I AM EXCEPTIONAL: WHY GRANDIOSE NARCISSISTS ENGAGE IN
ABUSIVE SUPERVISION
THE ROLE OF A PERSONALIZED- AND A SOCIALIZED POWER
MOTIVE
Master thesis, MSc Human Resource Management University of Groningen, Faculty of Economics and Business
ABSTRACT
I AM EXCEPTIONAL: WHY GRANDIOSE NARCISSISTS ENGAGE IN
ABUSIVE SUPERVISION
THE ROLE OF A PERSONALIZED- AND A SOCIALIZED POWER
MOTIVE
There are lots of narcissistic leaders (Brunell, Gentry, Campbell, Hoffman, Kuhnert, & DeMarree, 2008; Rosenthal & Pittinsky, 2006). They all have one thing in common: they use self-enhancement tactics to keep inflated self-views (Campbell, Brunell, & Finkel, 2006). Donald Trump is an example of a famous narcissistic leader (Ahmadian, Azarshahi, & Paulhus, 2017). During his cabinet meeting, he said: “Never has there been a president, with few exceptions, who has passed more legislation and has done more things than I have done” (Donald Trump, 2017). Donald Trump communicated to the world that he is one of the best presidents the United States ever had. Donald Trump made use of a self-enhancement tactic to keep inflated self-views. Furthermore, during the meeting he let every cabinet member individually praise him. Most people would get embarrassed, but Donald Trump seemed to enjoy it. Some persons would view this behavior as an act of confidence, while others would view this behavior as a way of disdaining the previous presidents. By disdaining the previous presidents and indirectly the civil servants who now work under Trump, he made use of a form of abusive supervision (Tepper, 2000), since he was only talking about himself and therefore did not give these civil servants credits for jobs requiring a lot of effort.
Grandiose narcissism1 could be seen as a double-edged sword for the subordinates.
The positive outcomes of narcissistic leaders are the ability to inspire followers and the ability to communicate a clear vision (Maccoby, 2000). Narcissistic leaders can thus be beneficial for
1 This thesis is about grandiose narcissism and is not about narcissism as defined by the Diagnostic and
subordinates. Besides the positive behaviors of narcissists, however, narcissistic leaders also display behaviors that can harm the subordinates. Literature points out that grandiose narcissism is associated with abusive behaviors, such as bullying (Baughman, Dearing, Giammarco, & Vernon, 2012), exploiting others (Campbell, Hoffman, Campbell, & Marchisio, 2011), and aggression (Jones & Paulhus, 2010). Narcissism is also included in the dark triad of personalities (Paulhus & Williams, 2002), besides psychopathy and Machiavellianism, because of the potential anti-social behaviors. The relationship between leaders’ grandiose narcissism and abusive supervision has not been empirically investigated yet and it is known that abusive supervision has tremendous negative effects on the subordinates (Tepper, 2007). Therefore, I will focus on the relationship between leaders' grandiose narcissism and abusive supervision. I will use a power perspective to explain this relationship.
Currently, little is known why narcissists display abusive behaviors and this makes it hard to intervene. It is important to understand the motive(s) of narcissists to engage in abusive behaviors, since this would give possibilities to positively influence the previously described double-edged sword. Measures could be taken to prevent the leader from acting in abusive ways. One widely embraced explanation why narcissists engage in abusive behaviors is the theory of threatened egoism. Which states that narcissists display abusive behaviors, because their self-esteem is threatened (Bushman & Baumeister, 1998). In line with Bushman and Baumeister (1998), I will also offer an explanatory mechanism why narcissistic leaders engage in abusive behaviors. I will argue that the power motive of narcissists will decide whether he or she will engage in abusive supervision.
(Brunell et al., 2008; Emmons 1989; Lee, Ashton, Wiltshire, Bourdage, Visser, & Gallucci, 2012). Most authors, including Lee et al., (2012) and Emmons (1989) have treated the desire for power as a general tendency irrespectively of whether the narcissistic leader wants to have power for himself or for others. McClelland (1970) and Winter (1973) make this distinction within power and distinguish a personalized- (self-serving) and a socialized (others-serving) power motive. The distinction made by McClelland and Winter is relevant for abusive supervision, since several authors find positive behaviors resulting from a socialized power motive (Magee & Langer, 2008; McClelland, 1975) and negative behaviors resulting from a personalized power motive (Magee & Langer, 2008; McClelland & Burnham, 2008). I will therefore stick to the distinction of the power motives made by McClelland and Winter. Based on previous research, I argue that narcissists have both power motives, but that a personalized power motive may be positively related to abusive supervision, while a socialized power motive may be negatively related to abusive supervision.
behaviors intuitively outweigh the potential positive behaviors of these leaders. With this research, I give insights in the additional trait (moral identity) and power motives that increase or decrease the likelihood of negative behaviors (abusive supervision). When HR managers take this in mind in recruitment and promotion decisions, the potential positive outcomes of narcissistic leaders could be kept, while diminishing or avoiding the negative (abusive) behaviors of narcissistic leaders.
In the remainder of this thesis I will define the key concepts, discuss relevant antecedents and consequences of the concepts and I will make my arguments. This is done respectively for the relationships between grandiose narcissism and the personalized- and socialized power motive and for the relationship between the personalized- and socialized power motive and abusive supervision. After that, I will argue for the moderating influence of moral identity on the relationship between the personalized- and socialized power motive and abusive supervision. Then, I will visualize the hypotheses in a conceptual model. Furthermore, I will describe the methods that I have used and the results that I have obtained. Lastly, the discussion section will elaborate on the results obtained and will discuss the theoretical and practical implications, the limitations, and areas for further research.
THEORY AND HYPOTHESES
Grandiose narcissism and power motives
Grandiose narcissism. In this thesis, I will argue that grandiose narcissism is related to a personalized- and socialized power motive. In order to do this, I will first define and discuss relevant consequences of grandiose narcissism. Grandiose narcissism is defined as a stable personality characteristic characterized by inflated self-views (Campbell et al., 2006).
group, narcissists take credits for success and do not acknowledge the efforts of other group members (Campbell, Reeder, Sedikides, & Elliot, 2000). Furthermore, narcissists enjoy being better than others (Morf, Weir, & Davidov, 2000), use social comparison tactics to influence others (Jonasen & Webster, 2012), seek opportunities to demonstrate how good they are relative to others (Morf & Rhodewalt, 2001), and they want to be admired by other people (Campbell, 1999). The underlying motives of grandiose narcissists are that they want to gain support for their inflated self-views (Brunell & Davis, 2016), reinforce their ego (Morf & Rhodewalt, 2001; Jones & Paulhus, 2014), and show their superiority (Morf & Rhodewalt, 2001).
General power motive. I will first argue that narcissists have a general power motive. A general desire for power is reflected in both the personalized- and the socialized power motive, since both motives lead persons to want to have influence over other people. After that, I will specifically argue that narcissists have a personalized- and a socialized power motive. Power is defined as the individual’s capacity to modify other’s states by providing or withholding resources or administrating punishments (Keltner, Gruenfeld, & Anderson, 2003).
to regulate behaviors related to achievements (Keltner et al., 2003). Keltner, Gruenfeld, and Anderson also suggest that individuals who are inclined to approach related rewards, will have the tendency to construe others as a means to one’s own end.
Based on previous described literature, I have two arguments to suggest that grandiose narcissists have a general desire to have power. First, narcissists want to impact and influence other persons (Morf & Rhodewalt, 2001; Hodson, Hogg, & McInnis, 2009). Power gives narcissists the opportunity to have impact on and influence over other persons, because one could modify other’s states by providing or withholding resources or administrating punishments (Keltner et al., 2003). Secondly, narcissists tend to be overconfident (Campbell, Goodie, & Foster, 2004). This may make narcissists sensitive for powerful positions, because one may think he knows best and may therefore trigger the desire to have power to change existing things. This is in line with the reasoning by Anderson and Galinsky (2006), who state that optimism in perceiving future events leads people to have a desire for power.
Personalized- and socialized power motive. Having argued that narcissists have a general desire for power, I will now argue that narcissists have a personalized- and a socialized power motive. A personalized power motive is defined as the desire to have influence over others for self-serving motives. The desire to have power for using it for the benefit of other people belongs to people who have a socialized power motive (McClelland, 1970).
I have two arguments to suggest that grandiose narcissism is related to a personalized power motive. First, narcissists want to be admired by others (Campbell, 1999). They are likely to be admired by others, because powerful positions are associated with higher levels of status and fame than non-powerful positions. By gaining admiration, their underlying goals of reinforcing their ego (Morf & Rhodewalt, 2001) and gaining support for their inflated self-views (Brunell & Davis, 2016) could be accomplished. Second, a position of power is associated with increased resources (Keltner, Gruenfeld, & Anderson, 2003) and these give narcissists the opportunity to maximize their own gain. Several authors have shown the egoistic tendencies of narcissists within a group (Jonasen et al., 2010; Campbell et al., 2000). For example, in a position of power where narcissists were able to allocate dollars, narcissists choose for the self and not for others (Jonasen et al., 2010). Narcissists might be egoistic because they believe that they deserve more, and are better than other persons (Campbell et al., 2006). Based on previous reasoning, I hypothesize:
Hypothesis 1: Grandiose narcissism is positively related to a personalized power motive.
about themselves and will give them the opportunity to demonstrate superiority. Based on previous reasoning, I hypothesize:
Hypothesis 2: Grandiose narcissism is positively related to a socialized power motive.
Having discussed the antecedents of power, I will now focus on its consequences. Leaders with a personalized power motive think about the consequences of their actions in terms of their own welfare and leaders with a socialized power motive think about consequences of their actions in terms of welfare for other people (McClelland, 1970). Furthermore, leaders with a socialized power motive want to see themselves as responsible individuals (Winter & Stewart, 1978), help subordinates (Yukl, 2006), and are more cooperative and indirect in their use of power to subordinates (Bunderson & Reagans, 2011). While leaders with a personalized power motive do not strive to see themselves as responsible (Winter & Stewart, 1978), are ruder to subordinates (McClelland & Burnham, 2008), and are more direct and confrontational in their use of power to subordinates (Bunderson & Reagans, 2011). In several experiments, Magee and Langler (2008) show that a personalized power motive is associated with anti-social decision-making and that a socialized power motive is associated with prosocial decision-making.
Power motives and abusive supervision
harm. But that abusive supervision is demonstrated as willful behavior to reach a certain objective, such as increasing the performance of a team or warning subordinates that certain behaviors will not be tolerated (Tepper, 2007). I consider abusive supervision a form of immoral behavior and anti-social behavior, because it harms subordinates in terms of job satisfaction and life satisfaction (Tepper, 2000).
The literature on antecedents of abusive supervision is in an early stage as pointed out by Tepper (2007) and Zhang and Bednall (2015). However, several investigations have been executed to examine supervisor-related antecedents of abusive supervision. Zhang and Bednall (2015) conclude that abusive supervision is positively related to an authoritarian leadership style and negatively related to a participative leadership style. Kiazad, Restubog, Zagenczyk, Kiewitz, and Tang (2010) find that leaders’ Machiavellianism is positively related to abusive supervision. Thus, that leaders who are willing to display a range of behaviors to maximize their self-interest (Christie & Weis, 1970) are more inclined to use abusive supervision. Furthermore, abusive supervision often takes place, because subordinates are seen as relatively safe targets by supervisors (Tepper, Duffy, Henle, & Lambert, 2006).
In the next paragraph, I will argue and hypothesize that a personalized power motive is positively related to abusive supervision. After that, I will argue and hypothesize that a socialized power motive is negatively related to abusive supervision.
supervision, since behaviors displayed in abusive supervision are mostly direct, uninhibited, and confrontational. With direct, uninhibited, and confrontational abusive behaviors such as: “ridiculing someone when he or she has done a bad job” and “putting someone down in front of others”, the leader reaches the goals of showing domination, control, and gain prestige. This is because the leader can show everyone that he or she knows best and can set clear boundaries of what is acceptable. In this way, the leader’s direct and confrontational way of using his or her power to advance his or her personal concerns may cause him/her to display abusive supervision. This reasoning is in accordance to the results found by Zhang and Bednall (2015), who find that a direct leadership style characterized by concerns for the self, domination, and control leads to abusive supervision. Furthermore, leaders with a personalized power motive may engage in abusive supervision, because leaders with a personalized power motive are predominantly neglecting the welfare of others (McClelland & Burnham, 2008). Magee and Langler (2008) show this anti-social tendency of leaders with a personalized power motive in several experiments. When one is displaying abusive supervision, one is also neglecting the welfare of other people and displaying anti-social behaviors, since research has shown the tremendous negative effects of abusive supervision on subordinates (Tepper, 2007). Leaders with a personalized power motive are neglecting the welfare of other people, because they think about the consequences of their actions in terms of their own welfare (McClelland, 1970) and because they do not strive to be seen as responsible (Winter & Stewart, 1978) for subordinates. Based on previous reasoning, I hypothesize:
Hypothesis 3: A personalized power motive is positively related to abusive supervision.
power to reach their collective goals and interests (McClelland 1970; Bunderson & Reagans, 2011). It is not likely that an indirect, restrained, and cooperative way of using power leads to abusive supervision, since behaviors displayed in abusive supervision are not indirect, restrained, and cooperative. Rather, a socialized way of using power could be seen as a way of helping and supporting the subordinates by subtly approaching someone when he or she has made a fault or/and by refraining from hostile behaviors. Zhang and Bednall (2015) support this reasoning, since they find that a participative leadership style is negatively related to abusive supervision. Therefore, it is likely that a restrained and cooperative way of using power does not lead to abusive supervision. Additionally, leaders with a socialized power motive may not engage in abusive supervision, because leaders with a socialized power motive care about the welfare of others (McClelland, 1970; Yukl, 2006). Indeed, in a series of experiments, Magee and Langler (2008) find that persons with a socialized power motive are inclined to engage in pro-social behaviors which is the opposite of the negative effects caused by abusive supervision (Tepper, 2007). This is because people with a socialized power motive carry out behaviors to make the subordinate feel good (McClelland, 1975) and they want to see themselves as responsible individuals (Winter & Stewart, 1978). Based on previous reasoning, I hypothesize:
Hypothesis 4: A socialized power motive is negatively related to abusive supervision.
Power motives, abusive supervision, and moral identity
compassionate, fair, friendly, generous, hardworking, helpful, honest, and kind. Moral identity consists of two dimensions: internalization (moral traits central to self-concept) and symbolization (respondent’s actions in the world). I will not make this distinction in the theory section, however as an additional analysis I will also report the effects of the internalization- and symbolization dimension separately.
Because leaders with a personalized power motive are predominantly neglecting the welfare of others. I suggest that a person’s moral identity influences the relationship between a personalized power motive and abusive supervision. People who have a high moral identity see themselves as moral persons. For example, by seeing themselves and acting: honest, generous, and kind (Acquino & Reed, 2002). However, when a person is neglecting the welfare of other people and using ways of power to reach a personal goal with little restriction in how to use power, a person is not honest, generous, and kind and does not act as a moral person. To remain consistent between conceptions of the moral self and the actions in the world (Younis & Yates, 1999), a person with a higher moral identity will be more inclined to execute fewer immoral behaviors and more moral behaviors. This would cause a person with a personalized power motive to display fewer abusive behaviors. On the other hand, people with a low moral identity does not care that much to be seen and act as a moral person. Therefore, people with a lower moral identity are less inclined to withdraw themselves when considering these behaviors. Thus, I hypothesize:
Hypothesis 5: A higher level of moral identity decreases the positive relationship of a personalized power motive on abusive supervision.
behavior, the positive feelings derived from acting in a way consistent with their moral self-conception will increase the chances that a person with a high moral identity will act moral. Persons with a lower moral identity do not derive these positive feelings from acting in a way consistent with their moral self-conception (O’ Fallon & Butterfield, 2011). When, occasionally they consider immoral/abusive behaviors, such as: ‘ridiculing others’ and ‘breaking promises’ they do not have this ‘extra’ tendency to behave moral as reflected in high levels of moral identity. Therefore, it is more likely that people with a lower moral identity will execute more immoral behaviors than people with a higher moral identity when having a socialized power motive. Thus, I hypothesize:
Hypothesis 6: A higher level of moral identity strengthens the negative relationship of a socialized power motive on abusive supervision.
The before mentioned hypotheses are visualized in the conceptual model of FIGURE 1.
METHODS
I investigated the relationship between one leader and the subordinate(s), since I measured the traits and motives of the leader and measured the perceptions of abusive supervision at the subordinate(s). I asked leaders who were in my network and I did not search for a specific type of leader. I asked the leaders to find a maximum of five subordinates to fill in a different survey. The survey was sent to 63 leaders and 255 subordinates. 50 leaders and 177 subordinates filled in the survey. The response rate among the leaders was 79% and the response rate among subordinates was 69%. Due to incomplete responses and the fact that some leaders did not complete the survey while their subordinates did, the sample for data analysis consisted of 48 leaders and 134 subordinates.
In general, the organizations where these leaders were employed were diverse and consisted of both the profit- and non-profit sector. However, one remarkable observation was that around 20% of the leaders were employed in the ‘education and research’ sector. 64.6% of the leaders and 48.5% of the subordinates were male. The leaders had a mean age of 40.4 (S.D. = 12.0) and the subordinates had a mean age of 34.9 (S.D. = 13.1). The mean tenure of leaders in the organization was 8.6 years (S.D. = 8.1) and for subordinates 6.2 years (S.D. = 7.7). Leaders were on average 3 years (S.D. = 4.3) leader in this specific team and subordinates were on average 1.7 years (S.D. = 2.6) years employed under this specific leader.
Measures
Grandiose narcissism. Jones and Paulhus (2014) have developed a relatively short
scale to measure grandiose narcissism which consisted of nine statements. I measured nine items on a five point Likert scale which ranged from ‘strongly disagree’ to ‘strongly agree’. An example statement of narcissism was: ‘’I insist on getting the respect I deserve”. I did not choose for the most used measurement scale of narcissism. This was because there are still lots of debates about how many factors the NPI measured and the aim of this research was to measure grandiose narcissism. The measurement scale of Jones and Paulhus (2014) measured grandiose narcissism. The reported reliability of the grandiose narcissism scale was relatively low (α = 0.47). I executed a factor analysis to see whether a new factor could be formed with a higher reliability and which was still in line with the chosen definition of grandiose narcissism. According to the KMO criteria, factor analysis was allowed (.55 > .50) (Malhotra & Birks, 2007). The results of factor analysis (Principal Component Method, Oblimin Rotation) based on the Eigenvalue criteria are shown in TABLE 1.
TABLE 1
Factor analysis for grandiose narcissism
Item: Component
1 2 3 4
1. People see me as a natural leader. .13 .35 .78 .08
2. I hate being the center of attention. (R) -.00 .73 -.02 .42 3. Many group activities tend to be dull without me. .51* .24 -.25 .03 4. I know that I am special because everyone keeps telling me so. .79 .01 .12 .07 5. I like to get acquainted with important people. .25 .31 -.59 -.09 6. I feel embarrassed if someone compliments me. (R) -.00 -.08 .10 .94
8. I am an average person. (R) -.08 .87 .11 -.25 9. I insist on getting the respect I deserve. .75 -.15 -.25 -.01
*Loading > 0.5 for practical significance (Hair, Black, Tatham, Anderson, & Babin, 2009).
In their search for a short narcissism scale which focused on grandiose narcissism, Jones and Paulhus (2014) took items from different theoretical origins from the NPI, such as: grandiosity, leadership, entitlement, and exhibitionism. According to Jones and Paulhus, item 3, 4, 7, and 8 had the theoretical origin of ‘grandiose’. Based on the definition of narcissism as an inflated self-view and the theoretical origin of the items, I chose for the items of the first component. In comparison with the other components, a higher reliability was generated, less data was lost, and three out of the four items had already been labeled specifically as grandiose by the NPI. According to Bushman, Bonucci, van Dijk, and Baumeister (2003) do grandiose narcissism also involves a sense of entitlement, therefore item 9 also fits within the chosen definition. Based on the previous reasoning, I formed a new grandiose narcissism scale with these items. (α = 0.62).
Personalized- and socialized power motive. There was not a well-known and used
my ideas to help people”. The helping motivation subscale consisted of 4 items. I asked the respondents to what extent he or she agreed with the statements of the power motives on a five point Likert scale which ranged from ‘strongly disagree’ to ‘strongly agree’. (α = 0.79 for personalized power motive and α = 0.69 for socialized power motive).
Moral identity. I measured moral identity by 11 items developed by Acquino and Reed
(2002). I measured the items on a five point Likert scale which ranged from ‘strongly disagree’ to ‘strongly agree’. In accordance with the research of Acquino and Reed, I gave the following text to the respondents before they filled in the statements: “Listed below are some characteristics that may describe a person (caring, compassionate, fair, friendly, generous, hardworking, helpful, honest, and kind). The person with these characteristics could be you or it could be someone else. For a moment, visualize in your mind the kind of person who has these characteristics. Imagine how that person would think, feel, and act. When you have a clear image of what this person would be like, you might answer the following statements”. An example statement was: “It would make me feel good to be a person who has these characteristics”. The 11 items that measured moral identity consisted of two dimensions: internalization and symbolization. As an additional analysis, I also investigated the effects of the two dimensions separately. The first 5 reported items for moral identity were related to the internalization dimension. The reliability of the internalization dimension was low. The last 6 reported items for moral identity were related to the symbolization dimension. (α = 0.79 for moral identity; α = 0.33 for the internalization dimension; and α = 0.86 for the symbolization dimension).
Abusive supervision. I measured abusive supervision by 6 items developed by Tepper
behavior with me," 3 is "He/she occasionally uses this behavior with me," 4 is "He/she uses this behavior moderately often with me," and 5 is "He/she uses this behavior very often with me”. An example statement was: “My boss gives me the silent treatment”. (α = 0.77).
Control variables. I controlled for the leader’s gender, because men are more
much as a person”. (α = 0.71 for positive affect; α = 0.78 for negative affect; and α = 0.92 for LMX).
Data analysis
Up to five subordinates filled in the score for the supervisor. The items were asked on the dyadic level and often more than one subordinate filled in the score for the supervisor. To be able to calculate with these different dyadic scores, I aggregated the data of the subordinates. I calculated the mean of the scores from the different subordinates, employed under the same leader, for each construct. In this way, the results from the dyadic level were aggregated to an average score on the team-level. I replaced missing data by giving it the mean of the series. Additionally, I calculated the mean of the different statements from the same measured construct for both the aggregated data of the subordinates and for the data from the leaders. After that, I standardized the constructs.
To test the hypotheses I made use of regression analysis. I used PROCESS, developed by Hayes, in SPSS for the model where both the mediator and the moderator effect had to be estimated. The benefit of PROCESS was that it offered the opportunity to estimate a model with two mediators and one moderator simultaneously (Hayes, 2012) which matched the purpose of this thesis. I used model 14 with 5000 bootstrap samples and a confidence level for confidence intervals of 95%.
I interpreted the results by means of p-values and bootstrapped confidence intervals. Additionally, I reported the standard errors.
RESULTS
TABLE 2
Descriptive Statistics and Intercorrelations
M SD 1 2 3 4 5 6 7 8 1.Leader-Member Exchange 3.91 0.59 (.92) 2.Negative Affect 1.49 0.31 .00 (.78) 3.Positive Affect 3.60 0.48 -.12 .17 (.71) 4.Grandiose Narcissism 2.68 0.49 -.03 -.47** -.13 (.62) 5.Moral Identity 3.35 0.50 -.11 -.14 -.01 .22 (.79) 6.Personalized Power Motive 3.01 0.66 -.03 -.39** .02 .50** .02 (.79) 7.Socialized Power Motive 3.57 0.49 -.07 -.19 -.12 .43** .27† .39** (.69) 8.Abusive Supervision 1.35 0.30 -.03 .43** -.03 -.04 .17 -.25† -.19 (.77)
Notes. Cronbach’s Alpha’s are positioned between parentheses on the diagonal. Cronbach’s
Alpha above .6 are considered as acceptable (Malhotra & Birks, 2007). † p < 0.10, * p < 0.05, ** p < 0.01.
Below, the results are given per hypothesis. In footnotes, I reported the results according to the original nine items of the grandiose narcissism scale by Jones and Paulhus (2014). Furthermore, I report the aforementioned additional analyses of the two dimensions of moral identity in text.
Test of hypotheses 1 and 2
To analyze if grandiose narcissism influences a personalized power motive, I executed a regression of personalized power motive on grandiose narcissism, as TABLE 3 shows. Grandiose narcissism influences a personalized power motive (model 2: β = 0.45, t = 3.18, p = 0.00)2. Thus, H1 is supported.
To analyze if grandiose narcissism also influences a socialized power motive, I executed a regression of socialized power motive on grandiose narcissism, as TABLE 3 demonstrates. Grandiose narcissism also influences a socialized power motive (model 4: β = 0.41, t = 2.82, p = 0.01)3. Thus, H2 is supported.
TABLE 3 Regression analyses
Personalized Power Motive Socialized Power Motive
Model 1 Model 2 Model 3 Model 4
Intercept 0.00 (0.14) 0.00 (0.13) -0.00 (0.14) 0.00 (0.13) Gender of the leader -0.30 (0.15)* -0.13 (0.14) -0.20 (0.15) -0.04 (0.15) Tenure leader in team -0.09 (0.15) -0.06 (0.13) -0.23 (0.15) -0.20 (0.14)
LMX -0.02 (0.15) -0.01 (0.13) -0.04 (0.15) -0.04 (0.14)
Narcissism 0.45 (0.14)** 0.41 (0.15)**
R² .09 .27 .08 .23
∆R² .18 .15
Notes. Standard errors are positioned between parentheses. † p < 0.10, * p < 0.05, ** p <
0.01.
Test of hypotheses 3 and 4
I executed a regression of abusive supervision on personalized power motive, as TABLE 4 shows, to analyze if a personalized power motive influences abusive supervision. A personalized power motive does not influence abusive supervision (model 6: β = 0.14, t = -0.80, p = 0.43; model 7: β = -0.16, t = 0.86, p = 0.21). Thus, H3 is not supported.
I also executed a regression of abusive supervision on socialized power motive, as TABLE 4 demonstrates, to analyze if a socialized power motive influences abusive supervision. A socialized power motive does influence abusive supervision marginally in model 6 and does not influence abusive supervision in model 7 (model 6, β = -0.28, t = -1.75,
p = 0.09; model 7: β = -0.25, t = -1.51, p = 0.14). Thus, H4 is partially supported.
Test of hypotheses 5 and 6
abusive supervision on the interaction of the two dimensions of moral identity (separately) and a personalized power motive. The internalization dimension does not influence the relationship between a personalized power motive and abusive supervision (β = 0.21, t = -1.06, p = 0.30). The symbolization dimension does not influence the relationship between a personalized power motive and abusive supervision (β = 0.28 t = 1.68, p = 0.10). Thus, H5 is not supported.
To analyze if moral identity influences the relationship between a socialized power motive and abusive supervision, I also executed a regression of abusive supervision on the interaction between moral identity and a socialized power motive, as TABLE 4 demonstrates. Moral identity does not influence the relationship between a socialized power motive and abusive supervision (model 7: β = 0.00, t = 0.01, p = 0.99). As an additional analysis, I executed a regression of abusive supervision on the interaction of the two dimensions of moral identity (separately) and a socialized power motive. The internalization dimension does not influence the relationship between a socialized power motive and abusive supervision (β = 0.12, t = 0.85, p = 0.40). The symbolization dimension does not influence the relationship between a socialized power motive and abusive supervision (β = -0.06 t = -0.38, p = 0.71). Thus, H6 is not supported.
TABLE 4 Regression analyses
Abusive supervision
Model 5 Model 6 Model 7
Intercept 0.00 (0.14) 0.00 (0.13) -0.00 (0.14)
Notes. Standard errors are positioned between parentheses. † p < 0.10, * p < 0.05, ** p <
0.01.
NA 0.48 (0.15)** 0.56 (0.16)** 0.58 (0.16)**
PA -0.13 (0.14) -0.13 (0.14) -0.13 (0.14)
Gender of the leader 0.03 (0.14) 0.02 (0.16) 0.00 (0.16) Tenure leader in team 0.22 (0.18) 0.21 (0.18) 0.20 (0.18) Tenure subordinate in team -0.17 (0.20) -0.21 (0.22) -0.26 (0.22)
Narcissism 0.31 (0.18)† 0.27 (0.19)
Personalized Power Motive -0.14 (0.17) -0.16 (0.18) Socialized Power Motive -0.28 (0.16)† -0.25 (0.16)
Mediating influence of a personalized- and a socialized power motive
Four criteria needs to be met before a mediation effect can occur (Baron & Kenny, 1986). (1) The effect of grandiose narcissism on abusive supervision needs to be significant without the effect of the mediators being measured. (2) The effect of grandiose narcissism on the power motives needs to be significant. (3) The effect of the power motives on abusive supervision needs to be significant. (4) Adding the power motives to the regression of grandiose narcissism on abusive supervision has to lead to a non-significant relationship between grandiose narcissism and abusive supervision (full mediation) or a strongly reduced relationship between grandiose narcissism and abusive supervision (partial mediation).
Based on the results of hypotheses 3 a mediation effect could not occur for the personalized power motive, because the regressions of abusive supervision on the personalized power motive are not significant. However, for the socialized power motive in model 6, a mediation effect could possibly occur. Although not estimated in one of the specified models without the mediation effect being measured, grandiose narcissism does not influence abusive supervision (β = 0.23, t = 1.33, p = 0.19). Thus, the mediation effects of a personalized- and a socialized power motive do not occur. Also, the moderated mediation effects do not occur, since the interaction effects and the mediation effects do not occur.
DISCUSSION
The main goal of this thesis was to investigate whether, and if applicable, why and when grandiose narcissists engage in abusive supervision. To achieve this goal, I formulated six hypotheses.
is related to a personalized power motive is in line with prior research (Rosenthal & Pittinsky, 2006). However, the analyses showed mixed results for the significant, positive relatedness of a socialized power motive. The nine-items-scale by Jones and Paulhus (2014) did not find a significant relatedness, while the four-item-scale did find a significant relatedness. I think this difference might be due to the theoretical origin of the statements. In their meta-analysis, Grijalva and Newman (2014) already concluded that combining different facets of narcissism could mask interesting facet-level effects. Although the main goal of Jones and Paulhus (2014) was to measure the grandiose aspect of narcissism, they also measured other facets.
The third hypothesis pertained to the positive relatedness of a personalized power motive to abusive supervision. The analyses revealed that this positive relatedness did not occur significantly. The fourth hypothesis pertained to the negative relatedness of a socialized power motive to abusive supervision. The analyses revealed mixed results. Adding the moderation effects to the regression of abusive supervision on socialized power motive made the relationship non-significant. Furthermore, the mediation effects of the power motives do not occur. A possible explanation for these non-significant results could be that I did not look at the situation, since one’s motive drives behavior based on the interaction with aspects of the situation (Maner, Gailliot, Butz, & Peruche, 2007). I assumed that one’s behavior could be derived directly from one’s motive, however this assumption might have been too simplistic.
dimension did not interact positively at a significant level. An explanation for these results might be that my assumptions were incorrect. I reasoned that abusive supervision must be anti-social and immoral, since it hurts subordinates (Tepper, 2007). However, one could also look at it from the perspective of the supervisors for whom it is willful behavior to reach a certain objective (Tepper, 2007) and therefore the intention to behave immoral or anti-social may not exist.
With regard to the main goal of the thesis, no relationship is found between grandiose narcissism and abusive supervision without the mediation effect being measured. This might by, because others facets of narcissism (e.g., exploitativeness) (Ackerman, Witt, Donnellan, Trzesnieuwski, Robins, & Kashy, 2010) might heavily account for the negative aspect narcissism is associated with. However, in one of the specified models, a marginal significant effect of grandiose narcissism on abusive supervision is found. Including the power motives to the regression with its high correlation with grandiose narcissism might have caused this marginal significant effect. This marginal significant result also disappeared when the moderation effects were added.
Theoretical implications
narcissism to be aware of the different facets of narcissism, since it could lead to different outcomes.
A small contribution to the power motives and abusive supervision literature might be that I found a marginal significant negative relationship between the socialized power motive and abusive supervision in one of the two estimated models. One has to be cautious with interpreting this result, because it is only marginal significant and the other specified model did not find a significant relationship. However, future researchers could use this result as a starting point and could investigate if the marginal significant result hold under other circumstances. I would advise others to be cautious with making direct links between motives and behavior. That it would be wise to also take the situation into account, since several authors stresses the importance (Maner et al., 2007; Magee & Langler, 2008). For example, Maner et al. (2007) show that people with high power motives act more conservative when they think their position is threatened. Controlling for several situational factors or using them as moderator variables might thus help to explain more variance in the estimated models.
It could be that the assumptions that I made do not hold for abusive supervision for the moral identity literature. One could argue that abusive supervision could not be considered anti-social and immoral, since the goal of abusive supervision is not to cause harm, but to reach a certain objective (Tepper, 2007) and that it might be useful for the leader to display direct and abusive behaviors to reach this objective. Furthermore, I would advise other authors in their research to abusive supervision to include the negative affect of the subordinates in the regression, because of its large effect on abusive supervision in all the reported models.
Practical implications
Therefore, I would advise HR managers to focus on changing the negative affect of subordinates to minimize perceptions of abusive supervision, since this has a large impact on perceptions of abusive supervision (Tepper et al., 2006).
The finding that grandiose narcissists also want to have a position of power to help others can be used as a strategic tool for HR managers. Grandiose narcissists can be located to places where they can be assigned a role of teacher (e.g., mentor). This can be a win-win situation for both the grandiose narcissists and the trainee, since the trainee can obtain new knowledge and the narcissists him or herself can feel superior by helping the trainee.
Limitations and future research directions
Other areas for future research
Future research can also investigate which facet of narcissism is associated with abusive supervision, since contradictory results exist in this field. Furthermore, I would recommend authors to investigate which situational factors are most influential in the relationship between (power) motives and abusive behaviors. Finding influential situational factors could expose significant different consequences of the power motives. As a first step, I would advise other researchers to include ‘perceived threat to the leader’s position’ into account as a moderator. Based on the results by Maner et al. (2008), I would expect that leaders will behave less abusive when their position is threatened, since they showed that leaders will behave more conservative when their position is threatened. Literature from antecedents of abusive supervision points out that abusive supervision often occurs because of ‘displaced aggression’ from the supervisor (Tepper, 2007). It is known from narcissists that they engage in negative behaviors, because their self-esteem is threatened (Bushman & Baumeister, 1998). Afterwards, I could have controlled for the self-esteem threat or I could have chosen a variable which is related to ‘displaced aggression’ of the supervisor. The before mentioned ‘displaced aggression’ and ‘threat to their self-esteem’ are especially interesting, since Waldman et al. (2018) find that supervisor’s self-control is a significant variable that influences the relationship between narcissism and abusive supervision. It could be interesting for future research to incorporated variables that involve a threat to the narcissists (e.g., negative feedback from their supervisor) and see whether they displace it to abuse their subordinates.
Conclusion
narcissistic leader would in general lead to more positive behaviors for the subordinates and the self-serving motive would in general lead to more negative behaviors for the subordinates. To the utmost extent I did not find evidence for the relationship between grandiose narcissism and abusive supervision. I did not find support for the relationship between the personalized power motive and abusive supervision, and mixed support for the relationship between the socialized power motive and abusive supervision. Much is still unknown why narcissism is related to abusive behaviors, but also to positive behaviors. I hope that other authors would be motivated to find the reason for this paradox.
REFERENCES
Ackerman, R., Witt, E., Donnellan, M., Trzesniewski, K., Robins, R., & Kashy, D. 2010. What Does the Narcissistic Personality Inventory Really Measure? Assessment, 18(1): 67-87.
Aquino, K., & Reed, A. 2002. The self-importance of moral identity. Journal of Personality and Social Psychology, 83(6): 1423-1440.
Ahmadian, S., Azarshahi, S., & Paulhus, D. 2017. Explaining Donald Trump via communication style: Grandiosity, informality, and dynamism. Personality and Individual Differences, 107: 49-53.
Ambrose, M. L., & Schminke, M. 1999. Sex differences in business ethics: The importance of perceptions. Journal of Managerial Issues, 11: 454–474.
Anderson, C., & Galinsky, A.D. 2006. Power, optimism, and risk-takin. European Journal of Social Psychology, 36: 511-536.
Baron, R., & Kenny, D. 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.
Baughman, H., Dearing, S., Giammarco, E., & Vernon, P. 2012. Relationships between bullying behaviors and the Dark Triad: A study with adults. Personality and Individual Differences, 52(5): 571-575.
Boneva, B.S., Frieze, I.H., Ferligoi, A., Jarasova, E., Pauknerova, D., & Orgocka, A. 1997. East-West European migration and the role of motivation in emigration desires. Migracijske Teme, 13(4): 247-254.
Brunell, A. B., Gentry, W. A., Campbell, W. K., Hoffman, B. J., Kuhnert, K. W., & DeMarree, K. G. 2008. Leader emergence: The case of the narcissistic leader. Personality and Social Psychology Bulletin, 34(12): 1663–1676.
Brunell, A., & Davis, M. 2016. Grandiose Narcissism and Fairness in Social Exchanges. Current Psychology, 35(2): 220-233.
Bunderson, J.S., & Reagans, R.E. 2011. Power, Status, and Learning in Organizations. Organization Science, 22(5): 1182-1194.
Bushman, B., & Baumeister, R. 1998. Threatened egotism, narcissism, self-esteem, and direct and displaced aggression: Does self-love or self-hate lead to violence? Journal of Personality and Social Psychology, 75(1): 219-229.
Bushman, B. J., Bonacci, A. M., van Dijk, M., & Baumeister, R. F. 2003. Narcissism, sexual refusal, and aggression: Testing a narcissistic reactance model of sexual coercion. Journal of Personality and Social Psychology, 84: 1027-1040.
Campbell, W., Reeder, G., Sedikides, C., & Elliot, A. 2000. Narcissism and Comparative Self-Enhancement Strategies. Journal of Research in Personality, 34(3): 329-347. Campbell, W.K. Goodie, A.S., & Foster. J.D. 2004. Narcissism, confidence, and risk attitude.
Behavioral Decision Making, 17(4): 297-311.
Campbell, W. K., Brunell, A. B., & Finkel, E. J. 2006. Narcissism, interpersonal self-regulation, and romantic relationships: an agency model approach. New York: Guilford.
Campbell, W. K., Hoffman, B. J., Campbell, S. M., & Marchisio, G. 2011. Narcissism in organizational contexts. Human Resource Management Review, 21(4), 268–284. Christie, R., & Weis, F. L. 1970. Studies in Machiavellianism. New York: Academic Press. DeCelles, K., DeRue, D., Margolis, J., & Ceranic, T. 2012. Does power corrupt or enable?
When and why power facilitates self-interested behavior. Journal of Applied Psychology, 97(3): 681-689.
Donald J. Trump (@realDonaldTrump) | Twitter. 2017. Twitter.com. https://twitter.com/realdonaldtrump, June 25, 2017.
Emmons, R. 1987. Narcissism: Theory and measurement. Journal of Personality and Social Psychology, 52(1): 11-17.
Emmons, R. A. 1989. Exploring the relations between motives and traits: The case of narcissism. New York: Springer-Verlag.
Frieze, I. H., & Boneva, B. S. 2001. Power motivation and motivation to help others. In The use and abuse of power, ed. J. A. Bargh, A. Y. Lee-Chai, pp. 75-89. Ann Arbor, MI: Sheridan Books.
Grijalva, E., Newman, D., Tay, L., Donnellan, M., Harms, P., Robins, R., & Yan, T. 2015. Gender differences in narcissism: A meta-analytic review. Psychological Bulletin, 141(2): 261-310.
Hair, J.F., Black, W.C., Tatham, R.L, Anderson, R.E., & Babin, B.J. 2009. Multivariate Data Analysis, New York: Pearson Education.
Hansbrough, K. T., & Jones, G. 2014. Inside the Minds of Narcissists: How narcissistic leaders' cognitive processes contribute to abusive supervision. Zeitschrift Für Psychologie, 222(4): 214-220.
Harvey, P., Harris, K., Gillis, W., & Martinko, M. 2014. Abusive supervision and the entitled employee. The Leadership Quarterly, 25(2): 204-217.
Hayes, A.F. 2012. PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modelling [white paper]. Retrieved from http://www.afhayes.com/public/process2012.pdf.
Hodson, G., Hogg, S., & MacInnis, C. 2009. The role of “dark personalities” (narcissism, Machiavellianism, psychopathy), Big Five personality factors, and ideology in explaining prejudice. Journal of Research in Personality, 43(4): 686-690.
Jonasen, P.K., Li, N.P., & Teicher, E.A. 2010. Who is James Bond? The Dark Triad as an Agentic Social Style. Individual Differences Research, 8(2): 111-120.
Jonasen, P., & Webster, G. 2012. A protean approach to social influence: Dark Triad personalities and social influence tactics. Personality and Individual Differences, 52(4): 521-526.
Jones, D., & Paulhus, D. 2010. Different Provocations Trigger Aggression in Narcissists and Psychopaths. Social Psychological and Personality Science, 1(1): 12-18.
Keltner, D., Gruenfeld, D.H., & Anderson, C. 2003. Power, Approach and Inhibition. Psychological review, 110(2): 265-284.
Kiazad, K., Restubog, S., Zagenczyk, T., Kiewitz, C., & Tang, R. 2010. In pursuit of power: The role of authoritarian leadership in the relationship between supervisors’ Machiavellianism and subordinates’ perceptions of abusive supervisory behavior. Journal of Research in Personality, 44(4): 512-519.
Konrath, S., Ho, M., & Zarins, S. 2016. The Strategic Helper: Narcissism and Prosocial Motives and Behaviors. Current Psychology, 35(2): 182-194.
Lee, K., Ashton, M., Wiltshire, J., Bourdage, J., Visser, B., & Gallucci, A. 2012. Sex, Power, and Money: Prediction from the Dark Triad and Honesty-Humility. European Journal of Personality, 27(2): 169-184.
Liden, R., & Maslyn, J. 1998. Multidimensionality of Leader-Member Exchange: An Empirical Assessment through Scale Development. Journal of Management, 24(1): 43-72.
Maccoby, M. 2000. Narcissistic leaders: The incredible pros, the inevitable cons. Harvard Business Review, January–February: 69−77.
Magee, J., & Langner, C. 2008. How personalized and socialized power motivation facilitate antisocial and prosocial decision-making. Journal of Research in Personality, 42(6): 1547-1559.
Malhotra N.K., & Birks, D.F. 2007. Marketing research: an applied approach, Harlow: Pearson education limited.
Martinko, M., Sikora, D., & Harvey, P. 2012. The Relationships between Attributions Styles, LMX, and Perceptions of Abusive Supervision. Journal of Leadership & Organizational Studies, 19(4): 397-406.
McClelland, D. C. 1970. The two faces of power. Journal of International Affairs, 24(1): 29–47.
McClelland, D.C. 1975. Power: The inner experience. New York: Irvington.
McClelland, D., & Burnham, D. 2008. Power is the great motivator. Boston, Mass: Harvard Business Press.
Megargee, E. 1969. Influence of sex roles on the manifestation of leadership. Journal of Applied Psychology, 53(5): 377-382.
Morf, C., Weir, C., & Davidov, M. 2000. Narcissism and Intrinsic Motivation: The Role of Goal Congruence. Journal of Experimental Social Psychology, 36(4): 424-438. Morf, C., & Rhodewalt, F. 2001. Unraveling the Paradoxes of Narcissism: A Dynamic
Self-Regulatory Processing Model. Psychological Inquiry, 12(4): 177-196.
O’Fallon, M., & Butterfield, K. 2011. Moral Differentiation: Exploring Boundaries of the “Monkey See, Monkey Do” Perspective. Journal of Business Ethics, 102(3): 379-399.
Paulhus, D., & Williams, K. 2002. The Dark Triad of personality: Narcissism, Machiavellianism, and psychopathy. Journal of Research in Personality, 36(6): 556-563.
Rosenthal, S., & Pittinsky, T. 2006. Narcissistic leadership. The Leadership Quarterly, 17(6): 617-633.
Tepper, B. 2000. Consequences of abusive supervision. Academy of Management Journal, 43(2): 178-190.
Tepper, B., Duffy, M., Henle, C., & Lambert, L. 2006. Procedural injustice, Victim precipitation, and abusive supervision. Personnel Psychology, 59(1): 101-123.
Tepper, B. 2007. Abusive Supervision in Work Organizations: Review, Synthesis, and Research Agenda. Journal of Management, 33(3): 261-289.
Thompson, E. 2007. Development and Validation of an Internationally Reliable Short-Form of the Positive and Negative Affect Schedule (PANAS). Journal of Cross-Cultural Psychology, 38(2): 227-242.
Waldman, D.A., Wang, D., Hannah, S.T., & Owned, B.P. 2018. Psychological and Neurological Predictors of Abusive Supervision. Unpublished paper, California State University, Sacramento.
Wisse, B., & Sleebos, E. 2016. When the dark ones gain power: Perceived position power strengthens the effect of supervisor machiavellianism on abusive supervision in work teams. Personality and Individual Differences, 99: 122-126.
Winter, D. 1973. The power motive (1st Ed.). New York: Free Press.
Winter, D. G., & Stewart, A. J. 1978. The power motive. New York: Wiley.
Younis, J., & Yates, M. 1999. Youth service and moral-civic identity: A case of everyday morality. Educational Psychology Review, 11: 361–376.
Yukl, G. A. 2006. Leadership in organizations. Upper Saddle River, NJ: Pearson–Prentice Hall.
APPENDIX A: MEASUREMENT ITEMS
Narcissism (Jones & Paulhus, 2014)
1. People see me as a natural leader. 2. I hate being the center of attention. (R)
3. Many group activities tend to be dull without me.*
4. I know that I am special because everyone keeps telling me so.
5. I like to get acquainted with important people. 6. I feel embarrassed if someone compliments me. (R)
7. I have been compared to famous people.
8. I am an average person. (R)
9. I insist on getting the respect I deserve.
*Statements in italic are used for the grandiose narcissism scale.
Personalized power motive (Schmidt & Frieze, 1997)
Prominence
6. I would like doing something important where people looked up to me. 7. I find satisfaction in having influence over others.
8. I enjoy debating with others in order to get them to see things my way. 9. I want to be a prominent person in my community.
10. I like to be admired for my achievements.
Socialized power motive (Schmidt & Frieze, 1997)
Helping
8. When people I know are trying to solve problems, my gut instinct is to offer them helpful suggestions.
9. It would be satisfying to be able to have impact on the quality of others' lives.
Moral identity (Aquino and Reed, 2002)
1. It would make me feel good to be a person who has these characteristics. 2. Being someone who has these characteristics is an important part of who I am. 3. I would be ashamed to be a person who has these characteristics. (R)
4. Having these characteristics is not really important to me. (R) 5. I strongly desire to have these characteristics.
6. I often buy products that communicate the fact that I have these characteristics. 7. I often wear clothes that identify me as having these characteristics.
8. The types of things I do in my spare time (e.g., hobbies) clearly identify me as having these characteristics.
9. The kinds of books and magazines that I read identify me as having these characteristics. 10. The fact that I have these characteristics is communicated to others by my membership in certain organizations.
11. I am actively involved in activities that communicate to others that I have these characteristics.
Abusive Supervision (Tepper, 2000)
1. Gives me the silent treatment. 2. Breaks promises he/she makes.
3. Expresses anger at me when he/she is mad for another reason. 4. Makes negative comments about me to others.
Control variables PANAS (Thompson, 2007) Positive Affect. 1. Determined. 2. Attentive. 3. Alert. 4. Inspired. 5. Active. Negative Affect. 1. Afraid. 2. Nervous. 3. Upset. 4. Ashamed. 5. Hostile.
LMX (Liden & Maslyn, 1998).
1. I like my supervisor very much as a person.
2. My supervisor is the kind of person one would like to have as a friend. 3. My supervisor is a lot of fun to work with.
4. My supervisor defends my work actions, even without complete knowledge of the issue in question.
5. My supervisor would come to my defense if I were "attacked" by others.
