• No results found

The effect of gender on the relationship between power attributes and leader perception : a guide on presenting power attributes as a leader

N/A
N/A
Protected

Academic year: 2021

Share "The effect of gender on the relationship between power attributes and leader perception : a guide on presenting power attributes as a leader"

Copied!
83
0
0

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

Hele tekst

(1)

The Effect of Gender on the Relationship Between

Power Attributes and Leader Perception

- A Guide on Presenting Power Attributes as a Leader -

MSc in Bedrijfskunde - Leadership and Management track Master Thesis Final Version

Britt van Gils (10373489)

January 31st, 2018

University of Amsterdam - Amsterdam Business School

Faculty of Economics and Business

Executive Program in Management Studies (EPMS) / Parttime Master Bedrijfskunde

Academic years: 2016 - 2018

(2)

1

Statement of Originality

This document is written by Student Britt van Gils who declares to take full responsibility for the contents of this document.

I declare that the text and 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.

Signature:

Britt van Gils (10373489) Amsterdam, January 31st 2018.

(3)

2

Table of content

1. Introduction ... 7

2. Literature review ... 12

2.1. Introducing power attributes ... 12

2.2. Leader perception ... 14

2.2.1. Effectiveness ... 15

2.2.2. Competence ... 17

2.2.3. Social skills ... 18

2.3. Gender ... 19

2.3.1. Gender and leader perception ... 20

2.3.2. Gender as moderating variable ... 21

2.4. Control variables ... 23

2.4.1. Gender of the follower ... 24

2.4.2. Age of the follower ... 24

2.4.3. Attractiveness ... 25

3. Method ... 27

3.1. Pilot study ... 27

3.1.1. Sample and procedure ... 27

3.1.2. Results and conclusion ... 28

3.2. Survey ... 29 3.2.1. Sample ... 29 3.2.2. Structure ... 30 3.2.3. Procedure ... 32 3.2.4. Demographics respondents ... 33 3.3. Measurements ... 33

(4)

3

3.3.1. Manipulated variables ... 34

3.3.2. Dependent variable ... 34

3.3.3. Control variables ... 35

4. Analysis and results ... 37

4.1. Statistical procedure ... 37

4.2. Descriptive statistics ... 37

4.3. Test of Hypotheses 1 and 2 ... 43

4.3.1. Test of Hypotheses 1a and 2a - The direct effects on perceived effectiveness .. 44

4.3.2. Test of Hypotheses 1b and 2b - The direct effects on perceived competence ... 46

4.3.3. Test of Hypotheses 1c and 2c - The direct effects on perceived social skills .... 48

4.4. Test of Hypothesis 3 ... 50

4.4.1. Test of Hypothesis 3a ... 51

4.4.2. Test of Hypothesis 3b ... 52

4.4.3. Test of Hypothesis 3c ... 53

5. Discussion ... 55

5.1. Strengths, limitations and suggestions for further research ... 60

5.2. Practical implications ... 63

5.3. Conclusion ... 64

6. References ... 65

7. Appendices ... 69

Appendix 1: Pilot study ... 70

Appendix 2: Social media message survey ... 73

Appendix 3: Pictures survey ... 74

(5)

4

List of tables and figures

Tables:

Table 1: Correlation matrix 40

Table 2: Effect of manipulated variables on leader perception variables 42 Table 3: Effect of combined manipulated variables on leader perception variables 43 Table 4: Results of power attributes and gender of the leader as predictor for perceived

effectiveness. 46

Table 5: Results of power attributes and gender of the leader as predictor for perceived

competence. 48

Table 6: Results of power attributes, gender leader in picture as predictor for perceived social

skills. 49

Table 7: Moderation analysis for perceived effectiveness 52

Table 8: Moderation analysis for perceived competence 53

Table 9: Moderation analysis for perceived social skills 54

Figures:

Figure 1: Research Model 26

Figure 2: Attributes in the pilot study that scored above 4 29

(6)

5

Acknowledgements

I would like to thank my supervisor Annebel de Hoogh for the constructive feedback and inspirational sessions throughout the process. Furthermore, I would like to thank the

respondents that have helped to retrieve the data that made this study possible and my close friends, brothers and boyfriend that have helped me look at this study from multiple

perspectives and giving me constructive feedback. Special thanks go out to my mother for being my inspiration and role model and for supporting me throughout my study together with the rest of my family.

(7)

6

Abstract

In business environments, male and female leaders express themselves verbally and non-verbally, consciously and unconsciously, by showing attributes with intended and unintended effects on their surroundings. In this paper, attributes are the center of attention, more

specifically power attributes. Power attributes are defined as attributes that give a leader more authority and that are considered to suit a leader. Current research, however, lacks guidance for male and female leaders on whether it is useful to display these power attributes more consciously to strengthen their leader’s authority. It provides insufficient attention to the effects power attributes could have on employees and their leader perception. This paper contributes to the existing literature through the investigation, by means of quantitative research in the form of an online experiment (N = 200), of the moderating effect of gender on the relationship between power attributes that are displayed by a leader and leader perception. Leader perception in this research is composed of three elements: perceived effectiveness, perceived competence and perceived social skills of the male or female leader. The results indicated that the gender of the leader did not moderate the relationship between power attributes and the leader perception. However, results did show significant positive

relationships between the displayed power attributes and both the perceived effectiveness and perceived competence. Furthermore, the results show that the perceived effectiveness and perceived competence are higher when the leader is male than when the leader is female. Results also indicate a positive relationship between attractiveness of the leader and leader perception and a negative relationship between age of the respondent and leader perception. Implications for the theory and findings are discussed.

(8)

7

1.

Introduction

The car keys of the latest Porsche, a high-end laptop, a certificate or a luxurious pen, attributes are constantly present and used in everyday life, in private as well as in working environments. People surround themselves by attributes, consciously and unconsciously, for different purposes, such as practical need, to increase their self-esteem or to impress others by presenting the newest, most technologically advanced and/or most luxurious attributes. The Oxford dictionaries defines attributes as “a material object recognized as symbolic of a person” (Oxford University Press, 2018). In other words, the material objects a person is surrounded by become symbolic for that person in the eyes of the people around them.

The amount of these material objects, or attributes, these days is tremendous, from luxurious, design and technological attributes to prestige attributes. Following Moore’s law, for example, when it comes to technology attributes, the number of innovations will continue to rise (Moore, 1995). People desire to have the most recent and sophisticated attributes on the market and carry these attributes with them constantly. Global trends in luxury goods show that “global luxury goods accounted for $175 billion in retail sales in 2008 […] and expectations are that the fast-growing markets of China and India will sustain the luxury market expansion for the next 10 years” (Truong & McColl, 2011, p.555).

These attributes may become symbolic for or create an image of a person, as

perceived by the people that surround them. The other way around, attributes can be used to someone’s advantage by showing or strengthening a style, wealth or authority (Khalil, 2000). For instance, when people show prestige attributes (e.g. certificates, medals and other signs of success) they send a message to their surrounding since they “symbolize the sense of one’s rank” and “ celebrate self-ability as an ongoing potential of learning and development” (Khalil, 2000, p.59). Luxurious and impressive attributes, for example, are shown in order to signal wealth, status and power (Veblen, 1899).

(9)

8 Based on the definitions and descriptions from earlier theories, among others Winter (1988) and Truong and McColl (2011), in this study the power attributes are defined as attributes that give a leader more authority and that are considered to suit a leader. These power attributes are worn, used or placed in plain sight in working environments. However, with what effect? What are the associations accompanying these power attributes? Especially for leaders? Do these associations influence the view people have of leaders when they use them? Or, more specifically, do these attributes contribute to the perceived effectiveness, perceived competence and perceived social skills of the leader? In what way, positively or negatively? What effect do they have on their followers, when leaders display these power attributes? Should leaders leave these power attributes at home or should they use and display them, when practicing a leader function? Should they use them more consciously to

strengthen their leadership? Is there a significant difference when the leader is male or female? To what extent does the gender of the follower influence this relationship between power attributes and the leader perception?

The underlying motivations, behaviors and traits that cause a leader to display these power attributes have been the center of attention for various previous researchers (Kusterer, Lindholm & Montgomery, 2013; Pratch & Jacobowitz, 1996; Truong & McColl, 2011; Winter, 1988), who suggested a relationship between the agentic and communal attributes and the (stereotypical) behavior and traits of males and females. They argue that agentic attributes correspond highly to male, or ‘instrumental’, behavioral traits and communal attributes correspond highly to female, or ‘nurturant’, behavioral traits (Trapnell & Paulhus, 2012; Winter, 1988). Up until now, the research enacted on power attributes is limited to the internal cues, behavior and traits or motivations for using them (Assor, 1989; Kusterer et al., 2013; Pratch and Jacobowitz, 1996). However, current research now lacks guidance for male and female leaders on the consequences of displaying the power attributes, whether it has a

(10)

9 strengthening or weakening effect on the perceived leader effectiveness, competence and social skills of their followers. Even though research suggests that the use of power attributes might influence their future as a leader and, on a higher level, the performance of the firm (Pratch & Jacobowitz, 1996).

This study will add another dimension to the current gender research, that is concentrated either on the significant differences in the relationship between gender and leader perception (Pratch & Jacobowitz, 1996), or on the significant differences in the relationship between gender and power attributes or power motivations (Trapnell & Paulhus, 2012; Truong & McColl, 2011; Yukl, 2009). Research is solely focused on the influences of behavioral and personal characteristics on leader perception (Assor, 1989; Day, 2001; Winter, 1988) and not on the influences of power attributes on this leader perception. It lacks

information on the effects power attributes might have on the leader perception in the eyes of the employees or followers. Since power attributes are playing an important role in society, the subject and the outcome of this research could have a significant relevance for leaders.

This paper aims to fill the gap in current literature by investigating the moderating effect of gender of the leader on the relationship between power attributes and the leader perception. To provide a detailed conclusion, additional control variables will also be examined to guide a leader when confronted with different genders and ages of followers.

For this reason, this study will contribute to the existing literature by investigating the impact of power attributes on the leader perception, by zooming in on the influence of

differences in gender on this relationship. The different types of power attributes are

determined through the execution of a pilot study, whereafter an online survey will question 200 respondents on their perception of the leader that is shown in a picture in this survey, using four different perspectives. Four different pictures are added to the surveys, in such a way that each picture reaches 50 respondents. The surveys are randomly distributed among

(11)

10 respondents. These surveys contain the same questions, but four different pictures are

included. The pictures shown in the surveys are manipulated for gender and power attributes, while the rest of the variables is kept as similar as possible, in order to examine if gender influences the relationship between power attributes and leader perception. In line with prior research (Assor, 1989; Day, 2001; Pratch & Jacobowitz, 1996; Trapnell & Paulhus, 2012 Truong & McColl, 2011; Veblen, 1899; Winter, 1988) it is expected that this influence is positive and substantial.

The aim of this study is to provide leaders with a guide on whether to display power attributes in order to increase a positive leader perception as experienced by the employees, depending on the gender of the leader. Therefore, the research question of this study is stated as follows: ‘what is the effect of gender on the relationship between power attributes and the

leader perception?’.

This research attempts to prove the positive influence of power attributes on leader perception and the differences in degree of influence between a male and a female leader. Wherein the leader perception is divided in three variables; the perceived effectiveness, perceived competence and perceived social skills of the leader. The hypotheses and the research question of this study are displayed in the research model in Figure 1.

In the next chapter this study continues with a description of the current findings on power attributes and the perception of a leader, which is build out of the leader’s

effectiveness, the leader’s competence and the leader’s social skills. Moreover, the relationship between these subjects, gender and various other influencing variables is discussed. Which leads to the research model of this study. Subsequently, the third chapter focuses on the data collection and research method. Whereafter, chapter four outlines the results based on this data. In the last chapter, chapter five, the results, implications and

(12)

11 limitations of this study are discussed and suggestions for further research are proposed. Hereafter the chapter is ending with the conclusion of this study.

(13)

12

2.

Literature review

In order to answer the previously mentioned research question, ‘what is the effect of gender on the relationship between power attributes and the leader perception?’, the existing literature concerning power attributes and leader perception in relation to gender has to be investigated. Numerous studies focused on gender and the impact on leader perception e.g. Judge and Cable (2004), Pratch and Jacobowitz (1996), Yukl (2009). The available research on the effect of power attributes on the variables gender and leader perception, however, are limited. To ensure this study will contribute to the current literature and to show how the research question is developed, the information and theories that currently are available on the subject are examined.

2.1.

Introducing power attributes

The visibility and the amount of attributes is continuously increasing (Moore, 1995). The number of potential power attributes people are surrounded by throughout their lives is tremendous. People desire the most impressive version to signal wealth, status and power and try to impress each other with these power attributes (Truong & McColl, 2011).

This “concern for having impact on others, arousing strong emotions in others or maintaining reputation and prestige” is defined by Winter (1988) as “the power motive” (p.510). This definition resulted from the research wherein Winter (1988) conducted 27 motive arousal experiments to link the power motive to various power-related actions, such as leadership, social power and prestige. As Winter (1988) states, “the first step in getting power is thus simply to become visible to others” and “the right possessions--whatever objects are valued by the group--can confer prestige on the owner” (p.512). These prestigious possessions influence the owner’s self-perception, by making him or her feel powerful and impactful (Winter, 1988). In his study, the results showed that the power motivation is significantly related to the number of prestigious possessions that are owned by a person (r = 0.23, p <

(14)

13 0.05) (Winter, 1988). Truong and McColl (2011) confirm these findings by stating that people use “highly visible goods to display their wealth and gain social status” (p.556). Khalil (2000) continues on the theories of Winter (1988) and Truong and McColl (2011) by mentioning that people display specific attributes, e.g. automobiles, in order to show success, the possession of the desired ability by means of so called prestige goods, e.g. prices or certificates, or one’s ability and rank by means of so called vanity goods, e.g. newest technological gadgets. The motivation for displaying these attributes is explained by Inesi, Lee and Rios (2014), who mention that “people are attracted to power because it promises the possibility of pursuing one’s own goals, unfettered by compromise and acquiescence to the desires of others” (p.19). As a result of the definitions and descriptions from the theories of Khalil (2000), Truong and McColl (2011) and Winter (1988), the possessions or power attributes in this study are defined as physical objects or attributes that give a leader more authority and that are considered to suit a leader.

These power attributes generally have the effect to make a person look wealthier (Truong & McColl, 2011), more professional or powerful and could provoke a significant reaction or response from a leader’s followers (Inesi et al., 2014; Khalil, 2000). Winter (1988) states in his study on the power motive that “power predicts a wide range of actions involving leadership and social power, acquiring prestige” (p.510). In other words, as stated by Truong & McColl (2011), the power attributes increase the professionality or power of a person and power in turn predicts leadership. No theories that are currently available however, have studied what the effect of gender might be on the relationship between the power attributes and perception followers have of their leader.

The visibility, popularity and significant relevance of these power attributes (Moore, 1995; Truong & McColl, 2011) and the minimal attention given to these power attributes in

(15)

14 current research in relation to leader perception and gender make them a relevant and

interesting subject for investigation.

To examine what attributes are considered to be power attributes, a pilot study is conducted. Herein different attributes, which evoke an increase in power and authority for the person displaying them, will be rated to ensure the right power attributes are displayed in the survey. Further elaboration on the pilot study can be found in the method section.

2.2.

Leader perception

The leader perception is influenced by the expectation people have of the behavior and traits that are considered to be appropriate for a leader (Pratch & Jacobowitz, 1996), such as effectiveness, competence and social skills (Day, 2001). This expectation is then used as an example of the perfect leader and as a benchmark when evaluating a leader in order to form the leader perception. Positive evaluations or perceptions of the followers are crucial for the success of a leader, since they evoke positive, favorable and necessary behaviors (Drescher, 2017). In this paper, the power attributes might be an indicator for the followers that the leader is competent and could help the firm reach a higher performance.

Otara (2011) confirms the importance of the leader perception by mentioning that “in organizations, perceptions of leaders, managers and employees shape the climate and

effectiveness of the working environment” and “perception therefore is just another tool, underutilized, that requires an administrator to develop the true desire to be the best he/she can be. Using all of the tools available to us we can achieve great things with our employees” (Otara, 2011, p.21).

Moreover, Truong and McColl (2011) state in their study on luxury consumption motivations, that luxury goods could evoke reactions within other people, such as the perception of social status or place in social hierarchy. This perception, as Otara (2011) mentions, is crucial, since people interpret and react upon the visual and external causes or

(16)

15 influences and “we react to specific situations based on what we see rather than on what it really is” (Otara, 2011, p.21). These reactions and perceptions might influence the leader perception. The existence and direction of the relationship between power attributes and the leader perception, however, is still to discover. Since (power) attributes could evoke reactions within people (Truong & McColl, 2011) and these reactions follow from the perceptions that people have (Otara, 2011), it is expected that followers will have a more positive leader perception, when the leader is displaying power attributes. Up until now, however, researchers have not focused on the effects of power attributes on the leader perception.

Since leader perception is a very broad term, in this study it is narrowed down into three areas based on the theory of Day (2001) on leadership development. Herein he suggests that there are three areas or capabilities that could influence leadership development and the evaluation of a leader, defined as the effectiveness, competence and social skills of a leader (Day, 2001). These areas are considered the main capabilities that influence the leader perception. Previous research from Otara (2011), Pratch and Jacobowitz (1996) and Yukl (2009) confirms that these three capabilities could influence the overall leader perception. In the following chapters a detailed description of the three areas that shape the leader perception will be given.

2.2.1.

Effectiveness

In recent literature leader effectiveness is a popular term. Various researchers investigated the relationship between leader effectiveness and personality traits or behaviors, among others Pratch and Jacobowitz (1996) on individual characteristics (such as assertiveness,

decisiveness and ambition) and De Hoogh, Den Hartog & Koopman (2005) on the ‘Big Five’ model in which they study the effects of the factors extraversion, openness to experience, agreeableness, conscientiousness and neuroticism on leadership. The studies on external

(17)

16 influences in the surroundings of a leader that could influence the leader effectiveness,

however, are limited.

According to Drescher (2017), leader effectiveness is crucial in the perception and evaluation of a leader. He defines leader effectiveness as the “overall satisfaction with the leader, leader prototypicality, and the perception of strong leadership” (Drescher, 2017, p.3). Pratch and Jacobowitz (1996) add to this that “recent researchers have reported positive correlations between effective leadership and a variety of individual characteristics, including internal locus of control orientation [...], problem-focused coping […], stress tolerance […], self-confidence, […] social perceptiveness and response flexibility” (p.205). They state that an effective leader will perform with the right balance between assertiveness towards and cooperation with its followers (Pratch & Jacobowitz, 1996). Moreover, they mention that a leader has to be independent, decisive and ambitious to receive a high leader effectiveness rate of their followers (Pratch & Jacobowitz, 1996). Yukl (2009) has a more externally focused definition and states that the leader’s effectiveness is a measurement of its contribution to a group. In this study the perceived effectiveness of the leader is a

combination of the right balance of Pratch and Jacobowitz (1996), the external focus of Yukl (2009) and the overall satisfaction perceived by the followers (Drescher, 2017). This study defines the leader’s effectiveness as the view of the employees on the functioning and capability of their leader.

To ensure the perception of the followers on the leader’s effectiveness is high, the leadership must be perceived as strong (Drescher, 2017) and the leader must be ambitious and decisive (Pratch & Jacobowitz, 1996). As explained in the previous section on power

attributes, this might be accomplished through the display of power attributes that evoke a perception of social status and power, such as luxury or prestige goods (Truong and McColl, 2011; Winter, 1988). When these attributes are objects that are perceived as valuable by the

(18)

17 group, they could result in a higher prestige of the leader (Winter, 1988). In line with these theories, it is expected that followers will have a more positive leader perception when the leader is displaying power attributes. Which is translated in the first hypothesis:

H1a: Power attributes positively influence the perceived effectiveness of a leader.

2.2.2.

Competence

To conduct a detailed research on the perception followers have of their leader, the leader’s competence is studied. Yukl (2009) states that “followers use information about leader actions, changes in the performance of the team or organization, and external conditions, to reach conclusions about responsibility for success or failure” (p.246). Furthermore, Yukl (2009) specifies that “a leader is usually judged more competent if his or her unit is successful than if it is unsuccessful” (p.247), and “leaders who are judged to be competent gain more power and have more discretion to make changes” (p.248). This in turn would lead to an improved leader perception. The leader’s competence involves the relevant visual actions of a leader that improve the performance of a team or firm (Yukl, 2009). Moreover, Day (2001) confirms the theory of Yukl (2009) by mentioning that the leader’s competence is visualized as strategic actions that could improve the performance of a team and therewith the

performance of a firm. Consequently, this study defines the competence of a leader as the overall performance as a leader, as judged by the followers, which originates from the visual signs.

Preceding studies have focused on internal influences, such as traits and behaviors (Winter, 1988) and external influences, such as smell (Otara, 2011) or voice pitch (Klofstad, Anderson & Peters, 2012). The studies that focus on power attributes that the leader surround themselves by, however, are limited. When combining the researches of Yukl (2009) and Day (2001), wherein they mention that the leader’s competence is focused on the visual actions of the leader, with the research of Truong and McColl (2011), who comment that this visual

(19)

18 effect could be reached through the display of certain power attributes, and the research of Khalil (2000), who argues that people are showing prestige attributes to symbolize their rank or potential or power, it is expected that followers will rate the perceived competence of their leader higher when the leader is displaying power attributes. As they might evoke and

symbolize power or rank, which can be seen as the leader’s competence (Yukl, 2009). Therefore, this suggestion is translated in the following hypotheses:

H1b: Power attributes positively influence the perceived competence of a leader.

2.2.3.

Social skills

The third important variable, when evaluating the leader perception, are the social skills of the leader. To achieve a high leader perception, leaders have to communicate in a good manner with their followers. To accomplish this, a leader needs the right social skills. Day (2001) describes these skills as “social awareness, such as empathy, service orientation, political awareness” and “collaboration and cooperation, building bonds, and conflict management” (p.585). This definition was confirmed and complemented by Yukl (2009) by stating that it is crucial to master communication techniques and interpersonal skills in order to build and lead a team. Following from this, social skills are interrelated with the likeability or admiration of a person, which is important when a leader wants to build a bond (Day, 2001; Yukl, 2009).

In conjunction with these findings, Khalil (2000) argues that this “admiration becomes more tangible when the agent buys appropriate prestige symbols” (p.72). Meaning that, if a leader displays prestige symbols, which can appear in the form of power attributes (Khalil, 2000; Truong & McColl, 2011; Winter, 1988), the admiration or social likeability may increase. This in turn might increase the perception of the social skills of a leader, according to Day (2001) and Yukl (2009). In line with the assumptions of earlier research it is expected that the power attributes positively influence the perception of the social skills, resulting in the third hypothesis:

(20)

19 H1c: Power attributes positively influence the perceived social skills of a leader.

2.3.

Gender

Numerous studies have focused on power behavior, the motives for this behavior and

leadership styles, as mentioned in the previous sections. Various researchers continued on this base and studied the concept of leadership in relation to gender (e.g. Kusterer et al., 2013; Pratch & Jacobowitz, 1996; Trapnell & Paulhus, 2012). Gender has become a well-known and appreciated subject in academic research and could evoke interesting effects in a business environment. Pratch and Jacobowitz (1996), Trapnell and Paulhus (2012) and Winter (1988), for example, suggest that the behaviors and traits that evoke a perception of good leadership are closely related to agentic attributes. Moreover, they argue that these agentic attributes (stereotypically) correspond closer to males than females (Pratch & Jacobowitz, 1996; Trapnell & Paulhus, 2012; Winter, 1988). As Winter (1988) indicates, “men’s power actions are said to be ‘instrumental’ while women’s power behavior is ‘interdependent’ or

‘nurturant’”(p.510). In line with Winter (1988), Trapnell and Paulhus (2012) mention that the social roles are divided into agentic life goals for males, e.g. “leadership, power, expertise, success, and economic interests” and communal life goals for females, e.g. “fulfilling relational and religious obligations and commitments, seeking ‘purpose’ in life, and

sacrificing for others” (p.40). This suggests, that the leader perception might be higher when the leader is male.

The cause could root in history, wherein the man possessed the final authority and the women occupied a subordinate role in society (Larsen & Long, 1988).Pratch and Jacobowitz (1996) mention that, in the business environment, the expectations people hold of the

behavior and traits that a certain gender should show, influences the evaluations of both male and female leaders. These prejudgments about men and women are defined as the gender role stereotype (Brogan & Kutner, 1976; Kusterer et al., 2013) and might be strengthening the

(21)

20 moderating influence of gender (Kusterer et al., 2013; Larsen & Long, 1988; Pratch &

Jacobowitz, 1996; Trapnell & Paulhus, 2012).

The research on gender role stereotypes is extensive. Which is the reason why this study will not expand on the stereotype itself, mere on the effect of gender in general in order to provide practical guidance to leaders. Nonetheless, the theories on gender roles and

stereotype will be involved in this study to elaborate on the influence that gender might have. Considering the studies of Trapnell and Paulhus (2012) and Winter (1988) that

elaborate on the increase in leader perception provoked by a sense of power and the

researches of Khalil (2000) and Winter (1988) that discuss the effect of certain attributes that might evoke this sense of power, this study will examine the influence of gender on the relationship between power attributes and the leader perception. Current literature lacks guidance for both genders on the use of power attributes and the effects they might have on the leader perception that people hold on them. This, however, is crucial for the performance of a leader and ultimately the performance of the firm. To emphasize the relevance of this research, the following section will elaborate on previous research that has been conducted on the relationship between gender and leader perception, followed by an elaboration of gender as a moderator. Whereafter attention is given to the current gap in literature that in turn leads to the next hypotheses.

2.3.1.

Gender and leader perception

Previous research shows that “expectations about behavior that is appropriate for a leader coincide largely with beliefs about the behavior that is appropriate for men” (Pratch & Jacobowitz, 1996, p.205). Moreover, they continue by stating that men could perform in a male and female style without influencing their evaluation, since their leadership is

considered to be legitimate at start (Pratch & Jacobowitz, 1996). This however is not the case for females. “The details of female leaders' behaviors may be scrutinized because of their role

(22)

21 conflict, male leaders are not ordinarily constrained by the attitudinal bias of their coworkers” (Pratch & Jacobowitz, 1996, p.205). Furthermore, they state that “women in leadership roles tended to be evaluated negatively relative to men when their leadership style was

characterized by stereotypically masculine features, particularly when their style was autocratic or directive” (Pratch & Jacobowitz, 1996, p.204). The grounds for these findings, according to Trapnell and Paulhus (2012) and Winter (1988) are the opposite life goals or actions that males and females might have. These opposite life goals are, on the one hand, male or agentic goals and actions (instrumental, such as expertise, power and success) and, on the other hand, female or communal goals and actions (interdependent and nurturant, such as purpose seeking and sacrificing themselves for others) (Trapnell & Paulhus, 2012; Winter, 1988).

Since the characteristics of successful leaders in general have a higher similarity to the (expected) characteristics of males than of females (Kusterer et al., 2013; Pratch &

Jacobowitz, 1996), this study also investigates the direct relationship between gender of the leader and leader perception. Wherein there is expected that all three capabilities or

characteristics that shape the leader perception of a male leader will be higher than those of a female leader. In line with this expectation, the following hypotheses will be tested:

H2a: The perceived effectiveness of a male leader will be higher than a female leader. H2b: The perceived competence of a male leader will be higher than a female leader. H2c: The perceived social skills of a male leader will be higher than a female leader.

2.3.2.

Gender as moderating variable

“The right possessions—whatever objects are valued by the group—can confer prestige on the owner” (Winter, 1988, p.512). Various studies are done on the motivation behind the use and display of the power attributes, for example by Winter (1988), but none of them focuses

(23)

22 on the actual effects that might be evoked by the display of power attributes on the people around them.

Trapnell and Paulhus (2012) investigated the influence of different types of attributes on a leader’s followers and found a critical difference in the effect that certain attributes evoke when used by males or females. They differentiate attributes in agentic and affiliative (or communal) attributes (Trapnell & Paulhus, 2012). The concept agentic attributes is derived from the word agency, which is “the meta-concept associated with self-advancement in social hierarchies”, e.g. “leadership, power, expertise, success and economic interests” (Trapnell & Paulhus, 2012, p.39). Affiliative attributes are related to communal values, which are derived from communion, defined as “ the partner concept associated with maintenance of positive relationships”, e.g. “fulfilling relational and religious obligations and commitments, seeking purpose in life, and sacrificing for others” (Trapnell & Paulhus, 2012, p.39). Power attributes, therefore, in this study are similar to the agentic attributes, as defined by Trapnell and Paulhus (2012). Moreover, Pratch and Jacobowitz (1996) use these terms and state that “men are expected to have high levels of agentic (instrumental) attributes, including being independent, masterful, assertive, and instrumentally competent. These qualities reflect a tendency to promote the self, especially self-competence” (p.204). They continue by stating that women are expected to contain more communal or affiliative attributes (Pratch & Jacobowitz, 1996).

Kusterer et al. (2013) continue on the previously mentioned theories by mentioning that “agentic attributes are often deemed more important than communal attributes and more in congruence with successful management characteristics” (p.573). Trapnell and Paulhus (2012) confirm this statement. The latter state that if this is not the case, a role conflict could arise and this might influence the evaluation people receive (Trapnell & Paulhus, 2012).

(24)

23 In other words, when males are not surrounded by power attributes, but affiliative attributes, this could evoke a negative or less positive evaluation of the leader.

Correspondingly, when females are not surrounded by affiliative attributes, but surrounded by power attributes, this could result in a less positive evaluation.

As a result of the theories on the relationship between gender and power attributes and the relationship between gender and leader perception, this study investigates the gap in current literature which focuses on the effect of the gender of the leader as a moderating variable in the relationship between the power attributes and the leader perception. Based on a combination of the previously mentioned theories (e.g. Khalil, 2000; Kusterer et al., 2013; Pratch & Jacobowitz, 1996; Trapnell & Paulhus, 2012; Winter, 1988) there is expected that the relationship between power attributes and the leader perception is stronger for males than for females. From these expectations, the following hypotheses are constructed:

H3a: The relationship between power attributes and the perceived effectiveness is stronger for

male than for female leaders.

H3b: The relationship between power attributes and the perceived competence is stronger for

male than for female leaders.

H3c: The relationship between power attributes and the perceived social skills is stronger for

male than for female leaders.

2.4.

Control variables

The variables gender of the respondent, age of the respondent and attractiveness of the leader are included in the analysis as control variables to ensure they do not cause a variance in the results. The variance that these three variables might cause in the examined relationships will be discussed from an explorative and controlling point of view. Although these variables are not the main aim of this research, they might contribute to the outcome of this paper and

(25)

24 provide additional guidance to leaders. The next section elaborates on the control variables, gender and age of the respondent and attractiveness of the leader.

2.4.1.

Gender of the follower

Beyond the gender of the leader, the gender of the follower might have a significant influence on the outcome. Previous research shows a significant effect between the gender of the follower and the leader perception (Kusterer et al., 2013; Pratch & Jacobowitz, 1996; Winter, 1988). The studies of Pratch & Jacobowitz (1996), for example, showed weaker leader perception ratings for female leaders, when the respondent was male. This is confirmed by Kusterer et al. (2013), who conducted a study in which they indicate that the magnitude of the difference in leader perception between male and female leaders is influenced by the gender of the respondent. They conducted cross-gender and same-gender analysis which resulted in greater cross-gender distances than same-gender distances (p < 0.001) (Kusterer et al., 2013). Further tests of Kusterer et al. (2013) revealed that this was the case for respondents with both male (p = 0.047) and female leaders (p < 0.001). However, this distance was greater for female respondents overall (Kusterer et al., 2013). In other words, this study suggests that the gender of the respondent might cause a variance in the influence of gender on the relationship between power attributes and leader perception.

As a result of previous studies, that show the gender of the follower might influence the leader perception, the gender of the follower is controlled for in the analysis.

2.4.2.

Age of the follower

Nowadays, people grow up with these (power) attributes, especially the latest generations. The number of power attributes available continues to increase (Moore, 1995; Truong & McColl, 2011). This suggests that attributes increasingly become the center of attention in lives. Therefore, since the latest generations grow up with these attributes, one could assume that the younger people might be more used to using and seeing power attributes in

(26)

25 comparison to older people (Statista, 2017a, 2017b). Additionally, Khalil (2000) mentions that when people progress in age the attention to the utility and display of goods is becoming less of a function of the approval of others. This might indicate that the importance and notice of power attributes in the environment becomes less influential to the leader perception when people get older. As mentioned above, Truong and McColl (2011), Winter (1988) and Yukl (2009) among others, argue that visual attributes could influence the perception people have of a leader. Thus, to influence the leader perception, the power attributes must be visible to the respondents. When this visibility or notice is decreasing, the effect of power attributes on the leader perception might also decrease.

Bearing this in mind, this study controls for the variance that the age of the respondent might produce in the results of the research. There is expected that the younger the follower is, the more used they are to using and seeing the power attributes. Thus, since previous research shows the age of the respondents might influence the notice of the power attributes (Khalil, 2000) and consequently the effect of power attributes on the leader perception

(Truong & McColl, 2011; Winter, 1988; Yukl, 2009), the age of the respondents is controlled for in the analysis.

2.4.3.

Attractiveness

The third control variable is the attractiveness of the leader. Previous research has proven that “greater influence and status are granted to physically attractive individuals” (Inesi et al., 2014). The underlying reason might be that the attractiveness of a person might evoke positive impressions on their surroundings (Inesi et al., 2014, p.22). Park and Lennon (2008) strengthen this assumption by mentioning that “people seem to use other people’s appearance as a source of personal-characteristics inferences” and “individuals tend to like others who appear to have similar personal characteristics in first-impression formation situations” (p.285). In line with these findings, the meta-analysis on physical attractiveness of Eagly,

(27)

26 Ashmore, Makhijani and Longo (1991) suggests that, in general, the beauty or attractiveness of a leader has a positive influence on the perception of the followers on the leader’s ability and competence.

As a result of the previously discussed studies of Eagly et al. (1991), Inesi et al. (2014) and Park and Lennon (2008), the attractiveness of the leader is included as a control variable to ensure this variable is not causing a variance in the results of this study.

Figure 1 shows an overview of the hypotheses in the research model of this study.

(28)

27

3.

Method

This section elaborates on the empirical method of this study. Starting with an explanation of the conducted pilot study. Hereafter, the gathered sample and procedure of the survey will be discussed, followed by the demographics of the respondents. Next, the measurements of the variables in this study are explained together with the statistical procedure that is used to research the previously mentioned hypothesis of this study.

3.1.

Pilot study

Previous to the development and distribution of the survey, a pilot study is executed. The pilot study is conducted in order to examine what specific attributes are considered as power

attributes by the target group of this study and to what degree the respondents perceive the attributes as power attributes. These power attributes are defined as attributes that give a leader more authority and that are considered to suit a leader. The following sections will elaborate on the sample and procedure, the results and conclusion of the pilot study.

3.1.1.

Sample and procedure

This pilot study is conducted online with a sample of 50 randomly selected respondents. To reach the preferred sample, the pilot study is conducted via stratified random sampling, on the online platforms LinkedIn (https://nl.linkedin.com/) and Facebook

(https://nl-nl.facebook.com/). The preferred sample of this pilot study includes people that have frequent contact with leaders and people who do not have frequent contact with leaders in the business environment. LinkedIn is a business platform, so there is expected that the people that are active on this platform are also active in business environments. Facebook is used by both people that are active in business environments and people that are not active in business environments. On these platforms, a message is posted with a link to the online

(29)

28 please help me out and fill in this questionnaire?’) and no reward or price is given to the respondents to ensure no bias will arise and the validity increases. The questionnaires are placed online to increase the speed of the data collection and ease for the respondent.

At the start of the pilot study, the respondents received a small introduction on the definition of power attributes, namely attributes that give a leader more authority and that are considered to suit a leader. Hereafter, the respondents were asked to ‘rate the following pictures on a scale of 1-5, one meaning not at all, five meaning definitely power attributes’. This five-point scale is used, in line with the research of Weathers, Sharma and Niedrich (2005) that has shown that this scale number is reliable and valid. The respondents were asked to rate 15 pictures of attributes in total, which included various attributes that can be found in offices of leaders in business environments. The pilot study consisted of a total number of 15 attributes that could be considered as power attributes, to make sure 5 of these attributes can be chosen from this pilot study, to display in the survey picture.

The 15 attributes involved an I-watch, a MacBook pro Touch Bar, glasses, a card protector wallet, a golden pen, a price cup, a certificate, a ribbon, a leather briefcase, an IPad, a medal, a Porsche car key, a leather file folder, a red wax letter stamp and a bronze statue of a horse (see Appendix 1).

3.1.2.

Results and conclusion

By analyzing this data, five attributes with the highest ratings (percentage that scored above 4) were selected as power attributes. These attributes involved a MacBook pro Touch Bar (44%), a golden pen (48%), a certificate (60%), a Porsche car key (62%) and a red wax letter stamp (54%) (see Figure 2 for the results). In the next section, there will be elaborated on the use of these selected power attributes and the composition and distribution of the survey.

(30)

29

Figure 2: Attributes in the pilot study that scored above 4

3.2.

Survey

After the completion of the pilot survey, which resulted in the selection of 5 power attributes, a survey was conducted. The data is gathered via a survey to rule out any socially desirable answers that might have been given, when the questions are probed by an interviewer. Moreover, a survey was chosen to accomplish a data set with the desired sample size within the limited time available for gathering the data. In the next sections, the sample and

procedure of the survey and demographics of the respondents are discussed.

3.2.1.

Sample

The sample is gathered via stratified random sampling. A small group is selected from the population, which includes employed women and men in between the age of 20 and 70 years. This is the age range wherein most people are working, that are employed in different sectors

(31)

30 and working in variating functions (Centraal Bureau voor de Statistiek, 2017). The sample exists of 200 completed surveys, which is seen as a reliable amount of respondents in comparison to the amount of predictors that is used (Green, 1991). This randomly gathered sample will be used to increase the diversity among the group of respondents, herewith the outcome of the research will be more detailed and serve as a guide for a diverse group of readers. The survey was structured in a way so that respondents could not leave any questions unanswered before moving on to the next question and the survey stayed available online up until a total of 200 finalized questionnaires was reached.

3.2.2.

Structure

In this section, the use of the power attributes, the survey administration and the structure of the questionnaire will be discussed.

Use of power attributes

The power attributes determined by the pilot study are displayed in two of the four pictures that will be visible in the surveys. To examine the effect of gender and the effects of the power attributes, four different pictures are made. These four different pictures display:

1. A men surrounded by power attributes 2. A women surrounded by power attributes 3. A men not surrounded by power attributes 4. A women not surrounded by power attributes

Each of the pictures is attached to 50 of the 200 questionnaires. All questionnaires contain exactly the same questions, however the picture attached to each of the questionnaires is variating per 50 surveys. These 200 questionnaires are randomly distributed among the

respondents by the use of the online survey platform Qualtrics (https://www.qualtrics.com/) to ensure exactly 50 questionnaires are completed for each of the 4 pictures. To ensure each picture is similar to a photo of a leader in the office, the men or women in the picture are

(32)

31 seated behind a desk. For all the pictures, the same male and same female leader is displayed. The power attributes, selected as a result of the pilot study, are placed on the desk and in the background of the photo (see Appendix 3).

This research design is chosen as this paper is centered on the visual impressions and attempts to exclude the highest amount of unwanted external influences as possible, for example smell (Otara, 2011) or voice pitch (Klofstad et al., 2012). The decision to show a picture of a person has been made in order to eliminate as much influencing variables as possible that could appear when showing a video or a person in real life. Moreover, in the survey a picture is shown to research the visual effects and responses.

This picture reduces the cognitive bias effects that could appear when only describing the situation(Hilbert, 2012). When describing the situation, respondents might experience a cognitive bias and depicture this description differently in their minds or the aim of the study might be revealed. When providing the respondents with a picture, every respondent has the same picture in mind and the aim of the study will stay unknown.

Questionnaire

The aim of the survey is to answer the question: ‘do you consider the person in the picture to be suitable in a leader position?’. The survey consists of 68 questions in total, from which 17 questions are main questions and 51 are sub-questions. The first 5 questions of the 68 are centered around the demographics of the respondent (to answer the control variables age and gender of the follower, among others), followed by 28 questions about the respondent’s current leader, among which the same questions that are asked about the leader in the picture, to distract the respondent from the aim of the study. Hereafter one of the four pictures of the leader is shown, followed by the 1 question concerning the attractiveness of the person

displayed in the picture and 25 out of 68 questions concerning the perceived effectiveness, the perceived competence and the perceived social skills of the leader displayed in the picture.

(33)

32 These questions are asked to gather data on the perception that the respondents have of the leader shown in the picture. To conclude, 9 out of 68 questions are asked on the powerfulness of the current leader and the leader in the picture, the gender stereotype and the notice of power attributes in the picture to distract the respondent from the goal of the survey.

The survey predominantly consists of Likert interval rating scales (1 up until 7, with 1 being ‘not at all’ and 7 being ‘definitely’), to make sure the degree of perception is measured and the results can be compared, and questions that can be answered by the use of a drop-down menu. The open questions are used for the numerical demographics, such as age and the number of people the respondent and leader lead.

Furthermore, the message that is placed on the social media platforms wherein the link to the questionnaire was attached, is kept as general as possible (see appendix 2) and no reward or price is given to the respondents to ensure no bias will arise and the answers are as honest as possible. Moreover, anonymity is ensured and the respondents are informed that the data gathered in the survey will exclusively be used for the purpose of the research of this master thesis.

3.2.3.

Procedure

The survey administration started on December the 3rd of 2017 and was closed on January the 3rd of 2018. To examine these variables, SPSS (Statistical Package for the Social

Sciences) (IBM Corp, 2016) is used. The data in this study was retrieved via an online survey via the online survey platform Qualtrics (2018). The link to the survey was spread via

LinkedIn and Facebook to reach the necessary target group and 200 completed surveys. The online platforms cause an increase in speed of the data collection and an increase in ease for the respondent. The survey is cross-sectional, instead of over-time, since the assumption is that the opinion of the respondent stays consistent over time. Each contestant completed one

(34)

33 of the four surveys, which are randomized among the respondents as explained earlier, in order to decrease the bias towards the aim of this study.

In total, 321 respondents started the survey, 200 completed the whole survey: response rate of 62.3%. Of the 121 respondents that did not finish the survey completely, 22% opened and immediately closed the survey before filling in any answer. Furthermore, the average completion time of the completed surveys was 646.44 seconds, or 10.77 minutes, due to some high outliers. After correcting for the outliers, this was 520.94 seconds, or 8.68 minutes.

3.2.4.

Demographics respondents

The mean age of the respondents was 36.8 years (SDage=12.2; age-range: 20 to 67 years) and 45% of the completed surveys was filled in by a male (55% female). Furthermore, the majority of 50% completed a college degree (13% high school degree, 37% postgraduate degree) and the respondents are managers over 32 employees on average (52% of the

respondents was manager of one or more employees and 5% of the total respondents managed 100 employees or more). The most popular sectors among the respondents were the business, consulting and management sector (around 12%), the hospitality sector (around 10%), the information technology sector (around 9%) and the marketing, advertising and public

relations (around 9%). The average age of the leaders that the respondents have is 45.3 years and 67% of the respondent’s leaders was a male (33% female) and managed over 539 employees on average (with a maximum of 60,000 employees).

3.3.

Measurements

The different variables are measured via the use of various established and corresponding scales of previous research. In the following section, the scales and measurements of all variables will be explained.

(35)

34

3.3.1.

Manipulated variables

The independent and moderating variable, power attributes and gender of the leader

respectively, of this study were manipulated in order to research the influence on the leader perception.

Power attributes

The power attributes used in the survey are a result of the pilot study, established on forehand. The effect of the power attributes is measured through comparing the results of the two

surveys with power attributes in the pictures to the results of the two surveys without power attributes. These outcomes are made visible in the manipulation table in the result section.

Gender of the leader

The gender is manipulated by displaying a male in two out of the four surveys and a female in the other two surveys. For both types an equal amount of respondents is gathered. The results are compared in the manipulation table to research the influence of gender on the relationship between power attributes and leader perception.

3.3.2.

Dependent variable

The dependent variable, leader perception, is measured via the variables; perceived

effectiveness, perceived competence and perceived social skills of the leader. To measure the perceived effectiveness of the leader the questions and scale of De Hoogh et al. (2005) (Cronbach’s α = 0.86) was used. This scale is based on three items: “To what extent is the overall functioning of the person you evaluate satisfactory?”, “How capable is the person you are evaluating as a leader?”, “How effective is the person you are evaluating as a leader?”(De Hoogh et al., 2005, p.851). The 7-point response Likert-scale is used, ranging from 1 (not at all) to 7 (very much so).

The leader’s competence and leader’s social skills or, as Assor (1989) defines the latter, sociability of the leader is researched based on the scale of Assor (1989). The

(36)

35 competence scale is composed of 9 adjectives (Cronbach’s α = 0.84) and the sociability scale of 12 adjectives(Cronbach’s α = 0.86). Assor (1989) has proven a correlation between the two scales of 0.09. The respondents are asked to rate two contrasting adjectives, ranging from 1 (being one adjective) to 7 (being the contrasting adjective). Examples of the sociality scale are ‘Friendly-Hostile’ or ‘Intolerant-Tolerant’. Examples of the competence scale are

‘Competent-Incompetent’ or ‘Unintelligent-Intelligent’.

3.3.3.

Control variables

Gender and age respondent

Since previous research has shown that gender and age might influence the leader perception and therefore might cause a variance in the outcome of this study (Khalil, 2000; Kusterer et al., 2013; Pratch & Jacobowitz, 1996; Truong & McColl, 2011; Winter, 1988), these variables are included in the analysis as control variables. Since gender and age differ among the

respondents, it will be interesting to analyze the effects on the leader perception. These results are gathered through questions concerning the demographics of the respondent and the gender stereotype. These variables will provide the readers with a detailed guide, matching the reader’s situation.

Attractiveness

Another variable controlled for is the attractiveness of the person displayed in the picture, which might influence the results (Eagly et al., 1991; Inesi et al., 2014; Park & Lennon, 2008). To exclude this possible effect, the participants will be asked to rate the attractiveness of the person in the picture via the question ‘Do you consider the person in the picture to be attractive?’, where the respondents rate the attractiveness on the Likert scale ranging from 1 (not attractive) to 7 (very attractive).

Moreover, to minimize the possible effects of other variables, such as the possible effect of the attractiveness the leader, the research surroundings will be kept as consistent as

(37)

36 possible. The male and female will wear a white blouse and black blazer and will not have prominent clothes, hair or make-up. Furthermore, the height of the persons in the picture will be kept as similar as possible, since previous research shows that height has an influence on the leader perception (Judge & Cable, 2004).

(38)

37

4.

Analysis and results

In this section the analysis of the gathered data and the corresponding results are discussed. As a start, the data is prepared for analysis in the statistical procedure section. In the second section, the correlation matrix will be discussed and the effects of the manipulated variables are visualized. Hereafter, a regression analysis will be enacted to discover if direct

relationships exist between the variables and to test the first hypotheses. Followed by a moderation analysis to analyze if the gender of the leader has a moderating effect on the relationship between power attributes and leader perception and to test the rest of the hypotheses.

4.1.

Statistical procedure

The data is first cleaned, which means the incomplete surveys (with missing values) were deleted from the collected data and the correct corresponding measurement scales were appointed to each question (nominal, ordinal or interval). No counter indicative values were reversed. Hereafter the scale reliabilities, or Cronbach’s alpha, are computed. These are visualized in the next section, where the results of the correlation analysis will be discussed, on the diagonal of the correlation matrix between brackets. Then the descriptive statistics and normality tests are calculated. The mean scores of the ordinal variables are calculated per variable (question 9 up to 15, which includes the perceived effectiveness, perceived competence and perceived social skills of the leader).

4.2.

Descriptive statistics

In order to visualize the correlations between the variables gender of the respondent, age of the respondent, attractiveness of the leader, perceived leader effectiveness, perceived leader competence and perceived social skills of the leader, a correlation analysis is executed. The

(39)

38 results of this analysis are displayed in the correlation matrix in Table 1, wherein the means, standard deviations, correlations and scale reliabilities are presented.

Dependent variable

The analysis of the dependent variable is divided into the analysis of perceived effectiveness, perceived competence and perceived social skills. Table 1 shows a strong positive

relationship (r = 0.82, p < 0.01) between perceived effectiveness (M = 4.60, SD = 0.94) and perceived competence (M = 4.80, SD = 0.80) of a leader. Furthermore, the results show a positive relationship between perceived social skills (M = 4.81, SD = 0.85) of the leader and perceived effectiveness (r = 0.31, p < 0.01) and perceived social skills of the leader and perceived competence (r = 0.34, p < 0.01). These relationships are positive, which indicates an augmentation or decrease in one of the variables enacts an augmentation or decrease of the other variable in the same direction. This means that the three dependent variables that

combine into the leader perception are interrelated and when one of the three leader perception variables increase or decrease, the other variables change in the same direction.

Control variables

The table shows a significant relationship between age of the respondent (M = 36.77, SD = 12.19) and perceived effectiveness of the leader (r = -0.14, p < 0.05). This relationship is negative, which means that when age of the respondent increases, perceived effectiveness of the leader will decrease. In other words, the older the respondent is, the lower the

respondent’s perceived effectiveness of the leader will be. The control variable age of the respondent does not show significant correlational relationships with perceived competence and perceived social skills.

The second control variable, attractiveness of the leader, shows significant positive relationships with perceived effectiveness (r = 0.42, p < 0.01), perceived competence (r = 0.47, p < 0.01) and perceived social skills (r = 0.29, p < 0.01). All three relationships are

(40)

39 interrelated, since the direction of all correlations is positive. This means that when

attractiveness of the leader increases, perceived effectiveness, perceived competence and perceived social skills will also increase. This is in line with the results from Inesi et al. (2014). This means that the attractiveness of the leader in the picture might influence the outcome of this study.

The third control variable gender of the respondent does not evoke a significant correlative relationship with the other variables of this study (p > 0.05).

(41)

Table 1: Correlation matrix

(42)

Manipulated variables

In the survey the independent and moderating variable, power attributes and gender of the leader respectively, are manipulated. The mean scores of perceived effectiveness, perceived competence and perceived social skills of the manipulated variables gender of the leader and power attributes are displayed in Table 2.

As can be seen in Table 2, leaders are generally rated higher on perceived

effectiveness and perceived competence when the leader displays power attributes (ΔM = 1.26 and ΔM = 1.26 respectively). With regard to perceived social skills, however, no differences are found (ΔM = 0.00). This suggests that generally leaders are rated similar on their social skills as perceived by their followers, whether they display power attributes or not.

Furthermore, when comparing the genders of the leaders, no differences are found with regard to perceived effectiveness for male and female leaders (ΔM = 0.00). The perceived

competence shows slightly higher results, on average, for males than for females (ΔM = 0.10) and a small difference is shown in perceived social skills, that are perceived to be higher when the leader is female (ΔM = 0.76).

(43)

42

Table 2: Effect of manipulated variables on leader perception variables

Additionally, the combination of the two manipulated variables is investigated. These results are displayed in Table 3. The table shows only small differences in the mean scores for male and female leaders with power attributes on perceived effectiveness and perceived competence (ΔM = 0.01, higher for males and ΔM = 0.08, higher for males respectively). Female leaders with power attributes are rated slightly higher on perceived social skills than male leaders with power attributes (ΔM = 0.81, higher for females).

Furthermore, as can be seen in Table 3, the ratings for leaders without power

attributes show only small differences in mean scores for perceived effectiveness (ΔM = 0.08, higher for females) and perceived competence (ΔM = 0.13, higher for females). Again, female leaders are generally rated somewhat higher on perceived social skills compared to male leaders, when the leader is not displaying power attributes (ΔM = 0.72, higher for females).

(44)

43

Table 3: Effect of combined manipulated variables on leader perception variables

4.3.

Test of Hypotheses 1 and 2

The direct relationships that could be apparent in this study are the direct relationships between the variables power attributes and gender of the leader and the variable leader perception. Since the leader perception exists of the perceived effectiveness, perceived competence and perceived social skills, these will be used as the dependent variables in the analyses.

Hierarchical multiple regression analysis is performed with SPSS (IBM Corp, 2016) to test the effect of power attributes (Hypotheses 1a,b,c) and gender of the leader (Hypotheses 2a,b,c) on leader perception (perceived effectiveness, perceived competence and perceived social skills). This multiple regression is repeated three times in total, in order to examine all three dependent variables. To test Hypotheses 1 and 2 the analyses are controlled for gender of the respondent, age of the respondent and attractiveness of the leader.

Referenties

GERELATEERDE DOCUMENTEN

Again, none of the hypotheses were statistically supported, which may indicate that a higher degree of autonomy granted to a subsidiary doesn’t necessarily affect the above-mentioned

De provincie Overijssel koos dus voor het stimuleren van burgerinitiatieven door middel van een wedstrijd om vervolgens de uitvoering van de meest kansrijke initiatieven

Zijn warme correspondentie met uitgesproken antimillitarist Berdenis van Berlekom – Hij begin zijn brief met een uitbreid dankwoord voor ‘uw zo heerlijke brieven’ – zijn belofte om

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

In this matrix, bordered by the virtual enterprise life cycle and the virtual product life cycles, the business functions of analyze, design deploy and operate are

The second hypothesis stated that power, gender and gender stereotype activation interact to influence impostor feelings, such that power is associated with increased

Based on the approach inhibition theory and former research about the relationship between power and risk-taking, we expect that leader power is associated with increased

various project characteristics, but rather will elaborate on earlier work by Bu-Bshait (2,3) in identifying which project ~haracteristics, as stated in Table 1, significantly