Leader traits and follower traits : the mediating effect of leader OCB on leader Honesty-Humility and follower Honesty-Humility and on leader Altruism and follower Altruism

(1)

UNIVERSITY OF AMSTERDAM

Amsterdam Business School

Bachelor Economics and Business

Specialization: Business Administration

Leader traits and follower traits

The mediating effect of leader OCB on leader Honesty-Humility and

follower Honesty-Humility and on leader Altruism and follower

Altruism

BSc Thesis by Omar Abdellatif

10740236

Supervisor: Merlijn Venus

Amsterdam, 26th June 2018

(2)

Past research has mainly focused on personality models like the Big five and personality traits that cause leaders to act in line with morals and ethics, to be friendly and to help others. Nonetheless, Confucius’ philosophy on leader and follower traits has not been tested before. This research addresses the mediating effect of leader OCB on leader Honesty-Humility and follower Honesty-Humility and on leader Altruism and follower. With no previous research on this topic, it was hypothesized that leader Honesty-Humility would positively influence leader OCB, and that leader OCB would in turn positively influence follower Honesty-Humility. It was also hypothesized that leader Altruism would positively influence leader OCB, and that leader OCB would in turn positively influence follower Altruism. By means of a cross-sectional questionnaire with 42 dyads, no significant results were found. Suggesting further research is needed.

Statement of Originality

This document is written by Student (Omar Abdellatif) who declares to take full

responsibility for the contents of this document. I declare that the text and the work presented in this document are 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.

(4)

1. Introduction

According to Tuncdogan, Acar & Stam (2017), “Observations about the link between an individual’s stable characteristics and behaviours have been made throughout the history of mankind”. Leaders and their characteristics have been the topic of research since ancient times. Interest has especially been on which characteristics and individual differences make some leaders more effective than others. The following quote by Machiavelli (1513) that portrays this interest:

“We note that [leaders] pursue the ends they have in view...by different ways. One uses caution while another is impetuous, one resorts to violence while another relies on craft, one acts patiently while another does the contrary; and each reaches his goal by a different route”.

Another philosopher that discussed and had ideas on leadership was Confucius (Fernandez, 2003). Just like Machiavelli, Confucius thought about leaders and their characteristics. Confucius created his own philosophy, namely Confucianism, wherein he argued that leaders with good characteristics would positively influence their nation and others with bad characteristics would cause corruption to their nation (McArthur, 2011). Throughout time the pool of knowledge on leader characteristics has only grown bigger. An example of this is, McMullen, Shepherd & Patzelt (2009) who have written on why some leaders are almost always more cautious than others. And Hamstra, Van Yperen, Wisse & Sassenberg (2011) who have written on why some leaders are patient while others behave somewhat recklessly.

However leadership research has slowed down somewhat over the last decade. As a countermeasure, numerous articles have been published with the intention of reviving leadership research on individual differences. Thus supplying us with a knowledge pool on leadership and individual differences that cover multiple dimensions.

One of these dimensions is genetics. The field of “Genetics and Leadership” has focused mainly on the effect of genetics and environmental factors on leadership emergence. One of these studies is the study by Arvey, Rotundo, Johnson, Zhang & McGue (2006) on twins. Arvey et al., (2006) found that variance in leadership role occupancy could be explained, between a quarter or third, by genetic factors. Another dimension is the field of “Physical attributes and leadership”. An example finding from the literature in this dimension

(5)

is that the width of a CEO’s face relative to its height predicts the financial performance of a firm (Wong, Ormiston & Haselhuhn, 2011). These dimensions all belong under a bigger domain, namely leadership research on individual physiological differences. Other

dimensions that belong to this domain are the biological, neurological and endocrinological dimensions. And all of these have interesting findings on individual differences between leaders.

Another domain is leadership research on individual psychological differences. This domain discusses dimensions such as intelligence, personality traits, background variables and self-regulation. An example finding from the literature is the finding by Lin, Ma & Johnsons (2016). Lin et al. (2016) argue that ethical leadership behaviours might result in subsequent unethical behaviours because they require the use of self-regulatory resources. Another finding, pulled from the research on personality traits, is that in dynamic

environments, neurotic leaders are perceived to be more charismatic (de Hoogh. den Hartog, & Koopman, 2005).

While we have this much literature and research on leadership and individual

differences of leaders, there are still gaps in the literature that have not been researched yet or have only received a small amount of attention. As mentioned before, personality traits have received a lot of attention. Models such as the Big-Five personality traits and dark triad personality traits have been used and have been the topic of many literary pieces. However other models, such as the HEXACO model of personality, haven't received the same amount of attention (Tuncdogan et al., 2017). The HEXACO model is believed to be a further improvement on the Big-Five personality model. The HEXACO model includes the same personality dimensions as the Big-Five but also has an extra dimension of personality, namely Honesty-Humility. Honesty Humility is a dimension that causes leaders to act in line with morals and ethics, to be friendly and to help others.

One of the gaps in the literature is the research on personality traits that have the opposite effect to those in the dark triad and cause leaders to act in line with morals and ethics, to be friendly and to help others.

(6)

2. Literature review 2.1 The model underlying this thesis

The model used in this thesis is based on Confucius his philosophy. Confucius was a philosopher back when china consisted of separate kingdoms. His philosophy spoke of the importance of leaders and their leadership (Fernandez, 2003). He drew society as a hierarchy of small relations of respect between different players. For example, at the bottom of the hierarchy was a son and above him his father. The son had to respect his father for being one of his parents however in turn the father had to respectfully treat his son. And so forth the father had an employer who he had to respect and the employer had someone above him who he had to respect and so forth and so forth (McArthur, 2011). At the top of this hierarchy stood the leader of the kingdom, namely the king. Confucius describes how when a leader had good values and ethics, these would slowly be adopted by his followers and that good actions from his followers would follow (Fernandez, 2003). However the same applied for a leader with bad values and ethics, his followers would adopt these and this would lead to bad actions (Fernandez, 2003). Confucius suggested that this caused corruption in the kingdom and that’s why he stressed the importance of good values and ethics with leaders (McArthur, 2011).

The current literature on follower effects is mainly focused on 2 different types of follower effects, namely; leader emergence and follower performance (Tuncdogan et al., 2017). Leader emergence is based on leadership endorsement, which is the acceptance of and cooperation with the leader (Van Vught & de Cremer, 2003). Another important category of follower effects is follower attributions and perceptions. Most trait effects that show up seem to be based on follower perceptions of their leader. A third category of follower effects is motivation of followers and them striving for goals. Tuncdogan et al. (2017) argue that although this effect is potentially related to leader traits, there seem to be individual

differences related to self-regulation. However we still don't know much about the effect of a leader’s traits on a follower’s degree of that same trait and whether followers adopt traits from their leaders.

2.2 Leadership traits: Honesty-Humility and Altruism

Personality traits fulfill 2 important roles in leadership research on individual differences. In addition to being a big antecedent of leadership behaviour, personality traits can help us understand how physiological traits affect leader behavior and personality traits are antecedents of many other trait-like factors, which may have an effect on leader

(7)

behaviour Tuncdogan et al., 2017). There is a big literature pool on leadership traits and on how personality should be divided into separate components and how many components there should be. An example of this is the Big Five personality model also known as the five-factor model.

The Big Five personality model consists of 5 factors, namely Extraversion,

Agreeableness, Conscientiousness, Neuroticism and Openness to experience. According to Tuncdogan et al (2017), extraversion can be defined as the degree to which a person is energetic, assertive and social, agreeableness is the degree to which a person helpful, selfless and tolerant, conscientiousness is the degree to which a person is organized, self-disciplined and reliable, neuroticism is the degree to which a person experiences negative emotions and openness to experience is the degree to which a person is flexible in thought, ready to imagine new things and open to new feelings and experiences.

There are a lot of findings regarding the 5 personality factors. All of these traits, except neuroticism, are positively related to transformational leadership (Bono & Judge, 2004). Extraversion is believed to have the strongest association with transformational leadership (Reichard, Riggio, Guerin, Oliver, Gottfried & Gottfried, 2011). Washington, Sutton & Field (2006) suggest that agreeableness is a precursor of servant leadership

behaviours. George & Zhou (2001) argue that one of the benefits of openness to experience is that it stimulates creative behaviour, at the other hand, conscientiousness is an important component of productivity for leaders and followers (Colbert & Witt, 2009). However not all direct effects of these personality factors are desirable. The direct effect of leaders’

neuroticism are undesirable most of the times, only under specific circumstances are they believed to have some benefit. Neurotic leaders tend to outperform others in busy

environments (Smillie, Yeo, Furnham, & Jackson, 2006), and are viewed as more charismatic in dynamic environments (De Hoogh, Den Hartog & Koopman, 2005).

Personality traits are also found to indicate the potential of an individual to become a leader. Extraversion, conscientiousness and openness to experience are positively related to leader emergence, whilst neuroticism is negatively related with leader emergence (Judge, Bono, Ilies, & Gerhardt, 2002). Another more recent study (Reichard et al., 2011) shows that leader emergence at the age of 29 could be predicted by extraversion scores at the age of 17.

However there is more than just the five factor model. Even though it has had a lot of research there are other models believed to better capture personality and do a better job at dividing personality into components. An example of such a model is the HEXACO model of personality. Ashton, Lee & de Vries (2014) argue that “the six-dimensional space (meaning

(8)

the HEXACO model) captures some personality variance not represented within the five-dimensional models while also allowing for a better theoretical interpretation of personality variance”.

The HEXACO model of personality is a model that discusses personality and consists of 6 constructs namely; Honesty-Humility (H), Emotionality (E), Extraversion (X), Agreeableness (A), Conscientiousness (C) and Openness to experience (O) (Ashton, Lee & de Vries, 2014). The HEXACO model has been studied in relation to different topics, for example religiousness, guilt and shame proneness and schizotypy (Ashton et al., 2014). Honesty-humility and the interstitial scale (which measures Altruism) are 2 of the dimensions of the HEXACO model. Honesty-humility discusses the sincerity, fairness, greed avoidance and modesty of a person (Ashton et al., 2014). Whilst the interstitial scale, altruism (vs Antagonism), measures “the tendency to be sympathetic and soft-hearted toward others” (Ashton et al., 2014).

2.3 Leadership behaviors and OCB

Most research on leadership behaviours seem to have been focused on leadership styles. The most studied leadership styles are transformational leadership and transactional leadership. An example of this we find in the field of genetics and leader behavior, it has almost exclusively been focused on self-reported transformational and transactional leadership (Tuncdogan et al., 2017).

However there isn't that much research on other behaviours of leaders. An example of a leadership behaviour that has received less attention is leading by example. Even though leading by example or role modelling hasn't received the same amount of research as

transactional leadership or transformational leadership, it has been recognized as an effective leadership method by multiple theories (Manz & Sims, 1980). A set of behaviours that are almost solely dependent on role modelling, are organizational citizenship behaviours (OCB).

OCB stands for individual behaviors that are not directly recognized by the formal reward system (Nahum-Shani & Somech, 2011). Borman and Motowidlo (1993) define OCB as employee behavior supporting the social and psychological fabric of the organization. Examples of OCBs are helping fellow employees or voluntarily promoting the organization. It is important to note that these behaviors are voluntary and employees choose themselves to engage in such activities. OCBs contribute to several aspects of performance in an

(9)

3. Model & Hypotheses 3.1 Honesty-Humility

Leaders with high scores of Honesty-Humility are known to be fair and unwilling to take advantage of others, sincere and unwilling to manipulate others, greed avoidant, not especially motivated by monetary or social-status considerations, modest and view themselves as ordinary. We believe leaders that have high scores of the Honesty-Humility component of the HEXACO personality model will engage in OCB. We believe this is the case because leaders with high scores of Honesty-Humility are modest and don’t view themselves any different from their followers. Furthermore their sincerity makes them genuine with interpersonal relations and their greed avoidance makes them uninterested in possessing lavish wealth. We believe this combination leads to OCB activity because leaders will be uninterested in monetary rewards while their sincerity makes them interested in helping others around them. Their modesty then helps to view everyone around them as the “same” which should mean that leaders with high levels of Honesty-Humility will help other leaders and followers around them with no interest in monetary rewards. We believe this is the same as OCB. Therefore we formulate the following hypothesis:

H1: The Honesty-Humility trait of leaders is positively related to OCB activity of leaders

3.2 Altruism

Altruism focuses on the tendency to be sympathetic and soft-hearted towards others. Individuals that have high levels of altruism tend to avoid causing harm and react with generosity towards those who are weak or in need of help. We believe that altruism will lead leaders to engage in OCB activity. OCB activities are defined as activities that do not belong to the required activities of a job. However a leader that has high scores of altruism will help others in need, whether it is part of the job or not. Therefore we formulate the following hypothesis:

(10)

3.3 Confucius’ model

As mentioned before, Confucius describes how leaders slowly influence their followers and how followers would adopt the traits of their leader. Confucius speaks about how followers would become more honest with an honest leader.

Following this model and considering the traits mentioned above, Honesty-Humility and Altruism, we believe that ultimately, followers will adopt the Honesty-Humility and Altruism trait of their leader. However we expect this relation to be mediated through some action. We believe OCB activity to act as a mediator between the leader traits and the follower traits.

OCB, is the collective name of individual behaviors that are not directly recognized by the formal reward system. An example of OCB is helping fellow employees. We suspect that leaders with Humility and Altruism will positively influence the Honesty-Humility and Altruism levels of followers because of an engagement in OCB activity. We therefore formulate the following hypotheses:

H2: OCB activity is positively related to the Honesty-Humility trait of followers

H4: OCB activity is positively related to the Altruism trait of followers

And the following hypotheses regarding mediation:

H5: The relation between leader Honesty-Humility and follower Honesty-Humility is mediated by leader OCB, such that leader Honesty-Humility influences leader OCB which influences follower Honesty-Humility

H6: The relation between leader Altruism and follower Altruism is mediated by leader OCB, such that leader Altruism influences leader OCB which influences follower Altruism

4. Methodology 4.1 Method

This thesis researches the effect of a leader’s personality traits on followers and the mediating role of OCB. Therefore a quantitative data collection is used for this research. We follow the same method of previous research that has investigated personality traits. To research the Humility and Altruism levels of a leader in relation to the

(11)

Honesty-Humility and Altruism levels of his/her followers a workplace study was conducted in Amsterdam, the Netherlands using questionnaires. The questionnaires contained a small introductory paragraph explaining the questionnaire and the goal of this research. Moreover, the distribution of the surveys was performed by 2 students to increase sample size and reliability of the data.

There were 2 different questionnaires, one for the leader and one for the follower. The questionnaire for the leaders consisted of a 2 parts: a small personality test and a couple of general questions regarding age, gender and work specifics. The questionnaire for the

followers consisted of 3 parts: a small personality test, 3 questions related to OCB activity of the leader and a couple of general questions regarding age, gender and work specifics.

The participants in this research were asked to participate in person or through email. Participants were collected through the network of the researchers, namely colleagues from 3 different companies were asked to participate. Followers were chosen by asking leaders which followers they had most frequent interaction with. Most of the times the followers and leaders were contacted directly, sometimes the follower was contacted through the leader. After getting the approval of participants, they were sent the questionnaire. All questionnaires were taken online and those done on paper were later put in the system. The participants stayed completely anonymous by filling out the questionnaire online without filling in any kind of personal information that could be retraced to the. The only way the questionnaires were marked by participants was by a code. Each group of participants had to fill in the same code to later match them together. Participants of this research were from different

organizations, had different management levels and worked in different branches.

4.2 Participants

For this research we approached dyads consisting of 12 leaders and 61 followers. The response rate with our participants was 75% with the leaders and 81.96% with the followers, leaving us with 9 responses from the leaders and 50 responses with followers. After

reviewing the data, some of the dyads had to be removed due to missing answers or being unable to match data to a code and thus match them with a leader. This process led us to remove 8 data sets from the follower data bundle. Eventually we managed to collect data from 42 dyads with leaders having multiple followers matched to them.

(12)

4.3 Measures

The Honesty-Humility and Altruism levels of leaders and followers were measured through a small personality test. The personality test consisted of questions from the

HEXACO 100-item self-report form, found on the original website of the HEXACO model (hexaco.org). The questions that were selected from the HEXACO 100-item self-report form were the questions in regard to the personality trait Honesty-Humility and Altruism. The Honesty-Humility dimension consists of 4 facets, namely sincerity, fairness, greed avoidance and modesty. Each of these facets consists of 4 items an example of an item from the greed avoidance facet is “Veel geld bezitten vind ik onbelangrijk”. Examples from the other facets are: “Als ik iets wil van iemand die ik niet mag, dan zou ik me erg vriendelijk gedragen om

het te krijgen” from the sincerity facet, “Ik zou nooit ingaan op een poging tot omkoping, zelfs niet als het om een erg hoog bedrag ging” from the fairness facet and “Ik ben een gewoon persoon, niet beter dan anderen” from the modesty facet. Each statement was scored

on a 5-item scale with 1 being strongly disagree and 5 being strongly agree. Both leader and follower answered the questions about themselves. The Cronbach Alpha for the Honesty-Humility dimension is .886 (α = 0.886) for the followers and .802 (α = 0.802) for the leaders.

The Altruism scale only consisted of 4 items. An example from the altruism scale is

“Ik probeer gul te geven aan hulpbehoevenden”. Each statement was scored on a 5-item

scale with 1 being strongly disagree and 5 being strongly agree. Both leader and follower answered the questions about themselves. However due to a low Cronbach’s Alpha we left out one item of the Altruism scale of the followers to increase reliability. The item that was left out is “Het zou me niets kunnen schelen om iemand die ik niet mag kwaad te doen”. The Altruism trait for followers originally had a Cronbach’s Alpha of .542 when all 4 items were used. However by removing one of the items we managed to increase the Cronbach’s Alpha to .602. Moreover, the Cronbach’s Alpha for the Altruism trait with leaders was .520 and could not be increased by excluding one of the items of the Altruism dimension. The final Cronbach’s Alpha was 0,603 (α = 0.603) for followers and .520 (α = 0.52) for the leaders.

Unlike the Honesty-Humility and Altruism trait, questions about OCB levels of the leader have been answered by the followers. OCB was measured through a series of 3 items that were taken from previous research (Liden, Wayne, Jaworski & Bennett, 2004). Items used in previous research have been selected for use in this research to increase the chance of valid outcomes and prevent biases (Podsakoff, MacKenzie, Lee & Podsakoff, 2003). An example of one of the 3 items is: “Deze leider staat mij bij met mijn taken wanneer dat nodig

(13)

1 being strongly disagree and 5 being strongly agree. In this research only the OCB levels of the leaders have been measured. The Cronbach Alpha for these items is reasonable (α = 0.737) and therefore we can assume this to be a reliable scale.

4.4 Analysis

The data was first used to compute the Cronbach’s Alpha and review the reliability of the scales. After that we computed the mean of the Honesty-Humility and Altruism values for the leaders and followers and the mean of the OCB values for the leader. From there we first performed a correlation analysis to check for any correlation between the variables.

Thereafter we focus on doing regression analysis to check if the relation between the variables is significant and whether we can support our hypothesis.

In the case of our model, which has a mediator, we have to do multiple regression analysis. Baron & Kenny (1986) argue that when testing for mediation you can do a regression analysis and check the significance of the relationship between leader Honesty-Humility or leader Altruism (variable a) and leader OCB(variable b) and the relationship between leader OCB (variable b) and follower Honesty-Humility or follower Altruism (variable c). First we test the relationship between leader Honesty-Humility (variable a) and follower Honesty-Humility (variable c) and the relationship between leader Altruism

(variable a) and follower Altruism (variable c). This tests whether there is an effect that may be mediated. The next step is testing for an effect between leader Honesty-Humility or leader Altruism (variable a) and leader OCB (variable b) and test for an effect between leader OCB (variable b) and follower Honesty-Humility or follower Altruism (variable c). When

performing the second test (between variable b and variable c) it is important to use leader Honesty-Humility or leader Altruism (variable a) as a criterion variable, thus controlling for the effect of leader Honesty-Humility or leader Altruism and establishing the effect of leader OCB on follower Honesty-Humility or follower Altruism. To establish the mediating effect of leader OCB, the effect of leader Honesty-Humility or leader Altruism on follower

Honesty-Humility or follower Altruism controlling for leader OCB (c’) should be zero. If all requirements are met then we can assume that the data is consistent with the hypothesis that leader OCB completely mediates the relationship between leader Honesty-Humility or leader Altruism and follower Honesty-Humility or follower Altruism. If the requirement: “(c’) should be equal to zero” is not met, then only partial mediation is indicated.

In our case we will test twice for mediation. First we will test the mediating effect of leader OCB on the relation between leader Honesty Humility and follower Honesty-Humility.

(14)

The second time we will test the mediating effect of leader OCB on the relation between leader Altruism and follower Altruism.

The regression analysis was done through SPSS. Before conducting the analysis an extension, namely “PROCESS v3.0 by Andrew F. Hayes”, was installed. The extension allowed us to do analysis with premade models and test for mediation. The model used in this research to test for mediation was model “4” of the PROCESS extension. This model allowed us to test all effects between leader Honesty-Humility or leader Altruism, leader OCB and follower Honesty-Humility or follower Altruism at once.

5 Results 5.1 Description of the leader group

The leaders that participated in this research had ages ranging from the age 26 to 35 with an average age of 30 years (SD: 3,621). From the leader, 66% are male and 33% are female (See Table 1 below). Most leaders have worked between 1 and 5 years (56%) at their current company, 2 of them worked for more than 5 years and 1 leader has been in

employment at their current company for over 10 years. While one leader has only been working at his current company for less than 1 year. Out of 9 leaders, 8 work fulltime and only 1 worked part-time.

N Age Male Female

9 30 (SD: 3,621) 6 (66%) 3 (33%) Table 1

5.2 Description of the follower group

The followers that participated in this research have ages ranging from the age of 18 to the age of 49 with an average age of 23 (SD: 5,9). From them, 28 (66%) followers are male and 14 (33%) are female. Most followers have worked between 1 and 5 years in employment of their current employer. Out of the 42 followers, 11 worked for less than a year and 1 worked between 10 to 15 years at his current job. Also, most followers work part-time (71,4%) in comparison to full-time (28,6%). Most of the followers (52,4%) have been working together with their leader for less than a year. The rest of them either worked with their leader for a period between 1 and 5 years (40,5%) or for a period between 5 and 10 years (7,1%).

(15)

N Age Male Female 42 23 (SD: 5,9) 28 (66%) 14 (33%)

Table 2

5.3 Hypothesis testing and correlation

To identify a possible link between the Honesty Humility trait of leaders, Altruism trait of leaders, OCB of leaders, Honesty-Humility trait of followers and the Altruism trait of followers a correlation analysis is performed. In the following Table (Table 3) an overview is given of the means, standard deviations of variables and correlations between the variables.

Below (Table 3) you can see the correlations between the tested variables

(Cronbach’s Alpha on the diagonal). Unexpectedly, OCB of leaders was not associated with

either the Honesty-Humility trait with followers (r=-.097, p>0.05), the Altruism trait with followers (r=-.006, p>0.05), the Honesty-Humility trait with leaders (r=-.610, p>0.05) or the Altruism trait with leaders (r=-.243, p>0.05). The only significant finding related to OCB of leaders is with Gender (r=.336, p<0.05) and Age (r=.345, p<0.05). From the results for gender (F), we deduce that followers are more likely to experience OCB from their leader when they are female (with value 1 = male and value 2 = female). We can also derive from the significant relationship between OCB of leader and Age of followers that older people are more likely to receive help in the form of OCB from their leader. Surprisingly, we found a correlation between the Altruism level of followers and the Honesty-Humility level of followers (r=.412, p<0.01). This may indicate that followers that have high levels of

Altruism will also have high levels of Honesty-Humility and vice-versa. Finally there was no relation found between the Honesty-Humility trait of leaders and the Honesty-Humility trait of followers (r=.409, p>0.05). We see a similar relation between the Altruism trait of

followers and the Altruism trait of leaders (r=-.227, p>0.05). This may infer that there is no relationship between the traits of leaders and followers, contradicting the framework.

(16)

M SD 1. 2. 3. 4. 5. 6. 7. 8. 9. 1. Gender (F) 1,33 .48 - 2. Age (F) 22,79 5.9 .234 - 3. HH trait (F) 3,66 .69 .055 .136 -(.886) 4. Altruism (F) 3,90 .68 .025 -.211 .412** -(.603) 5. OCB (L) 3,67 .66 .336* .345* -.097 -.006 -(.737) 6. Gender (L) 1,33 .5 .158 -.351 .306 -.542 -.318 - 7. Age (L) 30 3.6 .226 -.037 -.107 -.508 -.077 .322 - 8. HH trait (L) 3,63 .49 .242 -.317 .409 -.189 -.610 .511 .560 -(.802) 9. Altruism (L) 4,11 .45 -.291 -.233 -.419 -.227 -.243 .230 .583 .362 -(.520) Table 3 HH = Honesty-Humility

* Correlation is significant at the 0.05 level (2-tailed) (F) = Follower ** Correlation is significant at the 0.01 level (2 tailed) (L) = Leader

5.4 Median regression analysis

Originally, a mediated regression analysis was performed to analyse the mediating effect of leader OCB on the relationships between effects of leader Honesty-Humility and leader Altruism on follower Honesty-Humility and follower Altruism. Below we discuss the results of the regression analysis and test hypothesis 1, 2, 3 and 4.

5.4.1 Honesty-Humility and OCB activity

In Table 4.1 and 4.2 we see the coefficients between the variables leader Honesty-Humility, leader OCB and follower Honesty-Humility. The tables (Table 4.1 & 4.2) show multiple effects and how strong they are. For example, a (-.762) shows us that 2 leaders that differ 1 unit in Honesty-Humility are estimated to differ .762 in OCB activity. The sign in front of is a negative, which means that leaders relatively higher in Honesty-Humility are estimated to have a lower mark of OCB activity. However this coefficient is not significant (a= -.762, p>0.05). Honesty-Humility with leaders may not necessarily bring about a change in OCB activity. In addition, b (.159) describes the effect of OCB activity with leaders on Honesty-Humility levels with followers. Finally, c’ (.529) illustrates the direct effect of Honesty-Humility of leaders on the Honesty-Humility trait of followers, controlling for the effect of leader OCB. However, both b (p=.765) and c’ (p=.293) have a p larger than 0.05 and therefore we cannot call this effect statistically significant.

If we eliminate the mediator out of the equation we get coefficient c (.519). This indicates the amount of change will occur in Honesty-Humility of followers when a leader

(17)

has 1 extra unit of Honesty-Humility. Again, c (p=.275) has a p>0.05 and therefore we cannot call this effect statistically significant. Finally, we have the effect of ab (-.121). This describes the following effect; leaders that differ in one unit of Honesty-Humility are estimated to bring about a change of .121 in Honesty-Humility with followers as a result of the tendency of leaders to decrease OCB activity which in turn decreases Honesty-Humility levels with followers.

From a simple mediation analysis conducted using ordinary least squares path analysis, we cannot conclude that Humility of leaders indirectly influence Honesty-Humility of followers through its effect on OCB activity. As can be seen in Table 4.1, leaders relatively higher in Honesty-Humility levels are estimated to be lower in their OCB activity (a = - 0.762, p= .081), and leaders relatively higher in OCB activity are estimated to bring about a positive change in Honesty-Humility levels of followers (b = 0.159, p = .765).

However both effects cannot be statistically proven due to a p higher than 0.05. There was no evidence that Honesty-Humility of leaders influenced Honesty-Humility levels of followers independent of its effect on OCB activity (c′ = .519, p=.293)

Consequent

OCB (M) HH Follower (Y) Antecedent Coeff.. SE p Coeff. SE p HH Leader (X) a -.762 .374 .081 c’ .519 .45 .293 OCB leader (M) - - - b .159 .36 .765 constant i₁ 6.466 1.368 0.002 i₂ .753 2.670 .787

R₂=.372 R₂=.193

F(1,7)= 4,145 p=.081 F(2,6)= .719 p=.525 Table 4.1 Median regression analysis on Honesty-Humility and OCB

(18)

Effect SE p LLCI ULCI Direct effect c’ .519 .45 .293 -.584 1.622 Total effect c .398 .336 .275 -.396 1.192

Boot SE Boot LLCI Boot UCLI Indirect effect ab -.121 .85 -2.804 .438

Table 4.2 Median regression analysis on Honesty-Humility and OCB continued

5.4.2 Altruism and OCB activity

In Table 5.1 and 5.2 we see the coefficients between the variables leader Altruism, leader OCB and follower Altruism. The tables (Table 5.1 & 5.2) show multiple effects and how strong they are. For example, a (-.328) shows us that 2 leaders that differ 1 unit in Altruism are estimated to differ .328 in OCB activity. The sign in front of is a negative, which means that leaders relatively higher in Altruism are estimated to have a lower mark of OCB activity. However this coefficient is not significant (a= -.328, p>0.05). Altruism with leaders may not necessarily bring about a change in OCB activity. In addition, b (-.287) describes the effect of OCB activity with leaders on Altruism with followers. Finally, c’ (-.376) illustrates the direct effect of Altruism of leaders on the Altruism trait of followers. However, both b (p=.466) and c’ (p=.468) have a p larger than 0.05 and therefore we cannot call this effect statistically significant.

If we eliminate the mediator out of the equation we get coefficient c (-.283). This indicates the amount of change that will occur in Altruism of followers when a leader has 1 extra unit of Altruism. Again, c (p=.557) has a p>0.05 and therefore we cannot call this effect statistically significant. Finally, we have the effect of ab (.094). This describes the following effect; leaders that differ in one unit of Altruism are estimated to bring about a change of .094 in Altruism with followers as a result of the tendency of leaders to decrease OCB activity which in turn increases Altruism levels with followers.

From a simple mediation analysis conducted using ordinary least squares path analysis, we cannot conclude that Altruism of leaders indirectly influence Altruism of followers through its effect on OCB activity. As can be seen in Table 5.1, leaders relatively higher in Altruism levels are estimated to be lower in their OCB activity (a = - 0.328, p= .529), and leaders relatively higher in OCB activity are estimated to bring about a negative change in Altruism levels of followers (b = -.287, p = .466). However both effects cannot be statistically proven due to a p higher than 0.05. Thus there was no evidence that Altruism of

(19)

leaders influenced Altruism levels of followers independent of its effect on OCB activity (c′ = -.376, p=.468).

Consequent

OCB (M) Altr Follower (Y) Antecedent Coeff.. SE p Coeff. SE p Altr Leader (X) a -.328 .495 .529 c’ -.376 .486 .468 OCB leader (M) - - - b -.287 .36 .466 constant i₁ 5.051 2.046 0.043 i₂ 6.722 2.662 .049

R₂=.059 R₂=.142

F(1,7)= .438 p=.529 F(2,6)= .498 p=.631 Table 5.1 Median regression analysis on Honesty-Humility and OCB

Altr =Altruism

Effect SE p LLCI ULCI Direct effect c’ .-.376 .486 .468 -1.566 .813 Total effect c -.283 .459 .557 -1.368 .803

Boot SE Boot LLCI Boot UCLI Indirect effect ab .094 .659 -1.309 1.778

Table 5.2 Median regression analysis on Altruism and OCB continued

6. Discussion

Originally, this research aimed to research a couple of topics and variables that had not received much attention in the current literature. The HEXACO personality model, a model that is argued to capture some personality variance not represented in the Big Five model and also allowing for a better theoretical interpretation of personality variance (Ashton et al., 2014), is a personality model that has received less attention than its predecessor the Big Five model. Furthermore, we tried adding to existing research by testing Confucius’ model on leadership and personality traits. To the authors’ knowledge this is the first research testing personality traits using Confucius’ model. Not only were the Honesty-Humility and Altruism traits measured, the mediating effect of OCB activity from leaders on the

(20)

relationship between a leader’s and followers’ trait were also addressed. Surprisingly, in this study sample, there was no evidence to support such a relationship or the existence of a mediating effect.

Firstly, we expected Honesty-Humility with leaders to be positively related to OCB activity of leaders. This assumption was built on the notion that leaders that are “honest” and “humble” would help their colleagues and followers because their modesty views them as equals. Adding to this, their sincerity and fairness would not let them take advantage of others and would be unwilling to manipulate others. Finally, their greed avoidance would prevent them from only helping others when receiving a (monetary) reward because they are not especially motivated by monetary or social-status considerations (Ashton et al., 2014). However in comparison to our first assumption, our analysis has shown that, if there is any effect, the effect of Honesty-Humility of leaders on OCB activity of leaders to be negative and not significant. Therefore our first hypothesis is not confirmed.

This may be due to several reasons. It may be that our speculation is wrong and that Humility does not lead leaders to participate in OCB. It could also be that Honesty-Humility alone is not enough for leaders to engage in OCB, and that it’s a combination of Honesty-Humility and other traits that trigger OCB activity with leaders. However research shows a similar effect. Bourdage, Lee, Lee & Shin (2012) argue that individuals that are low in Honesty-Humility are driven to engage in OCB for impression management purposes. Therefore it could be that high levels of Honesty-Humility are related to lower levels of OCB.

Our second hypothesis discussed the effect of OCB activity on Honesty-Humility levels of followers. This hypothesis was based on Confucius’ model, which indicates that followers adopted personality traits from their leader. Also this hypothesis was not

confirmed. The analysis showed the effect of OCB activity of leaders on Honesty-Humility with followers to be positive, which is according our assumptions. However not significant. Thus, our second hypothesis is not confirmed.

It could be that the sample size used in this research is too small to observe this effect. Our research only had 42 participants, 9 leaders and only focused on 2 businesses. This could affect the results and it may be that a larger sample size will lead to a significant effect.

Furthermore, our third hypothesis suggests that Altruism with leaders will lead to more OCB activity with leaders. This hypothesis is based on the fact that high scorers of Altruism “avoid causing harm and react with generosity towards those who are weak or in need of help” (Ashton et al., 2014). The hypothesis assumes that high scorers of Altruism will help their followers/colleagues when in need of help and that this could take the form of

(21)

OCB. Again we were surprised to see that the effect observed in this research is negative and non-significant. The third hypothesis is not confirmed.

To repeat, a reason for this effect could be that Altruism does not trigger OCB activity. Or that the amount of participants is not enough to gather data that shows such a relationship. Another explanation could be individuals with high levels of Altruism translate “avoid causing harm” into “not being a burden unto others/the company”. Therefore if followers had adopted the Altruism trait they would instead not ask for any help or not show that they need any help. Thus it would make sense that higher levels of Altruism would cause lower levels of leader OCB. Finally, it could be that leaders only show Altruism whenever it serves them, and that they do not actually own the trait. And thus engage in the act of

impression management. Therefore an increase of observing Altruism with such leaders does not necessarily increase the OCB activity of leaders. However previous research has shown a positive effect between Altruism and OCB (Bashir, Sardar, Zaman, Swati & Fakhr, 2011). Therefore it is unlikely that this negative effect is in the “right” direction.

Finally, the 4th hypothesis discusses how OCB activity is positively related to Altruism levels of followers. This hypothesis was also based on Confucius’ model and his ideology wherein he argues that followers adopt the traits of their leader. This hypothesis is not confirmed due to non-significant evidence (β = -.287, p>0.05).

Confucius infers that followers adopt traits from their leader over time. He argues that it would take 4 years for a kingdom to adopt the traits of a king (McArthur, 2011). This could be a reason for not observing the assumed effect. It could be that followers need more time to adopt the Altruism trait from their leader. The respondents recorded to have spent an average of 1.55 time with leaders (with a dataset of 1 = less than one year, 2 = between 1 and 5 year). This suggests that followers may have not spent enough time with their leader to adopt their traits. Or that it takes longer for traits to be adopted in a business-like context or in certain branches. It could also be that the research method was wrong and that it would have been more suitable doing a longitudinal study by checking Altruism levels with followers over time.

To conclude, the last 2 hypotheses (H5 and H6) assume that there is a mediating effect of leader OCB on the relationship between leader Honesty-Humility or leader Altruism and follower Honesty-Humility or follower Altruism. These hypotheses are based on

Confucius’ model. However these hypotheses did not meet the “(c’) should be equal to zero” requirement. Therefore, if there is any form of mediation, only partial mediation could be

(22)

indicated. Nevertheless, all other hypotheses have been rejected. This leads us to reject these hypotheses, thus rejecting all six hypotheses.

7. Limitations

This study has some limitations that will be addressed in this section. Most of the limitations of this research are in regards to the study sample. This study was performed in a short time frame and therefore the students conducting this research could not collect

abundant data. Out of the 73 approached respondent, only 59 responded. After taking out missing data and eliminating certain responses only 42 dyads were formed, this resulted in a non-response bias. Adding to this, the sample consisted of people in the network and

surroundings of the students and thus a convenience sample was used. This lead to the sample not being representable for the Dutch population or a certain business branch in the

Netherlands, because generally leader-follower dyads do not only occur in the business branches that were studied or only in business-like contexts as the context that was studied in this research. Moreover, as the students are in the same phase of their lives, the approached respondents had similar characteristics. Due to this selection bias, it is unlikely that the results of this research are generalizable to other contexts, resulting in a low external validity.

External validity may also have been decreased through the response bias of

untruthful responses. The respondents asked were colleagues and people from the network of the researchers. Due to highly sensitive information and differences in hierarchy of leaders and followers, respondents may not have answered all of the questions truthfully.

Respondents may have feared being judged for their answers or have been suspicious of any after effects even though participants’ anonymity was guaranteed within the questionnaire.

Another limitation is found when looking at the descriptives of the variables. The mean of “For how long have you been employed by your current employer?” was 1.86 units indicating that the actual amount of years lies between the 1 and 5 years. However the mean of “How long have you and your current leader been working together for?” is 1.55 units. These numbers suggest that the average relationship with leaders is shorter than the average time employed at the company. Adding to this, 1.55 units of “How long have you and your current leader been working together for?” is between (1=less than one year) and (2=between 1 and 5 years). This indicated that the relationship between leader and followers is relatively young, and therefore followers were incapable to judge their dyadic partner correctly, resulting in biased responses. Moreover, the young relationship between leaders and

(23)

followers limits the ability of followers to adopt the trait from their leader. As mentioned before, Confucius suggested that this process takes time.

Finally, more limitations can be found regarding internal validity, due to the fact that the design of this study did not allow for ruling out alternative causal interpretations. Making use of a questionnaire may have allowed for extraneous variables to occur. One of these extraneous variable is the tendency of respondents to answer a questionnaire in extreme or moderate levels. For example, some respondents tend to answer 5 (strongly agree) on the Likert scale when they agree with a certain statement while other consider agreeing with a statement a 4 (moderately agree). This is an approach with which respondents fill out surveys in general. However the habit of respondents to answer questionnaires without any extremes may affect the results of research and lead to not finding significant results.

8. Suggestions for future research

There are multiple suggestions for future research, some of them focus on improving this research ad some of them concern following up on certain effects or questions.

Firstly, to improve the external validity of this research, future research could increase the sample size and diversify the sample. An increase of the number of leader and follower dyads may result in to significant results. Adding to this, collecting data from different (business-like) contexts is another way of improving the representativeness of the sample. This could be done through reaching out to businesses outside of the Netherlands or by asking respondents from businesses from different branches that were not covered in this research, such as NGOs or the medical branch. Due to different (business) cultures, different results may be found. Moreover, an enlarged and diversified sample decreases the chances of encountering biases of followers being unable to judge their leaders.

Another way to improve this research is by changing the method of research. Instead of opting for only a questionnaire, a longitudinal study may have been more appropriate. A longitudinal study would offer the chance to study leader and followers dyads over time and research whether there is a change in Honesty-Humility and Altruism trait with followers. Another way is to perform an experiment instead of a questionnaire. Experiments force participants to react in certain exercise-based contexts and therefore it may decrease the bias of participants not answering truthfully. To discover more, research may be expanded

through interviews or by doing observer reports instead of self-reports. These could give new insights and teach us more about the effect of leaders on a follower’s personality.

(24)

Another improvement is increasing the follower amount matched to one leader, and thereby increasing the chance on getting significant results. Also this will put more focus on the relationship between leader and follower and improve judgements made on variables such as leaders’ OCB.

Additionally, there are topics and questions that to this researcher’s belief have not been investigated yet. In this research we only studied Confucius’ model with the personality traits Honesty-Humility and Altruism, however other traits have not been looked at yet. It could be that only some traits are adopted by followers. Another interesting topic is the effect of the mediator. In this research we have chosen OCB as the mediator, it could be interesting to test other variables as mediators. OCB may not have been a proper mediator allowing for followers to adopt their leaders’ traits.

Lastly, another interesting topic for research is the moderating variables of the relationship between leader traits and follower’s traits. Examples of moderating variables could be the fit between leader personality and follower personality. For instance, followers that are really honest will probably less likely adopt traits from a leader that does not keep his word. Other examples of moderating variables are gender, age and ethnicity. Future research could research these moderators and their effect on the relationship between leader traits and follower traits.

Finally, there was a surprising correlation found between Altruism of followers and Honesty-Humility. This could mean that these traits are connected and individuals have either both these traits or no levels of both of them.

9. Conclusion

The literature on leadership and its relation to personality is abundant, however not all aspects of it are researched equally. Topics such as the HEXACO model that discuss

personality traits that make individuals act in line (Ashton et al., 2014) is one of them. The efforts of this paper try to discuss 2 of the HEXACO model traits using Confucius’ ideology on leadership and testing whether followers adopt the Honesty-Humility and Altruism trait from their leader. More specifically, this relationship mediated through OCB is investigated. However, there was no such effect found. There was no significant evidence found for the assumption that followers would adopt the Honesty-Humility and Altruism trait from their leader. Furthermore, there was no significant evidence found for the mediating effect of leaders’ OCB on this relationship. Nonetheless, this may infer a demand for future research with different traits or a different mediator.

(25)

Bibliography

Ahston, M.C., Lee. K., de Vries, R.E. (2014). The HEXACO Honesty-Humility, Agreeableness, and Emotionality Factors: A Review of Research and Theory. Personality and Social Psychology Review, 18(2), 139-152.

Arvey, R.D., Rotundo, M., Johnson, W., Zhang, Z., & McGue, M. (2006). The determinants of leadership role occupancy: Genetics and personality factors. Leadership Quarterly,

17(1), 1-20.

Bashir, N., Sardar, A., Zaman, K., Swati, A.K., & Fakhr, S. (2011). Determinants of organizational citizenship behavior: A case study of higher education institutes in Pakistan. Management Science Letters, 2(1), 329-338.

Baron, R.M, & Kenny, D.A. (1986). The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations.

Journal of Personality and Social Psychology, 51(6), 1173-1182.

Bono, J.E., & Judge, T.A. (2004). Personality and transformational and transactional leadership: A meta-analysis. Journal of Applied Psychology, 89(5), 901. Bourdage, J.S., Lee, K, Lee, J.H, & Shin, K.H. (2012). Motives for Organizational

Citizenship Behavior: Personality Correlates and Coworker Ratings of OCB.

Human Performance, 25(3), 179-200.

Colbert, A.E., & Witt, L.A. (2009). The role of goal-focused leadership in enabling the expression of conscientiousness. Journal of Applied Psychology, 94(3), 790.

De Hoogh, A. J., Den Hartog, D. N., & Koopman, P. L., (2005). Linking the Big-Five factors of personality to charismatic and transactional leadership; perceived dynamic work environment as a moderator. Journal of Organizational Behaviour, 26(7), 839-865. Ersoya, N.C., Derous, C., Born, M.P., van der Molen, H.T. (2015). Antecedents of

organizational citizenship behavior among Turkish white-collar employees in The Netherlands and Turkey. International Journal of Intercultural Relations, 49, 68-79. Fernandez, J.A. (2003). The Gentleman’s Code of Confucius: Leadership by Values.

Organizational Dynamics, 33(1), 21–31.

George, J.M., Zhou, J. (2001). When openness to experience and conscientiousness are related to creative behaviour: An interactional approach. Journal of Applied

(26)

Hamstra, M. R., Van Yperen, N. W., Wisse, B., & Sassenberg, K. (2011).

Transformational-transactional leadership styles and followers’ regulatory focus.

Journal of Personnel Psychology, 10, 187-191.

Judge, T.A., Bono, J.E., Ilies, R., & Gerhardt, M.W. (2002). Personality and leadership: A qualitative and quantitative review. Journal of Applied Psychology, 87(4), 765-780. Kalshoven, K. (2010). Ethical leadership: through the eyes of employees.

Kamdar, D., Van Dyne, L. (2007). The joint effects of personality and workplace social exchange relationships in predicting task performance and citizenship performance. Journal of Applied Psychology, 92, 1286–1298.

Liden, R. C., Wayne, S. J., Jaworski, R. A., & Bennett, N. (2004). Social loafing: A field investigation. Journal of Management, 30(2), 285-304.

Lin, S. H. J., Ma, J., & Johnson, R. E., (2016). When ethical behaviour breaks bad; How ethical behaviour can turn abusive via ego depletion and moral licensing. Journal of

Applied Science, 101(6), 815-830.

Machiavelli, N. (1513). The Prince, New York, United States, Dover Publications Inc. Manz, C. C., & Sims, H. P. (1980). Self-management as a substitute for leadership: A

social learning theory perspective. Academy of Management Review, 5, 361–367 McArthur, M. (2011). Confucius. London, United Kingdom, Quercus Publishing

McMullen, J. S., Shepherd, D. A., Patzelt, H. (2009). Managerial (in) attention to competitive threats. Journal of Management Studies, 46(2), 157-181.

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., and Podsakoff, N. P. (2003). Common Method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903.

Richard, R.J., Riggio, R.E., Guerin, D.W., Oliver, P.H., Gottfried, A.W., & Gottfried, A.E. (2011). A longitudinal analysis of relationships between adolescent personality and intelligence with adult leader emergence and transformational leadership. Leadership

Quarterly, 22(3), 471-481.

Smillie, I.D., Yeo, G.B., Furnham, A.F., & Jackson, C.J. (2006). Benefits of all work and no play: The relationship between neuroticism and performance as a function of resource allocation. Journal of Applied Psychology, 91(1), 139-155.

Tuncdogan, A., Acar, A.A., & Stam, D. (2017). Individual differences as antecedents of leader behavior: Towards an understanding of multi-level outcomes, The Leadership

(27)

Van Vugt, M., & de Cremer, D. (2003). Leader endorsement in social dilemmas: comparing the instrumental and relational perspective. European review of Social Psychology,

13(1), 155-184.

Washington, R.R., Sutton, C.D., & Field, H.S. (2006). Individual differences in servant leadership: The roles of values and personality. Leadership & Organization

Development Journal, 27(8), 700-716.

Wong, E. M., Ormiston, M. E., & Haselhuhn, M. P., (2011). A face only an investor could love: Ceos’ facial structure predicts their firms’ financial performance.