• No results found

Physical height in relation to leadership styles : a closer look at the height-leadership association through self-perceived height dissatisfaction and gender

N/A
N/A
Protected

Academic year: 2021

Share "Physical height in relation to leadership styles : a closer look at the height-leadership association through self-perceived height dissatisfaction and gender"

Copied!
46
0
0

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

Hele tekst

(1)

Physical Height in relation to Leadership

Styles: A closer look at the height-leadership

association through self-perceived height

dissatisfaction and gender

Carlo Mureau

11145021

Master Thesis

Business Administration (MSc)-Leadership & Management

Amsterdam Business School

Academic year: 2015/2016 Supervisor: Jill Knapen Second Reader: Wendelien Van Eerde Amsterdam, June 2016

(2)

2

Statement of Originality

This document is written by Student Carlo Mureau (11145021), 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)

3

Table of content

Abstract ... 4 Chapter 1 Introduction ... 5 Chapter 2 Theoretical Framework ... 8 2.1 Physical Height ... 8 2.2 Leadership styles ... 9 2.3 Gender ... 12 2.4 Self-perceived height dissatisfaction ... 14 2.5 Conceptual model ... 15 Chapter 3 Method ... 19 3.1 Design ... 19 3.2 Sample ... 20 3.3 Measures ... 22 3.3.1 Demographics ... 22 3.3.1 Circumplex Leadership Scan (CLS) ... 23 3.3.2 Height Satisfaction Questionnaire (HSQ) ... 24 Chapter 4 Results ... 26 4.1 Analysis & Findings ... 26 4.1.1. Multivariate analysis of Covariance (MANCOVA) ... 28 4.1.2. Regression analysis ... 32 4.2. Link with hypotheses ... 32 Chapter 5 Discussion ... 35 5.1 Key findings ... 35 5.2 Theoretical and practical implications ... 36 5.2 Limitations & Future Research ... 37 References ... 41

(4)

4 Abstract The current study aims to contribute to the knowledge concerning the association between physical attributes and leadership. Previous research indicates that the physical height-leadership association may contain unobserved factors (Lindqvist, 2012). Based on the existing literature, the current study investigates potential unobserved factors. The potential unobserved factors, in the physical height-leadership association, for the current study are self-perceived height dissatisfaction and gender. Through the use of a survey, quantitative data was collected amongst an international sample of leaders varying in gender, age, sector, country and educational level. In order for the physical height-leadership association to be measured, two verified questionnaires were used. The Circumplex Leadership Scan (Redeker, De Vries, Roukhout, Vermeren & De Fruyt, 2014) was used to measure leadership and the Height Satisfaction Questionnaire (Knapen & Pollet, 2016) was used to measure height and self-perceived height dissatisfaction. The results of the current study indicated that physical height is negatively related to self-perceived height dissatisfaction. Furthermore, the results showed that self-perceived height dissatisfaction is positively related to Yielding-, Withdrawn-, Distrustful- and Authoritarian leadership. Against expectations, we did not find a direct effect of physical height on leadership styles, nor did gender affect any of the above-mentioned relationships. Implications of the results are discussed. Keywords: Leadership, Leadership styles, Physical Height, Self-perceived height, Height dissatisfaction

(5)

5 Chapter 1 Introduction On the 18th of June 1815, Napoleon Bonaparte was defeated in the battle at Waterloo. This event marked the ending of an era where the short French general reigned large parts of Europe. During this period, Europe achieved great things and grew into the civilisation we know nowadays. Under the leadership of Napoleon, the infrastructure began to take shape and a central metric system was introduced in Europe. Up until this day, Napoleon is still seen as one of the most influential leaders the world has ever seen (Lyons, 1994). Mankind has always looked up to great leaders like Napoleon and told stories about them. Some leaders appear to be more influential than others. In the beginning of the 20th century, this resulted in the emergence of a field of research that focuses on leadership (Jago, 1982). Throughout the years, leaders and their leadership have been studied in different settings (Burns, 1978; Bass, 1985; Redeker et al. 2014). Although this has led to a variety of views and theories on leadership, most studies are in agreement that leadership plays a critical role in the effectiveness of an organisation (Yukl, 2010; Bennis, 2003). The early research on leadership focussed mainly on leadership traits (Stogdill, 1948). This focus has shifted from leadership traits towards styles of leader behaviour (Czarniawska-Joerges & Wolff, 1991). Effective leaders are distinguished from ineffective leaders by their behaviour when interacting with followers and potential followers (Jago, 1982). Currently, a lot of research focuses on different patterns -or ‘styles’- of leader behaviour and their characteristics (Jago, 1982). Some common leadership styles are transformational leadership, transactional leadership, charismatic leadership and narcissistic leadership (Redeker et al. 2014). Napoleon Bonaparte was a highly effective and successful, yet very short leader. One of the areas within the leadership research domain focuses on physical attributes of leaders, for instance their weight, height or physical attractiveness (Murray & Schmitz,

(6)

6 2011; Lindqvist, 2012; Re, DeBruine, Jones & Perrett, 2013). Within this line of research there is consensus that tall men are more likely to ascend into leadership positions than shorter men are (Judge & Cable, 2004; Stogdill, 1948). This is because tall men are perceived as more capable (Hensley, 1993), persuasive (Young & French, 1996), and impressive (Kurtz, 1969). The association works both ways, people that occupy high status positions, are perceived as taller (Lechelt, 1975; Wilson, 1968). On the other hand, research has shown that tall individuals, particularly men, are perceived as better leaders (Roberts & Herman, 1986). This indicates the potential influence of gender. According to Lindqvist (2012), it is open to question whether the height-leadership association contains unobserved factors. The current study aims to identify one of these unobserved factors in the concept of self-perceived height. As illustrated above, patterns of leader behaviour and leadership styles are currently the centre of attention. A commonly used concept by psychologists to explain behaviour is the self-concept. Markus & Kitayama (1991) argue that there are a lot of divergent views concerning the self. One of these views is the interdependent view of the self, where the self-in-relation-to-others is most important (Markus & Kitayama, 1991). In this view, the self is connected to the social context. According to Markus and Wurf (1987), height is a way to compare yourself to your social environment. In line with this, Knapen and Pollet (2016) found that the level of satisfaction with someone’s self-perceived height is an important factor of the feelings and behaviours of individuals in height-related social contexts. Consequently, self-perceived height could play a critical role in influencing leaders’ behaviour. However, the possible association between one’s self-perceived height and leadership behaviour has not yet been studied. The current study focuses on the association between one’s height, self-perceived height and his/her behaviour as a leader, in order to contribute to the knowledge concerning the relationship between physical attributes and leadership. The results of this study are expected to be useful for both companies and leaders in practice. On the one hand, companies in the process of developing future leaders may use the results. On the other hand, the result may raise

(7)

7 more awareness amongst (future) leaders concerning the origin of their leadership style. The following section consists of a theoretical framework for the current study. Each concept is defined and an overview of the literature is presented. Next, the hypotheses based on the literature are presented. The theoretical framework is followed by a description of the research design, sample and measures. Next, the data analysis will be described and the results of the current study will be presented. Finally, the discussion-chapter will sum up the findings of the study, and discuss implications and limitations.

(8)

8 Chapter 2 Theoretical Framework This section will focus on the various constructs that are part of this study. An overview of the current state of the literature for each construct is presented. The first construct is physical height. Second, an overview is presented for Leadership styles. The third construct is gender. In the fourth section, the self-perceived height dissatisfaction is discussed. The final section of this chapter presents the conceptual model and hypotheses for the current study. 2.1 Physical Height Physical attributes like weight, height and attractiveness make individuals stand out from the crowd. Physical attributes are crucial in the first impression of a person. When it comes to selecting leaders, humans look for cues that these individuals are physically ‘fit to lead’; this is related to properties like physical stamina, energy and health (Bass & Bass, 2008). According to et al. (2013), physical height is a good indicator of the above-mentioned fitness. A good example of the link between height and ‘fitness to lead’ is the fact that US presidents were found to be taller compared to other men in the same birth cohort (Stulp, Buunk, Verhulst & Pollet, 2013). Height has always been a way of comparing ourselves with our social environment (Markus & Wurf, 1987). Over the last centuries, the physical height of the population has increased as a result of improvements in nutrition, economic status, hygiene and health (Schönbeck et al, 2012). However, height variations will remain to exist as a result of differences in genetic background and socioeconomic status (Schönbeck et al, 2012). The variations in height were studied in relation to outcomes like social dominance (Melahmed, 1992) and earnings (Case & Paxson, 2006). The positive association between height and cognitive ability explains this height premium in earnings (Case & Paxson, 2006). Other studies found that tall men are perceived as more capable (Hensley, 1993), persuasive (Young & French, 1996) and impressive (Kurtz, 1969). Moreover, Hood (1963) linked physical height with happiness. The results indicated that tall men score lower on depression and feelings of inferiority.

(9)

9 All of the studies mentioned above support the ‘taller is better’-reasoning. Scholars have challenged the ‘taller is better’-reasoning in the past. For example, Hensley and Cooper (1987) recognise that taller than average persons indeed enjoy advantages in acquiring certain positions. In terms of the performance of these taller individuals, their actual performance is not significantly better compared to their shorter colleagues. Hensley and Cooper (1987) found no systematic evidence that taller individuals perform better in the academic-, police- and sales sector. Perhaps this is the case in other sectors or types of functions, for example leaders. 2.2 Leadership styles The leadership research domain we know today embodies a lot of different areas and has gone through a lot of changes over the years. Initially, leadership research focussed mainly on traits of a leader (Stogdill, 1948). In the 1960s, the focus shifted from the traits to behavioural styles and situational factors (Czarniawska-Joerges & Wolff, 1991). Nowadays, the focus is more on ‘New Leadership’, with emphasis on different leadership styles and topics like charisma and vision (Den Hartog & Koopman, 2001). When it comes to defining leadership, there is a lot of variation in terms of focus on abilities, personality traits, cognitive versus emotional orientation, relationships, group versus individual focus and appeal to collective versus self-interest (Den Hartog & Koopman, 2001). According to Bryman (1992), most of the existing definitions have three elements in common: group, goal and influence. Bryman (1992) therefore defines leadership in terms of processes of social influence whereby a leader steers members of a group towards a goal. This is a broad definition, which will suffice in the current study, since the study focuses on different leadership styles in particular. Throughout the years a lot of research has focussed on various styles of leadership (Burns, 1978; Bass, 1985; Howell & Avolio, 1993). Over the past 30 years, the transformational-transactional leadership theory has been subject of a large number of studies (Judge & Picollo, 2004). According to Burns (1978), these two leadership styles differ in terms of what followers and leaders can offer one another. Transactional leadership focuses on an exchange of resources between the leader and follower (Burns,

(10)

10 1978). Transformational leadership, on the other hand, goes beyond the short-term goals and aims to fulfil higher order intrinsic needs (Burns, 1978). According to Bass (1985), transactional – and transformational leadership are not opposite ends of a continuum; the best leaders are both transactional and transformational (Bass, 1985). Apart from the distinction between transaction – and transformational leadership, many other leadership styles are distinguished, like charismatic leadership (Bono & Ilies, 2006; Shamir et al, 1993) and authentic leadership (Gardner, Avolio, Luthans, Douglas & Walumbwa, 2005; Shamir & Eilam, 2005). Although most of these studies focus on the positive outcomes of leadership, research on the ‘dark side’ of leadership has recently increased (De Hoogh & Den Hartog, 2008; Hogan & Hogan, 2001). Examples of ‘dark’-leadership styles are Laissez-faire leadership, Machiavellian leadership and narcissistic leadership. First, Laissez-faire leadership is related to passive or absent leadership (Avolio & Bass, 1991). Second, Machiavellian leadership is named after the Italian politician Nicollo Machiavelli (Dahling, Kuyumcu & Librizzi, 2014). Machiavellian leaders are characterised by deception, emotional bitterness, blame shifting and manipulation of other in their own advantage (Wai & Tiliopoulus, 2012). Third, narcissistic leaders seek confirmation of their superiority, and find motivation in being admired by their followers (Judge, Scott & Ilies, 2006). Scholars have identified a growing need to integrate these positive and negative leadership styles (DeRue, Nahrgang, Wellman & Humphrey, 2011). Subsequently, Redeker et al. (2014) constructed the leadership circumplex, which captures many of the developed leadership models. The leadership circumplex is shown in figure 1.

(11)

11 Figure 1: Leadership Circumplex (Redeker et al. 2014) The integration of the variety of leadership styles is achieved by distinguishing two dimensions: ‘agency’ and ‘communion’. ‘Agency’ relates to the degree to which an individual is a differentiated person, this manifests itself in strivings for power and mastery. ‘Communion’, on the other hand, relates to the condition of being part of a larger entity, both socially and spiritually. ‘Communion’ manifests itself in solidarity within the larger entity and striving for union. Redeker et al. (2014) also developed a tool to measure these leadership styles, the Circumplex Leadership Scan (CLS), which is used in the current study. The CLS has proven to be an adequate instrument to measure leadership (Redeker et al., 2014). The use of a circumplex enables the researcher to measure combinations of the two dimensions. Once this measurement is done, an individual’s leadership behaviour can be placed on the circumplex. Each octant represents a different leadership style, for

(12)

12 example Inspirational leadership or Distrustful leadership. In Table 1, an overview of the leadership styles distinguished by Redeker et al (2014) is presented. Leadership style (octants) Characteristics Coaching Appreciates subordinates, stimulate subordinates through positive communication and listening to their opinion; Inspirational Persuade subordinate through clear vision, decisive when performance- and organisational problems arise, motivate subordinates to perform optimally; Directive Strict, try to reach success competitively, monitor and correct subordinates; Authoritarian Force subordinates to obey, harsh, not open to criticism; Distrustful Suspicious over motives of subordinates, quick and negative judgement, distant from subordinates; Withdrawn Personally and professionally absent, avoid confrontations and responsibility, act too late when problems arise; Yielding Flexible interaction with subordinates, hesitant to provide guidance, avoid being centre of attention, subordinate’s interest over company’s interest; Participative Include subordinates in all processes, accept and incorporate subordinate’s propositions, show understanding of feelings and emotions of subordinates. Table 1: Leadership Circumplex Octants (Redeker et al. 2014) 2.3 Gender From a young age, children learn about gender differences, stereotypes and roles (Eagly, Wood & Diekman, 2000). Gender is one of the central components of the self-concept (Ruble & Martin, 1998). A crucial part of the self-concept is the match of the actual

(13)

13 gender of an individual with the traits associated with the specific gender stereotype (Bem, 1981). For males this is masculinity; associated traits are aggressiveness, independence and dominance (Eagly et al., 2000). For females, on the other hand, this is femininity; associated traits are warmth, compassion and understanding (Eagly et al, 2000). The following phrase from Eddleston, Veiga and Powell (2006) demonstrates these gender stereotypes: “People accommodate the roles they are expected to emphasize (family for women, work for men) by developing skills that enhance success in these roles (communal for women, agentic for men) “ (p.438, 2006) The phrase above elaborates on a distinction between communal behaviour for women and agentic behaviour for men. This distinction is similar to the dimensions in the leadership circumplex (Redeker et al., 2014). Two dimension, ‘communion’ and ‘agency’ split the leadership circumplex. Since the dimensions from the leadership circumplex (Redeker et al., 2014) overlap with the skills that enhance success in gender stereotypical roles (Eddleston et al., 2006), gender is expected to influence the leadership style of a leader. Eagly and Johnson (1990) confirm this gender difference in leadership, they found that women have a tendency to adopt a more democratic or participative leadership style and a less directive and autocratic leadership style compared to men. However when it comes to female leaders in specific, Kirchmeyer (2002) suggested that the women that occupy a role of a leader or manager usually score high on masculinity, as a managerial role is traditionally a more masculine role. In terms of effective leadership, Burke and Collins (2001) argue that neither males nor females are superior; based on this finding gender neutral hiring is advised. In general the effect of height on leadership perceptions tends to be stronger for men than for women (Blaker et al., 2013; Re et al., 2013). This is probably due to the fact that status attainment, like supervision duties, power, influence, authority, earnings and promotional opportunity, is more important for men (Eddleston et al., 2006). Women, on the other hand, give more priority to interpersonal relationships. This is related to

(14)

14 working with other people, being a good co-worker and being able to make friends (Eddleston et al., 2006). 2.4 Self-perceived height dissatisfaction A common rule in nature is the bigger, the more dangerous (Freedman, 1979). Humanity has come a long way since our ancestral hunter-gatherer times. However, research has shown that we still pay attention to, and are influenced by, biological traits, like physical stature, fitness and strength (Murray & Schmitz, 2011). Indeed, there is a positive association between physical height and military rank (Masur, Masur, & Keating, 1984), income (Judge & Cable, 2004) and ascendance into managerial positions (Lindqvist, 2012). In addition, according to Wilson (1968) the way others perceive one’s height is influenced by his/her status, individuals with a higher academic status were perceived as taller compared to individuals with a lower academic status. Besides the perception of others, the individuals perceive their physical height in their own way. Physical height is a way to compare yourself to your social environment (Markus & Wurf, 1987). The comparison of physical height with the social environment is part of the self-perception. The self-in-relation-to-others, the interdependent view, influences one’s behaviour (Markus & Kitayama, 1991). Besides the behaviour of an individual, self-perception of height is linked to personality and self esteem. According to Melamed (1992), the correlation between height and personality is partly explained by mechanisms of self-perception and feedback. On the other hand, Prieto and Robbins (1975) conclude that one’s self-perceived height is for instance influential in his/hers self-esteem. These results indicate that it is not the actual height but the self-perceived height that influences behaviour. As indicated by the section above, self-perception of height is of great importance in explaining the behaviour, self-esteem and personality of an individual. In line with this finding, the level of satisfaction with someone’s self-perceived height is an important factor of the feelings and behaviour of individuals in height-related social contexts (Knapen & Pollet, 2016). According to Jacobi & Cash (1994), both males and

(15)

15 females from a Western sample indicate that they would rather be taller. Especially men in stereotypical masculine gender roles feel a need to be taller, as it is a socially

desirable masculine physical characteristic (Bogaert & McCreary, 2011). The position of a leader is a stereotypical masculine gender role (Kirchmeyer, 2002). The men in these stereotypical masculine gender roles feel the pressure, to fit the socially desirable masculine physical characteristics, to such a degree that they are inclined to distort this aspect of their body size (Bogaert & McCreary, 2011). This indicates the potential influence of satisfaction with one’s self-perceived height on the behaviour of a leader. In the last decade, several studies have focussed on the effects of self-perceived height dissatisfaction with mixed results (Knapen & Pollet, 2016). For the current study, we expect self-perceived height dissatisfaction to mediate the relationship between height and leadership styles. As leadership is a stereotypically masculine gender role, it is accompanied by socially desirable masculine physical characteristics, being tall is one of them. Therefore the self-perceived height dissatisfaction is expected to influence the behaviour of a leader. 2.5 Conceptual model The current study focuses on the association between one’s height, self-perceived height dissatisfaction and his/her behaviour as a leader, in order to contribute to the knowledge concerning the broader relationship between physical attributes and leadership. This research objective translates to the conceptual model presented in Figure 2.

(16)

16 Figure 2: Conceptual model Based on the literature and the conceptual model in figure 2, the following hypotheses are formulated: H1A: Physical height is positively related with leadership styles on the left side of the Leadership circumplex (Low on ‘communion’). According to Murray and Schmitz (2011), a taller male is more likely to deem himself fit to lead. This is explained by an increased sense of efficacy. Moreover, research has shown positive relationships between height and leader effectiveness and leader emergence (Roberts & Herman, 1986; Hensley & Cooper, 1987). In turn, Bass (1997) links leadership effectiveness to leadership styles, transformational leadership in particular. Moreover, according to Melamed (1992), taller people are more likely to be assertive and dominant or suspicious and hard to fool. Since the studies above indicate a relationship between height and leadership effectiveness on the one hand and leadership styles and leadership effectiveness on the other hand, we expect a direct, positive relationship between height and leadership styles on the left side of the circumplex. Based on this conclusion, the leadership style for taller leaders is more likely to be at the left side of the leadership circumplex. This would mean that taller leaders score higher on Directive-, Authoritarian-, Distrustful- and Withdrawn leadership. Physical Height Self-perceived Height Dissatisfaction (HD) Leadership style Gender

(17)

17 H1B: The direct effect from physical height on differences in leadership style is moderated by gender, in such a way that this effect is stronger for men. Melamed (1994) studied differences between men and women concerning career success. She concluded that for women a number of career paths are blocked in advance on the basis of their gender. This results in women not being promoted into the leadership positions. However, for men this isn’t the case, and consequently there is more room for other types of biases to influence their career success. These other types of biases could be physical attractiveness or physical height. Moreover, Stulp, Buunk, Verhulst and Pollet (2015) found that the positive relation between height and dominance was more significant for men. In the current study, we expect the relationships between physical height and leadership styles on the left side of the circumplex to be stronger for men compared to women. H2: Self-perceived height dissatisfaction is positively related with leadership styles on the left side of the Leadership circumplex (Low on ‘communion’). Leadership positions are stereotypically masculine gender roles (Kirchmeyer, 2002). Consequently, leaders are expected to feel the pressure, to fit the socially desirable masculine physical characteristics (Bogaert & McCreary, 2011). The leader is expected to adjust his behaviour as a result of this pressure. Adler (1956) described this adjustment of behaviour as the napoleon complex, where a leader’ short stature makes him feel inadequate and adopt overaggressive and dominant behaviour to compensate for a lack of physical height. This overaggressive and dominant behaviour of leaders translates to the left side of the leadership circumplex (Redeker et al., 2014), which means a low score on communion. H3A: Physical height is negatively related with self-perceived height dissatisfaction. Physical attributes influence one’s self-perception and the others’ perception of the individual. Melamed (1992) argues that, due to their physical height, short people are more likely to receive less positive feedback and feel inadequate. When it comes to leaders in specific, Duguid and Goncalo (2012) argue that people occupying powerful positions perceive themselves as significantly taller. For this study the focus is on

(18)

18 people in leadership positions, these positions are arguably powerful positions. . When it comes to selecting leaders, followers look for cues that these individuals are physically ‘fit to lead’ (Bass & Bass, 2008). Height is related to an individual appearing ‘fit to lead’; therefore a leader would like to be tall. In the current study the expectation is that physical height and self-perceived height dissatisfaction are negatively related, such that shorter leaders are more dissatisfied. H3B: The effect from physical height on self-perceived height dissatisfaction is moderated by gender, in such a way that the effect is stronger for men. Both males and females from a Western sample indicate that they would rather be taller (Jacobi & Cash, 1994). However, the proportion of males (62%) that would prefer to be taller is larger than the proportion of females (46%). This indicates that men are more likely to be dissatisfied with their height. Arguably, this is due to the traditional gender stereotypes and roles for males. In the stereotypes for males, traits like power and dominance are important (Eagly et al., 2000). Being taller shows natural superiority (Freedman, 1979). Since natural superiority is arguably of great importance for leader, the relationship between physical height and self-perceived height dissatisfaction is expected to be stronger for male leaders compared to female leaders.

(19)

19 Chapter 3 Method The previous chapter elaborated on the literature dealing with relevant concepts for the current study. A theoretical framework was provided to explain the possible relationship between physical height and various leadership styles portrayed. This chapter elaborates on the research design, sample, data collection, and data analysis, described in the paragraphs Design, Sample and Measures. 3.1 Design The current study was part of a project hosted by the Amsterdam Business School. The researchers were two Master students in Business Administration. The current study was exploratory in nature, as we aimed to explore the relationship between physical height and leadership styles, and how different constructs influenced this relationship. This was done by a cross-sectional survey design, the most common method used in this line of research. The cross-sectional survey design provides the researcher with the opportunity to contact a large number of potential respondents in a rather short period of time. The downside to the cross-sectional survey design is that a lengthy questionnaire may cause a low response rate (Lindell & Whitney, 2001). The survey was distributed digitally among the respondents, using the online software Qualtrics (Qualtrics, 2016). A short introduction to the study was sent to the respondents in combination with the hyperlink to the survey. In order to assure confidentiality, the survey started with an informed consent. The informed consent stressed the anonymity of the respondents and the voluntary nature of their response. After giving their informed consent, participants proceeded to the actual questions for the study. The main body of the survey consisted of demographic questions, the CLS (Redeker et al., 2014) and the Height Satisfaction Questionnaire (HSQ) (Knapen & Pollet, 2016); each of these sections will be discussed further on in this chapter. At the end of the questionnaire, a debriefing was included. In this section, respondents could choose to receive a summary of the results of the study.

(20)

20 In order for potential respondents to be incentivised to complete the survey, a (optional) leadership style report was sent afterwards. Since the respondents weren’t paid for their contribution, this feedback report could incentivise them (Redeker et al., 2014). Within four weeks after completing the survey, the respondent received a scorecard with their scores on the various leadership styles as identified by Redeker et al (2014). This scorecard was based on the input provided by the respondent. Moreover, respondents could choose to be informed about the results of the current study by providing their mail-addresses at the end of the survey. After the questionnaire was closed, a total of 150 reports on the leadership styles were distributed among the respondents within 2 weeks. Thus, this option was highly appreciated by the respondents. 3.2 Sample The population of the current study was the total of all individuals who currently occupy, or have occupied, a leadership position. This could be either a business or work related position, but it could be any other position as well. There was no clear overview or list of the entire population; therefore a non-probabilistic sample was used. Moreover this population was rather hard to contact, since they are mostly in the higher levels of companies, therefore convenience sampling is used. The current study strived for as many respondents as possible, however a minimal sample of 200 respondents is considered large enough for analyses. According to Finkelstein (1992), a response rate around 40% is considered good, given the level of manager questioned and the sensitivity of the topic. A total of 225 leaders started filling out the questionnaire. However, 49 respondents either did not complete the survey or not in an appropriate manner, therefore they were removed before analyses. The remaining sample of 176 responses was suited for analysis in the current study. The location of the respondents is presented in Table 2.

(21)

21

Country Number of Respondents Percentage of sample

Belgium 3 1,70% Denmark 8 4,55% Ethiopia 1 0,57% Germany 5 2,84% India 2 1,14% Israël 1 0,57% Lithuania 1 0,57% NL 97 55,11% Spain 1 0,57% Sweden 43 24,43% Switzerland 1 0,57% Taiwan 5 2,84% UK 6 3,40% USA 2 1,14% TOTAL 176 100% Table 2: Location of Respondents The current study had an international sample, the majority of the respondents were either located in the Netherlands (55,11%) or in Sweden (24,43%). However, there were a few respondents from other locations, for example the United Kingdom (3,40%), Switzerland (0.57%), Germany (2.84%), and Denmark (4.55%). A total of 49 respondents (27,8%) was female, the other 127 respondents (72,2%) were male. This uneven representation in gender was expected, as some career paths may be blocked in advance for women (Melahmed, 1994). The gender distribution is shown in Figure 3. The average age of the respondents is 46.6 years old (SD=11.49), the ages varied from 19 years old up to 70 years old. In total 86,4% participants had a high level of education, either University of Higher Vocational Education. The sectors in which the respondents operate are displayed in Figure 4. In terms of physical height, the mean height was 183.77 cm (SD=7.08) for males, and 170.13 cm (SD=6.81) for females.

(22)

22 Figure 3: Figure 4: Gender distribution Sector distribution 3.3 Measures The questionnaire contained the following sections: Introduction and informed consent, demographics, the CLS (Redeker et al, 2014), the HSQ (Knapen & Pollet, 2016), and a debriefing. The questionnaire was available in both Dutch and English, to ensure sufficient understanding from the respondents’ perspective. The CLS (Redeker et al., 2014) and HSQ (Knapen & Pollet, 2016) were already available in Dutch and English. Various readers checked the remaining sections and adjustments before the survey was published. 3.3.1 Demographics The first section of the questionnaire regarded the respondents’ demographics. This section addressed the respondent’s gender, age, physical height, education, sector, job description and leadership role. Due to limited resources and the large sample, actual height couldn’t be measured. Instead self-reported height was used. According to Spencer, Appleby, Davy and Key (2002), self-reported height and weight are valid data male Female Finance/Banking Communication Education Government Healthcare Hospitality Insurance Manufactoring Retail Transportation/ Logistics Other

(23)

23 for epidemiological studies. Moreover, according to Kuczmarski, Kuczmarski and Najjar (2001), self reported measures of height and measures only have limitations for older adults of 60 years and older. This group doesn’t overlap much with the population of leaders. Therefore we expected that this wouldn’t cause problems in the current study. Nonetheless, independent measurement of actual physical height would have been preferable. 3.3.1 Circumplex Leadership Scan (CLS) The CLS (Redeker et al., 2014) is a reliable, valid measure for leadership behaviours; it provides a comprehensive and integrative model of leadership styles. The CLS was designed in combination with the Leadership Circumplex (Redeker et al., 2014). This Leadership Circumplex is shown in Figure 1. The Leadership Circumplex consists of eight octants; each octant translates to a different leadership style. The CLS consists of 116 items, on a 5-point Likert scale from never to always. Each octant/leadership style is measured by an equal number of items. The variation between items for each octant is similar as well. Table 3 shows examples of items for each octant. Leadership style (octants) Example items Coaching “Asks for the staff’s opinion” Inspirational ‘‘Sets clear objectives for the staff’’ Directive ‘‘Supervises the work of the staff members carefully’’ Authoritarian “Avoids friendly relationships’’ Distrustful ‘‘Does not allow staff members to organize their work themselves’’ Withdrawn ‘‘Does not take responsibility’’ Yielding ‘Is able to subordinate the company’s interests to the staff’s interests’’ Participative ‘Allows staff members to do their work their own way’’ Table 3: Example items CLS The following statement accompanied the items in the CLS-section of the survey:

(24)

24 ‘Please indicate to what extend the following items apply to you’ The measure has been found to have good internal validity, with cronbach’s alpha varying between .70 and .83 for items on each octant (Redeker et al., 2014). Reliability for the current study for each octant varied between .69 and .86. The means for each octant vary between 2,03 for distrustful leadership and 4,19 for coaching leadership. Finally, the standard deviations are within the range of .33 and .45. An overview of reliabilities, means and standard deviation for each octant is displayed in Table 5 in the result-chapter. 3.3.2 Height Satisfaction Questionnaire (HSQ) The HSQ (Knapen & Pollet, 2016) measures a respondent’s satisfaction with his/her height and consists of three subscales: Height Dissatisfaction (HD), Social Expectations of Tallness (SET), and Height Benefits (HB). Reliability has been shown to be acceptable, as the cronbach’s alpha differs between subsets, subscales and gender from .609 to .871 (Knapen & Pollet, 2016). The HSQ uses a 7-point Likert scale, varying from ‘never’ to ‘always’ for the items of the subscales HB and HD. The subscale SET uses a picture with seven gender-similar silhouettes varying in height. Examples of for each subscale are displayed in Table 4. Subscale HSQ Example Item Social Expectations of Tallness (SET) Choose the person that you would most want to look like (Pick one of seven silhouettes) Height Benefits (HB) Have your friends ever told you that you are tall? Height Dissatisfaction (HD) Have you ever felt (too) small? Table 4: Example items HSQ

(25)

25 In the current study, data was collected on all three subscales. In order to test the hypothesis based on the conceptual model for this study, the HD subscale is most useful as it fits best with the theory presented in chapter 2. In the current study, the reliability of HD is proven to be good; since cronbach’s alpha is .79. Further information regarding the mean scores and standard deviation is presented in Table 5.

(26)

26 Chapter 4 Results This chapter elaborates on the results of the current study. The results are divided into two sections. First, the analysis of the data is discussed and results are presented. This section is split up in the Multivariate Analysis of Covariance and Regression analysis. The final section links the findings to the hypotheses from the current study. 4.1 Analysis & Findings First, the data was cleaned, explored, and recoded using IBM SPSS Version 23. Reversed items where reverse scored. However, there was one problematic item that negatively influenced the reliability and wasn’t reversed, HSQ14 (“Would you like to be taller?”). After removing the problematic item, reliability was acceptable for the subscale HD: α=.79(M=12,99, SD= 5.26). The acceptable range of values for asymmetry and kurtosis is -2<x<2, in order to prove normal univariate distribution (George & Mallery, 2010). In the current study, values on skewness for each construct varied from -.039 to .997, with a standard error varying from .183 to .186. The values for Kurtosis for each construct varied from -1.017 to 1.227, with a standard error varying from .364 to .369. Consequently, the value for skewness and kurtosis were within the acceptable range for gender, height, the eight leadership styles, and HD. In order to quantify the intensity and meaning of the relationships between the various constructs, a correlation table is shown in Table 5. This table shows the means, standard deviations, correlations and reliability for all the constructs that are subject of the current study.

(27)

27 Table 5: Correl ation Ta ble

(28)

28 The first notable numbers in Table 5 are the reliabilities. Each construct has an acceptable reliability to continue. Next, Table 5 showed a number of high correlations between different leadership styles. As some leadership styles are relatively close to each other on the Leadership circumplex (Redeker et al. 2014) and have some things in common, a high correlation was expected. Part of the analysis was to determine the most dominant leadership style for each respondent. The majority (81%) of the participants scored highest on Inspirational/Coaching, which is on the right part of the Leadership Circumplex (Redeker et al., 2014). Gender had some more notable significant correlations; it correlated significantly with Coaching (.162*), Participative (.202**) and Directive (-.181*) leadership. For Coaching and Participative leadership style this indicated that females score significantly higher. For Directive leadership on the other hand, this indicated that females score significantly lower on this leadership style. Other interesting correlations were Height in combination with Withdrawn leadership (-.174*) and HD (-.452**). Especially the latter one, height and self-perceived height dissatisfaction, is a relationship that was expected based on H3A. Since this correlation was negative it indicates that the taller a person is, the lower the score on self-perceived height dissatisfaction. Finally, HD seemed to correlate with 5 leadership styles, Inspirational (-.197**), Yielding (.228**), Withdrawn (.347**), Distrustful (.328**) and Authoritarian (.291**). Based on the literature, correlations would be expected for Withdrawn-, Distrustful-, Authoritarian- and Directive leadership. In sum, the correlation table showed some significant relations in the conceptual model, the next section describes the next steps in analysing the data. 4.1.1. Multivariate analysis of Covariance (MANCOVA) First, a multivariate analysis of covariance (MANCOVA) was used to test the model and the hypothesized relations. The independent variables were HD, actual height, and gender, and the depended variables were the eight leadership styles. MANCOVA is designed to look at multiple dependent variables simultaneously, (Field, 2013), in the current study the eight leadership styles. In the current study there is sufficient

(29)

29 theoretical basis for combining the dependent variables into the MANCOVA, as the relationship between the dependent variables should be taken into account. Box’s test showed that the assumption for homogeneity was not met (p=.041). For this reason, a bootstrapping procedure (2,000 resamples, 95% CI) was included in order to test for robustness of the outcomes. Since Pillai’s Trace test is the most reliable test to use if you may have violated the homogeneity assumption, this test is used in reporting the analysis (Field, 2013). There was no significant effect for Height, V=.039, F(8, 159)=.816, p=.590 or Gender, V=.069, F(8, 159)=1,464, p=.174 on leadership styles. However, there was a significant effect for HD on the leadership styles, V=.193, F(8, 159)=3,198, p=.002. The univariate ANOVA’s indicate robust statistically significant effects of HD on Yielding (t(1)= 2.185; p=.030; !!!=.028; 95% CI [.001 — .028]; bootstrap: p=.029; 95% CI [.002 — .028]); Withdrawn (t(1)= 3.925; p< .001; !!!=.085; 95% CI [.013 — .039]; bootstrap: p< .001; 95% CI [.012 — .041]), Distrustful (t(1)= 3.822; p< .001; !!!=.081; 95% CI [.013 — .040]; bootstrap: p< .001, 95% CI [.012 — .041]) and Authoritarian leadership styles (t(1)= 3.627; p< .001; !!!=.073; 95% CI [.012 — .041]; bootstrap: p=.001; 95% CI [.012 — .042]). Thus, individuals who scored higher on HD, also scored higher on Yielding-, Withdrawn-, Distrustful- and Authoritarian leadership. These results were mostly in line with the model and hypothesis; when a leader scores high on self-perceived height dissatisfaction he/she is more likely to have a leadership style on the left side of the Leadership circumplex. The left side of the Leadership circumplex translates to Withdrawn-, Distrustful-, Authoritarian– and Directive leadership. The exception is Directive leadership, situated on the left side of the Leadership circumplex, which didn’t show significant relation with self-perceived height dissatisfaction. On the other hand, Yielding leadership isn’t situated on the left side of the Leadership circumplex, yet it did show to have a significant relationship with self-perceived height dissatisfaction.

(30)

30 Furthermore, in contrast to the MANCOVA, the individual ANCOVA’s indicated statistically significant effects for gender on Coaching t(1)= -1.967; p=.051; !!!=.023; 95% CI [-.238 — .000]; bootstrap: p=.001; 95% CI [-.226 — -.011], Participative (t(1)= -2.530; p=.012; !!! =.037; 95% CI [-.257 — .032]; bootstrap: p=.002; 95% CI [-.237 — -.0.55]) and Directive (t(1)= 2.406; p=.017; !!!=.034; 95% CI [-.032 — .324]; bootstrap: p=.006; 95% CI [.049 — .306]). These results are in line with the significant correlations in Table 4. In the current study, females scored higher on Coaching- and Participative leadership, compared to men. On the contrary, female respondents scored lower on Directive leadership, compared to male respondents. A Discriminant analysis was conducted to follow up on the MANCOVA. HD was split up in 3 categories: Low HD (x<mean-1SD), Moderate HD (Mean–1SD < x < Mean+1SD) and High HD (x>Mean+1SD). The discriminant analysis showed two discriminant functions for this model. The discriminant function plot (Figure 5) indicated that the first function differentiates the High HD from the rest. The second function differentiates the group Low HD from the other groups. The first discriminant function explained 85,9% of the variance, canonical R²=0.119. Function two, on the other hand, explained only 14,1% of the variance, canonical R²=0,022. Figure 5: Plot Discriminant Functions

(31)

31 Neither one of the discriminant functions, nor combined, significantly differentiated the groups with different HD categories. For the combination of the discriminant functions this is shown in Λ=.862, χ² (16)=24,506, p=.079. Removing the first function resulted in Λ=.978, χ² (7)=3,628, p=.821 for the second discriminant function alone. This non-significance of Wilk’s Lambda shows that both discriminant functions have a low predictive capability. The discriminant analyses indicated that only 52% of the outcome variable could be classified. The MANCOVA showed that only HD had a significant effect on the different leadership styles. The univariate ANOVA’s suggested that the effect was robustly statistically significant for Yielding-, Withdrawn-, Authoritarian- and Distrustful leadership styles. However, the discriminant analysis suggests the group separation wasn’t significantly for either one or a combination of the discriminant functions. This could be due to the fact that not all assumptions of the discriminant analysis, like multivariate normality, homoscedasticity, multicollinearity and independence were met (Field, 2013). Moreover, the scale of HD was split up in three groups for the discriminant analysis; arguably splitting the scale in three groups instead of one continuum caused a loss of information. The literature, as discussed in chapter 2, suggested that gender would moderate the relations in the model. For the current study it is interesting to find out if gender moderates the relation between HD and the leadership styles. The existence of this moderating effect is tested by PROCESS in IBM SPSS Version 23. To test this Model 1 was used, the dependent variable (Y) was each leadership style, the independent variable (X) was HD and the moderator (M) was gender. For the eight models that were tested, one for each leadership style, the levels of significance varied from .060 to .907. Consequently, none of the moderation effects is significant. This means that gender is not a condition under which the relationship between HD and Leadership styles differs.

(32)

32 4.1.2. Regression analysis Since HD has a statistically significant effect on leadership styles, it is interesting to look into the relationship between physical height and HD. Regression analysis was done to look into this relationship. The effect for physical height on HD is R²=.250, F(1,173)=57,585, p<.000, with an unstandardized B of - 0,513. This means that the score on self-perceived height dissatisfaction becomes lower when a participant is taller (higher score on physical height). As stated in H3A gender was expected to moderate this relationship. In order to test this, an interaction term was constructed. First, both gender and height were centred on their respective mean by subtracting the mean from each score (Aiken & West, 1991). These variables were then multiplied to get the interaction term. First, Gender and Height were included in the regression analysis. These variables accounted for a significant amount of variance, R²=.268, F(2,172)=31,504, p<.001. Next, the interaction term of Gender and Height was added to regression model, R²=.275, F(3,171)=21,581, p<.000. This interaction term caused ΔR²=.007, p=.217, this indicated that there was no potentially significant moderation by gender. Finally a separate regression analysis was done to test if Gender had a direct relationship with HD. This resulted in R²=.009, F(1,174)=1,668, p=.198, this indicated that the effect of gender on HD, by itself, isn’t significant either. 4.2. Link with hypotheses In advance of the current study, a number of hypotheses were constructed based on the relevant literature. This section aims to connect the findings of the current study with these hypotheses. The first two hypotheses where regarding the direct effect of Physical Height on Leadership styles, as shown below: H1A: Physical height is positively related with leadership styles on the left side of the Leadership circumplex (Low on ‘communion’).

(33)

33 H1B: The direct effect from physical height on differences in leadership style is moderated by gender, in such a way that this effect is stronger for men. As explained in the previous section, the MANCOVA showed that this effect was not significant, neither was the potential direct effect from gender on leadership styles. This causes for both H1A and H1B to be rejected based on the results in the current study. This means that physical height doesn’t directly influence the behaviour portrayed by a certain leader. Based on previous studies on the physical height-leadership association, this was expected in advance. The second group of hypotheses were regarding the possible mediation from self-perceived height dissatisfaction. H2: Self-perceived height dissatisfaction is positively related with leadership styles on the left side of the Leadership circumplex (Low on ‘communion’). The MANCOVA showed significant results on the relationship between HD and Leadership styles in general. The results show robust statistically significant relations for HD on Yielding-, Withdrawn-, Distrustful- and Authoritarian Leadership. The left side of the Leadership circumplex, as stated in H2, translates to Authoritarian-, Directive-, Distrustful- and Withdrawn leadership. Therefore, on 3 out of the 4 leadership styles H2 is confirmed by the results. Based on the current study, Directive leadership is not significantly influenced by HD, where it was expected to in advance. On the other hand, Yielding turned out to be significantly influenced by HD; this was not expected in advance. Directive leadership and Yielding leadership are opposites on the Leadership circumplex. Where Directive leadership is related to a high score on the agency-dimension, Yielding leadership scores low on the agency-dimension. As a result of this, the area (Red) of the Leadership circumplex that is related to self-perceived height dissatisfaction is slightly turned to the left, as indicated in Figure 6.

(34)

34 Hypothesis Actual Figure 6: Leadership Circumplex area related to Self-Perceived Height Dissatisfaction Additionally, the discriminant analysis didn’t show significant results. Therefore, the exact nature of the relationship between HD and the leadership styles remains unclear. H3A: Physical height is negatively related with self-perceived height dissatisfaction. H3B: The effect from physical height on self-perceived height dissatisfaction is moderated by gender, in such a way that the effect is stronger for men. The regression analysis as discussed above indicated a significant negative relationship for physical height on HD. This means that taller leaders are in general less dissatisfied about their height, this is conforming H3A. This relation accounted for 25% of the variance in HD, it is relevant to see what other factors influence HD. As stated in H3B, Gender could moderate the relationship between physical height and self-perceived height dissatisfaction. In doing so, this may cause an increase in the explained variance for HD. However, based on the on the analysis above this wasn’t the case. This causes H3B to be rejected. Moreover, Gender alone is not proven to be of significant influence on HD.

(35)

35 Chapter 5 Discussion This section interprets the key findings of the current study, and discusses the implications from both a theoretical – and practical perspective. In addition, the limitations of the current study are outlined and possible directions for future research will be provided. 5.1 Key findings In the current study, we aimed to investigate the relationship between physical height and leadership styles. A direct positive effect of physical height on leadership styles on the left side (low on ‘communion’) was expected. However, this study showed no direct effect for physical height on leadership styles. According to Lindqvist (2012), it is open to question whether the height-leadership relation contains unobserved factors. Knapen and Pollet (2016) argue that level of satisfaction with someone’s self-perceived height is an important factor of the feeling and behaviour of individuals in height-related social contexts. In advance of the current study, we expected a negative relationship between physical height and self-perceived height dissatisfaction, in such a way that a taller leader is less dissatisfied with his self-perceived height. This was confirmed by the results from the study. Also, we expected a positive association between self-perceived height dissatisfaction and leadership styles on the left side of the Leadership circumplex (low on ‘communion’). The left side of the Leadership circumplex translates to Authoritarian-, Directive-, Distrustful- and Withdrawn leadership. This was partly confirmed, since significant relationships were found for Yielding-, Withdrawn-, Distrustful- and Authoritarian Leadership. An unexpected result here was the relationship from self-perceived height dissatisfaction on Yielding leadership instead of Directive leadership. Notably, Yielding– and Directive leadership are situated on the exact opposite sides of the Leadership circumplex (Redeker et al., 2014). As height has been shown to be more influential in well-being, social perception, workplace success, health and satisfaction with own body and height for men compared to women (Knapen & Pollet, 2016), we expected the effects to be stronger for men. However, no moderation effects for gender on any of the relationships in the conceptual

(36)

36 model were found. This is an unexpected result that is not in line with presented research. Notably, Kirchmeyer (2002) suggested that the women that occupy a role of a leader or manager score usually high on masculinity since adulthood. This could be an explanation for this unexpected result, since leadership roles and management role are traditionally more masculine roles. 5.2 Theoretical and practical implications First, from a theoretical point of view, the current study has added knowledge to the research domain on leadership. The concept of satisfaction with self-perceived height has, to the best of our knowledge, not been used while investigating the relationship between physical height and leadership styles. Our results show that a significant part of the variance in leadership styles among leaders is explained by their self-perceived height dissatisfaction. Leaders, who experienced more self-perceived height dissatisfaction, also reported more Yielding-, Withdrawn-, Distrustful- and Authoritarian leadership styles. A reason for this could be that the leaders that are dissatisfied with their self-perceived height try to compensate for this by displaying a less communion-oriented leadership style, in order to be perceived as more masculine. Indeed, as explained by Kirchmeyer (2002) the management – and leadership role is traditionally more masculine in nature; masculinity is associated with traits like aggressiveness, independence and dominance (Eagly et al., 2000). Secondly, the current study has provided the leadership research domain with more insight in the effect of gender. In general the effect of height on leadership perceptions is proven to be stronger for men than for women (Blaker et al., 2013; Re et al., 2013). Instead, the current study didn’t find significant differences in gender for this model. This could be explained by the suggestion that the women that occupy a role of a leader or manager score usually high on masculinity since adulthood (Kirchmeyer, 2002). Potentially, the explanation by Kirchmeyer (2002) could be linked to Burke and Collins (2001) arguing that neither males nor females are superior in terms of effective leadership. For the research domain focussing on gender in relation to leadership, the results of the current study and the potential explanation by Kirchmeyer (2002) could

(37)

37 have implications. Since the results of the current study are not in line with the current state of the research domain where the effect of height is proven to be stronger for men (Blaker et al, 2013; Re et al., 2000). In practice, this study could be of use for both organisations and leaders. The current study shows the effect of self-perceived height dissatisfaction on leadership styles. On the one hand, the organisation could take this relationship into account during the process of creating new leaders. Nowadays, there is a lot of focus on developing new, future leaders. Most of the multinationals have their own traineeship or program for this; examples are Unilever’s Future Leader Program (UFLP) and the Randstad Group Management Traineeship. When training and developing new leaders it appears to be useful to study the self-perceived height dissatisfaction of the potential leader. In this way, organisations and future leaders are aware of the potential leadership styles it causes. If this is measured in the development stage of the leader and his/her leadership style, you may still be able to influence this. Regarding effective leadership, self-rated Distrustful- and Withdrawn leadership are negatively related to effective leadership (Redeker et al., 2014). In the current study, both Distrustful- and Withdrawn leadership proved to be related to self-perceived height dissatisfaction. Consequently, it appears to be important for (potential) leaders to be aware of their potential leadership styles as it could influence their leadership effectiveness. 5.2 Limitations & Future Research The current study is, like any other study, subject to certain limitations. The first limitation is regarding the sampling method. In the ideal situation, the current study would use a probability sample. Due to the limited access to the population and limited resources of the current study, a convenience sample is used. The sample for the current study is mostly male, 72,2%. Furthermore, some branches like Government and Transportation/Logistics appear to be overrepresented in the sample. This leads to a reduced generalizability. Moreover, the current study is cross-sectional; consequently

(38)

38 the requirements for inferring causality aren’t met. However, the study is exploratory in nature and causality isn’t a requirement. The second limitation for the current study is the use of self-reported measures. Although Spencer et al. (2002) argue that self- reported measures are valid data; they leave room for biases and errors. By offering clear instructions in advance, these limitations were reduced to a minimum. In these instructions the anonymity of the respondents and the need for honest answers were explained. Nonetheless, since overestimation of height to fit the socially desirable masculine physical characteristics of a leader is at risk (Bogaert & Mccreary, 2011), exact measurements would make the measurements more truthful. The third limitation is regarding the source of the collected data. Data was only collected from leaders; this could lead to a possible single-source bias. This means that the data is collected at a single source, which could lead to biases influencing the results. Since self-perceived height dissatisfaction is part of the model, data had to be collected from the leaders. Nevertheless, future research could include multiple sources of data collection. Potentially, the use of dyad studies with surveys for both leaders and subordinates would increase the validity of the leadership style measurement. For instance, using subordinates to report on the leadership style of their leader. The fourth limitation is regarding the length of the survey. The survey turned out to be rather long, which could cause respondents to become sloppy or decide not to finish the survey. The fact that there were 49 incomplete or inadequate responses for the survey illustrates this. Future research could be conducted with a shorter survey to see if this influences the results. An example could be the use of the shorter version of the CLS (Redeker et al., 2014). This is relevant because it could be that the group of people that didn’t complete the survey, for whatever reason, is significantly different from the group that did complete the survey.

(39)

39 The final limitation is concerning the analysis. We conducted a bootstrap analysis for the MANCOVA, using sampling with replacement. The reason for this bootstrap was the insufficient sample size for statistical inference. However, the use of a bootstrap sample brings risk for misinterpretation to the analysis. A Bootstrap sample assumes that the original sample is representative for the population (Field, 2013). The use of a convenience sample could influence the representativeness of the sample in a negative way. Future research could focus on gathering a large enough sample, so bootstrapping isn’t necessary. Moreover, according to the discriminant analysis in the current study, neither discriminant function, nor combined, was able to significantly differentiate the groups with different HD categories. This may be due to the negative influence of the sample. Future research with a larger sample may overcome this. Another direction for future research is to look into the different constructs that cause self-perceived height dissatisfaction. The current study shows that 25% of the variance in self-perceived height dissatisfaction is explained by physical height. This leaves a large part of the variance unexplained. In addition to the current study, it would be interesting to see what other constructs might explain this variance. Therefore, other factors that influence self-perceived height dissatisfaction could be investigated. A construct like culture could play a part here, since some cultures are more focussed on physical attributes than others. Another possible construct could be the branch or sector. In the selection process of leaders, humans look for cues that a leader is physically ‘fit to lead’ (Bass & Bass, 2008). Arguably, being ‘fit to lead’ in, for example, a military setting is more related to physical attributes like physical height compared to financial institutions. Moreover, further research could focus on the nature of the relationship between self-perceived height dissatisfaction and Yielding leadership. This relation was not expected to be present, based on the existing literature. In the current study, both Yielding leadership and Withdrawn leadership proved to be positively related with self-perceived height dissatisfaction. At the same time, both leadership styles are related to a low score on the agency-dimension in the Leadership circumplex, are negatively related

(40)

40 to task-oriented leadership and are positively related to Laissez-faire leadership (Redeker et al., 2014). The similarities between both leadership styles could be the starting point for future research on the height-leadership association. Finally, future research could focus on the potential relationship of the current results with effective leadership. In terms of leadership effectiveness, self-rated Distrustful- and Withdrawn leadership are negatively related to effective leadership (Redeker et al., 2014). Both Distrustful- and Withdrawn leadership are proven to be related to self-perceived height dissatisfaction in the current study. Based on the overlap of Distrustful- and Withdrawn leadership in the current study and Redeker et al. (2014) a potential relationship exists between height and effective leadership. In conclusion, the current study has contributed to the literature by assessing the association between physical height and leadership styles. The results indicate that shorter leaders are more likely to be dissatisfied with their self-perceived height. In turn this specific group of leaders is more likely to display a Yielding-, Withdrawn-, Distrustful- or Authoritarian leadership style. The current study found no influence from gender on these relations. The current study was the first to study leadership styles in association with self-perceived height dissatisfaction. In order for a generalized framework to be constructed, further research is needed.

(41)

41 References Adler, A. (1956). The individual psychology of Alfred Adler. New York: Basic Books. Aiken, L. S., & West, S. G. (1991). Multiple regressions: Testing and interpreting interactions. Thousand Oaks, CA: Sage. Avolio, B. J., & Bass, B. M. (1991). The full range leadership development programs: Basic and advanced manuals. Binghamton, NY: Bass, Avolio & Associates. Bass, B. M. (1985). Leadership and performance beyond expectations. New York: Free Press. Bass, B. M. (1997). Does the transactional–transformational leadership paradigm transcend organizational and national boundaries?. American psychologist, 52(2), 130-139. Bass, B. M., & Bass, R. (2008). The Bass handbook of leadership: Theory, research, and managerial applications (4th ed.). New York, NY: Free Press. Bennis, W. G. (2003). On becoming a leader ([Rev. ed.). Cambridge, MA: Perseus Pub. Bem, S. L. (1981). Gender schema theory: A cognitive account of sex typing. Psychological Review, 4, 354–364. Blaker, N. M., Rompa, I., Dessing, I. H., Vriend, A. F., Herschberg, C., & Van Vugt, M. (2013). The height leadership advantage in men and women: Testing evolutionary psychology predictions about the perceptions of tall leaders. Group Processes & Intergroup Relations, 16(1), 17-27. Bogaert, A. F., & McCreary, D. R. (2011). Masculinity and the distortion of self-reported height in men. Sex Roles, 65(7-8), 548-556. Bono, J. & Ilies, R. (2006). Charisma, positive emotions and mood contagion. The Leadership Quarterly, 17(4), 317-334. Bryman, A. (1992). Charisma and leadership in organizations. London: Sage. Burke, S., & Collins, K. M. (2001). Gender differences in leadership styles and management skills. Women in Management Review, 16(5), 244-257. Burns, J.M. (1978). Leadership. New York: Harper & Row. Case, A., & Paxson, C. (2006). Stature and status: Height, ability, and labor market outcomes National Bureau of Economic Research, No. w12466.

Referenties

GERELATEERDE DOCUMENTEN

Onderzoeken of segregatie voor knolresistentie van de verkregen populaties is gebaseerd op hoofdgenen. Onderzoeken of blad- en knolresistentie zijn gecorreleerd in

Er is een regressie-analyse uitgevoerd met de ‘stemming van de afgelopen twee weken’ als moderator om te onderzoeken of de bereidheid om een alcoholaanbod in de angst versie

Characterization, height distribution, and diffusion with wall gap height, of spherical particles above the wall First, we measured the sphere dynamics above a planar wall both

Wanneer er reflecterend wordt teruggekeken naar de gebruikte theorieën in dit onderzoek, kan er gesteld worden dat de boa’s openbare ruimte in Nederland

They constrained the solution space for α and β of the individual pulsars, with β = ζ − α, using fits to these radio data as well as predictions for the value of the half opening

In section 3 we show how two feature detection algorithms (SIFT and SURF) can be used for im- age stitching, resulting in a panoramic image of the ground. These features can also

Steril, 96(2), 390-393. Thiol-disulfide status and acridine orange fluorescence of mammalian sperm nuclei. The hypothalamic GnRH pulse generator: multiple regulatory mechanisms.

Samenvattend adviseert de commissie ribociclib niet in de basisverzekering op te nemen vanwege de ongunstige kosteneffectiviteit, tenzij een acceptabele lagere prijs voor het middel