• No results found

Leader vulnerability : how leaders who purposefully display vulnerability influence followers’ perceptions of leader effectiveness

N/A
N/A
Protected

Academic year: 2021

Share "Leader vulnerability : how leaders who purposefully display vulnerability influence followers’ perceptions of leader effectiveness"

Copied!
55
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Leader vulnerability:

How leaders who purposefully display vulnerability influence

followers’ perceptions of leader effectiveness

MASTER THESIS

MSc Business Administration – Leadership and Management University of Amsterdam

Gabrielle de Jong – 11398167

Supervisor: Dr. Lisanne van Bunderen June 21, 2018

(2)

Statement of originality

This document is written by Gabrielle de Jong who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Table of Contents

1 Introduction ... 7

2 Literature Review ... 10

2.1 Leader Vulnerability ... 11

2.2 Trust in the Leader ... 13

2.3 Gender of the Leader ... 17

2.4 Perceived Leader Effectiveness ... 20

2.5 Conceptual Research Model ... 23

3 Method ... 24

3.1 Procedure and Sample ... 24

3.2 Measures ... 25 3.3 Data Analysis ... 27 4 Results ... 28 4.1 Descriptive Statistics ... 28 4.2 Hypothesis Testing ... 29 4.3 Exploratory Analysis ... 31 5 Discussion ... 33

5.1 Discussion of the Results ... 33

5.2 Limitations and Future Research Directions ... 38

5.3 Managerial Implications ... 43

6 Conclusion ... 44

7 References ... 45

8 Appendices ... 54

8.1 Appendix 1 – Scale Perceived Leader Effectiveness ... 54

8.2 Appendix 2 – Skewness and Kurtosis of Variables ... 54

(4)

List of Figures and Tables

Figure 1. Conceptual Research Model ... 24

Table 1. Means, Standard Deviations and Correlations ... 28

Table 2. Results of Regression Analysis ... 29

Table 3. Indirect Effects of Moderated Mediation Analysis ... 31

Table 4. Mediation Effects of Leader Vulnerability on Perceived Leader Effectiveness through Trust in the Leader ... 32

Table 5. Skewness and Kurtosis ... 54

(5)

Acknowledgements

This thesis is the final part of my Master in Business Administration at the University of Amsterdam. There are a number of people that helped and supported me through this thesis process, whom I would like to thank. First of all, my supervisor, Lisanne van Bunderen who provided me constructive feedback and support throughout the process. Second, the colleague students of my thesis group with whom I designed the survey and collected the data necessary for this thesis, and who did their very best to help me answer the questions I came up with throughout the process. Also, all people who invested their time and effort to fill in the survey, without which I would not be able to do this research. Finally, my family and friends, who not only provided me with feedback, but also supported me, motivated me and believed in me.

(6)

Abstract

The general idea that vulnerability is a weakness causes many leaders to be reluctant about displaying vulnerability in their function as a role model. However, increasing research on related topics, such as leader authenticity and humility, discussed positive effects of leaders who display vulnerability. Yet, empirical research on leader vulnerability, as a topic on itself, is still missing. The aim of this thesis is to put leader vulnerability on the map by showing its expected positive effects on trust in the leader and perceived leader effectiveness. By taking a quantitative approach and collecting data of 136 leader-follower dyads, leader vulnerability was found to be positively related to followers’ trust in their leader, which on its turn was found to be positively related to perceived leader effectiveness. Additional analysis showed that trust in the leader fully mediated the positive relationship between leader vulnerability and perceived leader effectiveness. Against the expectation that the proposed model would show stronger effects for female leaders than for male leaders, because leader vulnerability is seen as a rather feminine characteristic, gender of the leader was not found to have a moderation effect on the proposed model. The implications of the findings will be discussed.

Keywords: leader vulnerability, trust in the leader, perceived leader effectiveness, gender of the leader

(7)

A leader, first and foremost, is human. Only when we have the strength to show our

vulnerability can we truly lead - Simon Sinek (2010)

1 Introduction

In the current rapidly changing environment in which many organizations operate, characterized by learning and innovation, effective leadership is of high importance to ensure team performance (Bunker, 1997; Hirschhorn, 1990; Uhl-Bien, Marion, & McKelvey, 2007; Yukl, 2012). However, at the same time leaders are confronted with challenges, such as high performance standards, this environment brings. In order to deal with these challenges, leaders have to cope with the fact that they are becoming more vulnerable as they might not be able to deal with the high performance standards by themselves (George et al., 2007). Although the current work environment makes leaders thus more vulnerable, an interesting emerging line of research suggests that displaying vulnerability by leaders could be a manner to deal with this environment by improving the effectiveness of leaders (Bunker, 1997; DeLong & DeLong, 2011). In this thesis, (displayed) leader vulnerability will be addressed and a deeper understanding of its application in the workplace and its importance for leader effectiveness will be developed.

Leader vulnerability can be defined as a leader who displays vulnerability, by showing one’s weaknesses, and being open about one’s failures and insecurities (Bell, 2005; Bunker, 1997; Chambers, 1989; Hirschhorn, 1990; Lapidot, Kark, & Shamir, 2007; Ludeman & Erlandson, 2004; Mayer, Davis, & Schoorman, 1995; Nienaber, Hofeditz, & Romeike, 2015; Tsui-Auch & Möllering, 2010). Leaders who display vulnerability tend to be perceived as open, real, genuine, and authentic (Bell, 2005). They will acknowledge their imperfections and dependence on followers, and will ask them for help or advice if necessary. Based on related

(8)

literature, leader vulnerability is expected to have positive outcomes on followers’ trust in their leader and their perceptions of leader effectiveness (Avolio, Gardner, Walumbwa, Luthans, & May, 2004; Bunker, 1997; Gardner, Avolio, Luthans, May, & Walumbwa, 2005; Ladkin & Taylor, 2010; Lapidot et al., 2007; Nienaber et al., 2015; Tsui-Auch & Möllering, 2010; Walumbwa, Avolio, Gardner, Wernsing, & Peterson, 2008).

Leader vulnerability expectedly leads to followers’ trust in their leader because, as literature on authentic leadership states, followers will be able to identify with their leaders if their leaders open up about themselves (Avolio et al., 2004; Ladkin & Taylor, 2010). Literature on trust adds that leaders who display vulnerability tend to create an open climate in which they encourage their followers to behave the same way, and expose their own insecurities and weaknesses as well (Avolio et al., 2004; Norman, Avolio, & Luthans, 2010; Shamir & Eilam, 2005). Such climate encourages trusting relations between leaders and their followers. Some authors already discussed effects of vulnerability in leadership (Bell, 2005; Brown, 2010, 2012a, 2012b, 2015; Bunker, 1997; DeLong & DeLong, 2011; Farson & Keyes, 2002; George et al., 2007). For instance, Farson and Keyes (2002) argued that by displaying vulnerability, leaders earn empathy by their followers and make their followers admire them and identify with them, which results in closer ties between them. A regularly discussed outcome of followers’ trust in their leader is perceived leader effectiveness (Avolio et al., 2004; Burke, Sims, Lazzara, & Salas, 2007; Gillespie & Mann, 2004; Hogan, Curphy, & Hogan, 1994; Judge, Bono, Ilies, & Gerhardt, 2002; Norman et al., 2010). Hogan et al. (1994) stated that perceptions of leader effectiveness are mainly dependent on the relationships a leader has with his or her followers and followers’ perceptions of the credibility and trustworthiness of their leader. Although leader vulnerability is expected to have positive outcomes (i.e. trust in the leader and leader effectiveness), empirical research on leader vulnerability and its effects is still missing. This

(9)

thesis aims to show these expected positive effects of leader vulnerability by focusing on its link with perceived leader effectiveness, which showed to have different desirable organizational outcomes such as organizational performance and employee satisfaction (Avolio et al., 2004; Yukl, 2012).

Though displaying vulnerability is expected to have positive effects on follower’s trust in their leader and accordingly the perceived effectiveness of their leader, this may be contingent on the gender of the leader. Several studies found vulnerability to be a rather feminine characteristic (Alimo‐Metcalfe, 1995, 2010; Billing & Alvesson, 2000). This indicates that leader vulnerability might be subject to gender stereotyping, leading to possible different perceptions of leader vulnerability for male and female leaders, and therefore different effects on followers’ trust in their leader. It is expected that female leaders who display vulnerability would conform to the gender stereotypical expectations of females such as empathy, community and collaboration; while male leaders who display vulnerability expectedly act against the gender stereotypical expectations of males such as independence, autonomy, and control (Eagly, 2005; Eagly & Karau, 2002; Fletcher, 2004). Because female leaders who display vulnerability act in the way that others expect them to act, it is likely that they are perceived as more trustworthy and effective than male leaders who display vulnerability, and thus do not act in line with the expectations (Eagly & Karau, 2002). In order to see if gender-stereotyping indeed leads to more trust in female leaders who display vulnerability compared to male leaders who display vulnerability, this thesis will take gender of the leader into account in the proposed research model.

This study has several contributions to theory and practice. First, this thesis will contribute to literature on leader vulnerability by trying to gain deeper insights in the link between leader vulnerability and trust in the leader, its effects on leader effectiveness, and the influence of

(10)

gender of the leader on the model. Found relationships, could provide interesting directions for future research on leader vulnerability and have interesting implications for its application in practice. Second, this thesis will contribute to literature on trust by creating deeper insights in the effects of leaders who display vulnerability on the trust followers have in their leaders (Lapidot et al., 2007; Nienaber et al., 2015). Further, if gender of the leader is found to influence the model, this could give some interesting new insights into literature on gender, as it might show differences in what followers perceive as important when evaluating their male and female leaders. This could imply that male leaders should take different approaches, in order to be considered as trustworthy and effective, than female leaders. Finally, the results of this thesis could be of interest for leaders and managers in contemporary organizations. With a deeper understanding of leader vulnerability and its implications, organizations could provide training and education to leaders and managers about possible ways to integrate leader vulnerability in their daily practices. An influence of the gender of the leader would indicate that organizations should take different approaches in how they direct male and female leaders.

To start, a literature review will provide some deeper insights into existing research on the different variables. Based on existing literature, a conceptual model will be built. The corresponding hypothesis will be tested by using a quantitative approach. Based on the results, this thesis concludes with a discussion of the general findings, limitations, managerial implications and possible directions for future research.

2 Literature Review

In the following chapter the different topics will be introduced. With reference to existing research, hypothesis will be developed based on the expected interrelations between the topics. A conceptual theoretical framework will follow as a result.

(11)

2.1 Leader Vulnerability

Until now, literature mainly viewed vulnerability as an (emotional) state, especially in contexts such as the medical-, sociological and business sciences, and human-environment systems (Nienaber et al., 2015). There, vulnerability is seen as a rather negative state in which someone is “being exposed, open to attack or injury” (Bell, 2005, p.19) and not able to defend oneself (Nienaber et al., 2015). Leaders in these positions were often seen as ineffective leaders, because of their low levels of self-confidence and inability to develop trustworthy relations with followers (Costa & McCrae, 1995; De Hoogh, Den Hartog, & Koopman, 2005; Judge, Bono, Ilies, & Gerhardt, 2002). This thesis, however, does not approach vulnerability as a state, but rather as a consicously chosen behavior and refers to it as leader vulnerability. This approach was taken because, as leader vulnerability is a consciously chosen behavior rather than an uncontrollable event, it also has a strategic aspect. Besides, as leader vulnerability is a behavior, it is easier to measure than an emotional state. Leader vulnerability can be described as a leader who purposefully displays vulnerability by showing one’s weaknesses and insecurities (Bell, 2005; Bunker, 1997; Chambers, 1989; Hirschhorn, 1990; Lapidot et al., 2007; Mayer et al., 1995; Nienaber et al., 2015; Tsui-Auch & Möllering, 2010)(Bell, 2005). This implies that a leader who displays vulnerability is not afraid to admit his or her shortages and to acknowledge if he or she has fallen short in what was expected.

Displaying one’s vulnerability is often seen as a weakness, where leaders feel they do not act in accordance with the expectations of their positions as role models (Bell, 2005). Leadership used to be associated with masculinity and authority (Bell, 2005; Billing & Alvesson, 2000; Koenig, Eagly, & Mitchell, 2011; Paustian-Underdahl, Walker, & Woehr, 2014). Corresponding characteristics such as independence, control, rationality and objectivity are stereotypical for leaders, and displaying vulnerability does not belong to that list. As a result,

(12)

people tend to have a rather negative view towards displaying vulnerability, which might have caused leaders to be reluctant about doing it (Brown, 2012b, 2015; Bunker, 1997). Yet, a growing body of research argues that leader vulnerability has a lot of potential for effective leadership (Bell, 2005; Brown, 2012b, 2015; Bunker, 1997; Chambers, 1989; Hirschhorn, 1990; Lapidot et al., 2007; Mayer et al., 1995; Nienaber et al., 2015; Tsui-Auch & Möllering, 2010). This ambiguous image of vulnerability makes it relevant to explore the potential of leader vulnerability in literature. This thesis takes a positive approach to leader vulnerability by looking at possible positive outcomes, and aims to create a more open view towards leader vulnerability and the potential of its application in practice. Contrary to the difficult to measure vulnerability as a state, leader vulnerability can be measured through a scale and allows this thesis to bring new insights into literature by taking a quantitative approach.

Although there is a lack of empirical research on leader vulnerability as a topic on itself, this thesis based its expectations on related literature, such as literature on authentic leadership that frequently discusses displaying vulnerability as an important characteristic of authentic leaders (Agote, Aramburu, & Lines, 2016; Avolio et al., 2004; Ladkin & Taylor, 2010; Shamir & Eilam, 2005). Authentic leaders are characterized by four attributes. First, self-awareness, because authentic leaders are aware of their own values, beliefs, and identity (Diddams & Chang, 2012; Gardner et al., 2005; Ladkin & Taylor, 2010; Walumbwa et al., 2008). The second attribute is balanced processing, which refers to rationality, because authentic leaders objectively analyze information and data before they make decisions. Relational transparency, third, means that authentic leaders display openness, self-disclosure and trust in relationships. They are not afraid to openly share information and feelings. Finally, the internalized moral perspective refers to the ability of authentic leaders to let their true selves, their own internal moral standards, regulate their behavior. Leader vulnerability, as mentioned before, means that

(13)

leaders are aware of their vulnerabilities (i.e. self-awareness) and that they openly discuss their weaknesses and insecurities with their followers (i.e. relational transparency). Although leader vulnerability is part of authentic leadership, it is not limited to it. Where leader vulnerability mainly focuses on leaders displaying their weaknesses and insecurities, authentic leadership implies that leaders expose their true selves, not only by showing their weaknesses, but also by showing their strengths, values and beliefs. Even though authentic leaders need to be aware of their vulnerabilities and openly discuss them, they also need to adjust their behavior to their own values and beliefs and make rational choices. Moreover, any leader could display vulnerability, despite his or her leadership style.

So, leader vulnerability is seen as consciously chosen behavior. By displaying vulnerability, leaders, who act as role models in organizations, create an open climate in which mistakes and insecurities can be openly discussed and are widely accepted (Avolio et al., 2004; Gardner et al., 2005; Shamir & Eilam, 2005). By displaying vulnerability, leaders open up to their followers and invite them to do the same (Gardner et al., 2005; Shamir & Eilam, 2005). As a result, both the leader and followers will experience higher levels of congruency between actions and beliefs, which is likely to positively affect the relationship between them, as will be discussed in the following part.

2.2 Trust in the Leader

In literature on trust, vulnerability is a frequently recurring topic (Avolio et al., 2004; Bhattacherjee, 2002; Dirks & Ferrin, 2002; Lapidot et al., 2007; Mayer et al., 1995; Nienaber et al., 2015; Norman et al., 2010; Rousseau, Sitkin, Burt, & Camerer, 1998). A widely used definition of trust is given by Rousseau et al. (1998) who based their definition of trust on earlier work of Mayer et al. (1995): “trust is a psychological state comprising the intention to accept

(14)

vulnerability based upon positive expectations of the intentions or behavior of another”

(Rousseau et al., 1998, p. 395). In other words: “the willingness to be vulnerable” (Rousseau et al., 1998, p. 394). By displaying vulnerability, leaders show their followers that they dare to open up to them. By doing that, leaders signal a sign of trust to their followers which followers are likely to reciprocate (Gardner et al., 2005; Leroy, Palanski, & Simons, 2012; Shamir & Eilam, 2005). As a result, followers are expected to develop higher levels of trust in their leader. This suggests that leader vulnerability enforces follower’s trust in their leader. Followers’ trust in their leader is of high importance because leaders have the authority to make decisions that have an impact on their followers, it is therefore important that followers believe their leader to have the right intentions (Dirks & Ferrin, 2002). Followers’ trust in their leader was found to have different positive effects on them such as job satisfaction, job commitment and low intention to quit (Agote et al., 2016; Avolio et al., 2004; Burke et al., 2007; Dirks & Ferrin, 2002).

Trust in the leader can be categorized in two dimensions: cognitive trust and affective trust (Bhattacherjee, 2002; Burke et al., 2007; Dirks & Ferrin, 2002; Gillespie & Mann, 2004; Norman et al., 2010). Where affective trust reflects a special relationship based on concern for one’s welfare and emotions, cognitive trust refers to one’s trustworthiness and issues such as reliability, integrity, and honesty. This thesis decided to focus on the cognitive aspect of trust as it is expected that leader vulnerability influences the way people view a leader’s character rather than being an investment in relationships between leaders and followers. Mayer et al. (1995) found 3 factors that mainly determine the cognitive trust in a person, in this case the leader: integrity, benevolence, and competence (Avolio et al., 2004; Bhattacherjee, 2002; Burke et al., 2007; Mayer et al., 1995; Nienaber et al., 2015; Norman et al., 2010).

(15)

First, in order to develop trust in their leader, followers should perceive their leader as integer (Avolio et al., 2004; Bhattacherjee, 2002; Burke et al., 2007; Lapidot et al., 2007; Mayer et al., 1995; Nienaber et al., 2015; Norman et al., 2010). Integrity means that the leader complies to a set of principles that the follower finds acceptable (Lapidot et al., 2007; Mayer et al., 1995). Followers will perceive their leader as integer if they feel that their leader is honest, tells the truth, and does not break promises (Lapidot et al., 2007). Leaders who display vulnerability tend to be perceived as open and honest (Bell, 2005; Bunker, 1997; Chambers, 1989; Hirschhorn, 1990; Lapidot et al., 2007; Mayer et al., 1995; Nienaber et al., 2015; Tsui-Auch & Möllering, 2010). They share critical information about their feelings and perceptions with their followers and encourage their followers to do the same, leading to social and personal identification of followers with their leader, and consequently trust in the leader (Agote et al., 2016; Avolio et al., 2004; Leroy et al., 2012; Norman et al., 2010). Therefore, leader vulnerability is expected to have a positive impact on followers’ perceived integrity of their leader.

Second, trust in a leader can be associated with perceptions of a leader’s benevolence, the extent to which a leader’s intentions to the follower are believed to be good (Avolio et al., 2004; Bhattacherjee, 2002; Burke et al., 2007; Lapidot et al., 2007; Mayer et al., 1995; Nienaber et al., 2015; Norman et al., 2010; Wang, Chiang, Tsai, Lin, & Cheng, 2013). If followers perceive their leaders to be benevolent, they are likely to be more motivated, work harder and engage in extra-role behaviors (Burke et al., 2007). By acknowledging their weaknesses and insecurities, leaders can react consistent in different situations and will create transparency about their actions by their followers (Diddams & Chang, 2012). Although it cannot be excluded that some followers might perceive their leader’s intentions to be bad, it is likely that by being open and engaging their followers, leaders who display vulnerability are perceived to be benevolent by

(16)

their followers (Avolio et al., 2004). Moreover, leaders who display vulnerability will ask their followers for help or advice if necessary (Bell, 2005). By consulting their followers, their followers will perceive more autonomy and feel their voice is valued (Burke et al., 2007; Gillespie & Mann, 2004). According to Burke et al. (2007), this increases followers’ perceptions of leader benevolence. Therefore, leader vulnerability is likely to engender perceptions of leader benevolence.

The final factor that contributes to followers’ trust in their leader is the leader’s competence (Avolio et al., 2004; Bhattacherjee, 2002; Burke et al., 2007; Lapidot et al., 2007; Mayer et al., 1995; Nienaber et al., 2015; Norman et al., 2010). The competence of a leader is the extent to which he or she has skills, abilities and characteristics with which he or she can exert influence in a certain domain (Lapidot et al., 2007; Mayer et al., 1995). A leader can be trusted in one area, but might not be trusted in other areas that he or she has little knowledge or skills in. A leader’s competence is not only related to his or her knowledge about a certain domain, which leader vulnerability is not likely to affect as it is not expected to relate to technical knowledge, but also to (interpersonal) leadership skills such as communication (Butler, 1991; Lapidot et al., 2007). It is expected that leader vulnerability has a positive influence on a leader’s interpersonal competence, because a leader who displays vulnerability openly communicates with his or her followers, which is an important factor in interpersonal competence (Alimo‐Metcalfe, 1995; Bell, 2005; Lapidot et al., 2007). Besides, displaying vulnerability indicates a sufficient amount of self-confidence of leaders to do so in the first place. Hopkins and Bilimoria (2008) found this self-confidence to be an important characteristic of leader competency. In sum, although leader vulnerability might not influence a leader’s technical knowledge, it is expected to affect the interpersonal skills and self-confidence of a leader, and therefore followers’ perceptions of leader competence.

(17)

To conclude, leader vulnerability is likely to contribute to the perceived integrity, benevolence, and competence of a leader by his or her followers. With being completely transparent about one’s weaknesses and insecurities, an open climate can be created that stimulates followers’ trust in their leader. This suggests a positive relationship between leader vulnerability and trust in the leader, which leads to the following hypothesis:

Hypothesis 1. Leader vulnerability is positively related to trust in the leader.

2.3 Gender of the Leader

Although leader vulnerability is likely to be positively related to trust in the leader, this relationship may be moderated by the gender of the leader. Gender stereotyping is an often recurring topic in literature on leadership (Alimo‐Metcalfe, 1995, 2010; Billing & Alvesson, 2000; Eagly, 2005; Eagly & Karau, 2002; Koenig et al., 2011; Paustian-Underdahl et al., 2014). A well-known metaphor in the gender literature on leadership is “the glass ceiling”, which implies that even though women might have exactly the same background and education as men, they appear not to attain the highest positions in organizations (Billing & Alvesson, 2000; Eagly & Karau, 2002). A cause of this “glass ceiling” could be the traditional view on leadership. Traditionally, leadership was seen as a rather masculine position, with corresponding agentic characteristics such as autonomy, independence and control (Alimo‐ Metcalfe, 1995; Billing & Alvesson, 2000; Eagly & Carli, 2003; Eagly, Johannesen-Schmidt, & Van Engen, 2003; Eagly & Karau, 2002; Fletcher, 2004; Heilman, 2001). However, more recent studies also focus on the effectiveness of women in leadership positions in order to get rid of the widely held assumption that men fit better in leader positions than women. They state that communal, feminine, characteristics such as cooperation, acceptance, flexibility, empathy,

(18)

and vulnerability could also make an important contribution to effective leadership in contemporary organizations (Billing & Alvesson, 2000; Eagly & Carli, 2003; Eagly et al., 2003; Eagly & Karau, 2002).

Eagly and Karau (2002) investigated gender stereotyping with their research on role congruity theory of prejudice towards female leaders (Eagly, 2005; Eagly & Carli, 2003; Eagly et al., 2003; Eagly & Johannesen‐Schmidt, 2001; Koenig et al., 2011; Paustian-Underdahl et al., 2014). They found the prejudice to take two ways. First, female leaders used to be seen as incompetent leaders because of the dissimilarity between leadership expectations and female role expectations. This view of incompetency stems from descriptive norms and standards that were ascribed to gender. Competent leaders were considered to be competitive, self-confident, and ambitious; characteristics that are not in line with the typical communal characteristics that were ascribed to women. On the other hand, Eagly and Karau (2002) found that competent female leaders were often unfavorably evaluated because of the violation of female role behavior. This unfavorably evaluation finds its origin in injunctive norms that represent beliefs about how people, in this case women, should behave. If women showed agentic characteristics, such as autonomy and objectivity in their leadership, followers would see this as hostile and not rational, which was the same case for men showing communal characteristics, such as empathy and vulnerability. This prejudice about meeting expectations in behavior might explain different effects of leader vulnerability for male and female leaders.

Displaying vulnerability has been found to be a rather feminine characteristic (Billing & Alvesson, 2000). Linking this to the role congruity theory of Eagly and Karau (2002) would indicate that female leaders are, contrary to male leaders, expected to display vulnerability. Male leaders who display vulnerability could be perceived as violating their expected role behavior, while for female leaders this behavior is in line with their expected role behavior.

(19)

Because male leaders who display vulnerability do not act in line with the expectations of leader behavior that followers have, followers could perceive them as less integer compared to female leaders who display vulnerability.

Gender of the leader is also likely to influence the perceived benevolence of a leader. Once a leader is perceived to be benevolent, he or she is concerned about a follower’s wellbeing and his or her intentions are believed to be good (Avolio et al., 2004; Bhattacherjee, 2002; Burke et al., 2007; Lapidot et al., 2007; Mayer et al., 1995; Nienaber et al., 2015; Norman et al., 2010; Wang et al., 2013). Leader vulnerability is, as mentioned before, expected to influence the perceived benevolence of a leader, because leaders who display vulnerability are perceived as open and transparent. These are, together with caring important characteristics of benevolence (Burke et al., 2007; Wang et al., 2013). In line with Kelley’s (1972) attribution principles, if leaders display behavior that violates their gender role expectations, such as male leaders displaying vulnerability in this case, this behavior tents to be attributed to their intentions, while behavior that is in line with gender role expectations tends to be attributed to gender norms. As displaying vulnerability is incongruent with male leaders’ expected role behavior (Eagly & Karau, 2002), people could be more likely to doubt the intentions of male leaders who display vulnerability, compared to female leaders who display vulnerability. Therefore, female leaders who display vulnerability are expected to be perceived as more benevolent than male leaders who display vulnerability.

Brescoll, Dawson and Uhlmann (2010) argued that leaders who behave incongruent with their gender role behavior are more likely to be questioned on their competence than leaders who behave congruent with their gender role behavior. This results in the expectation that male leaders who display vulnerability, and thus violate gender role behavior, are more likely to be questioned on their competences when admitting their mistakes and weaknesses. Female

(20)

leaders who display vulnerability are less likely to be disadvantaged when admitting mistakes and weaknesses and will therefore score higher on competence than male leaders who display vulnerability. On top, contemporary organizations show more demand for communal leadership behaviors such as cooperation and empathy(Duehr & Bono, 2006; Eagly & Carli, 2003; Eagly et al., 2003; Eagly & Johannesen‐Schmidt, 2001; Paustian-Underdahl et al., 2014). Consequently, it can be assumed that perceptions of leader competence change. Because leader vulnerability is seen as a communal leadership behavior, which women are more likely to exhibit, it is expected that female leaders are perceived as more competent than male leaders. Taken together, it is expected that female leaders who display vulnerability are perceived as more competent than male leaders who display vulnerability.

To conclude, the gender of the leader is expected to influence the relationship between leader vulnerability and followers’ trust in their leader, leading to the following hypothesis:

Hypothesis 2. The gender of the leader moderates the relation between leader

vulnerability and trust in the leader, such that this relation will be stronger in case of a

female leader than in case of a male leader.

2.4 Perceived Leader Effectiveness

Dirks and Ferrin (2002) found having trust in one’s direct leader to have different desirable outcomes such as job satisfaction, organizational citizenship behavior, less intention to quit, organizational commitment, and job performance (Avolio et al., 2004; Burke et al., 2007; Connell, Ferres, & Travaglione, 2003; Dirks & Ferrin, 2002; Gillespie & Mann, 2004; Norman et al., 2010; Podsakoff, Mackenzie, Moorman, & Fetter, 1990). A commonly used factor that contributes to organizational effectiveness, and other desirable outcomes, is leader

(21)

effectiveness. Leader effectiveness is a widely researched topic that focuses on the contribution of a leader to organizational goals and outcomes, and refers to the extent to which a leader is able to influence his or her followers (Hogan et al., 1994; Norman et al., 2010; Paustian-Underdahl et al., 2014). Leader effectiveness could be measured by looking at a leader’s impact on organizational outcomes, such as profitability and market share (Hogan et al., 1994; Yukl, 2012). However, the exact impact of a leader on these outcomes is hard to measure because these factors are also influenced by external factors that leaders can’t influence. Therefore, this thesis focuses on subjective measures of leader effectiveness, also: perceived leader effectiveness, as rated by followers.

Although there are no universal traits of effective leaders, trust was found to be the most important characteristic in followers’ evaluations of leader effectiveness (Hogan et al., 1994; Judge et al., 2002). Different authors already showed a link between trust in the leader and perceived leader effectiveness (e.g. Avolio et al., 2004; Burke et al., 2007; Dirks & Ferrin, 2002; Gillespie & Mann, 2004; Judge et al., 2002; Knippenberg & Knippenberg, 2005; Norman et al., 2010). For instance, Dirks and Ferrin (2002) found having trust in one’s leader to lead to followers’ commitment to their leader’s decisions and to the extent in which they believe in the accuracy of information their leader provides, two factors that they also found to be important for followers’ perceptions of leader effectiveness. Furthermore, Van Knippenberg and Van Knippenberg (2005) argued that followers who trust their leader are more likely to positively evaluate their leader, because they feel their leader is concerned about their interests.

In line with the three subcomponents of trust in the leader (i.e. benevolence, competence, and integrity), different links to perceived leader effectiveness can be made. Hogan et al. (1994) found that, especially, integrity played an important role in followers’ ratings of the effectiveness of their leader, as integrity emerges from the way in which followers perceive

(22)

their leader as consistent and moral (Moorman, Darnold, & Priesemuth, 2013). Besides, leaders need to be perceived as integer by their followers in order to build and maintain relations with them (Kirkpatrick & Locke, 1991). Although Hogan et al. (1994) ascribed competence as a more important factor in bosses’ ratings about management, Burke et al. (2007) found that setting a compelling direction and enabling structure, which are two important markers of leader competence, often were viewed as effective leader behavior by followers. Further, interpersonal communication, which leads to perceptions of leader competence, was found to be an important predictor of perceived leader effectiveness (Riggio & Reichard, 2008; Riggio, Salinas, Riggio, & Cole, 2003). Benevolence, finally, was also found to lead to better evaluations of leader effectiveness (Niu, Wang, & Cheng, 2009). When followers perceive the intentions of their leader to be good, they tend to reciprocate this behavior and develop positive perceptions of their leader. On top, showing support for employees’ feelings and asking them for their opinions, two characteristics of benevolent leaders, were found to have an important influence on followers’ perceptions of leader effectiveness, because followers develop feelings of empowerment (Yukl, 2012).

Altogether, the link between (the subcomponents of) trust in the leader to the antecedents of perceived leader effectiveness, combined with existing research that already proved a link between trust in the leader and perceived leader effectiveness, suggests a positive relation between the two concepts. This results in the development of the following hypothesis:

(23)

2.5 Conceptual Research Model

Concluding, leaders who display vulnerability are open and honest about their weaknesses and insecurities (Bell, 2005; Bunker, 1997; Chambers, 1989; Hirschhorn, 1990; Lapidot et al., 2007; Mayer et al., 1995; Nienaber et al., 2015; Tsui-Auch & Möllering, 2010). As a result, they create an open climate in which their followers know what to expect from their leaders. In this way, followers are likely to perceive their leader as integer, benevolent, and competent, which are important components of followers’ trust in their leader (Avolio et al., 2004; Bhattacherjee, 2002; Burke et al., 2007; Mayer et al., 1995; Nienaber et al., 2015; Norman et al., 2010). At the same time, followers’ trust in their leader is an important prerequisite for their perceptions of their leader’s effectiveness (Hogan et al., 1994). Therefore, trust in the leader is expected to mediate the relation between leader vulnerability and perceived leader effectiveness. However, it is plausible that gender stereotypes influence this relation because they might affect followers’ expectations about their leaders. Since leader vulnerability is seen as a rather feminine characteristic, it is expected that male leaders who display vulnerability tend to be seen as less trustworthy and effective than female leaders who display vulnerability, because they do not meet their gender role expectations (Billing & Alvesson, 2000; Eagly & Karau, 2002). Combining these insights, the conceptual model in figure 1 was developed with the corresponding hypothesis:

Hypothesis 4. There is a moderated mediation effect of leader vulnerability and gender of

the leader on perceived leader effectiveness through trust in the leader, such that this

(24)

3 Method

3.1 Procedure and Sample

For this thesis, an explanatory research was carried out by using a quantitative approach. In order to collect data to test the hypothesis, two self-administered surveys were designed, one for leaders and one for their subordinates. All surveys were in Dutch, because this research was conducted within the Netherlands and the surveys were distributed amongst Dutch managers and subordinates that work at least three days a week. Due to limited time and budget available, a non-probability convenience sampling technique was used to collect cross-sectional data. With this technique participants that are easily accessible, available at a given time, and willing to participate are recruited (Etikan, Musa, & Alkassim, 2016). Although convenience sampling is often more likely to be biased than other types of sampling techniques, this effect was reduced because data was collected in a network of five students of the University of Amsterdam. A total of 160 leader-follower dyads was approached, of which in most cases either the leader or subordinate was approached initially. Once they admitted to participate, an E-mail was send Figure 1. Conceptual Research Model

(25)

with both a link to the survey and an unique code to match the leader and subordinate surveys. The approached participants on their turn forwarded the e-mail to their respective subordinate or leader. To increase the response rate, the participants were send a reminder, and a small incentive was raffled to every ten participants that provided their E-mail address.

Of the 160 approached dyads, 136 complete dyads and 12 incomplete dyads filled in the survey. Only all complete leader-follower dyads were filtered, resulting in a total response rate of 85%. All 136 dyads could be used for testing, because no answers were missing and no outliers were detected. Among the leaders, 49.3% was male. Their age ranged between 22 and 69 years (M = 43.4, SD = 11.56) and tenure between 2 months and 40 years (M = 79.26, SD = 93.19). Of the follower respondents the gender distribution was a bit less balanced, 39.7% was male. The age of the followers ranged between 18 and 66 years (M = 34.42, SD = 12.17) and their tenure between 2 months and over 41 years (M = 66.26, SD = 83.18). Both leaders (78,7%) and followers (70.6%) had higher education degrees. The amount of time the leaders and followers worked together ranged from 2 moths to over 25 years (M = 34.77, SD = 48.09). The respondents worked in different industries, of which health care (19.9%), retail (16.9%), and transport and logistics (14,7%) were most common.

3.2 Measures

The variables leader vulnerability, trust in the leader and perceived leader effectiveness were all measured on a 7-point Likert scale (completely disagree-completely agree). All items were translated to Dutch, using back-translation, because the surveys were spread on the Dutch labor market. All variables (i.e. leader vulnerability, trust in the leader, and perceived leader effectiveness) were follower rated, because the followers’ perceptions of all variables were

(26)

perceived as important for measuring them. The demographics gender of the follower, industry, age of the leader and education of the follower were used as control variables.

Gender of the leader. This variable is distracted from the demographics in the leader survey. With 49.3% male-leader and 50.7% female-leader responses it can be concluded that gender of the leader was equally distributed.

Leader vulnerability. Leader vulnerability was measured using the 7-item scale of Van Bunderen (2018). Example items include “My leader never shows a sign of weakness” and “My leader shows his/her vulnerable side”. The Cronbach’s alpha indicated sufficient internal reliability of the scale (α = .88).

Trust in the leader. Trust in the leader was measured using Blais and Thompson’s (2009) trust in leaders scale, which is an adjusted version of Adams and Sartori’s trust in leaders scale (2006). The scale consists of three subscales (i.e. benevolence, competence and integrity) each consisting of seven items1 (e.g., “My leader shows respect for me as a person”, “I have confidence in the abilities of my leader” and “I believe my leader is honest”). The corresponding Cronbach’s alphas of the subscales are respectively .86, .88, and .90, exhibiting sufficient reliability. By taking all subscales together, the trust in the leader scale was developed. This scale proved to be highly reliable (α = .93).

Perceived leader effectiveness. Perceived leader effectiveness was measured using a 7-item scale adjusted from Van Knippenberg and Van Knippenberg’s (2005) perceived leader effectiveness scale (e.g., “My leader is very effective as a leader” and “My leader leads in a

1 A factor analysis was performed to check validity of the trust in the leader scale (Table 6, Appendix 3). The three

factor model showed a deviation in the first two items of the subcomponent benevolence (i.e. “My leader shows respect for me as a person” and “I have confidence in the motivations of my leader”). These items load higher on respectively integrity and competence. Although the results show some problems in the validity of the benevolence

(27)

way that motivates people”). An overview of all seven items can be found in Appendix 1. The Cronbach’s alpha revealed a high internal consistency for the scale (α = .94).

Control variables. In order to rule out other possible effects on the hypothesis, some control variables were taken into account for this research. First the gender of the follower might have an influence on the expected model. Alimo‐Metcalfe (1995) found that both males and females have different perceptions of characteristics that leader should possess. This suggests that men and women could have different views on the effectiveness of leaders who display vulnerability. Also, the industry that people work in could have an effect on the model. For example healthcare is seen as a more feminine industry. Leader vulnerability might be seen as effective in a feminine industry, while being seen as the opposite in a masculine industry (Eagly, 2005; Eagly & Carli, 2003; Eagly et al., 2003; Eagly & Karau, 2002; Garcia-Retamero & López-Zafra, 2006; Heilman, 2001; Hopkins & Bilimoria, 2008; Koenig et al., 2011; Paustian-Underdahl et al., 2014). As is common in leadership research, also age of the leader and education of the follower were taken into account as control variables.

3.3 Data Analysis

Before testing the data, the data was checked on normality. The Kolmogorov-Smirnov and Shapiro-Wilk test indicated non-normality of all variables (i.e. leader vulnerability, trust in the leader and perceived leader effectiveness), since there were no non-significant values. However, the skewness and kurtosis of all variables lay between -2 and 2, as can be seen in Table 5 (Appendix 2), which are found to be acceptable values to assume normality of the sample (Field, 2009; Garson, 2012). Also, the sample size of this thesis (N=136) was large enough to assume normality according to the Central Limit Theorem.

(28)

4 Results

4.1 Descriptive Statistics

A first impression of the data was gained by looking at the descriptive statistics of the variables. The means, standard deviations and correlations of the data are presented in Table 1. To test the hypothesis, hierarchical regression analysis were performed. First, the independent variables were mean-centered to reduce multicollinearity between the independent variable and the interaction term (Hayes, 2012). Also, gender of the leader was recoded into a dummy variable.

Table 1. Means, Standard Deviations and Correlations

Variables M SD 1 2 3 4 5 6 7 8

1. Age of the Leader 43.40 11.56 - 2. Gender of the Follower 1.60 .49 -.08 -

3. Industry 4.87 2.97 .01 -.19* -

4. Education of the

Follower 5.82 1.14 .14 -.03 .01 -

5. Leader Vulnerability 4.67 1.11 -.03 .00 -.10 .26** (.88) 6. Trust in the Leader 5.65 .67 -.11 .05 -.16 .05 .20* (.93) 7. Gender of the Leader 1.51 .50 -.21* .28** -.29** -.16 .02 .04 - 8. Perceived Leader

Effectiveness 5.46 1.01 -.07 .06 -.13 -.05 .18* .85** .08 (.94)

Note. N = 136. * p < .05, ** p < .01. All tests are two-tailed. Internal consistency reliabilities

(29)

4.2 Hypothesis Testing

In Hypothesis 1, it was proposed that leader vulnerability would be positively related to trust in the leader. This hypothesis was supported by a regression analysis of which the outcomes can be found in Table 2 (β = .19, t = 2.09, p < .05, Adj. R2 = .02). Therefore, it can be concluded that leader vulnerability is positively related to trust in the leader, such that if leader vulnerability increases with one, trust in the leader increases with .19.

Table 2. Results of Regression Analysis

Trust in the leader Perceived Leader Effectiveness

Variable Step 1 Step 2 Step 3 Step 1 Step 2 Step 3 Step 4 Control variables

Age of the Leader -.11 -.10 -.10 -.06 -.05 -.05 .04

Gender of the Follower .03 .03 .03 .03 .03 .03 -.003 Education of the Follower .06 .01 .001 -.04 -.08 -.09 -.09

Industry -.11 -.09 -.08 -.13 -.10 -.10 -.03

Leader Vulnerability .19* .19* .19* .19* .02

Gender of the Leader -.02 -.01 .02 .02 .03

Gender of the Leader*Leader

Vulnerability .07 .03 -.03

Trust in the Leader .85**

R .17 .25 .26 .15 .24 .24 .86

Adjusted R2 -.001 .02 .01 -.01 .01 .01 .72

Change in R2 .03 .03 .01 .02 .03 .001 .68**

Overall F .95 1.38 1.27 .79 1.26 1.09 43.66**

df 4, 131 6, 129 7, 128 4, 131 6, 129 7, 128 8, 127

(30)

In Hypothesis 2, it was proposed that gender of the leader would moderate the relationship between leader vulnerability and trust in the leader. A regression analysis did not show support for this hypothesis, which therefore had to be rejected (β = .07, t = .83, p = .41, Adj. R2 = .01; see Table 2). Consequently, the relationship between leader vulnerability and trust in the leader was not found to be stronger for female leaders than for male leaders.

Hypothesis 3 proposed that trust in the leader was positively related to perceived leader effectiveness. The outcomes of the performed regression analysis, which are shown in Table 2, show support for the hypothesis (β = .85, t = 17.96, p < .001, Adj. R2 = .72). It can be concluded that trust in the leader is positively related to perceived leader effectiveness, such that if trust in the leader increases with one, perceived leader effectiveness increases with .85.

In Hypothesis 4, finally, a moderated mediation model was proposed, such that leader vulnerability was positively related to perceived leader effectiveness through trust in the leader. This positive effect would be stronger for female leaders than for male leaders. The rejection of Hypothesis 2 resulted automatically in the rejection of Hypothesis 4. Yet, the moderated mediation model was tested in order to gain insights in the indirect effects of the model. For this analysis, PROCESS macro (Model 7) was used with 5000 bootstrapped samples (Preacher & Hayes, 2004). The indirect effects did not show any significant results for both male leaders (b = .09, Bias and accelerated 95% CI: - 0.07, .26) and female leaders (b = .20, Bias and accelerated 95% CI: - 0.01, .44), as can be seen in Table 3. Consequently, there was no moderated mediation effect of leader vulnerability and gender of the leader on perceived leader effectiveness through trust in the leader.

(31)

Table 3. Indirect Effects of Moderated Mediation Analysis

Gender of the Leader Unstandardized

Boot Effects Boot SE Boot LLCI Boot ULCI

Conditional indirect effects of X on Y

Male .09 .09 -.07 .26

Female .20 .12 -.01 .44

Note. N = 136. PROCESS model 7 was used.

4.3 Exploratory Analysis

With only direct effects of leader vulnerability on trust in the leader and trust in the leader on perceived leader effectiveness proven, some additional analysis were performed. First, a regression analysis was performed to see whether leader vulnerability has a direct effect on perceived leader effectiveness. The regression analysis showed a significant positive effect of leader vulnerability on perceived leader effectiveness when controlled for age of the leader, gender of the follower, education of the follower and industry (β = .19, t = 2.07, p < .05, Adj.

R2 = .01; see Table 2). Interpreting this outcome, an increase of leader vulnerability with one, leads to an increase of perceived leader effectiveness with .19. Therefore, leader vulnerability is positively related to perceived leader effectiveness.

With the results of all three regression analysis in mind, PROCESS macro (Model 4) with 5000 bootstrapped samples at a confidence level of 95% was used to see whether the effect between leader vulnerability and perceived leader effectiveness was explained through trust in the leader (Preacher & Hayes, 2004). A bias-corrected bootstrap confidence interval showed a significant indirect effect when controlled for age of the leader, gender of the follower, education of the follower and industry (Indirect Effect = .14, SE = .07; see Table 4), as the interval excludes zero (CI: .01, .29). So, followers who perceive their leader to display vulnerability, tend to develop more trust in their leader, leading on its turn to better perceptions

(32)

of leader effectiveness. The model showed no significant direct effect of leader vulnerability on perceived leader effectiveness (Effect = .03, SE = .04, CI: -.06, .11; see Table 4), but does show a significant total effect (Effect = .17, SE = .08, CI: .01, .33; see Table 4), indicating a full mediation. So, trust in the leader fully mediates the relationship between leader vulnerability and perceived leader effectiveness.

Table 4. Mediation Effects of Leader Vulnerability on Perceived Leader Effectiveness through Trust in the Leader

Effect SE p LLCI ULCI

Direct Effect .03 .04 .57 -.06 .11

Total Effect .17 .08 .04 .01 .33

Boot SE Boot LLCI Boot ULCI

Indirect Effect .14 .07 .01 .29

Note. N = 136. PROCESS model 4 was used.

Finally, additional analysis were performed to explore the relationship between leader vulnerability and the subcomponents of trust in the leader, and between the subcomponents of trust in the leader and perceived leader effectiveness. Leader vulnerability only showed a significant positive relationship with benevolence when controlled for age of the leader, gender of the follower, education of the follower and industry (β = .26, t = 3.03, p < .01, Adj. R2 = .07). This indicates that when perceptions of leader vulnerability increase with one, perceptions of leader benevolence increase with .26. No significant effects were found for the effects of leader vulnerability on both competence (β = .06, t = .68, p = .50, Adj. R2 = -.02) and integrity (β = .17, t = 1.89, p = .06, Adj. R2 = .02). All three subcomponents of trust in the leader, namely benevolence (β = .68, t = 10.47, p < .001, Adj. R2 = .45), competence (β = .83, t = 17.13, p < .001, Adj. R2 = .69), and integrity (β = .65, t = 9.55, p < .001, Adj. R2 = .40); were found to be

(33)

positively related to perceived leader effectiveness. Indicating that increases in benevolence, competence, and integrity lead to higher perceptions of leader effectiveness.

5 Discussion

With leader vulnerability as an upcoming topic in literature, its potential for leadership has not been explored yet. Although vulnerability was typically viewed as a weakness, this thesis tries to bring new insights into literature by taking a positive approach to leader vulnerability, linking it to trust in the leader and perceived leader effectiveness (Brown, 2012b, 2015; Bunker, 1997). In addition, gender of the leader was added as a variable to the research model to see if leader vulnerability has different effects on followers’ trust in their leader for male and female leaders. This chapter will discuss the results of the performed analysis. Further, this chapter will elaborate on the limitations of this study, the managerial implications, and directions for future research.

5.1 Discussion of the Results

As proposed, leader vulnerability showed to be positively related to trust in the leader. In line with the expectations that were made based on related literature, followers tend to develop trust in their leaders when their leaders expose their weaknesses and insecurities (Avolio et al., 2004; Bunker, 1997; Gardner et al., 2005; Ladkin & Taylor, 2010; Lapidot et al., 2007; Nienaber et al., 2015; Tsui-Auch & Möllering, 2010; Walumbwa et al., 2008). Additional analysis on the relation between leader vulnerability and the subcomponents of trust in the leader only found leader vulnerability to be positively related to benevolence. An explanation for the insignificant relation between leader vulnerability and competence could be that when leaders display vulnerability, this raises followers’ doubts about the knowledge and abilities of

(34)

their leader (Lapidot et al., 2007). Besides, although leaders who display vulnerability are open and transparent about their weaknesses and insecurities, other factors such as their norms, values, and motives for certain behavior and decisions might be of more importance for followers perceptions of leader integrity and competence (Agote et al., 2016; Avolio et al., 2004; Burke et al., 2007; Mayer et al., 1995; Norman et al., 2010). Also, when leaders display vulnerability, but do not act in line with the expectations followers develop based on that, followers might not perceive their leader as integer (Leroy et al., 2012). Therefore, followers’ perceptions about the competence and integrity of their leader might be mainly dependent on other factors than leader vulnerability, explaining the insignificant outcomes of the regression analysis. Yet, the significant positive relation between leader vulnerability and benevolence shows to be strong enough to lead to a significant positive relation between leader vulnerability and trust in the leader.

In line with the proposition that was made based on existing literature, trust in the leader was found to be positively related to perceived leader effectiveness (Avolio et al., 2004; Burke et al., 2007; Dirks & Ferrin, 2002; Gillespie & Mann, 2004; Judge et al., 2002; Norman et al., 2010). All three subcomponents of trust in the leader (i.e. benevolence, competence and integrity) showed, as expected, positive relations with perceived leader effectiveness as well. Additional analysis were performed to show whether leader vulnerability was related to perceived leader effectiveness and if trust in the leader mediated this relationship. A regression analysis proved the former, indicating that leader vulnerability is positively related to perceived leader effectiveness. Using PROCESS macro it was shown that trust in the leader fully mediated the direct relationship between leader vulnerability and perceived leader effectiveness, implying that once trust in the leader was added to the model, the direct relationship between leader vulnerability and perceived leader effectiveness disappeared.

(35)

This mediation effect was not completely unexpected as related literature (e.g. on authentic leadership, consultative leadership and humility) often made links with leader effectiveness through trust. For instance, literature that studied the relation between authentic leadership and leader effectiveness, linked “exposing one’s true self” to followers’ perceptions of genuine leader actions (Avolio & Gardner, 2005). These genuine actions led to followers’ beliefs in the integrity and benevolence of their leader, which in turn is related to trust in their leader. Moreover, Avolio et al. (2004) and Agote et al. (2016) found that when a leader is transparent, this will deepen the level of trust between the leader and his or her followers, which on its turn enhances the perceptions of leader effectiveness (Norman et al., 2010). Literature on consultative leadership argued that by asking team members for their input, which leaders who display vulnerability tend do when they have doubts on a certain issue, followers feel their opinion is valued an appreciated (Burke et al., 2007; Gillespie & Mann, 2004). The resulting felt autonomy by followers explains increased trust of followers in their leaders. Finally, research on humility found humble behavior to lead to perceptions of leader effectiveness through supportiveness, socialized power and participative leadership (Morris, Brotheridge, & Urbanski, 2005; Nielsen, Marrone, & Slay, 2010; Owens & Herman, 2012; Owens, Johnson, & Mitchell, 2013; Owens, Rowatt, & Wilkins, 2011; Weick, 2001). For instance, Morris et al. (2005) found that leaders’ support for follower well-being (i.e. supportiveness) and involving followers in decision making (i.e. participative leadership), which suggests a link with leader benevolence and thus trust in the leader, led to positive organizational outcomes. On top, Weick (2001) argued that leaders who admitted their shortcomings, compared to leaders who did not, were often perceived as stronger and more effective, as it fostered climates of learning and trust.

It was proposed that gender stereotyping would influence the relation between leader vulnerability and trust in the leader. Yet, the insignificant findings did not support this

(36)

proposition, thus, the relation between leader vulnerability and trust in the leader is not stronger in case of a female leader. This outcome could be explained by the changing demands for leadership styles. Effective leadership used to be associated with masculine qualities such as dominance, independence, confidence and competitiveness (Eagly & Johannesen‐Schmidt, 2001). Since women were seen as more communal, they were not seen as effective leaders. However, contemporary organizations showed increasing demands for more communal, transformational, leader characteristics such as cooperation, mentoring, and collaboration; characteristics that are typically ascribed to women (Duehr & Bono, 2006; Eagly & Carli, 2003; Eagly et al., 2003; Eagly & Johannesen‐Schmidt, 2001; Paustian-Underdahl et al., 2014). Although this suggests that female leaders would be seen as more effective leaders, research found that both males and females who showed communal characteristics were seen as almost equally effective (Duehr & Bono, 2006; Eagly et al., 2003). If followers would base their evaluations of leaders less on expectations based on gender, and more on expectations based on desirable behavior, this could explain the insignificant effect of gender of the leader on the model. On top, in Western cultures gender stereotyping was found to have less effect than in Eastern cultures (Wang et al., 2013). The sample of this thesis was the Dutch – a typically Western country – labor market, which could also be an explanation of why gender of the leader did not have a significant effect on the research model.

Another explanation for the insignificant moderation effect of gender of the leader was given by Wang et al. (2013) who argued that the role congruity theory of Eagly and Karau (2002) does not apply to leaders whose behavior positively deviates from role expectations (Anderson, Lievens, Van Dam, & Born, 2006; Johnson, Murphy, Zewdie, & Reichard, 2008). Wang et al. (2013) based their findings on earlier studies of Anderson et al. (2006) and Johnson et al. (2008) who found that leaders whose behavior positively deviated from their expected

(37)

gender role behaviors received better evaluations than leaders whose behavior negatively deviated from their gender role expectations. In other words, females who show agentic behaviors are likely to be evaluated negatively, but males displaying communal characteristics, such as leader vulnerability, are likely to be evaluated favorably. In case the leader’s behavior positively deviates from gender role expectations, followers will have positive perceptions about their leader’s intentions (Anderson et al., 2006; Johnson et al., 2008; Wang et al., 2013). In this case, female leaders who display vulnerability will be positively evaluated because they behave in line with the gender norms, while male leaders who display vulnerability will be positively evaluated because their intentions are believed to be good. If both male and female leaders who display vulnerability will be evaluated positively, this might explain the absent moderation of gender on the relation between leader vulnerability and trust in the leader.

Altogether, this thesis made several contributions to existing literature. First, literature on leader vulnerability was extended by showing its link to trust in the leader and perceived leader effectiveness. The full mediation showed that trust in the leader is an important outcome of leader vulnerability, and explains the positive relation between leader vulnerability and perceived leader effectiveness. Second, related literature (e.g. authentic leadership, consultative leadership and humility), that discussed effects of leaders who showed their weaknesses and insecurities, took a more general view on transparency and openness of leaders, more explicit, not only transparency about one’s vulnerable side, but also about one’s values, beliefs, intentions and general information (Agote et al., 2016; Avolio & Gardner, 2005; Avolio et al., 2004; Burke et al., 2007; Gillespie & Mann, 2004; Morris et al., 2005; Nielsen et al., 2010; Norman et al., 2010; Owens & Herman, 2012; Owens et al., 2011, 2013; Weick, 2001). This thesis contributes to this related literature by showing the effects of leader vulnerability as a specific behavior. Finally, this thesis found no support for effects of gender stereotyping on the

(38)

research model. This extends more recent research on gender stereotyping that suggests that – especially in Western cultures – the effects of gender stereotyping are diminishing (Ayman & Korabik, 2010; Duehr & Bono, 2006; Eagly et al., 2003).

5.2 Limitations and Future Research Directions

This thesis has several limitations that inspire possible directions for future research. To start, this thesis measured all variables (i.e. leader vulnerability, trust in the leader, and perceived leader effectiveness) as rated by followers. Both the variable names trust in the leader and perceived leader effectiveness already indicate the importance of the followers perceptions of the scales. Also, it was decided that followers’ perceptions of leader vulnerability were of importance for this research, because leaders might be biased by rating themselves on leader vulnerability. For instance if leaders have a negative view towards leader vulnerability, they might not rationally rate themselves on the scale. The existing common source bias might affect the accuracy of the findings by inflating correlations (Spector, 2006), although some authors argued that the problem of a common source bias is overstated (Crampton & Wagner lll, 1994; Lindell & Whitney, 2001; Spector, 1987, 1994). Future research could avoid a common source bias by measuring leader effectiveness objectively through profitability and market share (Hogan et al., 1994; Judge et al., 2002; Yukl, 2012), or could explore the difference in outcomes if leader vulnerability is measured as rated by leaders.

A second limitation is related to the trust in the leader scale. The factor analysis showed that the first two items of benevolence (i.e. “my leader shows respect for me as a person” and “I have confidence in the motivations of my leader”) loaded higher on respectively the integrity and competence factor (Table 6, Appendix 4). Blais and Thompson (2009) also found some problems with the benevolence scale. They even found only the third, fifth and seventh item of

(39)

the benevolence scale to load on the benevolence factor. Although it was decided not to remove any items due to the high reliability of the scale, future research can revise the benevolence scale in order avoid non-validity of the scale. Added to that, the trust in the leader scale showed a high significant correlation (r = .85) with the perceived leader effectiveness scale, indicating that these scales are quite identical and, therefore, might measure the same construct (Garson, 2012). Taking different scales for trust in the leader and/or perceived leader effectiveness could explore whether this high correlation was caused by the scales used in this thesis, or could further explore the identicalness between the variables.

Third, this thesis has some limitations that can be attributed to the sampling technique and research design. Due to the limited time and budget available, a non-probability convenience sampling technique was used. As a result, the outcomes of this thesis cannot be generalized to the entire population (Etikan et al., 2016). Even though it has been tried to reduce this effect by collecting the data with four other students, future research could avoid this limitation by using random sampling techniques. For the same reasons as for which the choice for the sampling technique was made, this thesis decided to use cross sectional data. To be able to establish causal links between the variables, future research should use an experimental design (Bono & McNamara, 2011). Also, although anonymity and confidentiality of the survey responses were guaranteed, a social desirability bias might exist, because followers might have answered the survey in favor of their leader (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). This bias might be diminished if leaders and followers are approached separately and are not aware of each other’s participation.

Further, it was found that leader vulnerability leads to trust in the leader, but this thesis has not explored whether this relation also works the other way around, in other words, if leaders who are trusted by their subordinates are more likely to display vulnerability. Trust is

Referenties

GERELATEERDE DOCUMENTEN

In deze bachelorthesis wordt onderzoek gedaan naar de relatie tussen interetnisch contact en attitude en beeldvorming onder autochtone jongeren.. Het doel van

De interviewronde legde focus op de eerste twee deelvragen: “Hoe verplaatsen thuiswonende ouderen in krimpregio’s zich, en naar welke bestemmingen?” en “Wat beperkt

Furthermore, this study is the first study to show a positive moderating effect of internationalization on the relationship between both gender diversity as

Opening up the doors of a dementia care home, has the possibility of residents going outside without being noticed or getting lost.. This is the worst case scenario of something

Wanneer 'n persoon ander vergewe vir die pyn en seer wat hulle homlhaar aangedoen het, beteken dit dat so 'n persoon self verantwoordelikheid vir sylhaar lewe

There are also facial indicators discovered for immaturity: a high forehead, soft chin and round eyes, and for dominance: having a large head compared to the body and a broad and

The detected landslide polygon information can be used for disaster planning, mitigation measures for the affected areas.. They will gain knowledge about landslides information in

Government debt is a growing trend among different countries with the public’s needs growing as well as other factors. There is concern that a repeal of VAT zero rating on