Employees with high self-esteem suffer most from the detrimental effects of abusive supervision and exploitative leadership

(1)

Employees with high self-esteem suffer most from the

detrimental effects of abusive supervision and

exploitative leadership

Julia Einarson

11001097

Bachelor thesis BSc Management and Leadership in the Digital Age

University of Amsterdam

Faculty Economics and Business

Thesis supervisor: dr. Armin Pircher Verdorfer

Word count: 7.164

(2)

Statement of Originality

This document is written by Julia Einarson 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.

(3)

Abstract

Every day, many employees all over the world are confronted with supervisors engaging in destructive leadership. In this research, two variants of destructive leadership are covered: the inherently hostile “abusive supervision” and the more subtle and indirect “exploitative leadership”. Research has shown that being faced with any of these two variants is detrimental to employee well-being and as such can have very negative consequences. This study aims to discover whether self-esteem acts like a buffer against these harmful effects and whether self-esteem buffers differently against the effects of abusive supervision than those of exploitative leadership. By conducting an online-survey among 133 employee-significant other dyads, current research was supported by finding a correlation between aforementioned two destructive leadership styles and reduced well-being. Additionally, an interaction effect was found not only between abusive supervision and self-esteem on emotional exhaustion, but also between exploitative leadership and self-esteem on emotional exhaustion. However, not in the direction that was expected. Instead of acting as a buffer, the results suggest that self-esteem actually enhances the detrimental effects of the two variants of destructive leadership. Thus, abusive supervision and exploitative leadership are indeed detrimental to employee well-being by leading to higher emotional exhaustion. Additionally, I found that self-esteem does not serve a buffering purpose, but that employees high on self-esteem actually suffer more from being the victim of abusive supervision and exploitative leadership compared to those with lower self-esteem.

(4)

List of Tables and Figures

Figure 1: Visualization of the research question ... 7

Figure 2: Plotted interaction of abusive supervision x self-esteem on emotional exhaustion . 17

Figure 3: Plotted interaction of exploitative leadership x self-esteem on emotional exhaustion

... 18

Table 1: Means, standard deviations and correlations ... 16

Table 2: Results for the interaction effect between abusive supervision and self-esteem on

emotional exhaustion ... 17

Table 3: Results for the interaction effect between exploitative leadership and self-esteem on

emotional exhaustion (when not controlling for abusive supervision) ... 19

Table 4: Results for the interaction effect between exploitative leadership and self-esteem on

emotional exhaustion (when controlling for abusive supervision) ... 19

(6)

1. Introduction

Leadership has been the subject of many studies and its definitions vary greatly in terms of emphasis on, among other things, leader personality or traits, or whether the definition is primarily descriptive or normative in nature (Den Hartog & Koopman, 2001). In the past, research has mainly revolved around the constructive side of leadership, focusing on how to establish effective leadership, such as by identifying certain traits and/or effects which make supervisors more effective and provide favourable outcomes in order to be able to develop strategies and recommendations on how to improve management styles and techniques. Examples of such research are the research of Dobre (2013), who investigated how to motivate employees, and Luthans (2000), who researched how to increase employee performance through supervisor recognition. Ineffective leadership, on the other hand, was often assumed simply to be a reflection of the absence of leadership altogether (Ashforth, 1994).

However, nothing could be further from the truth. There do, in fact, exist leadership styles that can arguably be classified as destructive leadership, as they can have very negative consequences. As a matter of fact, there is solid evidence to be found in the literature for the negative and costly effects of sustained leader mistreatment on key workplace outcomes, including work attitudes, performance, and well-being (Schyns & Schilling, 2013). Notably, the influence of negative leader behaviours has consistently been found to outstrip the impact of co-worker or customer mistreatment (Herschcovis & Barling, 2010). Additionally, Schat, Frone, and Kelloway (2006) found that approximately 13.5% of American workers are faced with abusive behaviours from their boss, which, in the medium and/or long-run, can have severe psychological and somatic consequences, among other negative outcomes (Schat and Kelloway, 2005). While this percentage suggests that the majority of employees are in fact not faced with destructive leadership, the effects on those that are, and thus the consequences of such destructive leadership styles in general, are not to be underestimated. This is based on the “bad is stronger than good” logic, implying that negative consequences are particularly severe (Baumeister, Bratslavsky, Finkenauer & Vohs, 2001).

Over approximately the last two decades, the focus has gradually shifted to include this more malevolent side of leadership. Most of the research has covered the topic of abusive supervision, which is uniquely defined by its hostility and aggressiveness (Tepper, 2000; Tepper, 2007; Mitchell & Ambrose, 2007; Shyns and Shilling, 2013; Barnes, Lucianetti, Bhave and Christian, 2015). Lately, researchers have also begun to look at less hostile but still harmful forms of destructive leadership. Most notably, the concept of exploitative leadership has been introduced, which leadership style is not inherently aggressive but revolves mostly around leaders pursuing their own self-interest at the expense and exploitation of their subordinates (Schmid, Pircher Verdorfer & Peus, 2019).

Subsequently, the question is whether the abusive supervision and exploitative leadership behaviours are redundant, or whether there are any meaningful differences between the two. Moreover, does exploitative leadership have something to offer that goes further than abusive supervision?

(7)

So, while the dark side of leadership has increasingly been gaining attention, and multiple researchers such as Tepper (2000) and Harvey, Stoner, Hochwarter and Kacmar (2007) have observed the negative effects of destructive leadership on employee-related-outcomes, there are still multiple areas that have remained untouched by research. One such area that we still have a limited understanding of is whether aforementioned variants of destructive leadership, abusive leadership and exploitative leadership, have different effects on employee related work outcomes such as well-being, and how these effects might be mediated or moderated by other factors. This is important to know, because we need to show that related destructive leadership constructs work through different processes or have different effects to really determine their unique versus redundant contributions (Lemoine, Hartnell & Leroy, 2019).

This knowledge is subsequently of value, because having a proper understanding of the traits and behaviours associated with these variants of destructive leadership and how these interact with each other, as well as their impact on work outcomes, is invaluable for companies when deciding which candidate to hire for a leadership position. Moreover, since so many people are impacted severely by these behaviours, it is only logical to improve and build upon our understanding of the drivers and factors behind these destructive leadership styles, in order to maybe one day be able to reduce or even eliminate these destructive leadership styles altogether, to improve follower well-being as well as organizational outcomes.

With this research, I aim to contribute to this understanding as well as address this literature gap by researching whether employee self-esteem acts like a buffer of sorts against the harmful effects of destructive leadership on employee well-being, resulting in lower emotional exhaustion. Moreover, I will research whether employee self-esteem might buffer differently against the effects of abusive supervision than exploitative supervision on employee well-being. The research questions are visualized in Figure 1.

I will begin by defining the different variables whose interaction will be measured in this research based on the existing literature, starting with destructive leadership in general, followed by its two components: abusive- and exploitative leadership. Reviewing of the definitions is particularly important when it comes to the variable abusive leadership, as there has been some trouble with its

Destructive leadership Abusive leadership Exploitative leadership Emotional exhaustion Employee self-esteem

(8)

terminology in past research, and clarification in this regard is essential. Then, the variables employee self-esteem and emotional exhaustion will be defined and reviewed. I then proceed by describing the methods used to conduct this research and gather the required data, which, in short, is through an online survey. Lastly, I will analyze, state and then discuss the data and results gathered by the survey and describe their practical implications, their limitations and provide recommendations for future research.

2. Theoretical framework

2.1 Destructive leadership

Defining destructive leadership is quite the daunting task, for a multitude of reasons. First of all, there are different researchers, such as Einarsen, Aasland & Skogstad (2007), who have proposed varying definitions of the term, or those who have even argued that leadership can only be defined in a positive way (Yukl and Van Fleet, 1992) and as such there can be no such thing as “destructive” leadership.

Moreover, it is tricky to traverse the landscape that involves defining destructive leadership, as it can be very confusing. Not only is the amount of concepts that destructive leadership arguably incorporates, such as “abusive supervision” (Tepper, 2000), “petty tyrants” (Ashforth, 1994) and “exploitative leadership” (Schmid, Pircher Verdorfer and Peus, 2019) immense, but also because terms such as destructive leadership behaviour have been introduced (Thoroughgood, Tate, Sawyer & Jacobs, 2012), which seems similar to destructive leadership at first glance but in fact is very different. In a meta-analysis, Shyns and Shilling (2013, p. 141) actually set out on the quest to, for the first time ever, quantitively review the consequences of destructive leadership, and they began by doing away with all the previous inconsistencies and ambiguity around its definition, and ended up defining it as “a process

in which over a longer period of time the activities, experiences and/or relationships of an individual or the members of a group are repeatedly influenced by their supervisor in a way that is perceived as hostile and/or obstructive”, which is also the definition which will be followed in the current study.

2.2 Abusive supervision

When researchers first shifted their focus to the dark sides of leadership, most of the research covered the topic of abusive supervision. Interestingly however, in his article, Tepper (2007, p. 262) describes how there has been some trouble with its terminology. The literature surrounding this concept can be described as “fragmented and poorly integrated”, he states. He continues by writing that

“researchers have used different terminology and corresponding measures to investigate phenomena that overlap to varying degrees with abusive supervision, and extant research does not devolve from a unifying theoretical framework.” This is problematic because if left untangled, these embroiled concepts

(9)

and definitions have the potential to undermine the growth of the understanding of this important area of research.

Tepper (2000) defined abusive supervision as the extent to which subordinates view their supervisors as consistently behaving in a hostile and aggressive manner, both verbal and non-verbal (excluding physical abuse). The fact that it is based on the subordinates’ views makes it a subjective assessment. Abusive supervision can be characterised as enduring, and there are multiple features of the abusive relationship that add to its enduring nature. First of all, one of the reasons behind abusive-behaviours-victims remaining in the abusive situation is that they in some way or form are dependent on their abuser (Walker, 1979). In the case of abusive supervision, this dependence would be economically. Secondly, abusers often don’t recognize and/or take responsibility for their own behaviour, and those that do, often don’t modify it (Wolfe, 1987).

Mitchell and Ambrose (2007), by performing exploratory and confirmatory factor analysis using the original 15-item measure by Tepper (2000), actually found that abusive supervision exists of two different factors: One reflecting active interpersonal abuse and one that reflects passive acts of abuse. My research is focused on the core of abusive supervision, which would be its aggressive nature. This is most in line with what Mitchell and Ambrose (2007) describe as “active aggressive abusive supervision”.

What is interesting, is that abusive supervision’s definition matches the description of destructive leadership as described by Shyns and Shilling (2013), as the two are both enduring in nature, and individuals are impacted by their supervisor’s hostile behaviours.

2.3 Exploitative leadership

Not long ago, researchers have broadened their interest to not only include aforementioned direct, hostile style of leadership, but also a more indirect form of destructive leadership: exploitative leadership. This concept was recently introduced and defined by Schmid, Pircher Verdorfer and Peus (2019) as a form of leadership that is not inherently violent but revolves mostly around leaders pursuing their own self-interest at the expense and exploitation of their subordinates. The nature of exploitative leadership is reflected in five dimensions. The first dimension is “genuine egoistic behaviours”, which includes descriptors such as “my leader takes it for granted that my work can be used for his or her personal benefit. The second factor is “taking credit”, which is, for example, characterized by leaders benefiting from their subordinates’ work without sharing the praise. The next dimension is “exerting pressure”, the nature of which is overburdening followers. The fourth factor is “underchallenging followers”, which in practice translates to leaders not allowing their followers to develop themselves further professionally, because their own goals take priority above all else. The final dimension is “manipulating followers”, which involves items such as “my leader plays my colleagues and me off against each other to reach his or her own goals”.

(10)

Interestingly enough, exploitative leadership also matches the description of destructive leadership as given by Shyns and Shilling (2013) very well. However, as opposed to abusive supervision, the concept of exploitative leadership does not express itself in direct hostility, but rather in a more subtle exploitation, and as such provides great potential for explaining the “obstructive” part of the destructive leadership definition.

2.4 The effects of abusive supervision vs. exploitative leadership on employee

well-being

The harmful effects of abusive supervision on employee well-being is a well-researched phenomenon. It revolves around long-lasting emotional and/or psychological mistreatment of employees (Zellars, Tepper & Duffy, 2002) and multiple studies link abusive supervision, as a workplace stressor, to poor employee well-being, suggesting, among other things, psychological and somatic consequences (Schat and Kelloway, 2005), depression (Kessler, Spector, Chang, & Parr, 2008; Tepper, 2000), burnout (Grandey, Kern & Frone, 2007; Tepper, 2000; Yagil, 2006) and emotional exhaustion (Hobman et al., 2009; Tepper, 2000). The Job Demands Resources Model has potential for explaining these phenomenon, as abusive supervision depletes individual resources, and by doing so reduces well-being (Bakker & Demerouti, 2007).

The research of Schmid, Pircher Verdorfer and Peus (2019) states that exploitative leadership has a negative effect on employee well-being. Their research suggests that there is a negative relationship between exploitative leadership and job satisfaction, whilst it increases followers’ burnout. As the concept of exploitative leadership has only recently been introduced, there is no further research available to support their findings and provide further proof for the observed effects on employee-related outcomes that were found.

While the definitions of both abusive supervision and exploitative leadership are in line with that of destructive leadership, there certainly are differences to be found between the two styles. As has been mentioned before, abusive supervision is direct, hostile and aggressive, while exploitative leadership is more subtle, sneaky and clandestine. Based on all that is mentioned above, I expected both abusive supervision and exploitative leadership to have negative effects on well-being. Accordingly, I hypothesize:

Hypothesis 1a: Abusive supervision is positively related to emotional exhaustion.

Hypothesis 1b: Exploitative leadership is positively related to emotional exhaustion.

However, due to the lack of available research, it was impossible to predict which of these two variants of destructive leadership is most harmful to employee well-being and results in higher

(11)

emotional exhaustion. On the one hand, abusive supervision, characterized by its inherent direct acts of hostility and aggressiveness, can really take massive hits on an employees’ well-being each time he or she is confronted with these abusive behaviours. On the other hand, the subtle, ongoing nature of exploitative leadership allows it, over time, to slowly but surely wear down on employees’ tenacity. Finding sound arguments that substantiate the harmful effects of not just one, but for both these destructive leadership styles and lacking further research on the matter, I could not predict which style was worse, and thus instead of developing a hypothesis, I decided to frame it as a research question:

Research question 1: Which of the two variants of destructive leadership, abusive supervision or exploitative leadership, is most detrimental to employee well-being?

2.5 The moderating role of employee self-esteem

Employee self-esteem has been a subject of research for quite some time now. In their study, Pierce and Gardner (2004, p. 592) speak of “organization-based self-esteem”, which is “a

conceptualization of self-esteem largely determined by an individual’s work and organizational experiences”. While this certainly does touch upon the subject of employee self-esteem, it is important

to mention that this is not in line with the definition of self-esteem that will be followed in the current research. An important nuance in Pierce and Gardner’s (2004) definition is namely that the individual’s self-esteem is largely determined by his or her work and organizational experiences. Self-esteem as investigated in this research, on the other hand, should be interpreted more globally, and its origin is not to be sourced exclusively from an individual’s work-life.

Rosenberg (1965) stated that self-esteem refers to an individual’s overall self-evaluation of his or her competencies, suggesting that self-esteem mainly revolves around how people feel and think about themselves. For Lyubomirsky, Tkach and Dimatteo (2005) self-esteem is the global feeling of worth or adequacy as an individual, or generalized feelings of acceptance, goodness and self-respect. Campbell (1990) conceptualizes self-esteem as a global self-reflective attitude that addresses how an individual feels about themselves, when the self is seen as the subject of evaluation. Pelham and Swann (1989) mention that self-esteem also consists of an affective part, which is about liking or disliking the self.

Research shows that self-esteem is affected more heavily by negative rather than positive events (Schroth and Shah, 2000), and employees’ self-esteem can be undermined particularly well by a supervisor’s behaviour (Brockner, 1988). This is partly due to the fact that supervisors are in charge and responsible for the distribution of resources which are relevant to employees, such as promotions or bonuses. Moreover, supervisor treatment indicates whether employees are seen to be of added value to the company (Brocker, 1988; Burton & Hoobler, 2006; Vogel & Mitchell, 2015). Knowing this explains

(12)

why abusive supervision, which signals to the victim that they are a source of annoyance, is particularly hurtful to employee self-esteem (Tepper, Moss & Duffy, 2011).

Because abusive supervision tends to diminish self-esteem by signalling to an individual that he or she is perceived to be worthless and annoying, I suspect that if an individual’s self-esteem is high enough to begin with, he or she may be able to better cope with these abusive behaviours. This is based on the fact that a person with high self-esteem sees him or herself as a worthy, competent individual, no matter what somebody else might say or think. They will not be fazed or hurt by their supervisor’s hostile actions and mean words, at least not as much as somebody with low self-esteem would be, because they have accepted and are happy with who and what they are, and nothing and nobody will change that. Accordingly, I hypothesize:

Hypothesis 2: Self-esteem will moderate the relationship between abusive supervision and employee well-being, such that the relationship will be weaker for employees who have higher self-esteem.

Unfortunately, little is known about exactly how self-esteem relates to exploitative leadership. Some logic described above with regards to abusive supervision also holds for exploitative leadership, such as that employees’ self-esteem is particularly susceptible to a leader’s treatment (Brockner, 1988). In that sense it can be argued that exploitative leadership, which is also a source of horrible employee treatment, is detrimental to their self-esteem. Moreover, narcissists are generally known to engage in exploitative behaviours (Sedikides and Campbell, 2017), and employees with low self-esteem were found to suffer most (compared to those with higher self-esteem) from being in the grasp of a narcissistic leader, as they perceive them to be more abusive (Nevicka et al. 2018). Furthermore, Barelds et al. (2018) found that leader psychopathy relates positively to self-serving (exploitative) behaviours, but only when follower self-esteem is low.

This implies that leaders actually select the victims of their exploitative behaviours based on the latter’s self-esteem levels, and that high employee self-esteem might not only buffer against the harmful effects of exploitative leadership, but allow an employee to avoid being exploited by their supervisor all together. As such, I hypothesize:

Hypothesis 3: Self-esteem will moderate the relationship between exploitative leadership and emotional exhaustion, such that the relationship will be weaker for employees who have higher self-esteem.

However, because of a lack of research on the subject, I am not able to predict whether self-esteem, in the case that it does in fact buffer against the effects of destructive leadership, might buffer differently against the effects of exploitative leadership than those of abusive supervision. As such, I ask the following question:

(13)

Research question 2: Does self-esteem buffer differently against the effects of abusive supervision on well-being than against the effects of exploitative leadership on well-being?

3. Methods

3.1 Procedures

In my research, I sought to study effects that had not been researched previously, and as such it was necessary to gather primary data. Moreover, in order to be able to measure and analyse the effects and interactions between the variables in accordance with the research questions, I needed to gather data in numerical form (quantitative). Furthermore, due to time constraints that were related to the fact that this research was conducted as part of a Bachelor’s Thesis, the choice of data-gathering-instrument ultimately fell on an online survey, as this allowed me to gather quantitative data in a relatively short amount of time.

Potential participants were contacted by text, and those that showed an interest in filling in the survey were first faced with a consent form, providing instructions for the study, assuring confidentiality and providing them with a contact for any further questions. The surveys were provided in Dutch, which was the native language of the participants.

Data was collected from two different sources: Employees and significant others. While employees are most suited to rate their self-esteem and their perception of their supervisor, I strived to get a more objective assessment of well-being. As such, significant-other-ratings were used, reducing the risk of common method bias (Conway & Lance, 2010; Podsakoff et al., 2003).

For this study, I distributed two different surveys. Participants that were assigned the role of employee filled in a survey asking them about their experiences with their supervisor. Those who fulfilled the role of significant-other of that particular employee filled in a survey about them and were, among other things, asked to estimate their job-related well-being.

The first thing that the participants had to do, after agreeing to the consent form, was filling in a self-generated anonymous identification code, which code allowed me to link the employee and significant other responses that belonged together to each other.

Participants were asked to answer the survey whenever they had time for it, at any time of the day, but not later than two weeks after agreeing to participate. The data was, as planned, gathered in approximately two weeks.

3.2 Sample

Participants were drawn in collaboration with my thesis supervisor, who has been working on research in a similar field and thus was able to help and provide me with data he has gathered over the course of his research project. After having succeeded in gathering my own data, I merged it with the

(14)

data of my supervisor, which has allowed me to gather as much respondents as possible in this short time frame.

Personally, I drew my participants from family, friends and acquaintances. I contacted them and informed them about my study. The fact that I know these people personally made it so that they were immediately willing to help, and no further incentive, besides my eternal gratitude, was necessary to make them participate. My supervisor drew his participants through a network of students as well as via LinkedIn.

Respondents were gathered in the form of employee/significant-other dyads, and participants were contacted in pairs, as such. On the one hand there is the employee, for whom the requirements were that they were employed and frequently interacted with their supervisor. On the other hand, there is the significant-other, who had to be a person in close proximity with that particular employee, such as a partner, a family member or a close friend. The participants (employees) worked in a variety of industries and were of different occupations, including education, accounting, help desks for education systems as well as technology (phones, laptops etc.), marketing in the field of nutrition, PR consultancy, electrical and mechanical engineering, retail, the catering industry and animal rescue.

Between me and my supervisor, 133 employees and their significant-others completed the surveys, resulting in 133 employee/significant-other dyads.

Of the employees, 39.1% of the respondents were female; 64% were between 17 and 30 years old, 7.5% between 31 and 40, 12% between 41 and 50 and 16.5% were older than 51 (mean 33 years, SD 12.6). Two point three percent of employees had worked for the organization that currently employs them for less than 2 months, 11.3% for between 2 and 6 months, 12% for between 6 months and 1 year, 14.3% for between 1 year and 2 years, 28.6% for between 2 and 5 years, 11.3% for between 5 and 10 years and 20.3% for longer than 20 years. Mean employee tenure within their organization was 4.7 years (SD 1.69). In the sample, on average, employees interacted with their supervisors “once or twice in the last 1 to 3 months” (2.3%), “once or twice every month” (9.8%), “once or twice every week” (35.3%), “3 to 5 times per week” (18%), “once or twice per day” (11.3%) or “several times per day” (23.3%). On average, zero-point-eight percent of the employees interact for less than a minute with their supervisor, 9% interact between 1 and 2 minutes, 9% between 3 and 5 minutes, 21.8% between 5 and 10 minutes, 20.3% between 10 and 20 minutes, 16.5% between 20 and 30 minutes and 22.6% interact for longer than 30 minutes with their supervisors.

Given the fact that frequent interaction between employee and supervisor was one of the requirements, employees that indicated they interact for less than a minute with their supervisor at a time as well as those that responded they interacted only once or twice with their supervisor in the last 1 to 3 months have been removed from the data, leaving a final sample of 129 dyads.

In the sample, only a few outliers were detected by making use of the Explore tool in SPSS, in which case an outlier is a case that falls more than 1.5 box lengths from the lower or upper hinge of the boxplot. However, in this case there is no theoretical justification for eliminating those outliers, and

(15)

there is no reason to believe that these outliers are not in fact credible responses. Thus, I decided to remove none of the outliers.

3.3 Measures

Abusive supervision. To measure abusive supervision, I used the shortened version of Tepper’s

(2000) abusive supervision measure, as introduced by Mitchell and Ambrose (2007), which measures the active dimension of abusive supervision and consists of 5 items. In the present research a 5-point scale was used to indicate abusive leadership perceptions (1 = “not at all”, 5 = “frequently, if not always”). Example items are “my boss ridicules me” and “my boss tells me my thoughts or feelings are stupid”. Cronbach’s  for this scale was .77.

Exploitative leadership. To measure exploitative leadership, I used Schmid, Pircher Verdorfer

and Peus’ (2019) 15-item scale which scale is subcategorized into five factors, with three items per factor. The factors are the following: (1) egoistic behaviours, (2) exerting pressure, (3) underchallenging followers, (4) taking credit and (5) manipulating followers. A five-point response scale was used (1 = “not at all” to 5 = “frequently, if not always”). Example items are: “sees employees as a means to reach his or her own goals” and “manipulates others to reach his or her goals”. Cronbach’s  for this scale was .91.

Self-esteem. To measure employee self-esteem, I used the single item measure mentioned in

the works of Robins, Hendin and Trzesniewski (2001): “I have high self-esteem”, which measure is rated on a 5-point scale (1 = “totally disagree” to 5 = “totally agree”).

Emotional exhaustion. To study the negative effects of destructive leadership on well-being, I

will measure employee emotional exhaustion, using the emotional exhaustion subscale as described by Demerouti, Bakker, Kantas & Vardakou (2003). It consists of eight items and was measured on a 5-point scale (1 = “totally disagree” to 5 = “totally agree”). Example items are “after my work, I regularly feel worn out and weary” and “after working, I have enough energy for my leisure activities”. Cronbach’s  for this scale was .82. However, as opposed to abusive supervision, exploitative leadership and self-esteem, which were self-rated by the employee, emotional exhaustion was rated by the significant other. Thus, the items incorporated in the survey were adapted from the original Oldenburg Burnout Inventory so that they were no longer self-rating, but rather describing someone else. Example items were thus not “after my work, I regularly feel worn out and weary” but instead “after his/her work, he/she regularly feels worn out and weary”. This from self-describing to other-describing adaption applies for all items.

(16)

4. Results

Table 1 presents the means, standard deviations and bivariate correlations of the variables.

When subordinates perceived their supervisors to be more exploitative, they also perceived them to be more abusive and they reported significantly higher emotional exhaustion. When employees saw their leaders as more abusive, they reported significantly higher emotional exhaustion and significantly lower self-esteem. Lastly, emotional exhaustion and self-esteem are significantly negatively related.

4.1 Hypothesis testing

Hypothesis 1a, the relationship between abusive supervision and emotional exhaustion, was tested using linear regression analysis, in which abusive supervision was the predictor of employee emotional exhaustion. Results showed a positive effect between abusive supervision and emotional exhaustion, which suggests that abusive supervision negatively impacts well-being (b = .76, t(125) = 2.34, p = 0.02, 95%CI[.12, 1.41]). In conclusion, Hypothesis 1a is supported; abusive supervision increases emotional exhaustion, and by doing so reduces well-being.

To test Hypothesis 1b, the relationship between exploitative leadership and emotional exhaustion, linear regression was utilized with emotional exhaustion as the dependent variable and exploitative leadership as the independent variable. Results showed a positive effect between exploitative leadership and emotional exhaustion, which implies that exploitative leadership negatively impacts well-being (b = 0.11, se = .05, t(125) = 2.35, p = 0.02, 95%CI[.02, .21]). Thus, Hypothesis 1b is supported; employee emotional exhaustion is increased when confronted with exploitative leadership.

This leads to the first research question: Which variant of destructive leadership leadership increases emotional exhaustion most? To answer this question, it is necessary to interpret the standardized B coefficient for both the impact of abusive supervision as for exploitative leadership on emotional exhaustion. For abusive supervision, this coefficient is .19, while for exploitative leadership

Table 1: Means, standard deviations and correlations

Variable M SD 1 2 3 4 1. Exploitative leadership 23.27 8.01 - - - - 2. Abusive supervision 5.54 1.43 .50** - - - 3. Emotional exhaustion 18.68 4.65 .18* .19* - - 4. Self-esteem 3.88 .72 -.01 -.22** -.32** - N=129. *p < 0.05, **p < 0.01

(17)

the coefficient’s value is .18. This difference is very small and does not seem to be meaningful. Thus, the answer to Research Question 1 would be that abusive supervision and exploitative leadership are equally detrimental to employee well-being.

For Hypothesis 2, the interaction effect between abusive supervision and self-esteem on emotional exhaustion, the PROCESS macro by Hayes (2018) Model 1 was used with abusive supervision as the independent variable, self-esteem as the moderating variable and employee emotional exhaustion as the dependent variable. Results showed a significant interaction effect, which are summarized in Table 2 (b = .44, se = .20, t(125) = 2.16, p = .03, 95%CI[.04, .84]). Interestingly enough, when examining the influence of abusive supervision on emotional exhaustion for different levels of esteem, in which “low esteem” is one standard deviation below the mean and “high esteem” is one standard deviation above the mean, it became apparent that for people with low self-esteem there is no interaction between abusive supervision and self-self-esteem on emotional exhaustion (b = .45, se = .28, t(125) = 1.61, p = .11, 95%CI[-.10, .99]). For people with high self-esteem, on the other hand, there is a relationship between abusive supervision and emotional exhaustion in the way that emotional exhaustion is higher when self-esteem is high, which relationship can be seen in Figure 2 (b = 1.08, se = .42, t(125) = 2.56, p = .01, 95%CI[.24, 1.92]). Thus, Hypothesis 2 is not supported; self-esteem does not act as a buffer against the detrimental effects of abusive supervision on well-being, but instead seems to amplifiy those effects.

Figure 2: Plotted interaction of abusive supervision x self-esteem on emotional exhaustion

Table 2: Results for the interaction effect between abusive supervision and self-esteem on emotional exhaustion

b se t p

Constant 18.78 .39 48.70 .00

Abusive supervision .76 .33 2.34 .02

Self-esteem -2.20 .57 -3.87 .00

Abusive supervision x Self-esteem .44 .20 2.16 .03

Dependent variable is emotional exhaustion

R square = .15 N = 129

(18)

To test Hypothesis 3, the interaction effect between exploitative leadership and self-esteem on emotional exhaustion, the PROCESS macro by Hayes (2018) Model 1 was utilized with exploitative leadership as the independent variable, self-esteem as the moderating variable and employee emotional exhaustion as the dependent variable. Results did not show a significant interaction effect (see Table 3) as the p-value was > .05, although the effect was almost significant, as the exact p-value was .06. For now, however, Hypothesis 3 is not supported. Interestingly enough, when I controlled for abusive supervision by including it as a covariate, a significant interaction was found, which can be seen in Table

4 (b = .14, se = .07, t(125) = 2.09, p = .04, 95%CI[.01, .27]). The inclusion of abusive supervision as a

predictor altered the estimation and by doing so, caused previously insignificant results to become significant. This might be caused by overlap between exploitative leadership and abusive supervision. When studying the influence of exploitative leadership and self-esteem on emotional exhaustion for different levels of self-esteem, once again with “low self-esteem” being one standard deviation below the mean and “high self-esteem” being one standard deviation above the mean, it came forward that for people with low self-esteem there is no relationship between exploitative leadership and emotional exhaustion (b = -.01, se = .08, t(125) = -.18, p = .85, 95%CI[-.17, .14]). For people with high self-esteem, the data suggests that there is in fact a relationship between exploitative leadership and emotional exhaustion such that emotional exhaustion is higher when self-esteem is high, as can be seen in Figure

3 (b = .19, se = .07, t(125) = 2.64, p = .01, 95%CI[.05, .32]). So, Hypothesis 3 is not supported.

Based on the results, the answer to my second research question: “Does self-esteem buffer differently against the harmful effects on well-being of abusive supervision than of exploitative leadership?” would be that self-esteem buffers against neither, but instead seems to increase both their effects. Thus, self-esteem does in fact appear to have a moderating effect regarding both destructive leadership styles, just not in the direction I had predicted. Overall, the data suggests that the self-esteem’s moderating influence is stronger and more meaningful pertaining to abusive supervision than to exploitative leadership.

(19)

Table 3: Results for the interaction effect between exploitative leadership and self-esteem on emotional exhaustion (when not controlling for abusive supervision)

b se t p

Constant 18.69 .38 49.15 .00

Exploitative leadership .11 .05 2.35 .02

Self-esteem -2.05 .53 -3.85 .00

Exploitative leadership x Self-esteem .12 .06 1.90 .06 Dependent variable is emotional exhaustion

R square = .16 N = 129

Table 4: Results for the interaction effect between exploitative leadership and self-esteem on emotional exhaustion (when controlling for abusive supervision)

b se t p

Constant 16.96 1.89 8.96 .00

Exploitative leadership .09 .06 1.53 .13

Self-esteem -1.91 .55 -3.45 .00

Exploitative leadership x Self-esteem .14 .07 2.09 .04

Abusive supervision .31 .33 .93 .36

Dependent variable is emotional exhaustion

R square = .16 N = 129

5. Discussion

The goal of this study was to research whether abusive supervision and exploitative leadership are (differently) detrimental to employee well-being, and if self-esteem might act like a buffer against their harmful effects. Additionally, the question was whether self-esteem would buffer differently against abusive supervision than it would against exploitative leadership.

The data suggests that abusive supervision as well as exploitative leadership negatively impact employee well-being, as they increase emotional exhaustion, which is in line with Hypothesis 1a and 1b. These results are supported by and build on existing research, such as the findings of Tepper (2000, 2007), Schat and Kelloway (2005), Bakker and Demerouti (2007), Hobman et al. (2009) and Schmid, Pircher Verdorfer and Peus (2019).

Before performing the analysis, based on the (lack of) existing literature, it was impossible to predict whether it would be abusive supervision or exploitative leadership that would negatively affect well-being the most. Remarkably, the results indicated that abusive supervision is worse than exploitative leadership in this regard, but only by a negligible amount. This implies that the answer to Research Question 1 is that on average, employees find these two treatments to be approximately equally horrible to endure, even though these two variants of destructive leadership express themselves in an entirely different way, abusive supervision being direct and hostile (Tepper, 2000; 2007; Mitchell and

(20)

Ambrose, 2007), and exploitative leadership being more indirect and subtle (Schmid, Pircher Verdorfer and Peus, 2019).

In contradiction to Hypothesis 2, the analysis suggests that instead of acting as a buffer, high self-esteem actually increases the harmful effects of abusive supervision. This was surprising and not in line with the expectations set based on existing literature. However, the results caused my perspective to change, and from this new point of view one might actually argue that individuals who have high self-esteem and consequently perceive themselves to be worthy, competent individuals, are, instead of being “immune”, actually hurt very deeply by a supervisor treating them badly and telling them they are worthless and incompetent, because the individual did never see that coming and did never expect to be treated that way. These abusive, condescending behaviours cause the individual to fall off the pedestal their initial high esteem had placed them on, which is a long, painful way down. People low on self-esteem on the other hand may not mind being treated badly that much, perhaps because they might even feel they deserve to be treated in such a fashion, as they perceive themselves to be incompetent, anyway. Their supervisor confirming that feeling does not change much in that regard, resulting in relatively lower emotional exhaustion. These results provide new insights into the relationship between abusive supervision, employee well-being and self-esteem, by suggesting that in the case an employee is high in self-esteem, the detrimental effects of abusive supervision on well-being are actually enhanced.

At first, it seemed no support for Hypothesis 3 would be found due to the fact that initially, the interaction effect between exploitative leadership and self-esteem on emotional exhaustion was not significant, but only barely so; the effect was on the brink of significance. Then, when adding abusive supervision as another predictor to the mix, the interaction became significant and moderation was confirmed, but not in the way which was initially predicted. Contrary to the expectations adopted in Hypothesis 3, the results imply that high self-esteem actually enhances the positive effect of exploitative leadership on emotional exhaustion, which would result in lower employee well-being. Once again, this contradicted my literature-based predictions. However, similarly to abusive supervision, an explanation for the found results could be that individuals high in self-esteem, who find themselves to be competent and valuable to the company, are hugely affected by their bosses’ exploitation endeavours, including taking credit for the employee’s work and not allowing them to develop themselves professionally. Conversely, employees with low self-esteem might not really mind their boss taking credit for their work or undermining their promotion opportunities, as they never saw themselves deserving of credit or promotions in the first place.

The data indicates that the answer to Research question 2 is that self-esteem does not buffer against the harmful effects of either abusive supervision or exploitative leadership, but instead seems to enhance the harmful effect, which effect is stronger for abusive supervision than for exploitative leadership. This is fascinating, and further research is required to determine the exact roots of these results.

(21)

5.1 Limitations and future research

This study does have its limitations. First of all, the data gathered by myself was collected in the short time span of approximately two weeks, and my respondents were family, friends or acquaintances of mine, possibly resulting in selection bias as proper randomization was not achieved. However, this bias does not apply to the participants gathered by my supervisor, whose data accounts for the majority of dyads. I would recommend future researchers to make sure proper randomization is accounted for.

Moreover, the questions in the survey have been translated back and forth between English, Italian and Dutch, meaning some nuances may have gotten lost in translation. This might have resulted in respondents answering questions differently than they would have if they were provided with the original or properly translated survey.

Additionally, for the most part, the answer-options were scale-based. However, one of my respondents indicated that she would have liked to see an “other” answer-option in the survey, in which one could write a response in case he or she would have liked to add a nuance to their answer, or even give a whole other response than the scale on its own allowed. For my study, this was impossible to add, because originally, I and two of my fellow students also working on their Bachelor thesis were to gather data separately via the same survey, and then later merge our data. As a result, altering the survey called for a tremendous amount of communication, as well as approval, and so I decided to leave it as it is. However, for future research I would definitely recommend adding an “other” option, allowing for increased answer reliability.

A potential area for future research would be whether individual differences makes employees more susceptible to abusive supervision rather than to exploitative leadership, or vice versa. Whereas this study found that on average the two destructive leadership styles are equally harmful, it would be interesting to see whether individuals are equally harmed by the two styles, or whether perhaps one individual is harmed only be exploitative leadership, and another by abusive supervision and that only on average the two are equally detrimental.

Another possible subject for future research that requires more insight would be what the underlying causes of self-esteem enhancing the harmful effects of abusive supervision and exploitative leadership are. I have already made some suggestions, but it would be interesting to see what actual empirical research into this subject would reveal.

5.2 Practical implications

My study makes several contributions to practice. To begin with, while it might be the case that the destructive leadership behaviours covered in this research negatively impact employee well-being, some supervisors showing these behaviours might excel at other points, and overall still be well-performing leaders. In practice this might imply for top management that they have to avoid hiring individuals with high self-esteem to work for aforementioned abusive / exploitative supervisors, as these

(22)

people are most affected. Such a norm might avoid the otherwise potential removal of an on other aspects well-performing leader due to complaints or an above average burn-out rate of subordinates under this leader’s command.

In addition to this, the findings of this study can be of importance to individuals deciding who or what company to work for. If an individual knows themselves to be high in self-esteem, it might be wise to avoid organizations that are well known for treating their employees badly, because those individuals suffer most working under destructive leaders. In the Netherlands, “De Zuidas” might be a fitting example, as most companies in this area are notorious for their harsh and merciless treatment of employees.

6. Conclusion

This research set out to answer the question whether self-esteem might buffer against the harmful effects of abusive supervision and exploitative leadership on well-being, and if so, whether the strength of this buffering effect might be different for any of these two destructive leadership styles. Surprisingly, the results indicated that high self-esteem actually increases the detrimental effects of both styles, and that this effect is stronger for abusive supervision than for exploitative leadership. However, future research is required to determine the underlying causes of this finding.

In conclusion, my findings add to existing knowledge by showing a relationship between different destructive leadership styles, self-esteem and their effects on well-being, as well as by providing practical implications and highlighting potential areas for future research.

(23)

7. Sources

Ashforth, B. (1994). Petty tyranny in organizations. Human Relations, 47, 755−778.

Bakker, A. B., & Demerouti, E. (2007). The job demands‐resources model: State of the art.

Journal of Managerial Psychology.

Barnes, C.M., Lucianetti, L., Bhave, D.P., & Christian, M.S. (2015). “You wouldn’t like me when I’m sleepy”: Leader’s sleep, daily abusive supervision, and work unit engagement. Academy of

Management Journal, 58(5), 1419-1437.

Barelds, D.P.H., Wisse, B., Sanders, S., & Laurijssen, L.M. (2018). No regard for those who need it: the moderating role of follower self-esteem in the relationship between leader psychopathy and leader self-serving behaviour. Frontiers in Psychology.

https://doi.org/10.3389/fpsyg.2018.01281

Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than good.

Review of General Psychology, 5(4), 323.

Brockner, J. (1988). Self-esteem at work. Lexington, MA: Lexington Books.

Burton, J. P., & Hoobler, J. M. (2006). Subordinate self-esteem and abusive supervision. Journal of

Managerial Issues, 18, 340-355

Cambell, J.D. (1990). Self-esteem and clarity of the self-concept. Journal of Personality and Social

Psychology, 59(3) 538-549.

Conway, J., & Lance, C. (2010). What Reviewers Should Expect from Authors Regarding Common Method Bias in Organizational Research. Journal of Business and Psychology, 25(3), 325-334. doi: 10.1007/s10869-010-9181-6.

Den Hartog, D.N., & Koopman, P.L. (2001). Leadership in organizations. In N. Anderson, D.S. Ones, H.K. Sinangil & C. Viswesvaran (Eds.), Handbook of Industrial, Work & Organizational Psychology (pp. 166-187). Thousand Oaks, CA: Sage.

Demerouti, E., Bakker, A.B., Kantas, A., & Vardakou, I. (2003). The convergent validity of two burnout instruments: a multitrait-multimethod analysis. European Journal of Psychological

Assessment, 19(1), 12-23.

Dobre, O. (2013). Employee motivation and organizational performance. Review of Applied

Socio-Economic Research, 5(1), 53-60.

Einarsen, S., Aasland, M. S., & Skogstad, A. (2007). Destructive leadership behaviour: A definition and conceptual model. The Leadership Quarterly, 18, 207–216

Grandey, A.A., Kern, J.H., & Frone, M.R. (2007). Verbal abuse from outsiders versus insiders: Comparing frequency, impact on emotional exhaustion, and the role of emotional labor.

Journal of Occupational Health Psychology, 12, 63–79.

Harvey, P., Stoner, J., Hochwarter, W., & Kacmar, C. (2007). Coping with abusive supervision: The neutralizing effects of integration and positive affect on negative employee outcomes. The

Leadership Quarterly, 18(3), 264-280.

Herschcovis, M., & Barling, J. (2010). Towards a multi-foci approach to workplace aggression: A meta-analytic review of outcomes from different perpetrators. Journal of Organizational

Behavior, 31(1), 24-44.

Hobman, E.V., Restubog, S.L.D., Bordia, P., & Tang, R.L. (2009). Abusive supervision in advising relationships: Investigating the role of social support. Applied Psychology: An International

Review, 58, 233–256.

Kessler, S.R., Spector, P.E., Chang, C.H., & Parr, A.D. (2008). Organizational violence and aggression: Development of the three-factor violence climate survey. Work & Stress, 22, 108– 124.

(24)

Korman, A. K. (1970). Toward an hypothesis of work behavior. Journal of Applied Psychology, 54, 31–41.

Korman, A. K. (1971). Organizational achievement, aggression and creativity: Some suggestions toward an integrated theory. Organizational Behavior and Human Performance, 6, 593–613. Korman, A. K. (1976). Hypothesis of work behavior revisited and an extension. Academy of

Management Review, 1, 50–63

Lemoine, G. J., Hartnell, C. A., & Leroy, H. (2019). Taking Stock of Moral Approaches to Leadership: An Integrative Review of Ethical, Authentic, and Servant Leadership. Academy of

management Annals, 13(1), 148–187. doi: 10.5465/annals.2016.0121

Luthans, K.W. (2000). Recognition: A powerful, but often overlooked, leadership tool to improve employee performance. The Journal of Leadership Studies, 7(1), 31-39.

Lyubomirsky, S., Tkach, C., & Dimatteo, M.R. (2005). What are the differences between happiness and self-esteem? Social Indicators Research, 78, 363-404.

Mitchell, M.S., & Ambrose, M.L. (2007). Abusive supervision and workplace deviance and the moderating effects of negative reciprocity beliefs. Journal of Applied Psychology, 92(4), 1159-1168.

Nevicka, B., De Hoogh, A.H.B., Den Hartog, D.N., & Belschak, F.D. (2018). Narcissistic leaders and their victims: followers low on self-esteem and low on core self-evaluations suffer most,

Frontiers in Psychology.

https://doi.org/10.3389/fpsyg.2018.00422

Pelham, B. W., & Swann, W. B., Jr. (1989). From self-conceptions to self-worth: On the sources and structure of global self-esteem. Journal of Personality and Social Psychology, 57, 672–680. Pierce, J.L. & Gardner, D.G. (2004). Self-Esteem within the work and organizational context: A

review of the organization-based self-esteem literature. Journal of Management, 30(5) 591-622.

Podsakoff, P. M., MacKenzie, S., Jeong-Yeon, L., & Podsakoff, N. P. (2003). Common Method Biases in Behavioral Research: A Critical Review of the Literature and Recommended Remedies. Journal of Applied Psychology, 88(5), 879-903.

Robins, R.W., Hendin, H.M., & Trzesniewski, K.H. (2001). Measuring global self-esteem: construct validation of a single-item measure and the Rosenberg self-esteem scale. Personality and

Social Psychology Bulletin, 27(2), 151-161.

Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

Schat, A.C.H., Kelloway, E.K. (2005). Workplace Aggression. In J. Barling, E.K. Kelloway, & M.R. Frone (Eds.), Handbook of Work Stress (pp. 189-218). Thousand Oaks, CA: Sage

http://dx.doi.org/10.4135/9781412975995.

Schat, A.C.H., Frone, M.R., & Kelloway, E.K. (2006). Prevalence of workplace aggression in the U.S. workforce: Findings from a national study. In E. K. Kelloway, J. Barling, & J. J. Hurrell (Eds.), Handbook of workplace violence (pp. 47-89). Thousand Oaks, CA: Sage.

Schmid, E.A., Pircher Verdorfer, A., & Peus, C. (2019). Shedding light on leaders’ self- interest: theory and measurement of exploitative leadership. Journal of Management, 45(4), 1401-1433.

Schroth, H. A. & P. P. Shah. (2000). Procedures: Do We Really Want to Know Them? An Examination of the Effects of Procedural Justice on Self-esteem. Journal of Applied

Psychology, 85, 462-471.

Schyns, B., & Schilling, J. (2013). How bad are the effects of bad leaders? A meta-analysis of destructive leadership and its outcomes. The leadership quarterly, 24(1), 138-158.

(25)

Sedikides, C., & Campbell, W. K. (2017). Narcissistic force meets systemic resistance: the energy clash model. Perspectives on Psychological Science, 12, 400–421

Thoroughgood, C. N., Tate, B. W., Sawyer, K. B., & Jacobs, R. (2012). Bad to the bone: Empirically defining and measuring destructive leader behavior. Journal of Leadership and Organizational Studies, 19, 230–255.

Tepper, B.J. (2000). Consequences of abusive supervision. Academy of Management Journal, 43(2), 178-190.

Tepper, B.J. (2007). Abusive supervision in work situations: Review, synthesis and research agenda.

Journal of Management, 33(3), 261-289.

Tepper, B. J., Moss, S. E., & Duffy, M. K. (2011). Predictors of abusive supervision: Supervisor perceptions of deep-level dissimilarity, relationship conflict, and subordinate performance.

Academy of Management Journal, 54: 279-294

Vogel, R.M., & Mitchell, M.S. (2015). The motivational effects of diminished self-esteem for employees who experience abusive supervision. Journal of Management, 43(7), 2218-2251. Walker, L. (1979). The battered woman. New York: Harper & Ro.

Wolfe, D. A. (1987). Child abuse: Implications for child development and psychopathology. Beverly Hills, CA: Sage

Yagil, D. (2006). The relationship of abusive and supportive workplace supervision to employee burnout and upward influence tactics. Journal of Emotional Abuse, 6, 49–65.

Yukl, G., & van Fleet, D. D. (1992). Theory and research on leadership in organizations. In M. D. Dunette, & L. M. Hough (Eds.), Handbook of industrial and organizational psychology (pp. 147–199). Palo Alto: Consulting Psychologists Press.

Zellars, K.L., Tepper, B.J., & Duffy, M.K. (2002). Abusive supervision and subordinates’ organizational citizenship behavior. Journal of Applied Psychology, 87, 1068–1076.

(26)

8. Appendix

Run MATRIX procedure:

**************** PROCESS Procedure for SPSS Version 3.2 ****************** Written by Andrew F. Hayes, Ph.D. www.afhayes.com

Documentation available in Hayes (2018). www.guilford.com/p/hayes3 ************************************************************************** Model : 1 Y : Well_bei X : Abusive_ W : Self_est Sample Size: 129 ************************************************************************** OUTCOME VARIABLE: Well_bei Model Summary R R-sq MSE F df1 df2 p ,3831 ,1468 18,9076 7,1667 3,0000 125,0000 ,0002 Model

coeff se t p LLCI ULCI

constant 18,7820 ,3856 48,7045 ,0000 18,0188 19,5452 Abusive_ ,7643 ,3261 2,3439 ,0207 ,1190 1,4096 Self_est -2,2000 ,5679 -3,8740 ,0002 -3,3240 -1,0761 Int_1 ,4394 ,2036 2,1579 ,0329 ,0364 ,8423 Product terms key:

Int_1 : Abusive_ x Self_est

Covariance matrix of regression parameter estimates: constant Abusive_ Self_est Int_1

constant ,1487 ,0080 -,0067 ,0094 Abusive_ ,0080 ,1063 ,0090 ,0351 Self_est -,0067 ,0090 ,3225 -,0295 Int_1 ,0094 ,0351 -,0295 ,0415

(27)

Test(s) of highest order unconditional interaction(s): R2-chng F df1 df2 p

X*W ,0318 4,6563 1,0000 125,0000 ,0329 ---

Focal predict: Abusive_ (X) Mod var: Self_est (W)

Conditional effects of the focal predictor at values of the moderator(s): Self_est Effect se t p LLCI ULCI

-,7182 ,4488 ,2779 1,6147 ,1089 -,1013 ,9988 ,0000 ,7643 ,3261 2,3439 ,0207 ,1190 1,4096 ,7182 1,0798 ,4221 2,5582 ,0117 ,2444 1,9152 Moderator value(s) defining Johnson-Neyman significance region(s): Value % below % above

-,4379 20,9302 79,0698

Conditional effect of focal predictor at values of the moderator: Self_est Effect se t p LLCI ULCI

-2,8760 -,4993 ,4971 -1,0043 ,3172 -1,4832 ,4846 -2,6760 -,4114 ,4638 -,8869 ,3768 -1,3294 ,5066 -2,4760 -,3235 ,4318 -,7492 ,4552 -1,1782 ,5311 -2,2760 -,2357 ,4014 -,5871 ,5582 -1,0301 ,5588 -2,0760 -,1478 ,3730 -,3962 ,6926 -,8860 ,5904 -1,8760 -,0599 ,3470 -,1727 ,8632 -,7467 ,6268 -1,6760 ,0280 ,3240 ,0863 ,9314 -,6134 ,6693 -1,4760 ,1158 ,3048 ,3800 ,7046 -,4875 ,7191 -1,2760 ,2037 ,2901 ,7021 ,4839 -,3705 ,7779 -1,0760 ,2916 ,2806 1,0392 ,3007 -,2637 ,8468 -,8760 ,3794 ,2767 1,3711 ,1728 -,1683 ,9271 -,6760 ,4673 ,2789 1,6757 ,0963 -,0846 1,0192 -,4760 ,5552 ,2868 1,9356 ,0552 -,0125 1,1229 -,4379 ,5719 ,2890 1,9791 ,0500 ,0000 1,1438 -,2760 ,6431 ,3002 2,1424 ,0341 ,0490 1,2371 -,0760 ,7309 ,3182 2,2973 ,0233 ,1012 1,3606 ,1240 ,8188 ,3401 2,4074 ,0175 ,1457 1,4919 ,3240 ,9067 ,3653 2,4820 ,0144 ,1837 1,6296 ,5240 ,9945 ,3931 2,5300 ,0127 ,2165 1,7725 ,7240 1,0824 ,4230 2,5589 ,0117 ,2452 1,9196 ,9240 1,1703 ,4546 2,5744 ,0112 ,2706 2,0700 1,1240 1,2582 ,4875 2,5807 ,0110 ,2933 2,2230

*********************** ANALYSIS NOTES AND ERRORS ************************ Level of confidence for all confidence intervals in output:

95,0000

W values in conditional tables are the mean and +/- SD from the mean. NOTE: The following variables were mean centered prior to analysis: Self_est Abusive_

(28)

Shorter variable names are recommended. --- END MATRIX ---

(29)

Run MATRIX procedure:

**************** PROCESS Procedure for SPSS Version 3.2 ****************** Written by Andrew F. Hayes, Ph.D. www.afhayes.com

Documentation available in Hayes (2018). www.guilford.com/p/hayes3 ************************************************************************** Model : 1 Y : Well_bei X : Exploita W : Self_est Sample Size: 129 ************************************************************************** OUTCOME VARIABLE: Well_bei Model Summary R R-sq MSE F df1 df2 p ,3982 ,1586 18,6461 7,8516 3,0000 125,0000 ,0001 Model

coeff se t p LLCI ULCI

constant 18,6856 ,3802 49,1477 ,0000 17,9332 19,4381 Exploita ,1124 ,0478 2,3499 ,0203 ,0177 ,2070 Self_est -2,0454 ,5315 -3,8483 ,0002 -3,0973 -,9935 Int_1 ,1210 ,0636 1,9037 ,0592 -,0048 ,2468 Product terms key:

Int_1 : Exploita x Self_est

Covariance matrix of regression parameter estimates: constant Exploita Self_est Int_1

constant ,1445 ,0000 ,0000 ,0001 Exploita ,0000 ,0023 ,0001 ,0002 Self_est ,0000 ,0001 ,2825 ,0003 Int_1 ,0001 ,0002 ,0003 ,0040

Test(s) of highest order unconditional interaction(s): R2-chng F df1 df2 p

X*W ,0244 3,6242 1,0000 125,0000 ,0592 ---

Focal predict: Exploita (X) Mod var: Self_est (W)

Conditional effects of the focal predictor at values of the moderator(s): Self_est Effect se t p LLCI ULCI

(30)

,0000 ,1124 ,0478 2,3499 ,0203 ,0177 ,2070 ,7182 ,1993 ,0686 2,9050 ,0043 ,0635 ,3350 Moderator value(s) defining Johnson-Neyman significance region(s): Value % below % above

-,1437 20,9302 79,0698

Conditional effect of focal predictor at values of the moderator: Self_est Effect se t p LLCI ULCI

-2,8760 -,2357 ,1854 -1,2714 ,2060 -,6026 ,1312 -2,6760 -,2115 ,1731 -1,2216 ,2242 -,5542 ,1312 -2,4760 -,1873 ,1610 -1,1637 ,2468 -,5058 ,1313 -2,2760 -,1631 ,1489 -1,0956 ,2754 -,4577 ,1315 -2,0760 -,1389 ,1369 -1,0147 ,3122 -,4098 ,1320 -1,8760 -,1147 ,1250 -,9172 ,3608 -,3621 ,1328 -1,6760 -,0905 ,1134 -,7980 ,4264 -,3149 ,1339 -1,4760 -,0663 ,1020 -,6498 ,5170 -,2681 ,1356 -1,2760 -,0421 ,0909 -,4626 ,6445 -,2221 ,1379 -1,0760 -,0179 ,0804 -,2222 ,8245 -,1770 ,1412 -,8760 ,0063 ,0706 ,0899 ,9285 -,1333 ,1460 -,6760 ,0305 ,0618 ,4944 ,6219 -,0917 ,1528 -,4760 ,0547 ,0546 1,0030 ,3178 -,0533 ,1628 -,2760 ,0790 ,0496 1,5902 ,1143 -,0193 ,1772 -,1437 ,0950 ,0480 1,9791 ,0500 ,0000 ,1899 -,0760 ,1032 ,0477 2,1633 ,0324 ,0088 ,1975 ,1240 ,1274 ,0491 2,5966 ,0105 ,0303 ,2244 ,3240 ,1516 ,0535 2,8335 ,0054 ,0457 ,2574 ,5240 ,1758 ,0603 2,9133 ,0042 ,0564 ,2952 ,7240 ,2000 ,0689 2,9040 ,0044 ,0637 ,3363 ,9240 ,2242 ,0785 2,8546 ,0050 ,0688 ,3796 1,1240 ,2484 ,0890 2,7917 ,0061 ,0723 ,4245

*********************** ANALYSIS NOTES AND ERRORS ************************ Level of confidence for all confidence intervals in output:

95,0000

W values in conditional tables are the mean and +/- SD from the mean. NOTE: The following variables were mean centered prior to analysis: Self_est Exploita

NOTE: Variables names longer than eight characters can produce incorrect output. Shorter variable names are recommended.

--- END MATRIX ---

Employees with high self-esteem suffer most from the detrimental effects of abusive supervision and exploitative leadership