIA Openness von Wittch & Antonakis 2011 53 0.18 0.68 0.81 1.0 IM Openness

(continued on next page)

525G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

(continued)

Author Year n r ryy rxx u Leader V. Other V.

Walumbwa & Schaubroeck 2009 222 0.28 0.73 0.80 1.0 II Agreeableness Walumbwa & Schaubroeck 2009 222 0.17 0.84 0.8 1.0 II Conscientiousness Walumbwa & Schaubroeck 2009 222 −0.18 0.79 0.8 1.0 II Neuroticism Wang 2011 204 0.10 0.75 0.78 1.0 IM Task performance Whittington 2004 174 0.28 0.95 0.90 1.0 IM OCB Whittington 2004 174 0.28 0.92 0.88 1.0 Charisma Task performance Whittington 2004 174 0.21 0.92 0.90 1.0 IM Task performance Wilson-Evered & Hartel 2000 43 0.05 1.0 0.85 1.0 II Gender Wilson-Evered & Hartel 2000 43 0.06 1.0 0.75 1.0 IIA Gender Wilson-Evered & Hartel 2000 43 0.03 1.0 0.73 1.0 IIB Gender Wilson-Evered & Hartel 2000 43 0.02 1.0 0.83 1.0 IM Gender Wofford et al. 1998 89 0.18 0.89 0.89 1.0 IM Group or firm performance Wu & Wang 2012 260 0.32 0.82 0.88 1.0 CK Task performance Yammarino & Bass 1990 186 0.38 0.86 0.94 1.0 Charisma Task performance Yammarino & Bass 1990 186 0.25 0.86 0.82 1.0 IM Task performance Zacher et al. 2014 75 0.22 0.65 0.89 1.0 II Agreeableness Zacher et al. 2014 75 0.09 0.65 0.91 1.0 IM Agreeableness Zacher et al. 2014 75 0.07 0.85 0.89 1.0 II Conscientiousness Zacher et al. 2014 75 0.13 0.85 0.91 1.0 IM Conscientiousness Zacher et al. 2014 75 0.11 0.85 0.89 1.0 II Extraversion Zacher et al. 2014 75 0.24 0.85 0.91 1.0 IM Extraversion Zacher et al. 2014 75 −0.25 0.82 0.89 1.0 II Neuroticism Zacher et al. 2014 75 −0.10 0.82 0.91 1.0 IM Neuroticism Zacher et al. 2014 75 −0.13 0.82 0.89 1.0 II Openness Zacher et al. 2014 75 −0.08 0.82 0.91 1.0 IM Openness

526 G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

References2

Ackerman, P. L., & Heggestad, E. D. (1997). Intelligence, personality, and interests: Evidence for overlapping traits. Psychological Bulletin, 121, 219–245. http://dx.doi. org/10.1037/0033-2909.121.2.219.

Ames, D. R., & Flynn, F. J. (2007). What breaks a leader: The curvilinear relation between assertiveness and leadership. Journal of Personality and Social Psychology, 92, 307–324.

Antonakis, J., Bendahan, S., Jacquart, P., & Lalive, R. (2010a). On making causal claims: A review and recommendations. The Leadership Quarterly, 21(6), 1086–1120. Antonakis, J., Bendahan, S., Jacquart, P., & Lalive, R. (2010b). On making causal claims: A review and recommendations. The Leadership Quarterly, 21, 1086–1120. Antonakis, J., Fenley, M., & Liechti, S. (2011). Can charisma be taught? Tests of two interventions. Academy of Management Learning & Education, 10(3), 374–396. Antonakis, J., Day, D. V., & Schyns, B. (2012). Leadership and individual differences: At the cusp of a renaissance. The Leadership Quarterly, 23, 643–650. Antonakis, J., Bendahan, S., Jacquart, P., & Lalive, R. (2014). Causality and endogeneity: Problems and solutions. In D. V. Day (Ed.), The oxford handbook of leadership and

organizations (pp. 93–117). New York, NY: Oxford Press. Antonakis, J., Bastardoz, N., Jacquart, P., & Shamir, B. (2016a). Charisma: An ill-defined and ill-measured gift. Annual Review of Organizational Psychology and

Organizational Behavior, 3(1), 293–319. ⁎Antonakis, J., Simonton, D. K., & House, R. J. (2016b). Can super smart leaders suffer from too much of a good thing? The Curvilinear Effect of Intelligence on Perceived

Leadership Behavior Unpublished manuscript. Awamleh, R., & Gardner, W. L. (1999). Perceptions of leader charisma and effectiveness: The effects of vision content, delivery, and organizational performance. The

Leadership Quarterly, 10(3), 345–373. Babcock-Roberson, M. E., & Strickland, O. J. (2010). The relationship between charismatic leadership, work engagement, and organizational citizenship behaviors. The

Journal of Psychology, 144(3), 313–326. Banks, G. C., Kepes, S., & McDaniel, M. A. (2015). Publication bias: Understand the myths concerning threats to the advancement of science. In C. E. Lance, & R. J.

Vandenberg (Eds.), More statistical and methodological myths and urban legends (pp. 36–64). New York, NY: Routledge. Banks, G. C., Davis McCauley, K., Gardner, W. L., & Guler, C. E. (2016). A meta-analytic review of authentic and transformational leadership: A test for redundancy. The

Leadership Quarterly, 27, 634–652. Barlow, K. M. (2013). A meta-analysis of transformational leadership and subordinate nursing personnel organizational commitment, job satisfaction, and turnover inten-

tions (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (Accession Order No. 3563301). Bascle, G. (2008). Controlling for endogeneity with instrumental variables in strategic management research. Strategic Organization, 6, 285–327. ⁎Bass, B. M. (1985). Leadership and performance beyond expectations. New York, NY: Free Press. Bass, B. M., & Avolio, B. J. (1995). Manual for the multifactor leadership questionnaire: Rater form (5× short). Palo Alto, CA: Mind Garden. Beal, D. J., Corey, D. M., & Dunlap, W. P. (2002). On the bias of Huffcutt and Arthur's (1995) procedure for identifying outliers in the meta-analysis of correlations.

Journal of Applied Psychology, 87, 583–589. Berson, Y., Shamir, B., Avolio, B. J., & Popper, M. (2001). The relationship between vision strength, leadership style, and context. The Leadership Quarterly, 12, 53–73. Bono, J. E., & Judge, T. A. (2004). Personality and transformational and transactional leadership: A meta-analysis. Journal of Applied Psychology, 89(5), 901–910. Borenstein, M., Hedges, L. V., Higgins, J. P., & Rothstein, H. R. (2009). Introduction to meta-analysis. West Sussex, UK: Wiley. Bosco, F. A., Aguinis, H., Singh, J., Field, J. G., & Pierce, C. A. (2015). Correlation effect size benchmarks. Journal of Applied Psychology, 100, 431–449. Bryman, A. (1992). Charisma and leadership in organizations. London, UK: Sage Publications. Buckman, B., Crawford, E., LePine, J., & Zhang, Y. (2015, October). A meta-analytic investigation of job engagement's role in the transformational leadership–job sat-

isfaction–job performance relationship. Paper presented at the annual meeting of the Southern Management Association, St. Petersburg Beach, Florida. Burke, C. S., Stagl, K. C., Klein, C., Goodwin, G. F., Salas, E., & Halpin, S. M. (2006). What type of leadership behaviors are functional in teams? A meta-analysis. The

Leadership Quarterly, 17(3), 288–307. Burns, J. M. (1978). Leadership. New York, NY: Harper and Row. Carson, C. (1987). Martin Luther King, Jr.: Charismatic leadership in a mass struggle. The Journal of American History, 74(2), 448–454.

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

527G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

⁎Chen, C. C., Belkin, L. Y., McNamee, R., & Kurtzberg, T. R. (2013). Charisma attribution during organizational change: The importance of followers' emotions and con- cern for well-being. Journal of Applied Social Psychology, 43, 1136–1158.

Cho, J., & Dansereau, F. (2010). Are transformational leaders fair? A multi-level study of transformational leadership, justice perceptions, and organizational citizenship behaviors. The Leadership Quarterly, 21(3), 409–421.

Cobb-Clark, D. A., & Schurer, S. (2012). The stability of big-five personality traits. Economics Letters, 115, 11–15. Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20, 37–46. Conger, J. A., & Kanungo, R. N. (1987). Towards a behavioral theory of charismatic leadership in organizational settings. Academy of Management Review, 12, 637–647. Conger, J. A., & Kanungo, R. N. (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 Publications. Connelly, B. L., Certo, S. T., Ireland, R. D., & Reutzel, C. R. (2011). Signaling theory: A review and assessment. Journal of Management, 37(1), 39–67. De Hoogh, A. H., Den Hartog, D. N., & Koopman, P. L. (2005). Linking the big five-factors of personality to charismatic and transactional leadership; perceived dynamic

work environment as a moderator. Journal of Organizational Behavior, 26(7), 839–865. DeGroot, T., Kiker, D. S., & Cross, T. C. (2000). A meta-analysis to review organizational outcomes related to charismatic leadership. Canadian Journal of Administrative

Sciences, 17(4), 356–372. Dirks, K. T., & Ferrin, D. L. (2002). Trust in leadership: Meta-analytic findings and implications for research and practice. Journal of Applied Psychology, 87(4), 611. Dulebohn, J. H., Bommer, W. H., Liden, R. C., Brouer, R. L., & Ferris, G. R. (2012). A meta-analysis of antecedents and consequences of leader-member-exchange: Inte-

grating the past with an eye toward the future. Journal of Management, 38, 1715–1759. Dumdum, U. R., Lowe, K. B., & Avolio, B. J. (2002). A meta-analysis of transformational and transactional leadership correlates of effectiveness and satisfaction: An up-

date and extension. In B. J. Avolio, & F. J. Yammarino (Eds.), Transformational and charismatic leadership: The road ahead (pp. 35–66). Oxford, UK: Elsevier Science. Duncan, G. J., Magnusson, K. A., & Ludwig, J. (2004). The endogeneity problem in developmental studies. Research in Human Development, 1, 59–80. Eagly, A. H., Johannesen-Schmidt, M. C., & van Engen, M. L. (2003). Transformational, transactional, and laissez-faire leadership styles: A meta-analysis comparing

women and men. Psychological Bulletin, 129, 569–591. Fanelli, A., Misangyi, V. F., & Tosi, H. L. (2009). In charisma we trust: The effects of CEO charismatic visions on securities analysts. Organization Science, 20, 1011–1033. Fleiss, J. L. (1981). Statistical methods for rates and proportions (2nd ed.). New York, NY: John Wiley. Frese, M., Beimel, S., & Schoenborn, S. (2003). Action training for charismatic leadership: Two evaluations of studies of a commercial training module on inspirational

communication of a vision. Personnel Psychology, 56, 671–697. Halverson, S. K., Murphy, S. E., & Riggio, R. E. (2004). Charismatic leadership in crisis situations: A laboratory investigation of stress and crisis. Small Group Research,

35(5), 495–514. Hamilton, B. H., & Nickerson, J. A. (2003). Correcting for endogeneity in strategic management research. Strategic Organization, 1, 51–78. Hampson, S. E., & Goldberg, L. R. (2006). A first large cohort study of personality trait stability over the 40 years between elementary school and midlife. Journal of

Personality and Social Psychology, 91, 763–779. Harms, P. D., & Credé, M. (2010). Emotional intelligence and transformational and transactional leadership: A meta-analysis. Journal of Leadership and Organizational

Studies, 17(1), 5–17. Harrison, J. S., Banks, G. C., Pollack, J. M., O'Boyle, E. H., Jr., & Short, J. C. (2016). Publication bias in strategic management research. Journal of Management (in press). Heath, J. D. (2011). An examination of the relationship between leadership style and sense of humor: A correlation study of federal probation officers. (Doctoral dissertation).

Retrieved from ProQuest Dissertations and Theses. (Accession Order No. 3517061). Hertzog, C., & Schaie, K. W. (1986). Stability and change in adult intelligence: Analysis of longitudinal covariance structures. Psychology and Aging, 1, 159–171. House, R. J. (1977). A 1976 theory of charismatic leadership. In J. G. Hunt, & L. L. Larson (Eds.), The cutting edge (pp. 189–207). Carbondale, Southern Illinois: University Press. House, R. J. (1999). Weber and the neo-charismatic leadership paradigm: A response to Beyer. The Leadership Quarterly, 10, 563–574. House, R. J., Spangler, W. D., &Woycke, J. (1991). Personality and charisma in the U.S. Presidency: A psychological theory of leader effectiveness. Administrative Science

Quarterly, 36, 354–396. Howell, J. N., & Shamir, B. (2005). The role of followers in the charismatic leadership process: Relationships and their consequences. Academy of Management Review,

30, 96–112. Huffcutt, A. I., & Arthur, W. (1995). Development of a new outlier statistic for meta-analytic data. Journal of Applied Psychology, 80, 327–334. Jacquart, P., & Antonakis, J. (2015).When does charismamatter for top-level leaders? Effect of attributional ambiguity. Academy ofManagement Journal, 58, 1051–1074. Jick, T. D. (1979). Mixing qualitative and quantitative methods: Triangulation in action. Administrative Science Quarterly, 24, 602–611. Johnson, J. W. (2001). Computer program exchange: Rweight: Computing the relative weight of predictors in a regression. Applied Psychological Measurement, 25, 342. Johnson, J. W., & LeBreton, J. M. (2004). History and use of relative importance indices in organizational research. Organizational Research Methods, 7, 238–257. 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. Kepes, S., Banks, G. C., McDaniel, M. A., & Whetzel, D. L. (2012). Publication bias in the organizational sciences. Organizational Research Methods, 15, 624–662. Kepes, S., McDaniel, M. A., Brannick, M. T., & Banks, G. C. (2013). Meta-analytic reviews in the organizational sciences: Twometa-analytic schools on the way to MARS.

Journal of Business and Psychology, 28, 123–142. Kepes, S., Banks, G. C., & Oh, I. -S. (2014). Avoiding bias in publication bias research: The value of “null” findings. Journal of Business and Psychology, 29, 183–203. Kirkpatrick, S. A., & Locke, E. A. (1991). Leadership: Do traits matter? The Executive, 5(2), 48–60. Kraft, P. S., & Bausch, A. (2015). How do transformational leaders promote exploratory and exploitative innovation? Ameta-analysis. In J. Humphreys (Ed.), Academy of

Management Proceedings (pp. 12777). Vancouver, BC: Academy of Management Meeting. Lord, R. G., Binning, J. F., Rush, M. C., & Thomas, J. C. (1978). The effect of performance cues and leader behavior on questionnaire ratings of leadership behavior.

Organizational Behavior and Human Performance, 21(1), 27–39. Lord, R. G., Brown, D. J., Harvey, J. L., & Hall, R. J. (2001). Contextual constraints on prototype generation and their multilevel consequences for leadership perceptions.

The Leadership Quarterly, 12, 311–338. Lowe, K. B., Kroeck, K. G., & Sivasubramaniam, N. (1996). Effectiveness correlates of transformational and transactional leadership: A meta-analytic review of the MLQ

literature. The Leadership Quarterly, 7(3), 385–425. MacKenzie, S. B. (2003). The dangers of poor construct conceptualization. Journal of the Academy of Marketing Science, 31(3), 323–326. Mio, J. S., Riggio, R. E., Levin, S., & Reese, R. (2005). Presidential leadership and charisma: The effects of metaphor. The Leadership Quarterly, 16(2), 287–294. Nohe, C., Michaelis, B., Menges, J. I., Zhang, Z., & Sonntag, K. (2013). Charisma and organizational change: A multilevel study of perceived charisma, commitment to

change, and team performance. The Leadership Quarterly, 24, 378–389. O'Boyle, E. H., Jr., Forsyth, D. R., Banks, G. C., & McDaniel, M. A. (2012). A meta-analysis of the dark triad and work behaviors: A social exchange perspective. Journal of

Applied Psychology, 97, 557–579. Paustian-Underdahl, S. C., Walker, L. S., &Woehr, D. J. (2014). Gender and perceptions of leadership effectiveness: A meta-analysis of contextual moderators. Journal of

Applied Psychology, 99(6), 1129–1145. Pillai, R. (1996). Crisis and the emergence of charismatic leadership in groups: An experimental investigation. Journal of Applied Social Psychology, 26, 543–562. Plomin, R., DeFries, J. C., Knopik, V. S., & Neiderhiser, J. M. (2016). Top 10 replicated findings from behavioral genetics. Perspectives on Psychological Science, 11(1), 3–23. Podsakoff, P. M., Mackenzie, S. B., Moorman, R. H., & Fetter, R. (1990). Transformational leader behaviors and their effects on followers' trust in leader, satisfaction, and

organizational citizenship behaviors. The Leadership Quarterly, 1, 107–142. Polderman, T. J., Benyamin, B., De Leeuw, C. A., Sullivan, P. F., Van Bochoven, A., Visscher, P. M., et al. (2015). Meta-analysis of the heritability of human traits based on

fifty years of twin studies. Nature Genetics, 47, 702–709. Rafferty, A. E., & Griffin, M. A. (2004). Dimensions of transformational leadership: Conceptual and empirical extensions. The Leadership Quarterly, 15, 329–354. Reed, J. G., & Baxter, P. M. (2009). Using reference databases. In H. Cooper, L. V. Hedges, & J. C. Valentine (Eds.), The handbook of research synthesis and meta-analysis

(pp. 74–101). New York, NY: Russell Sage Foundation.

528 G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

Sackett, P. R., & Larson, J. R., Jr. (1990). Research strategies and tactics in industrial and organizational psychology. In M. D. Dunnette, & L. M. Hough (Eds.), Handbook of industrial and organizational psychology (pp. 419–489). Palo Alto, CA: Consulting Psychologists Press.

Schmidt, F. L., & Hunter, J. E. (2015). Methods of meta-analysis: Correcting error and bias in research findings (3rd ed.). Newbury Park, CA: Sage. Schmidt, F. L., Shaffer, J. A., & Oh, I. -S. (2008). Increased accuracy for range restriction corrections: Implications for the role of personality and general mental ability in

job and training performance. Personnel Psychology, 61, 827–868. http://dx.doi.org/10.1111/j.1744-6570. 2008.00132.x. Shamir, B. (1992). Attribution of influence and charisma to the leader: The romance of leadership revisited. Journal of Applied Social Psychology, 22, 386–407. Shamir, B., House, R. J., & Arthur, M. B. (1993). The motivational effects of charismatic leadership: A self-concept based theory. Organization Science, 4, 577–594. Shamir, B., Zakay, E., Breinin, E., & Popper, M. (1998). Correlates of charismatic leader behavior in military units: Subordinates' attitudes, unit characteristics, and su-

periors' appraisals of leader performance. Academy of Management Journal, 41(4), 387–409. Silvia, P. J., & Beaty, R. E. (2012). Making creative metaphors: The importance of fluid intelligence for creative thought. Intelligence, 40, 343–351. Simonton, D. K. (1985). Intelligence and personal influence in groups: Four nonlinear models. Psychological Review, 92, 532–547. Sosik, J. J., Juzbasich, J., & Chun, J. U. (2011). Effects of moral reasoning and management level on ratings of charismatic leadership, in-role and extra-role performance

of managers: A multi-source examination. The Leadership Quarterly, 22, 434–450. Sun, J. (2010). A review of transformational leadership research: A meta-analytic approach (Doctoral dissertation). Available from ProQuest Dissertations & Theses Global.

(Accession Order No. NR97781). Tonidandel, S., & LeBreton, J. M. (2014). RWA-Web—a free, comprehensive, web-based, and user-friendly tool for relative weight analyses. Journal of Business and

Psychology, 30, 207–216. Tosi, H. L., Misangyi, V. F., Fanelli, A., Waldman, D. A., & Yammarino, F. J. (2004). CEO charisma, compensation, and firm performance. The Leadership Quarterly, 15, 405–420. Towler, A. J. (2003). Effects of charismatic influence training on attitudes, behavior, and performance. Personnel Psychology, 56, 363–381. Van der Linden, D., te Nijenhuis, J., & Bakker, A. B. (2010). The general factor of personality: A meta-analysis of Big Five intercorrelations and a criterion-related validity

study. Journal of Research in Personality, 44, 315–327. van Knippenberg, D., & Sitkin, S. B. (2013). A critical assessment of charismatic-transformational leadership research: Back to the drawing board? The Academy of

Management Annals, 7, 1–60. ⁎von Wittich, D., & Antonakis, J. (2011). The KAI cognitive style inventory: Was it personality all along? Personality and Individual Differences, 50(7), 1044–1049. Waldman, D. A., & Yammarino, F. J. (1999). CEO charismatic leadership: Levels-of-management and levels-of-analysis effects. Academy of Management Review, 24, 266–285. Walter, F., & Bruch, H. (2009). An affective events model of charismatic leadership behavior: A review, theoretical integration, and research agenda. Journal of

Management, 35, 1428–1452. Wang, G., Oh, I. -S., Courtright, S. H., & Colbert, A. E. (2011). Transformational leadership and performance across criteria and levels: A meta-analytic review of 25 years

of research. Group & Organization Management, 36, 223–270. Wood, J. A. (2008). Methodology for dealing with duplicate study effects in a meta-analysis. Organizational Research Methods, 11, 79–95. Yukl, G. (1999). An evaluation of conceptual weaknesses in transformational and charismatic leadership theories. The Leadership Quarterly, 10(2), 285–305. Zaccaro, S. J., Kemp, C., & Bader, P. (2004). Leader traits and attributes. In J. Antonakis, A. T. Cianciolo, & R. J. Sternberg (Eds.), The nature of leadership (pp. 101–124).

Thousand Oaks, CA: Sage.

Further reading

⁎Alimo-Metcalfe, B., & Alban-Metcalfe, R. J. (2001). The development of a new transformational leadership questionnaire. Journal of Occupational and Organizational Psychology, 74, 1–27.

⁎Antonakis, J., & House, R. J. (2004). On instrumental leadership: Beyond transactions and transformations. Unpublished manuscript. ⁎Antonakis, J., & House, R. J. (2014). Instrumental leadership: Measurement and extension of transformational-transactional leadership theory. The Leadership

Quarterly, 25, 746–771. ⁎Ayman, R., Bast, J. D., Korabik, K., & Morris, S. (2000). Transformational leadership and leader's performance: The moderating effect of the gender composition of the leader-

subordinate dyad. Unpublished manuscript Chicago, Illinois: Illinois Institute of Technology. Bass, B. M., & Avolio, B. J. (1993). Transformational leadership: A response to critiques. In M. M. Chemmers, & R. Ayman (Eds.), Leadership theory and research: Perspec-

tives and directions (pp. 49–88). San Diego, CA: Academic Press. ⁎Bass, B. M., & Yammarino, F. J. (1991). Congruence of self and others' leadership ratings of naval officers for understanding successful performance. Applied Psychology,

40, 437–454. ⁎Bass, B. M., Avolio, B. J., & Atwater, L. (1996). The transformational and transactional leadership of men and women. Applied Psychology. An International Review, 45, 5–34. ⁎Bommer, W. H. (1995). Contextual influences on transformational leader behavior's effectiveness: Transformational leader behavior in a substitutes for leadership frame-

work (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (Accession Order No. 9544402). ⁎Bono, J. E., & Judge, T. A. (2000, August). Responses to transformational leadership: Are some followers immune? Paper presented at the annual meeting of the Academy

of Management, Toronto, Ontario, Canada. ⁎Boomer, K. M. (1994). The perceived transformational leadership traits of principals of restructuring elementary schools in California (Doctoral dissertation). Retrieved

from ProQuest Dissertations and Theses Global. (Accession Order No. 9417944). ⁎Burbach, M. E. (2004). Testing the relationship between emotional intelligence and full-range leadership as moderated by cognitive style and self-concept (Doctoral disser-

tation). Retrieved from ProQuest Dissertations and Theses. (Accession Order No. 3126944). ⁎Cavazotte, F., Moreno, V., & Hickmann, M. (2012). Effects of leader intelligence, personality and emotional intelligence on transformational leadership and managerial

performance. The Leadership Quarterly, 23, 443–455. ⁎Center for Leadership Studies. (2000). Multifactor leadership questionnaire: Norms. Retrieved June 22, 2000, from http://cls.binghamton.edu/mlq.htm. Website now

located at www.gallupleadershipinstitute.org. ⁎Chen, I. H., & Chung, A. (2014). Self-managers: Social contexts, personal traits, and organizational commitment. Asia Pacific Journal of Management, 31, 621–642. ⁎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. The Leadership Quarterly, 20, 689–707. ⁎Colbert, A. E. (2004). Understanding the effects of transformational leadership: The mediating role of leader-follower value congruence (Doctoral dissertation). Retrieved

from ProQuest Dissertations and Theses. (Accession Order No. 3157961). ⁎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 & Sales Management, 15, 17–32. ⁎Courtright, S. H. (2012). Fired up or burned out? Exploring the effects of leadership challenge demands on leadership behaviors through engagement and burnout (Doctoral

dissertation). Retrieved from ProQuest Dissertations and Theses. (Accession Order No. 3628471). ⁎Crant, J. M., & Bateman, T. S. (2000). Charismatic leadership viewed from above: The impact of proactive personality. Journal of Organizational Behavior, 21, 63–75. ⁎Cuadrado, I. (2002). Estilos de liderazgo y ge´nero: Una perspectiva psicosocial [Leadership styles and gender: A psychosocial perspective] (Unpublished doctoral disserta-

tion). Almerı´a, Spain: Universidad de Almerı´a. ⁎Davidson, B. J. (1996). Leadership styles of successful male and female college choral directors (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses

Global. (Accession Order No. 9611661). ⁎Deluga, R. J. (1995). The relationship between attributional charismatic leadership and organizational citizenship behavior. Journal of Applied Social Psychology, 25,

1652–1669. ⁎Ernst, L. C. (1998). An empirical investigation of the relationship between visionary leadership and psychological androgyny (Doctoral dissertation). Retrieved from

ProQuest Dissertations and Theses Global. (Accession Order No. 9806395).

529G.C. Banks et al. / The Leadership Quarterly 28 (2017) 508–529

⁎Gerstner, C. R. (1998). Leadership relationships and work group effectiveness: A multi-level empirical examination (Doctoral dissertation). Retrieved from ProQuest Disser- tations and Theses. (Accession Order No. 9836673).

⁎Gillespie, N. A., & Mann, L. (2000, August). The building blocks of trust: The role of transformational leadership and shared values in predicting teammembers' trust in their leaders. Paper presented at the annual meeting of the Academy of Management, Toronto, Ontario, Canada.

⁎Golden, E. B. (1999). A study of the relationship between superintendents' perceived leadership practices and socioeconomic status of school districts in New Jersey (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses Global. (Accession Order No. 9930551).

⁎Goodwin, V. L., Wofford, J. C., & Whittington, J. L. (2001). A theoretical and empirical extension to the transformational leadership construct. Journal of Organizational Behavior, 22, 759–774.

⁎Graham, C. M. (2013). An examination of leadership types among generation y and its impact on virtual team effectiveness (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (Accession Order No. 3602278).

⁎Groves, K. S. (2005). Linking leader skills, follower attitudes, and contextual variables via an integrated model of charismatic leadership. Journal of Management, 31, 255–277.

⁎Hill, S. A. (2000). Leadership style, behaviors and spheres of influence of university leaders: Moving beyond position (Doctoral dissertation). Retrieved from ProQuest Dis- sertations and Theses Global. (Accession Order No. NQ54417).

⁎Hobman, E. V., Jackson, C. J., Jimmieson, N. L., & Martin, R. (2011). The effects of transformational leadership behaviors on follower outcomes: An identity-based anal- ysis. European Journal of Work and Organizational Psychology, 20, 553–580.

⁎Jantzi, D., & Leithwood, K. (1996). Toward an explanation of variation in teachers' perceptions of transformational school leadership. Educational Administration Quarterly, 32, 512–538.

⁎Judge, T. A., & Bono, J. E. (2000). Five-factor model of personality and transformational leadership. Journal of Applied Psychology, 85, 751–765. ⁎Komives, S. R. (1991a). Gender differences in the relationship of hall directors' transformational and transactional leadership and achieving styles. Journal of College

Student Development, 32, 155–165. ⁎Komives, S. R. (1991b). The relationship of same- and cross-gender work pairs to staff performance and supervisor leadership in residence hall units. Sex Roles, 24,

355–363. ⁎Kouzes, J. M., & Posner, B. Z. (1995). The leadership challenge: How to keep getting extraordinary things done in organizations. San Francisco, CA: Jossey-Bass. ⁎Kuchinke, K. P. (1999). Leadership and culture: Work-related values and leadership styles among one company's U.S. and German telecommunication employees.

Human Resource Development Quarterly, 10, 135–154. ⁎Leithwood, K., & Jantzi, D. (1997). Explaining variation in teachers' perceptions of principals' leadership: A replication. Journal of Educational Administration, 35,

312–331. ⁎Manning, T. T. (2000). Gender, managerial level, transformational leadership, and work satisfaction. Unpublishedmanuscript Baltimore, Maryland: Notre Dame ofMary-

land University. ⁎Mawn, L. (2012). Transformational leadership in higher education lecturing (Unpublished doctoral dissertation). Bangor, Wales: Bangor University. ⁎Meibaum, A. J., III (1993). The antecedents and consequences of transformational leadership (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses.

(Accession Order No. 9419911). ⁎Monroe, M. J. (1997). Leadership of organizational change: Antecedents and implications (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (Ac-

cession Order No. 9729239). ⁎Moss, S. A., & Ritossa, D. A. (2007). The impact of goal orientation on the association between leadership style and follower performance, creativity andwork attitudes.

Leadership, 3, 433–456. ⁎Nahum-Shani, I., & Somech, A. (2011). Leadership, OCB and individual differences: Idiocentrism and allocentrism as moderators of the relationship between trans-

formational and transactional leadership and OCB. The Leadership Quarterly, 22, 353–366. ⁎Oreg, S., & Berson, Y. (2015). Personality and charismatic leadership in context: The moderating role of situational stress. Personnel Psychology, 68, 49–77. ⁎Pollock, K. S. (1998). The relationship between leadership style and subordinate satisfaction and performance in public accounting firms (Doctoral dissertation). Retrieved

from ProQuest Dissertations and Theses Global. (Accession Order No. 9831374). ⁎Rosen, J. E. (1993). The ways women lead school districts in New York, New Jersey, and Connecticut (Doctoral dissertation). Retrieved from ProQuest Dissertations and

Theses Global. (Accession Order No. 9321002). ⁎Rozier, C. K., & Hersh-Cochran, M. S. (1996). Gender differences in managerial characteristics in a female-dominated health profession. The Health Care Supervisor, 14,

57–70. ⁎Schriesheim, C. A., Castro, S. L., Zhou, X. T., & DeChurch, L. A. (2006). An investigation of path-goal and transformational leadership theory predictions at the individual

level of analysis. The Leadership Quarterly, 17, 21–38. ⁎Shao, L., & Webber, S. (2006). A cross-cultural test of the ‘five-factor model of personality and transformational leadership’. Journal of Business Research, 59, 936–944. ⁎Shen, W., Yoon, D., & Bono, J. E. (2012). Personality and transformational and transactional leadership: Differing profiles by gender. Paper presented at Academy of

Management Annual Meeting, Boston, Massachusetts. ⁎Smith, J. P. (2011). The effects of self-efficacy and spirituality on the job satisfaction and motivation to lead among redeploying soldiers as moderated by transformational

leadership (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (Accession Order No. 3447893). ⁎Soh, S. (2000). Organizational socialization of newcomers: A longitudinal study of organizational enculturation processes and outcomes. The Ohio State University, Colum-

bus, OH: Unpublished PhD. ⁎Sosik, J. J. (2005). The role of personal values in charismatic leadership of corporate managers: A model and preliminary field study. The Leadership Quarterly, 16,

221–244. ⁎Stepp, P. L. (2000). Leadership styles of male and female research scientists: Transformational or transactional? Cornell University, Ithaca, New York: Unpublishedmanuscript. ⁎Taylor, A. M. (2012). Cultivating an engaged workforce: The roles of leader personality, motivation, and leadership style (Doctoral dissertation). Retrieved from ProQuest

Dissertations and Theses. (Accession Order No. 3505150). ⁎van Engen,M. L. (2000). Self-ratings ofmale and female leaders in differently gender-typed contexts. Unpublished document Tilburg, the Netherlands: Tilburg University. ⁎Vecchio, R. P., Justin, J. E., & Pearce, C. L. (2008). The utility of transactional and transformational leadership for predicting performance and satisfaction within a path-

goal theory framework. Journal of Occupational and Organizational Psychology, 81, 71–82. ⁎Walumbwa, F. O., & Schaubroeck, J. (2009). Leader personality traits and employee voice behavior: Mediating roles on ethical leadership and work group psycholog-

ical safety. Journal of Applied Psychology, 94, 1275–1286. ⁎Wang, L. (2011). An alternative explanation for transformational leadership effectiveness: The role of leader proactive feedback-seeking behavior in the relationship between

transformational leadership and subordinate performance (Doctoral dissertation). Retrieved from ProQuest Dissertations and Theses. (Accession Order No. 3460812). ⁎Whittington, J. L., Goodwin, V. L., & Murray, B. (2004). Transformational leadership, goal difficulty, and job design: Independent and interactive effects on employee

outcomes. The Leadership Quarterly, 15, 593–606. ⁎Wilson-Evered, E., & Hartel, C. E. J. (2000). A longitudinal study of work group innovation: The importance of transformational leadership and morale. Queensland Health,

Queensland, Australia: Unpublished manuscript. ⁎Wofford, J. C., Goodwin, V. L., & Whittington, J. L. (1998). A field study of a cognitive approach to understanding transformational and transactional leadership. The

Leadership Quarterly, 9, 55–84. ⁎Wu, M., & Wang, J. (2012). Developing a charismatic leadership model for Chinese organizations: The mediating role of loyalty to supervisors. International Journal of

Human Resource Management, 23, 4069–4084. ⁎Yammarino, F. J., & Bass, B. M. (1990). Long-term forecasting of transformational leadership and its effects among naval officers: Some preliminary findings. In K. E.

Clark, & M. B. Clark (Eds.), Measures of leadership (pp. 151–169). West Orange, NY: Leadership Library of America. ⁎Zacher, H., Pearce, L. K., Rooney, D., & McKenna, B. (2014). Leaders' personal wisdom and leader–member exchange quality: The role of individualized consideration.

Journal of Business Ethics, 121, 171–187.