Followers’ HEXACO personality traits and preference for charismatic, relationship-oriented, and task-oriented leadership

The aim of the current study was to examine the HEXACO personality traits in relation to followers’ preference for charismatic, relationship-oriented, and task-oriented leadership. Based on the similarity perspective, we expected followers high on Honesty-Humility, Extraversion, and Openness to Experience to prefer a charismatic leader, and those followers high on conscientiousness and low on Openness to Experience to prefer a task-oriented leader. In addition, from a need fulfillment perspective, we expected followers high on Emotionality to prefer a task- and a relationship-oriented leader. We examined these expectations using paper vignette methodology in a sample of 272 undergraduates. The results showed that most participants preferred a relationship-oriented leader over a charismatic or task-oriented leader. In addition, we found support for all our hypotheses, with the exception of the relations between Honesty-Humility and preference for charismatic leadership, and Conscientiousness and preference for task-oriented leadership. Our findings contribute to the nomological network of the role of follower characteristics in the leader-follower relationship. Implications and suggestions for research on charismatic leadership are provided.

With its transformative potential and many successful real-life examples, it is not surprising that research on charismatic leadership, and similarly, research on transformational leadership, has flourished. In fact, there is a strong belief that such inspirational leaders are universally impactful (see e.g., Bass, 1990;Conger, 1999;House, Howell, Shamir, Smith, & Spangler, 1991;DeGroot, Kiker, & Cross, 2000). Yet, at the same time, it is theorized that charismatic leaders may not appeal to every follower (Shamir, House, & Arthur, 1993), which is supported by meta-analytic evidence showing that charismatic leadership is not always as strongly related to follower outcomes as one would expect (e.g., Hoch, Bommer, Dulebohn, & Wu, 2018;Wang, Oh, Courtright, & Colbert, 2011). This begs the question which followers appreciate a charismatic leader and, if they do not prefer charismatic leadership, what kind of leadership they do prefer. To answer these questions, we examine individual differences in followers' personality related to their preference for charismatic, relationship-oriented, and/or task-oriented leadership. Together, these leadership styles capture major leadership styles (Yukl, 1998), while having sufficient distinctiveness (Howell & Frost, 1989).
Our study contributes to the literature in three ways. First, despite the widely accepted idea that leadership results from the interaction between a leader and his or her followers (e.g., Jermier, 1993;Klein & House, 1998;Meindl, 1990;Shamir et al., 1993), most leadership research is still very much leader centered (Oc & Bashshur, 2013). Therefore, as Avolio and Yammarino (2013) state, "there is a critical need to explore the dynamics between the leader and follower" (p. xxxii). The current study sheds a light on the dynamics between leaders and followers by examining personality characteristics related to followers' preference for a charismatic, relationship-oriented, and/or task-oriented leader. Second, we expand the existing literature on individual differences related to followers' preference for leadership (Ehrhart & Klein, 2001;Thoroughgood & Sawyer, 2018) with our focus on the HEXACO model of personality. The HEXACO model of personality distinguishes six personality dimensions-i.e., H o n e s t y -H u m i l i t y, E m o t i o n a l i t y, E x t r a v e r s io n , Agreeableness, Conscientiousness, and Openness to Experience-that are based on the same lexical paradigm that has earlier led to the emergence of the Big Five model of personality Ashton, Lee, & De Vries, 2014). The most obvious difference between the two models of personality is the inclusion of six (HEXACO) rather than five (OCEAN) personality dimensions. It is due to this sixth factor, Honesty-Humility, that the HEXACO model of personality explains additional variance over the Big Five model of personality in important workplace outcomes such as workplace deviance (Lee, Ashton, & De Vries, 2005;Pletzer, Bentvelzen, Oostrom, & De Vries, 2019) and unethical business decisions (Ashton & Lee, 2008). Finally, we test two underlying theoretical rationales to explain the relations between followers' HEXACO personality traits and their preference for leadership. On the one hand, from a similarity perspective, we expect followers to prefer a leader who is similar to them. On the other hand, from a need fulfillment perspective, we expect that followers may not always prefer a leader who is similar to them but may rather prefer a leader who is able to provide for their needs.

Preference for leadership
From a similarity perspective (Byrne, 1971), Klein and House (1998) argued that charismatic leaders are especially appealing to followers with similar characteristics (socalled strong followers). In support of this perspective, Keller (1999) showed that even implicit leadership theories (Lord, Foti, & De Vader, 1984) are affected by the principle of similarity, in such a way that extraverted and conscientious individuals described their ideal leader as being charismatic. Additionally, drawing from path-goal theory (House, 1971), Ehrhart and Klein (2001) proposed a need-fulfillment perspective, stating that followers will be motivated by leaders who are able to meet their needs. In their study, Ehrhart and Klein (2001) examined the relations between a range of follower work values, as well as follower personality (i.e., achievement, risk-taking, self-esteem, and need for structure) and their preference for charismatic, relationship-oriented, and task-oriented leadership. Based on their results, Ehrhart and Klein concluded that although the similarity perspective may play a role in explaining the relations between follower characteristics and leader preference, "based on the predictors that had the strongest relation with leader preference, need fulfillment appears to be the dominant mechanism." Specifically, they found that followers' work values, especially their worker participation, extrinsic rewards, and security work values, distinguished followers' preference for either one of the three types of leadership. With regard to followers' personality characteristics, the only relations that were found were that followers with a higher selfesteem and more need for structure had a stronger preference for a task-oriented leader.
More recently, Thoroughgood and Sawyer (2018) used discriminant function analyses to compare personality (Big Five, authoritarianism, cognitive rigidity, rational mindedness, and temporal focus) and work value profiles (autonomy, teamwork, competition, and stability) of those preferring a charismatic, ideological, and pragmatic leader. So rather than looking at the extent to which followers prefer all three types of leaders, these authors focused solely on the follower characteristics related to the forced choice of one out of three (i.e., charismatic, ideological, and pragmatic) leaders. They performed two discriminant function analyses, one on personality and one on work values and their results mainly supported the similarity perspective. For personality, they found that individuals who preferred to work with a charismatic leader scored higher on Extraversion, Agreeableness, and future focus (labeled as interpersonal concerns), while scoring lower on rational mindedness, neuroticism, and past focus (labeled as pragmatism) compared with those individuals preferring an ideological or pragmatic leader. Those preferring a pragmatic leader were most pragmatic and least focused on interpersonal concerns. Work values were particularly useful in differentiating those followers preferring a charismatic leader from those preferring a pragmatic leader. Individuals preferring a charismatic leader had more collectivistic work values (i.e., low on competition and autonomy and high on teamwork), whereas those preferring a pragmatic leader had more individualistic work values (i.e., high on competition and autonomy and low on teamwork).
The current study adds to the abovementioned literature in several ways. First, we draw upon the HEXACO model of personality to further explore follower characteristics related to leadership preferences. We hereby increase the nomological net surrounding the role of follower characteristics in the leader-follower relationship. Next, we draw upon the similarity perspective of Klein and House (1998) and the need fulfillment perspective introduced by Ehrhart and Klein (2001) to explain the relations between follower characteristics and their preference for leadership. Finally, we do not focus exclusively on forced choices between different leaders (i.e., charismatic, relationship-oriented, or task-oriented leader) but acknowledge that followers may differ in the extent to which they prefer to work with different types of leaders (i.e., charismatic, relationship-oriented, or task-oriented leader) and examine how follower personality is related to both.

Follower preference for charismatic leadership
Charismatic leaders are visionaries who motivate their followers to perform beyond expectations by transforming followers' self-interest to the collective interest (Bass, 1988;Burns, 1978;House, 1977;Weber, 1921). With their ideological goals, they make work meaningful and create a sense of purpose in followers' work (Shamir et al., 1993). Based on the similarity perspective, we expect that followers who prefer a charismatic leader are characterized by high Honesty-Humility, Extraversion and Openness to Experience. First, individuals who score high on Honesty-Humility are honest and sincere, which matches with charismatic leaders' moral righteousness and trustworthiness (Conger, 1999), something which is reflected in the high correlation (.63 uncorrected) between transformational/charismatic leadership and ethical leadership (Hoch et al., 2018). Similarly, applicants low on Honesty-Humility were more likely to prefer a morally questionable leader or a leader whose ethicality was unknown (Ogunfowora, 2014). In addition, charismatic leaders express the sentiment of the group to motivate employees for their cause, and employees high in Honesty-Humility are less egoistic (De Vries, De Vries, De Hoogh, & Feij, 2009) and therefore more likely to appreciate the charismatic leader's group focus. Finally, leaders who score high on Honesty-Humility are more likely to show charismatic leadership (De Vries, 2008), supporting the similarity perspective between followers high in Honesty-Humility and charismatic leaders.
Second, it seems likely that extravert followers prefer a charismatic leader. That is, extraverts, similar to charismatic leaders, are positive, energetic, enthusiastic, and feel comfortable in social situations. Further supporting the similarity argument for Extraversion, research shows that charismatic leaders are more likely to be extravert themselves (Bono & Judge, 2004;De Vries, 2012). Finally, we expect that followers who score high on Openness to Experience prefer to work with a charismatic leader. Followers high on Openness are characterized by openness to new experiences and change, and charismatic leaders are more likely to emerge in work environments that offer many opportunities for change (Shamir & Howell, 1999). Additionally, charismatic leaders communicate an attractive vision of the future for the organization and encourage followers to come forward with new and creative ideas (Yukl, 1999;Kark, Shamir, & Chen, 2003), which should be especially appealing to creative followers with high levels of imagination and divergent thinking skills (i.e., followers high on Openness to Experiences). Lastly, followers who are high on Openness to Experience are interested in unconventional ideas and people, and charismatic leaders often look for new opportunities and use unconventional methods to reach their ideological goals (e.g., Conger & Kanungo, 1994).
Hypothesis 1: (a) Honesty-Humility, (b) Extraversion, and (c) Openness to Experience are positively related to followers' preference for charismatic leadership.

Follower preference for relationship-oriented leadership
Relationship-oriented leaders are leaders who are focused on the general well-being and needs of their followers by listening to and communicating with followers, showing trust and confidence in followers, and by recognizing their performance (Fleishman, 1953;Halpin & Winer, 1957;Likert, 1961;Yukl, 1998). From a need fulfillment perspective, we expect followers high on Emotionality to prefer working for a relationship-oriented leader, because these followers experience a range of negative emotions such as anxiety, worry, and fear, and are therefore in greater need of emotional support and comfort from others (Ashton et al., 2014), and because they value emotional connections to others. In support of this argument, previous research has shown that employee Emotionality is associated with a higher need for leadership (De Vries, Roe, Taillieu, & Nelissen, 2004) and that employees with a higher need for leadership rate their leader as more relationship-oriented (De Vries, 2000). Following this reasoning, we hypothesize: Hypothesis 2: Emotionality is positively related to followers' preference for relationship-oriented leadership.

Follower preference for task-oriented leadership
Task-oriented leaders have a strong task focus and consequently, structure followers' tasks in terms of what needs to be done and when it needs to be finished (Fleishman, 1953;Halpin & Winer, 1957;Likert, 1961;Yukl, 1998). These leaders motivate followers by setting high, but attainable goals, and provide assistance where needed to achieve these goals. From a needs perspective, we expect followers high in Emotionality to prefer such as leader, because task-oriented leaders reduce risks by clearly outlining what followers need to accomplish and how to accomplish this, thereby reducing any negative emotions that these followers may experience (e.g., worry, anxiety). That is, when followers follow a taskoriented leader, they do not have to worry about not getting their job done, because they receive all the necessary support from their leader. Empirical support for the relation between Emotionality and task-oriented leadership is found in research by De , who found that followers' Emotionality was positively related to their need for leadership, which in turn was positively related to perceptions of task-oriented leadership.
From a similarity perspective, one would expect conscientious followers to prefer a task-oriented leader. That is, both conscientious followers and task-oriented leaders are organized, disciplined, and goal-oriented. They both work hard and make informed decisions rather than being guided by impulses. Moreover, task-oriented leaders can be characterized as conscientious themselves too (De Vries, 2012). Finally, we expect followers' Openness to Experience to be negatively related to their preference for a task-oriented leader, because followers high on Openness to Experience and taskoriented leaders are very dissimilar. Whereas task-oriented leaders set clear goals with standards and instructions to achieve these goals and allow no room for deviations from that plan and/or experimentation, followers who are open to experience are creative and therefore like to be original and inventive. They are curious and unconventional, something that task-oriented leaders may not appreciate. Following these arguments, we hypothesize: Hypothesis 3: (a) Emotionality and (b) Conscientiousness are positively related to followers' preference for task-oriented leadership, whereas (c) Openness to Experience is negatively related to followers' preference for task-oriented leadership.
Finally, we did not hypothesize a relation between follower Agreeableness and the preference for or any kind of leadership, because we expect that agreeable followers work well with different types of leaders, whether they are high or low on relationship-oriented leadership, task-oriented leadership, and/or charismatic leadership. That is, agreeable followers are highly tolerant of other people, willing to compromise when necessary, forgive quickly, do not hold grudges, and are able to stay calm when they are unfairly treated by others (such as their leaders).

Participants and procedure
The participants in our study were 272 first year Psychology students at a Dutch university. Students received an e-mail with the link to the online questionnaire. Three hundred and twenty-two students started the questionnaire, of which two hundred seventy-two (84.5%) finished it. The sample included 55 (20.2%) men and 217 (79.8%) women with an average age of 20.30 (SD = 3.63), ranging from 18 to 48. At the end of the questionnaire, students could fill out their e-mail address to receive a personalized personality report. This e-mail was sent to them automatically by Qualtrics.
Leader descriptions For the descriptions of the charismatic, task-oriented and relationship-oriented leaders, we used the materials developed by Ehrhart and Klein (2001). These authors developed a hypothetical situation in which participants were told that they had been hired to work as a company manager at a company called "Not Just Java" (see Appendix 1 in the Ehrhart & Klein, 2001 article). Participants were also told that they had the opportunity to choose the district manager they would like to work with, and that three different district manager provided a description of their management styles to aid them in making a decision. These descriptions reflected either a charismatic, task-oriented, or relationshiporiented management style (see Ehrhart & Klein, 2001 for more information on the development of these descriptions), which were offered one by one in random order to prevent order effects from biasing our results. Each leader description consisted of nine sentences; one introductory sentence and eight sentences that captured four defining behaviors of the specific type of leadership (i.e., charismatic, task-oriented, or relationship-oriented). Please see Appendix 1 for the descriptions of all types of leaders.
Preference for leadership We measured followers' preference for leadership in two ways. First, for each description, participants had to indicate to which extent they preferred working for that specific leader by answering six questions such as "I would enjoy working with this district manager" and "I would get along with this district manager". Participants could answer these questions on a 5-point scale, ranging from "to little or no extent" to "to a great extent". The preference for leadership questions showed good internal consistencies for charismatic leadership (α = .91), task-oriented leadership (α = .92), and relationship-oriented leadership (α = .91). Second, after rating all three individual district managers, participants were asked to choose one of the three different managers they would most like to work for. We used this forcedchoice question for our discriminant function analysis. Table 1 shows the means, standard deviations, reliabilities, and inter-correlations between all variables included in the study. The zero-order correlations show that the preference for relationship-oriented leadership is positively correlated with both the preference for charismatic (r = .19) and taskoriented (r = .38) leadership.

Leader ratings
Most participants (54.4%) opted for the relationship-oriented leader when we asked them to choose between the three district managers, followed by the charismatic (24.3%) and the task-oriented (21.3%) leader. These findings matched the mean preference score for the relationship-oriented leader, which was higher (M = 4.23) compared with either the charismatic (M = 3.46; t (271) = 12.24, p < .001) or the taskoriented (M = 3.63; t (271) = 10.92, p < .001) leader. The results for the preference for charismatic and task-oriented leadership were somewhat different from the forced-choice question responses, showing that the difference was in the opposite direction, with a higher mean for preference for task-oriented (M = 3.63) compared with charismatic leadership (M = 3.46; t (271) = −2.32, p < .05).

Hypotheses testing
We tested all our hypotheses in one model using Mplus (Muthén & Muthén, 1998. This model included the six HEXACO dimensions and age as predictors of the three preferences for leadership (i.e., charismatic, task-oriented, and relationship-oriented). We did not include gender in this model because gender was unrelated to all our outcomes variables (see Table 1). We controlled for measurement error by fixing the residual variance of all continuous variables at (1 − internal consistency) * variance (Antonakis, Bendahan, Jacquart, & Lalive, 2010;Ree & Carretta, 2006).

Discriminant function analysis
We used discriminant function analysis (DFA) to examine what constellation of personality dimensions maximally distinguishes those followers who prefer a charismatic leader from those who prefer either a relationship-oriented or a task-oriented leader. DFA is especially useful when the dependent variable is categorical and has more than two categories (Cohen & Cohen, 1983, see also Ehrhart &Klein, 2001 andThoroughgood &Sawyer, 2018), which is why we used this analysis to distinguish follower characteristics based on their forced choice for one of the three abovementioned types of leaders. The DFA resulted in two discriminant functions, one less than the amount of categorical dependent variables. The first function explained 66.6% of the variance, canonical R 2 = .06, and the second function explained 33.4% of the variance, canonical R 2 = .03. Together, these functions significantly discriminate between the groups (X 2 = 23.506 (12), p < .05), but the second function alone does not significantly differentiate the three different groups (X 2 = 7.920 (5), p = .161). The structure matrix provides the correlations between the independent variables (i.e., HEXACO dimensions) and the functions, and showed that Extraversion (.616 vs. − .211), Emotionality (− .594 vs. .243), and Openness to Experience (.538 vs. .073) loaded more strongly on the first than the second function. Conscientiousness loaded more strongly on the second (.849) than the first (.176) function. Honesty-Humility (.108 and − .187) and Agreeableness (− .247 and − .203) had weak loadings on both functions. The discriminant functions plot (see Fig. 1) shows that the first function discriminates between the preference for charismatic leadership group (M = .409), the preference for relationship-oriented leadership group (M = −.075), and the preference for task-oriented leadership group (M = −.274). Furthermore, the second function also distinguished between preference for charismatic leadership (M = .097), preference for relationship-oriented leadership (M = −.149), and preference for task-oriented leadership (M = .269). To increase the understanding of the interpretation of the DFA, we also included the mean personality scores for each personality trait per group-i.e., those preferring a charismatic, task-oriented, and relationship-oriented leader (see Fig. 2). Together, Figs. 1 and 2 show that the more extraverted and open to new experiences followers are, the more likely they are to prefer a charismatic leader compared with a relationship-and taskoriented leader, and the more likely they are to prefer a relationship-oriented leader compared with a task-oriented leader. The opposite is true for Emotionality-i.e., emotional individuals are more likely to prefer a task-oriented leader, followed by a relationship-oriented leader and last, a charismatic leader. Finally, Conscientious individuals are more likely to prefer either a charismatic or a task-oriented leader over a relationship-oriented leader. Although DFA does not allow to draw conclusions about the significance of the mean differences such as those presented in Fig. 2, it does show that the individuals' preference for either a charismatic, relationshipor task-oriented leader is best predicted by the personality traits Conscientiousness, Extraversion, Openness to Experience, and Emotionality.

Discussion
The aim of the study was to examine how the six HEXACO personality domains are related to followers' preference for charismatic, relationship-oriented, and task-oriented leadership. First, based on the similarity perspective (Klein & House, 1998), we expected Honesty-Humility, Extraversion, and Openness to Experience to be positively associated with a preference for charismatic leadership, whereas low Openness to Experience and high Conscientiousness were expected to relate to a preference for task-oriented leadership. Second, from a need fulfillment perspective (Ehrhart & Klein, 2001), we expected Emotionality to be positively related to the preference for both a relationship-oriented and a task-oriented leader. We found support for all expectations, with two exceptions; (1) there was no significant relation between Conscientiousness and preference for task-oriented leadership, and (2) Honesty-Humility was negatively related to preference for charismatic leadership.

Theoretical implications
Our study contributes to the literature, showing that followers who prefer a charismatic leader can be characterized by low Honesty-Humility and high Extraversion and Openness to Experience. Whereas Ehrhart and Klein (2001) focused on followers' work values to characterize followers with a preference for charismatic leadership from a need fulfillment perspective, our study seems to support the similarity argument: followers who are extraverted and open to experience prefer a charismatic leader. We found the unexpected finding Our finding seems to suggest that charismatic leaders fulfill the needs of those low on Honesty-Humility, because charismatic leadership is likely to instigate changes, bringing about uncertainty. People low on Honesty-Humility may thrive under these circumstances, because uncertainty allows them to exploit the situation. In contrast, people high on Honesty-Humility may have a lower preference for charismatic leadership, because such leadership may be associated with greater status differences between the leader and his/her followers, something which runs counter to the need for relationships that are characterized by equality and fairness among those high on Honesty-Humility. Interestingly, Schyns and Sanders (2007) found that Honesty-Humility was positively related to followers' perceptions of transformational leadership. It would be interesting for future research to further investigate the relation between follower Honesty-Humility and charismatic leadership, paying special attention to status differences and personalized ("bad") versus socialized ("good") charisma distinctions (House & Howell, 1992). That is, whereas followers high on Honesty-Humility may appreciate the social and power-sharing side of charisma, they may have an aversion for the personalized side of charisma, which is about dominance and exploitation. That is, some charismatic leaders may be seen as self-confident, dominant, and with a need for influence (Conger & Kanungo, 1987), whereas followers high on Honesty-Humility value modesty and fairness and generally dislike greediness and pretentiousness (Ashton & Lee, 2001). Building on this study, future studies might like to further explore the relation between Honesty-Humility and the preference for charismatic leadership using socialized and personalized charismatic leadership vignettes that allow for a clearer differentiation between different expressions of charismatic leadership. Our findings regarding the preference for task-and relationship-oriented leadership seem to fit better with the need fulfillment rather than the similarity perspective. That is, whereas relationship-oriented leaders are able to provide comfort and emotional support to followers high on Emotionality, task-oriented leaders may prevent followers from experiencing emotions such as anxiety and worry. Whereas Ehrhart and Klein (2001) conclude that need fulfillment is the "dominant perspective" to explain the relation between follower personality and their preference for leadership, Thoroughgood and Sawyer (2018) state that followers "tended to choose the leader who they perceived to be most similar to them". Based on these and our findings, we may conclude that more research is needed to understand the characteristics of followers of charismatic, relationship-or task-oriented leaders. It also seems possible that because leaders are similar to their followers, they are better able to understand what their followers need and consequently, to provide in those needs. Thus, more research is needed to clarify which of two perspectives, − i.e., similarity and need fulfillmentis dominant for each personality dimension to get a better grasp of the underlying mechanisms and interpersonal processes there are at play here. One way to achieve this is by drawing from the person-organization fit literature. Kristof (1996) distinguished fit as a result of similarity in for example work values (i.e., similarity perspective) and fit as a result of complementary demands and supplies (i.e., need fulfillment perspective). These two kinds of fit can be measured either directly, indirectly, or both. Direct measures are indicators of perceived fit, for example asking participants very explicitly to what extent they perceive themselves to be similar to a leader (see also Thoroughgood & Sawyer, 2018) and to what extent they expect that leaders will fulfill their needs. Indirect  Fig. 2 Average scores on the HEXACO personality dimensions per group measures more explicitly measure fit and can be considered indications of actual rather than perceived fit. An example of explicit fit is asking both leaders and followers to rate their own personality and use difference scores and/or polynomial regression analyses to examine to what extent (dis)similarity in personality predicts followers' preference to work with that leader. Similarly, asking followers what they need from their leader and the leader to what extent (s)he fulfills these needs is a direct measure of demands-supply fit. More exploratory, future research could include open ended questions, asking why followers would or would not prefer to work with a specific leader (Ehrhart & Klein, 2001). Surprisingly, we did not find a relation between Conscientiousness and preference for task-oriented leadership. From a similarity perspective, we expected that conscientious followers would prefer a task-oriented leader, because both are organized and disciplined, and because task-oriented leader tend to be conscientious too (De Vries, 2012). Yet, conscientious individuals are also perfectionists who carefully execute and check their work in order to not make mistakes, and make well-advised decisions, which may be hindered by a task-oriented leader who wants to get things done. Additionally, conscientious individuals may be able to adapt to work with any type of leader, considering their levels of adaptability to many situations in life (e.g., Pulakos et al., 2002;Shiner & Masten, 2002). For instance, conscientious individuals are noted to have a better work-family balance (Wayne, Musisca, & Fleeson, 2004;Witt & Carlson, 2006); they may similarly also be better able to balance different demands in the workplace and to better adapt to different styles of leadership by better organizing-and taking responsibility for-their own tasks.
Similar to Ehrhart and Klein (2001), we found that most followers preferred a relationship-oriented leader rather than a charismatic leader (see also Thoroughgood & Sawyer, 2018). That is, the mean preference for relationship-oriented leadership was highest (M = 4.23) compared with the preference for charismatic and task-oriented leadership, and when forced to choose between the three types of leaders, 54.4% chose the relationship-oriented leader. Together with our findings on followers' personality traits related to their preference for charismatic leadership, our study shows the importance of a more balanced approach to leadership, including the role of both leaders and followers when studying charismatic (and other styles of) leadership.
Similar to previous studies (Ehrhart & Klein, 2001;Thoroughgood & Sawyer, 2018), we performed discriminant function analyses to examine whether, when forced to choose, those preferring a charismatic, relationship-oriented, or taskoriented leader could be distinguished based on their HEXACO personality traits. We showed that, similar to our regression analyses, those preferring a charismatic leader were more extraverted and open to experiences compared with those preferring a relationship-or task-oriented leader. Similarly, those preferring a relationship-or task-oriented leader were more emotional compared with those preferring a charismatic leader. Additionally, the DFA showed that conscientious individuals were more likely to prefer either a charismatic or task-oriented leader over a relationship-oriented leader. Thoroughgood and Sawyer (2018) compared employees preferring a charismatic leader with employees preferring either an ideological or pragmatic leader and showed that those preferring a charismatic leader were more extraverted and agreeable, and less neurotic. It is important to realize that DFA assumes that employees preferring one style of leadership do not prefer any other styles of leadership (i.e., forced choice between leadership styles), whereas in reality, employees may prefer different styles of leadership and leaders may even behave differently-i.e., both charismatic and taskoriented, depending on the situation. In support of this claim, we found (small to medium) positive correlations between the preferences for all three styles of leadership.

Practical implications
Our study shows that is it not just important to select those who are suited for a leadership position, but that it is equally important to pay attention to the selection and placement of employees working with that leader. That is, there should be a match based on personalities, work values, and a good fit between what followers need and what leaders are able to provide. Leadership research has been very much leader-centered, which has led to the belief that charismatic leadership is universally appealing and a key to success. Yet, our study, and previous studies, shows that when given the option to choose between different leaders, most participants would choose a leader other than a charismatic leader (e.g., a relationshiporiented leader). Thus, it is important for organizations to consider what kind of leaders they wish to have and what kind of followers get along best with such leadership.

Strengths, limitations, and implications for future research
The different strengths and limitations of the current study may serve as an inspiration for further research on follower characteristics and their preferences for charismatic, relationship-oriented, and task-oriented leaders and more broadly, on the boundary conditions of the effectiveness of different leadership styles. A first strength of the current study is the inclusion of not only the preference for a single leadership style, but the preference for three different styles of leadership, so our results are not biased by a general preference for leadership. Furthermore, the paper vignette methodology ensured that all participants had the same understanding of the different leaders, and this perception was not affected by their relationship with the leader and/or paraverbal (e.g., tone of voice) and non-verbal (e.g., appearance or gestures) cues.
Our study also has its limitations. Similar to the Ehrhart and Klein (2001) study, we made use of an undergraduate sample, which may differ from an employee sample in terms of experience working with leaders, and consequently, reduce the engagement in the study and limit the generalizability of the findings. Yet, there are several indications that our results are unlikely to be different in a sample of working adults. First, leadership is an influence process in which the leader encourages someone to achieve a common goal (Northouse, 2012). Accordingly, leadership does not only exist within formal organizations, but also within schools (e.g., teachers), sports (e.g., team leader), peers (e.g., emergent leadership), and even within parenting. In addition, many students nowadays work alongside their studies. In the Netherlands, 50% of all students in pre-university training have a side job (equally divided among men and women; CBS, 2019) and 67% of all students attending university have a side job or a paid internship (Van der Werf, Schonewille, & Stoof, 2017), working 17 h a week on average. These numbers are quite high and show an increase over the past years due to major changes in the financial support offered to students by the Dutch government ( Van der Werf et al., 2017). Third, although we do not have information about the work experience of our specific sample, we included age as the closest proxy to work experience. Our results were highly similar to those of Ehrhart and Klein (2001), who did control for students' actual work experience, most of whom (55%) had over 2 years of work experience. In both their and our study, most students preferred a relationship-oriented leader (54.4% in our study versus 51%), followed by a charismatic leader (24.3 versus 28.8%) and a task-oriented leader (21.3 versus 19.9%).
Finally, Thoroughgood and Sawyer (2018) studied the preference for charismatic, ideological, and pragmatic leadership among working adults, who on average had 15.50 (SD = 5.12) years of work experience. They included several demographic factors, including work experience, as controls, but found that these controls did not explain significant differences in participants' leader preferences, meaning that work experience does not seem to affect the style of leadership that followers prefer. One argument could be that regardless of one's work experience, participants may not have experience with each and every one of the leadership styles included in these studies (e.g., charismatic, relationship-oriented, pragmatic, etc.). Although work experience may not directly affect employees' preference for leadership, employees' preference for a specific type of leader may change over the course of their career depending on the experience they have with certain leaders. For example, charismatic leaders try to push followers out of their comfort zones, which may be uncomfortable, and thus not preferable, at a relatively young age. However, employees may start to appreciate and prefer this kind of leadership once they grow older and once they become more familiar with charismatic leaders. To make sure that our results are indeed generalizable to the general working population, future studies should employ samples from the working population and present them with more vivid leadership scenario's such as videos (although these have their own confounds and may therefore be used in addition to the paper vignettes). In addition, future studies might like to include the amount of exposure to each type of leader to see whether this affects leadership preferences.
While our study shows that followers' preference for leadership is dependent on followers' personality traits, we do not know what happens when followers work for a specific leader while they do not prefer working for such a leader. Interestingly, Ehrhart and Klein (2001) showed that followers with a low preference for charismatic leadership viewed the assumed positive characteristics of these leaders, such as team-oriented, committed, and empowering, in a negative light (e.g., overconfident, arrogant, and pushy). Following these results, one would actually expect that the well-being and performance of these followers may drop when "forced" to work with a charismatic leader. In addition, as followers gain experience working under a specific leader, this may change followers' perspective on leadership, changing their preference for leadership. Future research is needed to shed light on the role of follower characteristics in the effects of charismatic, relationship-and task-oriented leadership on outcomes such as follower well-being and performance. More broadly, it would be interesting to study the effects of leaderfollower (dis)similarity, because dissimilarity can be potentially beneficial to performance (e.g., different views).
Another meaningful avenue for future research would be to examine situational characteristics in addition to individual difference variables (e.g., personality, work values) related to followers' preference for charismatic leadership. That is, there may be situations in which followers prefer to have a leader who gives them hope for the future and who creates meaning and purpose in followers' work, such as when followers are facing a sustained high workload and/or when there is a lack of cohesion within workgroups.

Conclusion
We examined the idea that charismatic leaders are not universally appealing and that some followers may be more attracted to a relationship-or task-oriented leader. We showed that followers' (low) Honesty-Humility, (high) Extraversion, and (high) Openness to Experience were predictive of followers' preference for charismatic leadership, whereas followers' (high) Emotionality was predictive of their preference for a relationship-and task-oriented leader. In addition, and in line with previous research, it appeared that followers preferred a relationship-oriented leader over a charismatic leader. Our findings offer support for the idea that followers prefer a leader who is similar to them and at the same time fulfills their needs, which shows the importance of including the role of followers when studying leadership.
Appendix 1: Leader descriptions (literally transcribed from Ehrhart & Klein, 2001) Charismatic leader description "I have been a successful leader because I am committed to this company's future and I work hard to communicate my vision for this company to my store managers. I set high standards for my store managers. I expect them to work as hard as they can to reach those standards. However, I don't push them only for the sake of productivity; rather, I want them to reach their potential and do the best job they can. I want them to realize how good they can be and how much they have to offer. My goal is to do things differently than this organization has done them in the past, and I'm willing to take some chances to show them how things can be improved. I rely on my store managers to be creative in finding new ways to get the job done. I don't want my store managers to think of this as just another job. Instead, I try hard to make them feel like they're a part of something special here, something big, something that's going to make a difference in this organization." Relationship-oriented leader description "I attribute my success as a leader to my concern for my store managers' personal well-being. The first thing I try to do in all of my interactions with my store managers is to treat them with kindness and consideration. I am committed to being friendly and respectful, even when stress is high or there is a lot of work to be done. Another thing I emphasize with my store managers is communication. I keep them informed of progress on projects or any other organizational issues that might affect them, and I am always available to listen to my subordinates' problems, whether their problems are personal or work-related. In addition, I show trust and confidence in my store managers. I want them to feel involved in their work and to know that I think they can do a good job. The final thing I do with my store managers is that I recognize their contributions. If they work hard and do a good job, I go out of my way to make sure they know that their work is appreciated." Task-oriented leader description "I'm successful as a leader because I emphasize task accomplishment. I begin by working with my store managers to set goals for their work. I don't want to overwhelm my store managers with impossible standards, so I make sure their goals are realistic yet still challenging. I am very careful and detailed in laying out what my store managers need to get done. I don't want there to be any ambiguity; they need to know exactly what to do and when it needs to get done. Once they know what needs to get done, I make sure they have everything they will need to do it. I provide them with the necessary supplies, equipment, and technical assistance to insure that they can be successful at their jobs. Finally, I coordinate the work so that the store managers and their assistant managers know what their job is and there is no overlap between the two. I want everyone to know what their role is so that they can see how they are contributing to the accomplishment of our organization's goals." Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.