LEADER PERSONALITY, ABUSIVE SUPERVISION AND EMPLOYEE OUTCOMES: AN INTEGRATIVE MODEL By Dongyuan Wu A DISSERTATION submitted to Michigan State University In partial fulfillment of the requirements for the degree of Human Resources and Labor Relations—Doctor of Philosophy 2020 LEADER PERSONALITY, ABUSIVE SUPERVISION AND EMPLOYEE OUTCOMES: AN ABSTRACT INTEGRATIVE MODEL By Dongyuan Wu In this study, I propose an integrative model that examines both antecedents and outcomes of abusive supervision. I integrate personality theory into the abusive supervision literature, examining how both widely and narrowly defined leader personality traits are associated with abusive supervision. In addition, I investigate how employees’ power distance orientations and job type (white-collar and blue-collar) impact the abusive supervision-outcome linkage and potential underlying mechanisms. I tested this model using a sample of 1,009 workers from 136 work teams in four organizations in China. The study used a three-wave time- lagged design and obtained responses from individual employees and their team leaders. The results overall provide support for the proposed model and several of the hypotheses. The results indicate that leader agreeableness is an important predictor of abusive supervision and that ethical climate moderates the relationship between leader narcissism and abusive supervision. Regarding the relationship between abusive supervision and employee outcomes, this study supports that interactional justice is an important mediator. In addition, the findings show that the indirect effects via interactional justice are different for white-collar and blue-collar employees, highlighting the importance of considering job type as a key boundary condition in future studies. In contrast to prior research, this study does not provide strong evidence for the moderating role of power distance orientation on the relationship between abusive supervision and employee outcomes. Theoretical contributions and future directions are discussed. Copyright by DONGYUAN WU 2020 ACKNOWLEDGEMENTS I would like to express my gratitude to everyone who made my dream of becoming an organizational researcher come true. This would not have been possible without the support from all of you. I would like to give special thanks to my advisor, Dr. James Dulebohn. During my five years in the doctoral program, Professor Dulebohn spent a lot of time advising me. I benefited greatly from our formal meetings and informal conversations. Importantly, his enthusiastic attitude toward research highly impacts me and makes me passionate about doing research. I deeply believe that doing organizational research is a lot of fun and meaningful to the world. I also would like to thank my other dissertation committee members, Dr. Mark Roehling, Dr. John Schaubroeck, Dr. Jason Huang, and Dr. Chenwei Liao. I appreciate that they always keep their doors open to me whenever I have research questions for my dissertation and other research projects. Their insights and guidance greatly improved my critical thinking and passion toward research. I would like to thank other faculty members, especially those in our Ph.D. committee, in our department. They sincerely care for our well-being and make us feel supported and valued. At Michigan State University, I benefited from taking Ph.D. level classes from other departments including management, psychology, and education. The opportunity to take classes outside of our department greatly broadened my views and enhanced my scientific thinking. I also thank my friends from my department, Department of Geography, and Department of Psychology. The gatherings with them added additional happiness to my life at MSU. iv Finally, I would like to thank my husband and my best friend, Jiang Chang, for his unconditional support and love. It is unbelievable that we become alumni again after high school. Along the way of growing up as researchers, we support each other and provide each other with the strongest encouragement. I thank my son, An Chang, who adds an additional dimension and role to my life and always gives me amazing love. I thank my family members for their encouragement and instrumental support along the way. v TABLE OF CONTENTS LIST OF TABLES .................................................................................................................................. viii LIST OF FIGURES .................................................................................................................................. ix CHAPTER 1 INTRODUCTION ............................................................................................................. 1 1.1 Overview / Statement of the Problem ................................................................................ 1 1.2 Purpose and Contribution of the Research ........................................................................ 4 1.3 Organization of the Research ............................................................................................ 6 CHAPTER 2 LITERATURE REVIEW ................................................................................................. 7 2.1 Abusive Supervision .......................................................................................................... 7 2.1.1 Conceptualization of Abusive Supervision ............................................................. 7 2.1.2 Nomological Network of Abusive Supervision ...................................................... 9 2.2 Leader Personality and Abusive Supervision .................................................................. 10 2.2.1 Two Approaches to Examining Personality .......................................................... 12 2.2.2 Trait Activation Theory and Situational Factors ................................................... 16 2.3 Abusive Supervision and Cultural Values ....................................................................... 18 2.3.1 Implicit Leadership Theory and Cultural Values .................................................. 21 2.3.2 Level of Analysis for Cultural Values................................................................... 22 2.3.3 Justice as A Mediator ............................................................................................ 25 2.4 Abusive Supervision and Job Type ................................................................................. 26 CHAPTER 3 MODEL AND HYPOTHESIS DEVELOPMENT ..................................................... 28 3.1 Antecedents to Abusive Supervision ............................................................................... 30 3.1.1 Leader Personality and Abusive Supervision ....................................................... 30 3.1.2 Trait-Relevant Situational Factor as A Moderator ................................................ 35 3.2. Outcomes to Abusive Supervision ................................................................................. 37 3.2.1 Power Distance Orientation as A Moderator ........................................................ 37 3.2.2 Job Type as A Moderator ...................................................................................... 42 3.2.3 Interactional Justice as a Mediator ........................................................................ 46 3.2.4 A Moderated Mediation Model ............................................................................. 47 CHAPTER 4 METHOD .......................................................................................................................... 49 4.1 Participants and Procedures ............................................................................................. 49 4.2 Measures .......................................................................................................................... 50 4.2.1 Time 1 Survey ....................................................................................................... 50 4.2.2 Time 2 Survey ....................................................................................................... 52 4.2.2 Time 3 Survey ....................................................................................................... 52 4.3. Analytical Approaches ................................................................................................... 53 4.3.1 Data Screening and Preparation ............................................................................ 54 4.3.2 Descriptive Data, Correlation Analysis, and Reliability Analysis ........................ 54 4.3.3 Regression and Moderation Analysis .................................................................... 54 vi 4.3.4 Measurement Model, Hierarchical Linear Modeling Analysis, and Structural Model ............................................................................................................................. 55 CHAPTER 5 RESULTS ......................................................................................................................... 58 5.1 Data Aggregation, Descriptive Data, and Correlations ................................................... 58 5.2 Antecedents to Abusive Supervision ............................................................................... 58 5.2.1 Regression and Moderation Analysis .................................................................... 58 5.3 Outcomes to Abusive Supervision .................................................................................. 64 5.3.1 Moderation Analysis for Power Distance Orientation and Job Type.................... 64 5.3.2 Multilevel Confirmatory Factor Analysis ............................................................. 68 5.3.3 Multilevel Moderated Mediation Model ............................................................... 69 CHAPTER 6 DISCUSSION ................................................................................................................... 74 6.1 The Role of Leader Personality in Abusive Supervision ................................................ 74 6.2 Ethical Climate and Trait Activation Theory .................................................................. 77 6.3 The Relative Importance of Power Distance Orientation and Job Type ......................... 79 6.4 Generalizability Issues of Current Research with a Focus on White-Collar Employees 80 6.5 Strengths, Limitations, and Future Directions ................................................................. 81 6.6 Conclusions ..................................................................................................................... 85 APPENDIX ............................................................................................................................................... 86 REFERENCES ......................................................................................................................................... 91 vii LIST OF TABLES Table 1. Means, SDs, reliabilities, and correlations ..................................................................... 60 Table 2. Regression analyses predicting abusive supervision ...................................................... 62 Table 3. HLM analyses for power distance orientation and job nature as moderators ................. 65 Table 4. Indirect effects from abusive supervision to outcomes through interactional justice ..... 71 Table 5. Moderated indirect effects from abusive supervision to outcomes through interactional justice by power distance orientation ............................................................................................ 72 Table 6. Moderated indirect effects from abusive supervision to outcomes through interactional justice by job type ......................................................................................................................... 73 viii LIST OF FIGURES Figure 1. Conceptual model ..........................................................................................................29 Figure 2. The relationship between leader narcissism and abusive supervision with ethical climate as a moderator ..................................................................................................................63 Figure 3. The relationship between abusive supervision and job performance with power distance orientation as a moderator ..............................................................................................................66 Figure 4. The relationship between abusive supervision and interactional justice with job type as a moderator ....................................................................................................................................67 Figure 5. The relationship between abusive supervision and deviance with job type as a moderator .......................................................................................................................................68 Figure 6. A moderated mediation model .......................................................................................70 ix CHAPTER 1 INTRODUCTION 1.1 Overview / Statement of the Problem A primary focus in the leadership literature is on the positive influence that leaders may have on their followers (Hoch, Bommer, Dulebohn, & Wu, 2018; Yukl, 1989). Examples of these positive leadership forms include transformational leadership, transactional leadership (Bass, 1985), consideration and initiating structure (Fleishman, 1995), and, more recently, ethical leadership (Brown, Treviño, & Harrison, 2005), and servant leadership (Greenleaf, 2002). However, within the past three decades, other groups of researchers have turned their attention to destructive supervisory behaviors. To label these negative leader behaviors, researchers have developed a few constructs, such as workplace bullying (Hoel, Rayner, & Cooper, 1999), supervisor aggression (Neuman & Baron, 1998), petty tyranny (Ashforth, 1994), and abusive supervision (Tepper, 2000). Among these constructs, researchers have given more attention to abusive supervision. According to Tepper (2000), abusive supervision refers to “subordinates' perceptions of the extent to which supervisors engage in the sustained display of hostile verbal and nonverbal behaviors, excluding physical contact” (p. 178). Research on abusive supervision has primarily focused on its negative effects on outcomes at different levels. Empirical studies have indicated that abusive supervision has a negative impact on employee attitudes (Tepper, Henle, Lambert, Giacalone, & Duffy, 2008; Harvey, Stoner, Hochwarter, & Kacmar, 2007) and employee behaviors (Harris, Kacmar, & Zivnuska, 2007; Zellars, Tepper, & Duffy, 2002; Mitchell & Ambrose, 2007; Tepper, Henle, Lambert, Giacalone, & Duffy, 2008). Research also has indicated that the negative effects of abusive supervision occur at the team level, decreasing team creativity (Liu, Liao, & Loi, 2012) and team proactive behavior (Rousseau & Aube, 2018). Within organizations, it has been 1 estimated that abusive supervision affects 13.6% of workers in the United States, resulting in a cost estimate of $23.8 billion annually for U.S. corporations (Tepper, 2007; Tepper, Duffy, Henle, & Lambert, 2006). Besides outcomes, researchers have also paid some attention to the antecedents of abusive supervision. The major groups of antecedents include subordinate personality (e.g., Brees, 2012), subordinate behaviors (e.g., Walter, Lam, Van Der Vegt, Huang, & Miao, 2015), supervisor experience with abusive supervision (Taylor, Griffith, Vadera, Folger, & Letwin, 2019), family aggression (Garcia, Restubog, Kiewitz, Scott, & Tang, 2014), as well as team climates (Mawritz, Dust, & Resick, 2014; Taylor, 2004). Although some studies have focused on subordinate personality, very few studies have focused on leader personality traits to examine their association with leaders’ perpetration of abusive supervision. Yet, leadership scholars traditionally have examined how leader personality traits relate to leadership behaviors for other leadership styles, such as transformational leadership (e.g., Bono & Judge, 2004; Judge & Bono, 2000), ethical leadership (e.g., Kalshoven, Den Hartog, & De Hoogh, 2011; Walumbwa & Schaubroeck, 2009), and servant leadership (e.g., Washington, Sutton, & Field, 2006). These studies help answer the question of “what makes a leader great?” (Judge & Bono, 2000). Regarding abusive supervision, researchers know very little about why some leaders engage in these behaviors while others do not. Therefore, it is important to study subordinates’ perceptions of whether leaders with certain personality traits are more likely to be abusive. By exploring this question, I respond to Tepper’s (2007) call for future research to devote more attention to supervisor-level factors, such as personality traits. Furthermore, based on trait activation theory (Tett & Guterman, 2000), I argue that certain situational factors may strengthen the relationship between personality traits 2 and abusive supervision. Identifying personality traits as antecedents and situational factors as moderators has the potential to help organizations select and develop non-abusive leaders and foster more positive working relationships. Additionally, in the global context, it is important to acknowledge how cultural values impact the effectiveness of leader behaviors. Tepper (2007) proposed that people from high power distance countries, such as Japan and China, may find abusive supervision more acceptable than people from low power distance countries, such as the United States and Sweden. Tepper called for future researchers to examine the role of cultural values in abusive supervision. Responding to this call, some studies have explored how power distance orientation moderates the relationship between abusive supervision and outcomes (e.g., Lian, Ferris, & Brown, 2012; Lin, Wang, & Chen, 2013; Vogel et al., 2015). The findings of these studies have indicated that people with low power distance orientations tend to view abusive supervision as less fair than people with high power distance orientations. Moreover, for people with low power distance orientation, abusive supervision tends to exhibit a more negative impact on trust, job satisfaction, and work effort. However, additional research is needed to address the role of cultural values in abusive supervision. Existing studies on cultural values have only examined a limited number of employee attitudes and behaviors. Thus, it is not clear how cultural values moderate the relationship between abusive supervision and important employee behaviors, such as job performance, organizational citizenship behavior (OCB), and creativity. In addition, more research should examine the underlying mechanisms that explain why the abusive supervision- outcome linkage differs among people with different cultural values. 3 In this study, I also investigate the role of job type, either white-collar or blue-collar, as a moderator for the relationship between abusive supervision and employee outcomes. The abusive supervision literature has ignored job type as a moderator and has heavily relied on white-collar samples (e.g., Liu et al., 2012; Shoss, Eisenberger, Restubog, & Zagencyzk, 2013; Vogel et al., 2015). It is important to examine the role of job type to extend our understanding of abusive supervision. The results would shed light on the generalizability of existing findings on different occupations. In addition, abusive supervision researchers have typically studied cultural values by themselves, ignoring the influences of other factors such as job type (e.g., Lian et al., 2012; Vogel et al., 2015). In other words, how employees respond to abusive supervision is not simply influenced by their cultural values, such as the power distance orientation they hold, but also by the nature and characteristics of their jobs. Therefore, by examining both cultural values and job type as moderators simultaneously, I examine the relative importance of cultural values and job type on employees’ responses to abusive supervision. 1.2 Purpose and Contribution of the Research The overarching purpose of this dissertation is twofold: (1) to integrate personality theory into the abusive supervision literature, examining how leader personality trait antecedents are associated with abusive supervision, and (2) to investigate how employees’ power distance orientations and job type impact the abusive supervision-outcome linkage and potential underlying mechanisms. By addressing the above two questions, I test an integrated model that includes both antecedents and outcomes of abusive supervision, as well as potential moderation and mediation mechanisms. In my dissertation, I contribute to the leadership literature in several ways. First, I include both widely and narrowly defined personality traits of leaders as antecedents of abusive 4 supervision and therefore examine the predictive power of personality on abusive supervision. This approach expands the field’s understanding of leader personality traits represent one group of understudied antecedents. Personality antecedents may be uniquely suited for predicting abusive supervision because as a form of nonphysical aggressive behaviors, abusive supervision reflects leaders’ deep-seated values and behavioral tendencies. Second, I include a trait-relevant situational factor—ethical climate—to examine whether it moderates the personality-abusive supervision linkage. It is possible that certain situational factors may encourage personality expression in terms of abusive behaviors. In this way, I integrate trait activation theory with abusive supervision. Third, I examine how employee power distance orientation moderates the relationship between abusive supervision and outcomes, which include key employee behaviors such as job performance, OCB, creativity, and deviance. Generally, researchers assume that people with a high power distance orientation have less intensive responses to abusive supervision than their low power counterparts (Martinko, Harvey, Brees, & Mackey, 2013; Tepper, 2007). However, Lian et al. (2012) found that for people with a high power distance orientation, the relationship between abusive supervision and deviance is stronger than for people with a low power distance orientation. Thus, it is important to examine how power distance orientation impacts the abusive supervision-outcome linkage. Fourth, I examine how employee job type moderates the relationship between abusive supervisor and outcomes, and thereby extend our understanding of abusive supervision. Collectively, these additions improve our understanding of the relative influence of cultural values and job type on employees’ responses to abusive supervision. Fifth, based on the organizational justice theory (Tyler & Bies, 1990), I include interactional justice as a potential 5 mediator to examine whether it helps explain the relationships among abusive supervision, power distance orientation, job type, and outcomes. This moderated mediation model reveals the underlying mechanisms of moderated relationships by power distance orientation and job type. 1.3 Organization of the Research The dissertation is organized as follows. In Chapter 2, I present a review of the abusive supervision literature and identify gaps in the literature. In Chapter 3, I present my dissertation that delineates (1) the potential role of leader personality traits as propositional predictors of abusive supervision and a trait-relevant situational factor as a moderator, as well as (2) power distance orientation and job type as moderators for the abusive supervision-outcome linkage and potential mediator for such a linkage. For each relationship in this model, I describe its theoretical basis and corresponding hypothesis. In Chapter 4, I detail the method I use to test my proposed theoretical model and hypotheses. In Chapter 5, I present the results of my hypothesis testing. In Chapter 6, I discuss the implications of these findings and offer directions for future research. 6 CHAPTER 2 LITERATURE REVIEW 2.1 Abusive Supervision 2.1.1 Conceptualization of Abusive Supervision Leadership scholars have studied positive forms of leadership, such as transformational leadership and transactional leadership, to reveal how supervisors can positively influence subordinates and organizations (Bass, 1985). However, in recent decades, researchers have turned their attention to examine destructive supervisory behaviors, including sexual harassment, physical violence, and nonphysical aggressive behavior (Ashforth, 1997; Duffy, Ganster, & Pagon, 2002; Schat, Desmarais, & Kelloway, 2006). Among the identified behaviors, nonphysical behaviors, such as gossiping, withholding information, public ridicule, and ostracism have been most commonly studied. To study these nonphysical hostile behaviors, Tepper (2000) formally proposed the construct of “abusive supervision” to refer to “subordinates' perceptions of the extent to which supervisors engage in the sustained display of hostile verbal and nonverbal behaviors, excluding physical contact” (p. 178). After Tepper (2000) published this seminal paper on abusive supervision, many researchers began to pay attention to this new construct of negative leadership and conduct empirical research to examine its nomological networks (Martinko et al., 2013). Before Tepper (2000) proposed the construct of abusive supervision, in the workplace hostility and aggression literature scholars had proposed a few related constructs such as workplace bullying (Hoel et al., 1999), supervisor aggression (Neuman & Baron, 1998), and petty tyranny (Ashforth, 1997). However, abusive supervision is different from these constructs and covers different content domains. For example, workplace bullying occurs when individuals experience repeated exposure to hostile actions, such as attacks, abuse, and social isolation, in 7 the workplace (Hoel et al., 1999). This form of bullying is different from abusive supervision as it does not necessarily involve a downward target, and its perpetrators can possibly be coworkers and subordinates. In addition, bullies aim to cause harm, whereas abusive supervision does not. Abusive supervision does not encompass content other than hostility and does not have specific aim. Moreover, abusive supervision is different from aggression. While abusive supervision reflects indifference and willful hostility and may or may not be deviant (Tepper, 2000), aggression involves deviant, physical and nonphysical behaviors that cause harm (Neuman & Baron, 1998). Tepper (2007) summarized how abusive supervision is different from the existing constructs in four key ways. First, abusive supervision has a downward target—subordinates. Second, it excludes physical hostility but includes both verbal and nonverbal hostility. Third, it solely focuses on hostility. Fourth, it does not refer to any intended outcomes. Besides these differences, Tepper (2007) further emphasized that abusive supervision is a construct based on subordinates’ subjective assessment of leaders’ sustained and willful display of nonphysical hostility. A comprehensive review indicates that until now, abusive supervision has attracted more empirical examination than other constructs detailing destructive supervisor behaviors. Since I am interested in the supervisor-subordinate relationship specifically from a “dark” side, I find that abusive supervision is the most appropriate construct. Researchers have primarily and consistently operationalized abusive supervision with the most widely used unidimensional scale developed by Tepper (2000). Tepper (2000) drew on instruments that measure nonphysical abuse in other relationships, such as dating (Raymond & Bruschi, 1989), domestic abuse (Shepard & Campbell, 1992), and other management literature on nonphysical abusive behaviors (e.g., Robinson & Bennett, 1995). He identified an initial list 8 of 20 items. After performing a content analysis of these 20 items, Tepper interspersed these items with 20 other items adapted from physical abuse measures and created a checklist. He eventually kept 15 items that use a 5-point Likert scale. An example of an item in this scale is “(Boss) Tells me my thoughts or feelings are stupid.” 2.1.2 Nomological Network of Abusive Supervision As indicated by review papers and meta-analytical studies, researchers have a strong interest in abusive supervision. In Tepper’s 2007 review, he found 20 articles on abusive supervision published since 2000. Whereas in a more recent review, Martinko et al. (2013) identified 62 new studies in the six years since Tepper’s review. In a recent meta-analysis, Mackey, Frieder, Brees, and Martinko (2017) found 112 relevant studies based on a search conducted in March 2014. These primary studies and reviews have presented a large nomological network of abusive supervision that includes antecedents, attitudinal and behavioral outcomes, moderators, and mediators as explanatory mechanisms. I summarize the research on abusive supervision, focusing on major findings across the literature. However, this review is not intended to be an inclusive discussion of all published and unpublished studies. Research on abusive supervision has mostly focused on outcomes at multiple levels (Martinko et al., 2013; Tepper, 2007). Overall research has found that abusive supervision is related to undesirable employee outcomes, such as perceptions of injustice (Tepper, 2000), decreased job satisfaction (Lin et al., 2013), decreased organizational commitment (Tepper et al., 2008), psychological distress (Harvey et al., 2007), deviance (Mitchell & Ambrose, 2007; Tepper et al., 2008), as well as low levels of OCBs (Zellars et al., 2002) and job performance (Harris et al., 2007). A cross-over effect of abusive supervision is that it also has a negative impact on employees’ well-being outside of the workplace. For example, research has shown 9 that abusive supervision is related to work-to-family conflict (Carlson, Ferguson, Hunter, & Whitten, 2012), less family satisfaction (Carlson, Ferguson, Perrewé, & Whitten, 2011), and family undermining (Wu, Kwong Kwan, Liu, & Resick, 2012). At the team level, abusive supervision is negatively associated with team creativity (Liu et al., 2012) and team proactive behavior (Rousseau & Aube, 2018). At the organization level, abusive supervision is related to more deviance behavior toward the organization (Detert, Treviño, Burris, & Andiappan, 2007; Mitchell & Ambrose, 2007) and economic cost (Tepper et al., 2006). Research also has identified antecedents of abusive supervision at multiple levels (Zhang & Bednall, 2016). At the subordinate level, abusive supervision is related to subordinate personality (Brees, 2012), core self-evaluation (Wu & Hu, 2009), hostile attribution bias (Brees, 2012), and poor job performance (Walter et al., 2015). At the supervisor level, abusive supervision is related to leaders’ previous experience with abusive supervision (Taylor et al., 2019) and family aggression (Garcia et al., 2014). Lastly, at the team level, abusive supervision is related to a hostile climate (Mawritz et al., 2014) and unethical climate (Taylor, 2004). 2.2 Leader Personality and Abusive Supervision To date, a literature review indicates that the abusive supervision research on personality has been mostly from the subordinate perspective. Defined as a subordinate’s subjective assessment, abusive supervision can be colored by the subordinate’s personality (Brees, Martinko, & Harvey, 2016; Tepper, 2007). Primary studies have examined how subordinates with different personalities view abusive supervision differently. For example, research has indicated that perceived abusive supervision was positively related to neuroticism and negatively related to conscientiousness (Wang, Harms, & Mackey, 2015), and positively related to negative affectivity and trait anger (Brees et al., 2016). Researchers have also has examined how 10 subordinate personality moderates the relationship between abusive supervision and outcomes. In one example, Nandkeolyar, Shaffer, Li, Ekkirala, and Bagger (2014) found that the relationship between abusive supervision and job performance was weaker when employees were high in conscientiousness. A meta-analysis on abusive supervision indicates that most studies that examined personality have focused on the Big Five, positive affect and negative affect, and concluded that personality variables have weak to moderate associations with abusive supervision (Mackey et al., 2017). Surprisingly, only a few studies have examined how leader personality has an impact on abusive supervision, and this represents an understudied area (Zhang & Bednall, 2016). Although it is interesting to see how the personality of subordinates colors their perceptions of abusive supervision, it is important to study how the personality of supervisors is related to their abusive supervision behaviors because supervisors are the perpetrators. The relationship between leader personality variables and abusive supervision helps answer why some leaders actively engage in abusive supervision behaviors while others do not. Results from a limited number of studies have examined leader personality variables including the Big Five, the dark triad, positive and negative affect, and HEXACO (honesty-humility, emotionality, extraversion, agreeableness, conscientiousness, and openness to experience) personality. They indicated that abusive supervision is related to supervisor negative affect (Pan & Lin, 2018), Machiavellianism (Wisse & Sleebos, 2016; Kiazad, Restubog, Zagenczyk, Kiewitz, & Tang, 2010), narcissism (Waldman, Wang, Hannah, Owens, & Balthazard, 2018), conscientiousness in the Big Five (marginally significant in correlation analysis, Camps, Stouten, & Euwema, 2016), and agreeableness and honesty-humility in the HEXACO personality framework (Breevaart & de Vries, 2017). McGinnis (2010) examined the relationship between MBTI personality and 11 abusive supervision but did not find any relationship between them. In addition, one study also examined and supported the moderator roles of neuroticism, conscientiousness, and agreeableness between supervisor role overload and abusive supervision via frustration (Eissa & Lester, 2017). However, with few existing studies on leader personality, we cannot be confident in drawing overarching conclusions from the results. More research is needed to examine how leader personality traits, both broadly and narrowly defined traits, are associated with abusive supervision and how these relationships are moderated by situational factors. 2.2.1 Two Approaches to Examining Personality The Big Five Model. Researchers have recognized and appreciated the important role of personality in explaining individual behaviors. As an important aspect of individual differences, personality is usually stable overall time and across situations (Hough, Oswald, & Ock, 2015). Among different personality theories, the Big Five model has dominated research and applications (Hough et al., 2015). A search of “Big Five” in Google Scholar as of March 2019 shows 4.4 million results which include a large number of highly cited meta-analytical studies, such as personality and job performance (Barrick & Mount, 1991), personality and leadership (Bono & Judge, 2004), personality and entrepreneurial status (Zhao & Seibert, 2006), personality and job satisfaction (Judge, Heller, & Mount, 2002). Although some researchers are concerned with the precise meaning of the five personality factors, it is widely agreed upon which traits define each factor (Barrick & Mount, 1991). The first dimension is extraversion which is associated with traits such as sociability, gregariousness, assertiveness, talkativeness, and high energy. The second dimension is neuroticism or emotional stability, which is associated with anxiety, depression, and anger. The 12 third dimension is agreeableness, which indicates a tendency to be courteous, flexible, trusting, cooperative, and tolerant. The fourth dimension is conscientiousness, which refers to being dependable, responsible, hard-working, and achievement-oriented. The fifth dimension is openness to experience, which is associated with a tendency to be imaginative, cultured, curious, and intelligent (Barrick & Mount, 1991). Although there is no uniform agreement among researchers on this broad and inclusive framework of personality, one advantage of using the Big Five framework is that it enables the ability to refer and contribute to the cumulative results across studies. In general, research has supported the relationship between the Big Five and leadership behaviors. For example, the Big Five traits are associated with transformational leadership and transactional leadership, with extraversion as the strongest and most consistent predictor of transformational leadership (Judge & Bono, 2004). For ethical leadership, conscientiousness and agreeableness are significant predictors because these two traits are associated with being responsible and caring (Kalshoven et al., 2011). For servant leadership, leader agreeableness is positively associated with followers’ ratings of servant behaviors (Washington et al., 2006). A comprehensive meta-analysis on LMX has indicated that leaders high in agreeableness and extraversion tend to develop high-quality relationships with followers (Dulebohn, Bommer, Liden, Brouer, & Ferris, 2012). Overall, these relationships are small to medium in magnitude. However, only a few studies have examined the relationship between Big Five personality dimensions of supervisors and abusive supervision, and as a dominant personality framework, the Big Five deserves more attention to contribute to cumulative understanding. The existing studies on the Big Five and abusive supervision provide limited and confusing results. For example, research has supported that the neuroticism and agreeableness dimensions appear 13 to be particularly associated with aggressive behaviors (Costa, McCrae, & Dembroski, 1989; Gleason, Jensen-Campbell, & Richardson, 2004; Graziano, Jensen-Campbell, & Hair, 1996; Miller, Lynam, & Leukefeld, 2003; Suls, Martin, & David, 1998). However, using the Big Five, Camps and his colleagues (Camps et al., 2016) found only a marginally significant correlation between consciousness and abusive supervision. This is contrary to expectations, especially since the results did not indicate a significant relationship between neuroticism, agreeableness and abusive supervision. It would be interesting to explore whether the Big Five are useful in predicting abusive supervision behaviors. The results from such studies would be beneficial to both the research and practice. Narrowly Defined Personality Traits. As personality research has progressed, researchers have pointed out that the Big Five is not comprehensive, indicating that some important constructs such as those related to honesty and interpersonal interaction are missing from this framework (Hough et al., 2015). In addition, researchers have noticed that the Big Five only have small to moderate predictive validity in outcomes (e.g., Bono & Judge, 2004). In response to such criticisms for the Big Five, researchers have begun to pay attention to more narrowly defined personality variables such as positive and negative affect, the dark triad, and trait anger with the hope of better understanding the relationship between personality and important life and work outcomes. Research using these personality variables has yielded fruitful results. For example, Bettencourt, Talley, Benjamin, and Valentine (2006) examined how narrowly defined personality traits, such as trait aggression, and trait irritability, are associated with aggressive behaviors, and found moderate to strong relationships, supporting the predictive validity of these personality traits. Hershcovis, et al. (2007) conducted a meta-analysis on the predictors of workplace aggression and found out that individual personality including negative affectivity and trait anger have moderate 14 relationships with both individual and organization targeted aggressive behaviors. In addition, negative affectivity and trait anger generally have stronger relationships with aggression than situational factors such as injustice, job dissatisfaction, interpersonal conflict, situational constraints, and poor leadership. In a meta-analysis, O’Boyle and his colleagues (O’Boyle, Forsyth, Banks, & McDaniel, 2012) found that across studies Machiavellianism and psychopathy are negatively associated with job performance, and the three components of the dark triad are all associated with deviance behaviors. In sum, empirical studies have supported the predictive validity of these narrowly defined personality traits in explaining outcomes. Unfortunately, existing studies have ignored some narrowly defined yet important personality variables, such as trait aggressiveness, and the dark triad. Results from meta-analysis have supported the predictive power of narrowly defined personality variables (Judge, Rodell, Klinger, Simon, & Crawford, 2013), as well as the relationship between these dark personality variables and aggressive behaviors in general (Bettencourt et al., 2006). Research indicates that some personality variables, such as trait aggressiveness and trait irritability have strong relationships with aggressive behavior across conditions, and others have moderate relationships with aggressive behavior under certain conditions (Bettencourt et al., 2006). Hershcovis et al. (2007) found that trait anger and negative affectivity have moderate relationships with workplace aggression, and these effects are stronger than most of the situational antecedents. Therefore, it would be interesting to examine whether these narrowly defined personality variables have high predictive validity for abusive supervision. I expect that trait aggressiveness and the dark triad are predictive of abusive supervision, which is one type of aggressive and hostile behavior characterized by nonphysical and a downward direction in the workplace. 15 2.2.2 Trait Activation Theory and Situational Factors Although personality can explain behaviors, it also interacts with situations. Interactionist psychology acknowledges that individuals can behave similarly across situations, and situations can cause different people to behave in a similar way (Tett & Guterman, 2000). To explain this long-standing debate regarding traits and situations as sources of behavioral variance, Tett and Guterman (2000) proposed trait activation theory. They argued that the principle of trait activation reconciles the trait-situation relationship by holding that “the behavioral expression of a trait requires arousal of that trait by trait-relevant situational cues.” They further explained that this theory offers a link to classic behaviorism (i.e., stimulus-response theory) by framing traits as differential response tendencies. To incorporate the role of situation in explaining behaviors, they defined personality traits as “intraindividual consistencies and interindividual uniquenesses in propensities to behave in identifiable ways in light of situational demands” (p. 398). An example they discussed is the relationship between trait aggression and aggressive behaviors. People high in aggression do not always behave aggressively, they exhibit aggressive behaviors only in certain situations. With aggression-inducing stimuli, people high in aggression will show a quicker and stronger response or greater sensitivity to weak situational cues. This indicates that personality expression in behavior varies by situation type, and trait activation is a process underlying trait expression. This theory is in line with interactionism’s perspective that the expressions of personality traits require trait-relevant situations (Kenrick & Funder, 1988). One testable hypothesis based on the trait activation theory is “behavioral predictions based on trait measures should improve with knowledge of situation trait relevance” (Tett & Guterman, 2000). Researchers have tested this hypothesis and have provided general support for this theory. For example, Tett and Burnett (2003) proposed a theoretical model for a relationship between 16 personality-job performance based on trait activation theory that calls for the consideration of situational factors. This model has received empirical support, such that extraversion may better predict job performance in jobs requiring social skills (e.g., Judge & Zapata, 2015). A meta- analysis on the relationship between personality and aggressive behavior indicated that some personality variables such as trait anger and Type A personality better predict aggressive behaviors in the provoking condition, such as stressful actions or situations than the neutral condition (Bettencourt et al., 2016). In the field of leadership, researchers have also been aware of the importance of situational factors when examining the role of personality. Specifically, situational factors refer to “aspects of the social context that are perceived by people and are largely influenced by other members of the organization” (Hershcovis et al., 2007). Judge and his colleagues (Judge, Bono, Ilies, & Gerhardt, 2002) called future research to focus on the many situational factors that may moderate the validity of personality in predicting leadership; their proposal received support from empirical research. For example, De Hoogh, Den Hartog, and Koopman (2005) examined how the perceived dynamic work environment moderated the relationship between the Big Five and charismatic and transactional leadership. They found that perceived dynamic work environment moderated the relationship between all five personality factors except extraversion and the leadership forms. In a study of military leaders and their supervisors, Ng, Ang, and Chan (2008) found that job demands and job autonomy moderated the relationship between leader neuroticism, extraversion, conscientiousness, and leader effectiveness through leadership self- efficacy. From the employee perspective, Greenbaum, Hill, Mawritz, and Quade (2017) studied the relationship between employee Machiavellianism and unethical behavior and supported that abusive supervision was a trait activator for Machiavellianism. 17 In summary, the trait activation theory has received significant empirical support that warrants integrating this theory in the study of abusive supervision. However, the existing studies have failed to sufficiently account for context factors as moderators for the relationship between supervisor personality and abusive supervision as trait activation theory has suggested. Until now, researchers have only examined how employee organization-based self-esteem (Kiazad et al., 2010), leader-member exchange (LMX) (Pan & Lin, 2018), and supervisor’s perceived position power (Wisse & Sleebos, 2016) moderate the relationship between personality variables and abusive supervision. However, more situational factors need to be examined. As proposed by Tett and Burnett (2003), trait-relevant cues in the workplace can be multilevel, including organizational, social and task level cues. To be specific, organizational level factors can include organizational climate, culture, structure. Social factors capture trait- relevant cues that are embedded in interaction with others. Examples of these are needs and expectations from supervisors and peers regarding effort and communication, and related social behaviors. At the task level, situational factors stem from the nature of the work itself and include day-to-day tasks, responsibilities, and procedures of the job. Examining the interaction between leader personality and trait-relevant cues in explaining abusive supervision could provide new insights into the literature. 2.3 Abusive Supervision and Cultural Values With the surge of research on how cultural values impact human behaviors (e.g., Hofstede, 1980a), leadership scholars began to study leadership from a cross-cultural perspective (e.g., Jung, Bass, & Sosik, 1995; Resick, Hanges, Dickson, & Mitchelson, 2006; Walumbwa, Lawler, & Avolio, 2007). Hofstede’s seminal book on cultural dimensions (1980b) provides a 18 useful framework to broaden our understanding of how leadership works cross-culturally. Leadership scholars have shown great interest in examining how cultural values influence the effectiveness of different leadership forms. Leadership researchers have examined how cultural values, especially individualism-collectivisms and power distance orientation, have an impact on leadership behaviors, such as transformational leadership (e.g., Dulebohn, Wu, Liao, & Hoch, 2017; Kirkman, Chen, Farh, Chen, & Lowe, 2009), servant leadership (e.g., Hale & Fields, 2007; Schaubroeck, Lam, & Peng, 2011), and abusive supervision (Vogel et al., 2015; Wang, Mao, Wu, & Liu, 2012). Many researchers have called for future research on abusive supervision from a cultural perspective (Mackey et al., 2017; Martinko et al., 2013). Because abusive supervision was proposed in the U.S., many initial studies were conducted in Western cultures. Similarly, even after Tepper’s (2007) call for cross-cultural studies, a meta-analysis indicated that only a few studies incorporated national culture in studies of abusive supervision (Mackey et al., 2017). Some of these studies simply noted the potential influence of cultural differences in their introductions and discussions, but they did not further explore the potential impact of these differences empirically (e.g., Jian, Kwan, Qiu, Liu, & Yim, 2012; Lee, Yun, & Srivastava, 2013; Rafferty & Restubog, 2011). Among the varied cultural values or dimensions, one particularly relevant cultural dimension is power distance, which captures “The extent to which the less powerful members of institutions and organizations within a country expect and accept that power is distributed unequally” (Hofstede, 2001, p. 98). To distinguish between power distance at country and individual levels, I use the term power distance orientation to indicate an individual-level construct following practices used by other researchers (e.g., Kirkman et al., 2009; Lian et al., 19 2012). Empirically, Lian et al. (2012) found that how individual power distance orientation moderates the relationship between abusive supervision and its outcomes depends on the nature of outcomes. Subordinates with higher power distance orientation have been found to be more likely to show deviant behaviors and less likely to have injustice perceptions with abusive superiors than subordinates with lower power distance orientation. However, Hon and Lu (2016) found that for employees with low power distance orientation, they were more likely to exhibit abusive behaviors with abusive supervisors. These two studies found contradictory results regarding whether power distance orientation strengthens or weakens the relationship between abusive supervision and employee deviant behaviors. In addition, Vogel and his colleagues (Vogel et al., 2015) examined whether abusive supervision behaviors are perceived similarly by subordinates across different cultures. They found out that the negative effects of abusive supervision were stronger for subordinates within the Anglo than the Confucian Asian culture and subordinates from Anglo culture perceived abusive supervision as less fair. These differences can be explained by subordinates’ power distance orientation in these cultures. In another study, with a Chinese sample, researchers found that abusive supervision had a stronger negative relationship with interactional justice for employees with low power distance orientation than for employees with high power distance orientation (Wang et al., 2012). Similarly, Lin et al.’s (2013) findings supported the moderating role of power distance with two Chinese samples. They found that the negative relationships of abusive supervision with employee psychological health and job satisfaction were weaker for employees with higher power distance orientation. Besides power distance, researchers have examined other cultural dimensions such as traditionalism (Liu, Kwong Kwan, Wu, & Wu, 2010), achievement orientation, and benevolence (Kernan, Watson, Chen, & Kim, 2011). In 20 almost all these studies, cultural values were modeled as moderators to examine whether cultural values influence the relationship between abusive supervision and its correlates. However, because few abusive supervision studies have included cultural values, researchers believed that “opportunity for future research to investigate the impact of cultural differences on abusive supervision causes, perceptions, and reactions remains largely untapped” (Martinko et al., 2013). A review of the literature indicates that more research needs to be done. 2.3.1 Implicit Leadership Theory and Cultural Values Researchers have used implicit leadership theory to explain why cultural values may influence the acceptance and effectiveness of leadership (Den Hartog et al., 1999; House, Javidan, Hanges, & Dorfman, 2002). Implicit leadership theory argues that people have their own ideas about the nature of leaders and leadership, and they have their own beliefs and expectations about how leaders should behave in general (Eden & Leviathan, 1975). With a leader prototype (i.e., a collection of characteristic traits or attributes) in mind, people match the perceived attributes of potential leaders to their internal prototypes (Foti & Luch, 1992). The fit between the perceived individual and the leadership prototype is associated with the likelihood of perceiving that person as a leader (Offermann, Kennedy, & Wirtz, 1994; Foti & Luch, 1992). Researchers have identified tyranny, including characteristics such as pushy, conceited, dominant, and manipulative, as a negative prototype of implicit leadership theory (Epitropaki & Martin, 2004; Offermann et al., 1994). Research has suggested that people with different cultural orientations perceive and react to authority differently (Kirkman et al., 2009). For example, individuals with high power distance orientation accept the power differences between leaders and subordinates and believe that people in leader positions deserve respect, trust, and deference, as well as accept 21 subordinates’ limitations in the decision-making process (Javidan, Dorfman, de Luque, & House, 2006; Kirkman et al., 2009). Therefore, it is possible that people with high power distance orientation find abusive supervision more acceptable and respond less intensively (Tepper, 2007). Consistent with Tepper’s (2007) expectation, empirical studies have found that people with high power distance orientation tend to respond less intensively than people with low power distance (Lian et al., 2012; Vogel et al., 2015; Wang et al., 2012). One exception is interpersonal deviant behavior. Lian et al. (2012) found that based on social learning theory, when power distance orientation is high, abusive supervision is related to more deviant behaviors via the perception of the likelihood of rewards. This does not indicate that employees with high power distance react intensively with abusive supervision, rather it is because people with high power distance tend to treat their leaders as role models and are more likely to mimic their leaders’ behaviors. However, it is not clear whether this unexpected pattern is stable or not considering opposite conclusions from another study conducted by Hon and Lu (2016). They found that employees with lower power distance orientations are more likely to exhibit abusive behaviors when they experience abusive supervision. It is also not clear whether this unexpected pattern for deviant behaviors generalizes to other behaviors such as job performance, OCB, and creativity. In my model, I examine both attitudinal outcomes, such as trust, psychological safety, and behavioral outcomes such as job performance, OCB, creativity, and deviance. 2.3.2 Level of Analysis for Cultural Values Hofstede initially developed his cultural framework to measure cultural values at the country level (1980b). With numerous theoretical advancements in the field of cross-cultural studies, a growing number of studies have argued that “country” may not be the most appropriate 22 unit of analysis for these studies (Fischer, 2009; Kirkman, Lowe, & Gibson, 2006; Taras, Steel, Kirkman, 2016). Research has found that cultural values also vary significantly at the individual, group, organizational, socioeconomic status, state, and religious levels within countries (Daniels & Greguras, 2014; Dheer, Lenartowicz, Peterson, & Petrescu, 2014; Greenfield, 2014). In addition, contrary to the static views of culture, dynamic views have produced research on cultural frameshifting, which argues that in the global world individuals can dynamically integrate elements from other cultures and dissociate from elements of their culture (e.g., Benet- Martínez, Leu, Lee, & Morris, 2002; Hong, Morris, Chiu, & Benet-Martínez, 2000). With these advancements, researchers have called for additional future research to study culture beyond the national level (Gelfand, Aycan, Erez, & Leung, 2017). In addition, it is important to conduct studies at more micro-levels to avoid the ecological fallacy which refers to incorrectly generalizing results found at the group level to individuals that belong to that group (Robinson, 1950). Relatedly, Hofstede (2001) cautioned that researchers should not generalize results found at the individual level to the group level to avoid reverse ecological fallacy. Besides the conceptual differences of relationships at different levels, the reason to avoid the ecological fallacy is also that the relationship between two constructs can vary at different levels. For example, Spector et al. (2001) found no relation between collectivism and job satisfaction at the country level whereas Kirkman and Shapiro (2001) found a positive relation between collectivism and job satisfaction at the individual level. Therefore, it is important to consider the level of measurement and analysis when making theoretical expectations and drawing inferences from results (Daniels & Greguras, 2014; Gelfand, Erez, & Aycan, 2007; Kirkman et al., 2006). 23 In my model, I measure power distance orientation and conduct analysis at the individual level and use scales that are specifically designed to measure individual cultural values (i.e., Dorfman & Howell, 1988). This practice is in line with most cross-cultural studies conducted in the field of human resources, organizational behavior, and industrial psychology. For example, Tsui, Nifadkar, and Ou (2007) found that among the studies they reviewed 84% investigated cultural values at the individual level. In a meta-analysis, Tara, Kirkman, and Steel (2010) found that 76% of the data points were at the individual level. In addition, Kirkman et al. (2006) in his review highlighted the importance of direct measurement of cultural values at the individual level rather than using country scores as proxies when the study is at the individual level. Most importantly, the focus on the individual level can be employed to better answer my research questions which examine how cultural values influence employee attitudes and behaviors in the workplace at the employee level. In addition, although some meta-analytical studies incorporated power distance orientation as a moderator in the analysis for several outcomes, results cannot be fully generalized to the individual level. In one study, Park, Hoobler, Wu, Liden, Hu, and Wilson (2017) assigned cultural values to each sample based on the country where the sample was drawn as a proxy, and then dichotomized the power distance values and conducted subgroup analysis. Similarly, Zhang and Liao (2015) used geographic regions as a proxy for cultural values and conducted subgroup analysis for samples from Asia and samples from North America. Although this approach is appropriate for meta-analysis considering that this is one of the few ways that they could assign cultural values to each sample, there are some concerns with this practice. As Daniels and Greguras (2014) discussed, countries differ on variables other than cultural values, such as language, economic development, government systems, etc. Therefore, it 24 is impossible to disentangle the influences of other factors in these cases. With some insights from the meta-analysis, it is still important to conduct studies at the individual level to examine the corresponding research questions. 2.3.3 Justice as A Mediator Much of the existing cross-cultural research on abusive supervision has not addressed potential mechanisms through which abusive supervision and cultural values relate to outcomes. Little is known about why and how abusive supervision and individual cultural values affect individual attitudes and behaviors within cultures. Mediators in such a moderated mediation model may help explain why the interaction effects exist. Until now, researchers have only examined the likelihood of rewards, turnover intention, and feelings of shame as mediators in studies with cultural values as moderators. For example, Lian et al. (2012) examined the likelihood of rewards as a mediator based on the social learning theory and found an indirect positive relation of abusive supervision and interpersonal deviance through the likelihood of rewards is stronger for individuals with higher power distance orientation. In another study, Richard, Boncoeur, Chen, and Ford (2018) found that the indirect relationship between abusive supervision and interpersonal aggression via turnover intentions was stronger for high power- distance-oriented individuals when the HR support climate was perceived low. Daniels (2015) studied the relationship between abusive supervision and in-role performance and OCB, and examined feelings of shame as a mediator and power distance as the second-stage moderator (i.e., moderator for the relationship between mediator and outcomes), and found that the indirect effects at different levels of feelings of shame were only different for in-role performance but not OCB. However, they did not study how power distance moderates the direct relationship between abusive supervision and in-role performance, and OCB. 25 In response, I include interactional justice as a theory-based mediator to help explain why abusive supervision and power distance orientation affect employee behaviors. Interactional justice has been the primary mediator that explains how abusive supervision could have an impact on outcomes (Aryee, Chen, Sun, & Debrah, 2007; Tepper, 2000; Vogel et al., 2015). This responds to a call for future research to focus on the mediation mechanisms. The justice mediator could help explain the underlying mechanisms of the influence of the interaction between abusive supervision and power distance orientation on outcomes. 2.4 Abusive Supervision and Job Type Scholars have dichotomized occupations in terms of white-collar and blue-collar employees. Following the definitions from previous research, we define white-collar employees as professional and semi-professional employees who have more job autonomy and more challenging tasks, and blue-collar employees as those who perform physical work and have relatively restricted career paths (Hu, Kaplan, & Dalal, 2010; Toppinen-Tanner, Kalimo, & Mutanen, 2002). Research has also revealed the differences between white-collar and blue-collar employees regarding their expectations and preferences in the workplace. For example, scholars have indicated that compared with blue-collar employees, white-collar employees care more about the intrinsic values of their jobs but care less about extrinsic values (Harris & Locke, 1974; Locke, 1973; Weaver, 1975) and have different conceptualizations of job satisfaction facets (Hu, Kaplan, & Dalal, 2010). More specifically, with a national sample from Wright, Bengtsson, and Frankenberg (1994) discussed the differences in aspects of the physical work environment, medical symptoms, psychological stress, job satisfaction, and life satisfaction between white- collar and blue-collar employees. Other empirical studies using physiological methods have indicated that organizational justice only has an independent impact on white-collar employees 26 but not blue-collar employees (Herr, Bosch, Loerbroks, et al., 2015; Herr, Bosch, van Vianen, et al., 2015). Overall, research has supported that white-collar and blue-collar employees not only differ in their job characteristics but also differ in their expectations and preferences. Similar to the literature on other constructs in the organizational behavior field (e.g., organizational commitment, Riketta, 2002), the literature on abusive supervision has not paid much attention to the influence of job type. Most studies on abusive supervision have relied on white-collar samples to draw conclusions (e.g., Liu et al., 2012; Shoss et al., 2013; Vogel et al., 2015). For the few studies that have used blue-collar employees as samples, they did not examine the role of job type either (e.g., Bamberger & Bacharach, 2006; Haar, Fluiter, & Brougham, 2016; Lin et al., 2013; Kluemper, Mossholder, Ispas, Bing, Iliescu, & Ilie, 2019). Without discussing the role of job type as a boundary condition, researchers have implicitly assumed that findings based on white-collar samples are generalizable to blue-collar employees. This can be problematic and may limit our understanding of abusive supervision. In order to deal with this concern, I include two categories of job type, white-collar and blue-collar, as a moderator. I examine both how job type moderates the relationship between abusive supervision and whether the mediation mechanisms are the same or not for white-collar and blue-collar employees. 27 CHAPTER 3 MODEL AND HYPOTHESIS DEVELOPMENT The purpose of this dissertation is to contribute to the literature on abusive supervision by developing and testing a moderated mediation model. Overall, the model examines whether leaders with certain personality traits are more likely to exhibit abusive supervision behaviors, how cultural value and job type influence the abusive supervision and outcome relationship, and how underlying mechanisms explain such relationships. As described in Chapter 3, I develop a model that examines both antecedents and outcomes of abusive supervision. Specifically, this model incorporates leader personality dimensions as antecedents to abusive supervision. Based on trait activation theory, I propose that situational relevant factors moderate the leader personality-abusive supervision linkage. For outcomes, I incorporate cultural dimensions as proposed by Hofstede (1980b) and use justice theory (Bies & Moag, 1986) to explain the abusive supervision-outcome relationship. I focus on how power distance orientation as a cultural value moderates the relationship between abusive supervision and outcomes, as well as how the justice mechanism mediates the relationship between abusive supervision and outcomes. To examine the generalizability of research findings based on white-collar employees and to study the relative importance of cultural values, I also examine job type as a moderator. As presented in Figure 1, the first component in my model is leader personality that includes agreeableness, neuroticism, trait aggressiveness, narcissism, Machiavellianism, and psychopathy. These personality traits represent antecedents of abusive supervision. The second component in my model is ethical climate, a hypothesized moderator for the relationship between leader personality traits and abusive supervision. The third component is abusive supervision, the key construct of interest in my dissertation. The fourth and fifth components are power distance orientation and job type, hypothesized to moderate the relationship between 28 abusive supervision and employee outcomes. The sixth component is interactional justice, a variable hypothesized to mediate the relationship between abusive supervision and outcomes. The last component is employee outcomes, including job performance, OCB, creativity, and deviance. In this chapter, I describe each hypothesis based on my model. I first develop hypotheses concerning how leader personality traits, including both broadly and narrowly defined personality variables, are associated with abusive supervision. Based on trait activation theory, I hypothesize that ethical climate moderates the above-discussed relationships. Next, I hypothesize that power distance orientation and job type moderate the influence of abusive supervision on employee outcomes and that the relationship between abusive supervision and outcomes are mediated by interactional justice as an underlying mechanism. Finally, I hypothesize that power distance orientation and job type moderate these mediation effects in this moderated mediation model. Figure 1. Conceptual model 29 This model contributes to the abusive supervision literature in three ways. First, my review of the literature on abusive supervision in Chapter 2 indicates that extant research has largely ignored why some leaders engage in abusive supervision. This model integrates both individual (leader) personality traits and situational factors, based on trait activation theory to explore this understudied question. I expect that leaders high on certain personality traits will tend to exhibit abusive behaviors and some situational factors can strengthen/weaken such positive relationships. Second, this model integrates individual cultural orientation as a boundary condition and tests how it moderates the relationship between abusive supervision and outcomes including job performance, OCB, creativity, and deviance. To investigate the influence of cultural value, I integrate the justice mechanism to explain how abusive supervision and individual cultural orientation interactively influence employee outcomes. Third, my proposed model integrates job type as a boundary condition to examine potential differences between white-collar and blue-collar employees regarding how abusive supervision influences employee behaviors. The inclusion of job type as a moderator also helps test the relative influence of power distance orientation. 3.1 Antecedents to Abusive Supervision In this section, I describe the relationship between leader personality variables and abusive supervision, as well as the moderation effects of ethical climate. 3.1.1 Leader Personality and Abusive Supervision As discussed in Chapter 2, leader personality represents one group of understudied antecedents in the abusive supervision literature in the past. Research has generally supported the relationship between leader personality and leadership behaviors (e.g., Bono & Judge, 2004; Dulebohn et al., 2012). Revealing the relationship between personality and abusive supervision 30 can help researchers gain a better understanding of what traits make a leader abusive. In addition, researchers have also noted that the associations between widely defined personality variables, represented by the Big Five, and leadership behaviors are usually weak. This has caused scholars to suggest that future research should focus on more narrowly defined personality traits (Bono & Judge, 2004). Research has also supported the better predictive validity of lower-order personality traits in explaining some behaviors (Judge et al., 2013). As presented in Figure 1, I examine both widely defined personality traits, including neuroticism, agreeableness, and narrowly defined personality traits, including trait aggressiveness, and the dark triad. These personality variables are antecedents to abusive supervision as depicted in Figure 1. Neuroticism and agreeableness. It is reasonable to believe that people have different propensities to display aggressive behaviors (Tedeschi & Felson, 1994). Recent studies have focused on personality traits rather than situational factors to explain why people engage in aggressive behaviors (Bettencourt et al., 2006). Neuroticism and agreeableness in the five-factor model (Costa & McCrae, 1992) appear to be particularly related to aggressive behaviors (Costa et al., 1989; Gleason et al., 2004; Graziano et al., 1996; Miller et al., 2003; Suls et al., 1998). Agreeableness is associated with the motives to maintain positive relationships with other people (Gleason et al., 2004). The opposite of agreeableness is antagonism, which is associated with hostility and irritability— “they need to oppose, to attack, or to punish others” (Costa et al., 1989, p. 45). Antagonistic people tend to mistrust, lack concern for others, and may exclude those who are disliked or inferior. Neuroticism is different from antagonism and is characterized by a tendency to experience negative affectivity, psychological stress, and unstable emotions (Costa et al., 1989). Although both agreeableness and neuroticism are related to aggression and 31 hostility, Costa et al. (1989) distinguished between them and stated that “whereas neurotic hostility is exemplified by frequent and strong experiences of anger…antagonistic hostility is exemplified by cynicism, callousness, and lack of cooperation” (p. 53). In the workplace, I expect that leaders high in neuroticism, low in agreeableness, tend to display more abusive supervision behaviors. Trait aggressiveness. Besides widely defined traits, researchers also have studied a variety of specific personality traits without reference to these major dimensions in the Big Five. Research has found that trait aggressiveness is an important antecedent to aggressive behaviors and has moderate to strong relationships with aggressive behaviors in general (Bettencourt et al., 2006). Trait aggressiveness refers to one’s propensity to engage in physical and verbal aggression, to hold hostile cognitions, and to express anger (Buss & Perry, 1992). Trait aggression includes four dimensions: Physical aggression, verbal aggression, anger, and hostility. In the workplace, researchers have found that trait anger, a dimension of trait aggressiveness, shows a moderate to strong relationship with interpersonal targeted aggression (Hershcovis et al., 2007). Kant, Skogstad, Torsheim, and Einarsen (2013) warned people to beware of angry leaders. They found that leaders high in trait anger are more likely to display petty tyranny, which refers to a leader’s use of power and authority oppressively, capriciously, and vindictively (Ashforth, 1997). Similarly, researchers have found that trait aggressiveness is related to workplace interpersonal deviance and that it interacts with interactional justice and race in explaining deviance (Aquino, Galperin, & Bennett, 2004). In the organizational context, it is expected that leaders high in trait aggressiveness are more likely to use abusive supervision behavior, one of the nonphysical aggressive behaviors, as a way to express anger. 32 Dark triad. Besides focusing on the personality traits that are related to general aggressive behaviors, three other aversive personality traits also deserve attention. Among the socially aversive personalities, the dark triad, including Machiavellianism, narcissism, and psychopathy, has attracted the most empirical attention (Paulhus & Williams, 2002). These three personality traits all have a long history in the clinical and philosophical literature and migrated to the management literature with the publications of classic questionnaires in the 1970s and 1980s (Furnham, Richards, & Paulhus, 2013). First, the Machiavellian personality is defined by three sets of interrelated values: “an avowed belief in the effectiveness of manipulative tactics in dealing with other people, a cynical view of human nature, and a moral outlook that puts expediency above principle” (O’Boyle et al., 2012). People high in Machiavellian characteristics usually hold a negative view of people and are more likely to make ethically suspect choices such as cheating, lying, and betraying others, but they do not engage in extremely negative forms of antisocial behaviors such as violent crimes (Jones & Paulhus, 2009; Kish-Gephart, Harrison, & Trevino, 2010). Second, narcissism is characterized by extreme self-aggrandizement and is not necessarily a bad thing. Most people possess some level of narcissism that colors their perceptions and behaviors (Rhodewalt & Peterson, 2009). Different from healthy self-respect and confidence, narcissists exaggerate their achievements, block criticism, and refuse to compromise (Campbell, 1999; Resick, Whitman, Weingarden, & Hiller, 2009). They also appear arrogant, aggressive, and less likable (Buffardi & Campbell, 2008). Research also indicates that with the perception of ego threat, narcissists are likely to respond aggressively (Bushman, Baumeister, Thomaes, Ryu, Begeer, & West, 2009). Third, people with psychopathy are characterized by a lack of empathy and concern for other people and social norms, impulsivity, emotional coldness, and engagement 33 in antisocial behaviors including criminal activities to achieve their ends (Hare & Neumann, 2009). Although all aversive, these three personality traits are overlapping but distinct constructs. The correlations among these three traits are usually positive and from moderate to strong, but they are differentially correlated with other constructs, such as the Big Five, and cognitive ability (Furnham et al., 2013; O’Boyle et al., 2012; Paulhus & Williams, 2002). Researchers have identified five major outcome domains for the dark triad, one of which is workplace behavior. The dark triad is often associated with the notions of “toxic leadership” and “bad bosses” (Furnham et al., 2013). Research has indicated that psychopathy is positively related to passive leadership behaviors and is negatively related to consideration (Westerlaken & Woods, 2013). In addition, leaders high in Machiavellianism and psychopathy have detrimental effects on subordinates' career satisfaction and job satisfaction (Volmer, Koch, & Goritz, 2016). From a theoretical perspective, researchers have explained how narcissistic leaders’ cognitive processes can contribute to abusive supervision (Keller Hansbrough & Jones, 2014). A recent meta-analysis indicates that the three traits in the dark triad are all associated with counterproductive work behaviors (O’Boyle et al., 2012). Therefore, I expect a positive relationship between the dark triad and abusive supervision. Based on the above, I hypothesize the relationship between leader personality variables and abusive supervision: Hypothesis 1: Abusive supervision is negatively associated with leader (a) agreeableness, and positively associated with leader (b) neuroticism, (c) trait aggressiveness, (d) narcissism, (e) Machiavellianism, and (f) psychopathy. 34 3.1.2 Trait-Relevant Situational Factor as A Moderator Tett and Burnett (2003) proposed trait activation theory to explain the relationship between personality and situation in explaining behaviors. They argued that the expressions of personality traits require trait-relevant situations. In the model presented in Figure 1, I include one trait-relevant situation factor, ethical climate, as a moderator for the relationship between these “negative” personality traits and abusive behaviors. Based on the typology proposed by Howell, Dorfman, and Kerr (1986), I hypothesize that this moderator works as a neutralizer. Ethical climate. I conceptualize the ethical climate at the organization level consistent with its conceptualization and published studies (e.g., Martin & Cullen, 2006; Victor & Cullen, 1988). The study on work climate has more than half a century of history and has demonstrated that work climate exerts an influence on people’s work in the organization and the organization as a whole (Kuenzi & Schminke, 2009). Schneider and Reichers (1983) defined work climates as a set of shared perceptions regarding the policies, practices, and procedures that an organization or group rewards, supports, and expects. According to Victor and Cullen (1988), an ethical work climate encompasses the perceptions that help to answer “What should I do?” and reflects the normative patterns perceived by members in the organization with some degree of consensus. In their seminal article, they contended that organizations develop different ethical climates that in turn influence employees’ moral behaviors beyond individual characteristics. Researchers have identified five facets of ethical climates: Caring, law and code, rules, instrumental, and independence (Victor & Cullen, 1988). Among the five facets, caring is the one that characterizes benevolence and interpersonal relationships. Specifically, a caring ethical climate is an affective climate that is concerned with interpersonal and social relations among employees and subsumes four dimensions such as participation, cooperation, warmth, and social 35 rewards (Ostroff, 1993). The affective climate is one type of ambient stimulus that is available to all group members and can shape their behaviors (Hackman, 1992). With a caring climate, individuals perceive that decisions should be made based on the consideration of other people’s well-being (Martin & Cullen, 2006). Ethical climate can be considered as a trait-relevant situational factor for these above- discussed personality traits, including agreeableness, neuroticism, trait aggressiveness, and the dark triad. Results from empirical studies have supported that ethical climate has an impact on employees’ attitudes and behaviors. Meta-analysis has supported the positive influence of caring ethical climate on employee satisfaction, commitment, and well-being (Martin & Cullen, 2006). Through ethical climate, employees assess and diagnose their working environments by understanding what they should do and identifying what are unethical issues within their organizations (Cullen, Parboteeah, & Victor, 2003). In addition, Wang and Hsieh (2014) found that with a perception of low ethical climate, the relationship between perceived psychological contract breach and acquiescent silence is stronger. The positive effects of ethical climate can be understood from the social information processing theory (Salancik & Pfeffer, 1978). Individuals look around for cues regarding behavioral expectations and then adjust their behaviors accordingly. An unethical climate signals to supervisors that treating subordinates with less caring is acceptable and is less likely to be punished. Based on trait activation theory (Tett & Burnett, 2000), I expect that a lower level of ethical climate serves as a provocation condition that signals the existence and acceptance of hostile behaviors, and to some extent, it strengthens the association between these negative personality traits and abusive supervision. Therefore, I hypothesize: 36 Hypothesis 2: Ethical climate moderates the relationship between abusive supervision and leader (a) agreeableness, (b) neuroticism, (c) trait aggressiveness, (d) narcissism, (e) Machiavellianism, and (f) psychopathy, such as these relationships are stronger when the ethical climate is low. 3.2. Outcomes to Abusive Supervision In this section, I describe the relationship between abusive supervision and employee outcomes, as well as their moderators and mediator as presented in Figure 1. First, I expect that individual power distance orientation and job type moderated the relationship between abusive supervision and outcomes. Next, I expect the indirect effects of abusive supervision on outcomes via interactional justice. Finally, I expect moderated mediation relationships based on the above- discussed logic with power distance orientation and job type as moderators. 3.2.1 Power Distance Orientation as A Moderator In this section, I describe how power distance orientation moderates the relationship between abusive supervision and interactional justice, job performance, OCB, creativity, and deviance. Interactional justice. Researchers have identified interactional justice as the third type of justice after distributive justice and procedural justice (Cropanzano, Prehar, & Chen, 2002). Interactional justice refers to the quality of interpersonal interaction and is most likely to occur when supervisors treat employees with interpersonal respect and dignity and necessary explanations (Bies & Moag, 1986; Bies, 1989). As organization representatives, leaders largely determine employees’ perception of interactional justice (Cohen-Charash & Spector, 2001). Compared with distributive justice and procedural justice, interactional justice is especially predictive of employee reactions to supervisors and to the immediate work environment 37 (Malatesta & Byrne, 1997; Masterson, Lewis, Goldman, & Taylor, 2000). Tepper (2000) explained the negative effects of abusive supervision on employee attitudes and behaviors based on the justice theory. He argued that abusive supervision is perceived by employees as interpersonally unfair, and such perception, in turn, affects employees’ job satisfaction, commitment, and turnover decision. This association has been supported by empirical studies, indicating that abusive supervision negatively affects the perception of interactional justice (Aryee et al., 2007; Rafferty & Restubog, 2011; Tepper, 2000). However, this negative relationship between abusive supervision and interactional justice is moderated by individual power distance orientation. Research has demonstrated that for individuals with low power distance orientation, the relationship between abusive supervision and interactional justice is stronger (Lian et al., 2012; Vogel et al., 2015; Wang et al., 2012). This is because justice perception is inherently based on norms and values (Cropanzano, Byrne, Bobocel, & Rupp, 2001) which are associated with prevailing cultural standards. It has been argued that although concerns about justice are universal, justice standards can be highly particularistic (Greenberg, 2001). Specifically, what people believe to be fair depends on their repeated exposure to validated opinions regarding what is considered to be fair, and such repeated exposure shapes their expectations of fairness that serve as their basis of assessment (Greenberg, 2001). This explains that people may have different fairness perceptions because they have different values and norms. Among values and norms, power distance orientation is a relevant one (Hofstede, 1980b). People with high power distance orientation may find abusive supervision more acceptable because they believe that people in authority are superior and deserve compliance from subordinates, whereas people with low power distance may find 38 abusive supervision unfair because they think they should be treated with respect and dignity regardless of the relative status in the organization. Therefore, I hypothesize: Hypothesis 3: Power distance orientation moderates the negative relationship between abusive supervision and interactional justice, such that this relationship is stronger when power distance orientation is low. Job performance. Research has found that abusive supervision negatively affects job performance (Harris et al., 2007; Shoss et al., 2013; Xu, Huang, Lam, & Miao, 2012). Researchers have explained this negative relationship using social exchange theory. In other words, both supervisors and subordinates bring resources to the workplace for exchange. When subordinates have a poor relationship with supervisors, supervisors may not provide them with valuable resources and support, making them not willing to fully contribute to the organization. In addition, based on the conservation of resources theory, subordinates may perceive a threat in terms of resource loss and low anticipated return of effort with abusive supervision. As a result, they withdraw their efforts from work (Harris et al., 2007; Xu et al., 2012). In addition, this negative relationship varies with power distance orientation. For subordinates with high power distance orientation, they may not associate abusive supervision with the potential loss of resources or consider abusive supervision as something that indicates a poor relationship. Therefore, they may not restrain their efforts in the work as much as subordinates with low power distance orientation do. Therefore, I hypothesize: Hypothesis 4: Power distance orientation moderates the relationship between abusive supervision and job performance, such that this negative relationship is weaker when power distance orientation is high. 39 OCB. Different from job performance, OCB refers to individual behaviors that go beyond role descriptions but benefit organizational operations (Organ & Ryan, 1995). Research indicates that abusive supervision also decreases OCB (Aryee et al., 2007; Zellars et al., 2002). Research has found that employees are more likely to express their attitudes in extra-role behaviors which they have greater discretion than in-role behaviors (Organ, 1977; Smith, Organ, & Near, 1983). When subordinates perceive their supervisors as less supportive, they are likely to withhold their engagement in OCB. Power distance orientation may moderate this negative relationship. I expect that abusive supervision influences OCB for employees with low power distance orientation to a great extent. Besides the values associated with power distance, research has found that people from different cultures have different definitions of OCB. Lam, Hui, and Law (1999) found that participants from Hong Kong and Japan, with cultures of high power distance, were more likely to define some categories of OCB as their required job roles than participants from U.S. and Australia. Therefore, abusive supervision is likely to have a smaller influence on OCB for people with high power distance orientation. Therefore, I hypothesize: Hypothesis 5: Power distance orientation moderates the relationship between abusive supervision and OCB, such that this negative relationship is weaker when power distance orientation is high. Creativity. Employee creativity refers to the generation of novel and useful ideas and is often the starting point for organizational innovation (Zhou & George, 2001). Among the influential factors of creativity, leaders have been proposed to be an important situational factor that can cultivate or hinder employee creativity (e.g., George, 2007; Gong, Huang, & Farh, 2009; Oldham & Cummings, 1996). Research has demonstrated that abusive supervision, as a destructive leadership style, can hinder employee creativity (e.g., Liu et al., 2012; Zhang, Kwan, 40 Zhang, & Wu, 2014). Further, this negative relationship may be moderated by power distance orientation. Researchers have had discussions on how culture can have an impact on creativity and innovation. For example, Ahmed (1998) summarized that freedom and risk-taking can promote employee creativity. However, for employees with high power distance orientation, they tend to behave submissively and are more fearful of their leaders. As a result, they tend to perceive that they do not have much latitude in executing their own work and are less likely to take risks such as expressing novel ideas at work. In one empirical study, Farmer, Tierney, and Kung-McIntyre (2003) found that being exposed to U.S. culture (low power distance) could increase the creativity of Taiwanese employees because the educational system in the U.S. tends to stimulate personal expression rather than mimetic learning (Gardner & Hatch, 1989). Therefore, I hypothesize: Hypothesis 6: Power distance orientation moderates the relationship between abusive supervision and creativity, such that this negative relationship is stronger when power distance orientation is high. Deviance. Research indicates that abusive supervision is positively associated with employee deviant behaviors (e.g., Michell & Ambrose, 2007; Tepper, Carr, Breaux, Geider, Hu, & Hua, 2009). Bennett and Robinson (2003) found that negative work experiences such as perceptions of frustration and injustice are primary antecedents of deviance. With abusive supervision, subordinates are likely to have a feeling of frustration and injustice perceptions. As a result, they may retaliate against their supervisors with destructive behaviors such as deviance. However, the limited empirical studies have indicated contradictory results regarding how power distance orientation moderates the relationship between abusive supervision and deviance. Some researchers found that this positive relationship between abusive supervision and deviance is 41 stronger for people with low power distance orientation (Hon & Lu, 2016), whereas other researchers argued that this positive relationship between abusive supervision and deviance is stronger for people with high power distance orientation based on social learning theory (Lian et al., 2012). Hofstede (2001) used “fear of disagreement” in measuring power distance at the country level. This indicates that people in high power distance countries consider their supervisors as more fearful. When subordinates are treated with an injustice like abusive supervision, they are fearful and try to avoid subsequent actions that may trigger future abuse (Kiewitz, Restubog, Shoss, Garcia, & Tang, 2016; Kish-Gephart, Detert, Treviño, & Edmondson, 2009). In addition, related research on traditionality, a cultural value correlated with power distance orientation, indicates that traditional Chinese values encourage forgiveness, which may discourage retaliation to the source of injustice especially authority or other negative reactions to other parties (Liu, Kwong Kwan, et al. 2010; Wu, Zhang, Chiu, Kwan, & He, 2014). Therefore, I argue that people with high power distance orientation are less likely to display deviance under abusive supervision than people with low power distance orientation. Therefore, I hypothesize: Hypothesis 7: Power distance orientation moderates the relationship between abusive supervision and deviance, such that this positive relationship is weaker when power distance orientation is high. 3.2.2 Job Type as A Moderator In this section, I describe how job type, either white-collar or blue-collar, moderates the relationship between abusive supervision and interactional justice, job performance, OCB, creativity, and deviance. Although in Figure 1, I indicate that interactional justice mediates the relationship between abusive supervision and job performance, OCB, creativity, and deviance, in 42 this section, I am interested in how job type also moderates the total effects between abusive supervision and outcomes including job performance, OCB, creativity, and deviance. Interactional justice. Research has indicated that the social exchange process is more central to white-collar employees, whereas the economic exchange process is more central to blue-collar employees (Herr, Bosch, van Vianen et al., 2015; Littek & Heisig, 1989). Overall, white-collar employees care more about the intrinsic values of their jobs, and blue-collar employees care more about the monetary rewards of their jobs (e.g., Harris & Locke, 1974; Locke, 1973; Weaver, 1975). Specifically, white-collar employees tend to have better work experiences than blue-collar employees, such as having higher job satisfaction (e.g., Fisk & Friesen, 2012), higher interactional justice (e.g., Inoue et al., 2009), higher supervisor social support (e.g., Morris et al., 1999), and higher work engagement (e.g., Kanten & Sadullah, 2012). As some researchers have argued, in a relationship where social exchange is expected, “a violation of justice standard should have more serious ramifications than when the relationship is less communal” (Cropanzano, Rupp, Mohler, & Schminke, 2001). Research has demonstrated that white-collar and blue-collar employees have different justice expectations and are influenced by justice to different extents. For example, Herr and his colleagues found that justice perceptions were only related to chronic heart disease and musculoskeletal pain for white-collar employees but not blue-collar employees (Herr, Bosch, Loerbroks, et al., 2015; Herr, Bosch, van Vianen, et al., 2015). Therefore, when white-collar employees experience abusive supervision, they are more likely to consider it as a violation of justice standards and less acceptable. Therefore, I hypothesize: 43 Hypothesis 8: Job type moderates the relationship between abusive supervision and interactional justice, such that this negative relationship is stronger for white-collar employees than for blue-collar employees. Job performance. Researchers have argued that because white-collar employees have more job autonomy than blue-collar employees, white-collar employees may translate their attitudes into behaviors more easily than blue-collar employees (Randall, 1990). Meta-analytical results have also supported the stronger relationship between attitudes and job performance for white-collar employees than blue-collar employees (Riketta, 2002). In addition, as white-collar employees have different job preferences from blue-collar employees, they also have different expectations regarding their workplace treatment (Hu et al., 2010; Weaver, 1975). They expect to be treated by their supervisors with dignity and respect. Therefore, I expect that when white- collar employees experience abusive supervision, they are more likely to feel dissatisfied and decrease their work efforts. Therefore, I hypothesize: Hypothesis 9: Job type moderates the relationship between abusive supervision and job performance, such that this negative relationship is stronger for white-collar employees than for blue-collar employees. OCB. Compared with in-role behaviors, employees are more likely to express their attitudes in extra-role behaviors toward which they have greater discretion (Organ, 1977; Smith et al., 1983). Researchers have demonstrated in a meta-analysis that job satisfaction is related to OCB to a stronger extent for professional employees than for nonprofessional employees (rc = .41 vs. .20; Petty, McGee, & Cavender, 1984). This indicates that white-collar employees’ OCBs are more sensitive to their job attitudes. When they are satisfied with their jobs, they are more likely to exhibit more extra-role behaviors. Therefore, when white-collar employees 44 experience abusive supervision, they are more likely to express their dissatisfaction by decreasing their extra-role behaviors. Therefore, I hypothesize: Hypothesis 10: Job type moderates the relationship between abusive supervision and OCB, such that this negative relationship is stronger for white-collar employees than for blue- collar employees. Creativity. Employees are more likely to exhibit creative behaviors when they have intrinsic motivation (e.g., Da Costa, Páez, Sánchez, Garaigordobil, & Gondim, 2015; Zhang & Bartol, 2010). Compared with blue-collar employees, white-collar employees typically view themselves as valuable and consider having a sense of accomplishment as an important aspect in their jobs (Weaver, 1975; Locke, 1973; Centers & Bugental, 1966). Therefore, when abusive supervision happens, white-collar employees are more likely to have decreased intrinsic motivation and withdraw their creative behaviors. Therefore, I hypothesize: Hypothesis 11: Job type moderates the relationship between abusive supervision and creativity, such that this negative relationship is stronger for white-collar employees than for blue-collar employees. Deviance. Because white-collar employees have different job preferences from blue- collar employees (Hu et al., 2010; Weaver, 1975), their tolerance levels of abusive supervision are lower. When supervisors violate the professionalism expected in the white-collar occupations by treating employees without dignity and respect, white-collar employees would likely react more intensively and negatively. Therefore, I expect that the positive relationship between abusive supervision and deviance is stronger for white-collar employees. Therefore, I hypothesize: 45 Hypothesis 12: Job type moderates the relationship between abusive supervision and deviance, such that this positive relationship is stronger for white-collar employees than for blue- collar employees. 3.2.3 Interactional Justice as a Mediator In this section, I focus on the mediated relationship between abusive supervision and outcomes via interactional justice as presented in Figure 1. Interactional justice. Research has indicated that interactional justice is a significant predictor of employee attitudes and behaviors. When employees feel they are being treated with respect and dignity, they are more likely to have a higher level of job satisfaction, better job performance, more OCB, and less deviant behaviors (e.g., Cohen-Charash & Spector, 2001; Colquitt, 2001; Colquitt et al., 2013). Interactional justice has been viewed as a primary mediator and underlying mechanism explaining why abusive supervision is detrimental to employees (e.g., Aryee et al., 2007; Tepper, 2000; Lian et al., 2012). Tepper (2000) was the first scholar to propose the construct of abusive supervision and set forth a justice-based model that posits interactional justice as a key mediator for the relationship between abusive supervision and employee outcomes. He argued that abusive supervision is a significant source of interactional injustice, and the justice perception, in turn, translates the negative effects of abusive supervision into negative attitudes and behaviors. Aryee et al. (2007) using a sample of 47 supervisors and 178 subordinates from a telecommunication company demonstrated that subordinates’ perceptions of interactional justice, but not procedural justice, fully mediated the relationship between abusive supervision and employee work outcomes. In addition, leadership scholars have demonstrated justice perceptions as mediators for other leadership forms and outcomes, such as transformational leadership (e.g., Cho & 46 Dansereau, 2010; Pillai, Schriesheim, & Williams, 1999), servant leadership (e.g., Mayer, Bardes, & Piccolo, 2008; Walumbwa, Hartnell, & Oke, 2010), ethical leadership (e.g., Zehir, Akyuz, Eren, & Turhan, 2013). The overall argument is that justice explains the reciprocal nature of the leader-follower relationship and is one of the most important mediators that can explain the influence of leadership on subordinates (van Dierendonck, 2011). Therefore, I hypothesize: Hypothesis 13: Interactional justice mediates the negative relationship between abusive supervision and job performance, OCB, creativity and the positive relationship between abusive supervision and deviance. 3.2.4 A Moderated Mediation Model The logic I have outlined implies a moderated mediation relationship that the mediator can explain the predictivity of the interactive relationship between abusive supervision and power distance orientation in outcomes (Edwards & Lambert, 2007). A moderated mediation model tests whether a moderating effect is transmitted through a mediator variable (Baron & Kenny, 1986). In other words, in a moderated mediation model, the indirect effect through a mediator varies at different levels of the moderator. This moderated mediation model can provide insight into the “black box” regarding how cultural values and job type together with abusive supervision influence outcomes (Kirkman et al., 2006). Therefore, I hypothesize: Hypothesis 14: Power distance orientation moderates the indirect relationship between abusive supervision and job performance, OCB, creativity, and deviance via interactional justice, such that the indirect effects are weaker when power distance orientation is high. Hypothesis 15: Job type (white-collar vs. blue-collar) moderates the indirect relationship between abusive supervision and job performance, OCB, creativity, and deviance via interactional justice, such that the indirect effects are stronger for white-collar employees. 47 48 4.1 Participants and Procedures CHAPTER 4 METHOD I recruited participants from organizations in China. I collected data using surveys from both supervisors and subordinates with three waves in multiple medium companies. I used one month as the time interval between every two consecutive waves. The use of multiple sources and multiple wave data helped mitigate issues of common method bias. I also intended to collect data at multiple levels, with subordinates at level 1, and leaders at level 2, and measure variables at both levels. These structured data enabled me to examine the influence of leader behaviors on individual attitudes and behaviors. I estimated the sample size following the procedures outlined by Scherbaum and Ferreter (2009). The power analysis indicated that I needed to have complete data from 80 leaders and 320 employees to have at least a statistical power of 80%. In total, my sample include survey data from 1009 employees and 136 leaders, a much larger sample size than the required minimum sample size. The team size ranges from 3 to 10 employees per team. On average, each team has 7.26 employees. These employees and leaders were from four companies, including two urban designing and architecture companies, one high-tech company, and one textile company. The average age of leaders was 42.13 (SD = 6.65) and 36% were females. The average age of employees was 37.44 (SD = 8.69) and 62% were females. All the scales I used have been previously developed and validated. Participants answered the survey items using Likert scales. I received approval by MSU IRB regarding my study and survey instruments prior to collecting data. The use of multi-wave data collection also helped reduce survey fatigue and increased the quality of the data collection. To be specific, in the first wave, for leaders I measured leader personality variables, ethical climate in their 49 organizations, and their own demographic information; for subordinates, I measured their perceptions of abusive supervision and their demographic variables. In the second wave, I only collected data from subordinates. I measured interactional justice, and power distance orientation. In the third wave, I only collected data from leaders. I measured their evaluations of their subordinates’ behavioral outcomes. At the beginning of each survey, I asked the participants to read and sign the voluntary consent form. Participants were offered small gifts or cash as incentives to participate in the survey for each wave of the survey completed. 4.2 Measures All measures were assessed with 5-point Likert scales with anchors of “1 = strongly disagree” and “5 = strongly agree” unless noted otherwise. Participants needed to indicate to what extent they agree with each statement or survey item using this 5-point Likert scale. All the Cronbach alphas are found to be acceptable. 4.2.1 Time 1 Survey I measured the variables listed below in the first wave from the supervisors. Leader agreeableness and neuroticism. I measured leader agreeableness, neuroticism, and conscientiousness from the Big Five Inventory developed by John, Donahue, & Kentle (1991). This scale allows an efficient and flexible assessment of the five dimensions when individual facets of the big five are not the primary focus. This scale shows good psychometric properties. Agreeableness has 9 items, neuroticism has 8 items, and conscientiousness has 9 items. An example item for agreeableness is “Is considerate and kind to almost everyone” (α is .70), and an example item for neuroticism is “Worries a lot” (α is .78). Leaders needed to indicate to what extent they think each item is descriptive of themselves. Leader trait aggression. I use a refined version of self-report Aggression 50 Questionnaire (Bryant & Smith, 2001) to measure leader trait aggression. This refined version has 12 items including three items from each of the four dimensions originally included by Buss and Perry (1992). This shortened scale shows better CFA results than the original scale developed by Buss and Perry (1992) and good reliabilities from .72 to .80 for each dimension. One example item is “My friends say that I'm somewhat argumentative.” Leaders needed to indicate to what extent each item is descriptive of themselves. Leader dark triad. I used the scale developed by Jonason and Webster (2010) to measure leader narcissism, Machiavellianism, and psychopathy. This is a shortened version of the original scales developed to measure these three personality variables. This scale includes three items for each, and the reliabilities are acceptable, with .83 for narcissism (e.g., “I tend to want others to admire me”), .81 for Machiavellianism (e.g., “I tend to manipulate others to get my way”, and .70 for psychopathy (e.g., “I tend to be cynical”). Ethical climate. To measure supervisors’ perceptions of the ethical climate in their organization, I used the caring dimension of ethical work climate questionnaire developed by Victor and Cullen (1988). An example item is “What is best for everyone in the company is the major consideration here.” This scale has acceptable reliability, with α = .71. Control variables. I measured the age, gender, education level, and department tenure as control variables for supervisors as these variables may influence relational perceptions (Tsui and O’Reilly, 1989). I measured the variables listed below in the first wave from the subordinates. Abusive supervision. I used the 15-item scale developed by Tepper (2000) to measure subordinate’s perception of abusive supervision. One example is “(My supervisor) Ridicules 51 me.” This scale has good reliability, with α = .91. For this scale, 1 indicates never, 2 indicates seldom, 3 indicates occasionally, 4 indicates moderately often, and 5 indicates very often. Job type. I collected this variable by asking the job position that each participant had, and then categorized all the answers into two categories, white-collar, or blue-collar. All the subordinates remained in the same positions during the entire survey time window. Control variables. I measured the age, gender, education level, dyadic tenure, as control variables for subordinates as these variables may influence relational perceptions (Tsui and O’Reilly, 1989). 4.2.2 Time 2 Survey I measured the variables listed below in the second wave from the subordinates. Power distance orientation. I measured subordinate’s power distance orientation using the 6-item scale developed by Dorfman and Howell (1988). This scale was designed specifically to measure cultural values at the individual level. One example item is “Employees should not disagree with management decisions” (α = .76). Interactional justice. I measured subordinate’s perception of interactional justice based on their interaction with their direct leaders using the 9-item scale developed by Colquitt (2001). An example item is “Has he/she treated you in a polite manner?” This scale has good validity and reliability (α = .91). 4.2.2 Time 3 Survey I measured the variables listed below in the third wave from the supervisors. Job performance. I asked leaders to evaluate the job performance for each of their subordinates using the 4-item scale developed by Liden, Wayne, and Stilwell (1993). An 52 example item is “What is your personal view of your subordinate in terms of his or her overall effectiveness?” This scale has good reliability (α = .85). OCB. I asked leaders to evaluate the OCB for each of their subordinates using the 16- item scale developed by Lee and Allen (2002). An example item is “(This subordinate) Help others who have been absent.” This scale has good reliability of .92. Creativity. I asked leaders to evaluate the creativity for each of their subordinates using 5 items from the scale developed by Zhou and George (2001). An example item is “(This subordinate) Suggests new ways to achieve goals or objectives.” This scale has good reliability (α = .89). Deviance. I asked leaders to evaluate the deviant behaviors for each of their subordinates using the scale developed by Bennett and Robinson (2000). An example item is “(This subordinate) Made fun of someone at work.” This scale has good reliability (α = .85). For this scale, I used a 5-point Likert scale to measure the frequency of deviant behaviors. 1 indicates never, 2 indicates seldom, 3 indicates occasionally, 4 indicates moderately often, and 5 indicates very often. 4.3. Analytical Approaches With my integrated model, I examined both the antecedents of abusive supervision and outcomes of abusive supervision. Because I was interested in questions such as what personality traits were predictive of abusive supervision and what were the consequences of abusive supervision, I conducted the analyses in two steps using appropriate approaches. For example, I examined the relationship between antecedents and abusive supervision with regression analysis. For the relationship between abusive supervision and outcomes, I adopted HLM to consider the data dependence issue within teams and use path analysis to examine the mediation and 53 moderated mediation effects. In addition, in my dissertation I was not interested in how leader personality traits influenced employee outcomes through abusive supervision. Therefore, it was more appropriate to conduct analyses separately to examine the antecedent-abusive supervision relationship and abusive supervision-outcome relationship than testing the theoretical model all together and treating abusive supervision as a mediator. 4.3.1 Data Screening and Preparation Before conducting the analyses, I wanted to make sure the data were usable and of good quality. First, I checked for missing data. Because I used paper and pencil surveys and collected data face-to-face with the help of human resource employees, all the surveys were largely complete. If some participants only missed some items, I used the average of the corresponding scale of that item to replace the missing value of that item. Second, I checked the frequency and distribution of each item and each variable. If a variable was not normally distributed, I checked whether the skewed distribution had an impact on any analysis used. Checking the frequency and distribution of variables also revealed possible data entry errors. 4.3.2 Descriptive Data, Correlation Analysis, and Reliability Analysis I first calculated the mean and standard deviation of each variable. In addition, I also reported the zero-order correlations for all variables and Cronbach’s α for the reliability of each scale. 4.3.3 Regression and Moderation Analysis I used moderated regression analysis as the main statistical procedure for examining the relationship between leader personality traits and abusive supervision, as well as the proposed moderating effect of ethical climate. Moderation regression analysis allows for a comparison between alternative models with and without interaction terms, where an interaction effect only 54 exists if the interaction term contributes significantly to the variance explained in the dependent variable over the main effects of the independent variables (Aiken, West, & Reno, 1991). I centered predictors and moderators before creating the interaction terms and graphed interaction(s) following procedures set forth by Aiken et al. (1991). 4.3.4 Measurement Model, Hierarchical Linear Modeling Analysis, and Structural Model I used the analytical approaches described below to examine the relationship between abusive supervision and employee outcomes, as well as the proposed moderators and mediator. Before conducting the path analysis, I conducted multilevel confirmatory factor analyses on survey items to determine whether my measurement model captured distinct constructs. I performed these analyses in Mplus 8.2 (Muthén & Muthén, 2017), with the raw data entered. I first specified a model in which all the items loaded on their corresponding hypothesized latent constructs. I compared the above model to the two-factor model in which all the items loaded either on one attitudinal latent construct or on a behavioral outcome. In addition, a chi-square difference test indicated whether the hypothesized two-factor model provided a better fit to the observed data than the one-factor model. Besides comparing the models, I also used the following indices to evaluate the model fit, including the chi-square test, standardized root mean square residual (SRMR), root mean square error of approximation (RMSEA), and the comparative fit index (CFI). An SRMR of less than .09, an RMSEA less than .08, and a CFI greater than .90 indicate that the model fits the data well (Baumgartner & Homburg, 1996; Hu & Bentler, 1999; Iacobucci, 2010). The nested structure of my data, with employees nested within supervisors, violates the independence assumption of traditional ordinary least squares (OLS). The use of OLS in my situation may result in biased estimation of standard error (Hofmann, 1997). Since hierarchical 55 linear modeling (HLM) is an appropriate statistical-based analytical tool for dealing with non- independence problems caused by nested data, I used HLM 7.0 (Raudenbush & Bryk 2002) to test the cross-level main effect and moderation effect hypotheses and to obtain a robust standard error and estimates of greater accuracy. The HLM approach is a two-stage strategy that investigates variables occurring at two levels of analysis (Hofmann, Griffin, & Gavin, 2000). The outcome of this first stage is intercept and slope terms estimated separately for each group. These intercept and slope estimates from the level 1 analysis are then used as outcome variables in the level 2 analysis. In my data set, the employee is at level 1 and the supervisor or working team is at level 2. Although the level 1 and level 2 equations are discussed separately, it should be noted that they are estimated simultaneously. The key terms are fixed effects, random effects, and variance components. Fixed effects are parameter estimates that do not vary across groups. The variance of the level 1 residuals and the variance-covariance of the level-2 residuals comprise the variance components. The HLM procedure uses the EM algorithm to produce maximum-likelihood estimates of the variance components. Random coefficients are those that are allowed to vary across groups. The HLM procedure does not provide any statistical tests for these parameters. Centering is another important issue in HLM data analysis. The choice between grand- mean centering or group-mean centering depends on what research question is of interest (Enders & Tofighi, 2007). For hypotheses regarding the relationship between abusive supervision, individual cultural values, and outcomes, because level-1 predictor, abusive supervision, was of substantial interest, the use of group-mean centering made the results easier to interpret. 56 Several hypotheses are associated with the moderation effects in HLM. The basic logic and procedure of examining the moderation effects in HLM are similar to that in OLS regression. Researchers have outlined specific steps of examining moderation effects in HLM (Preacher, Curran, & Bauer, 2006). With the statistical software HLM 7.03, it is possible to generate output regarding the variances and covariances among the predictors in the model. I would need these statistics in calculating the simple slopes of each regression line. I used the structural model to test the hypotheses for the mediation and moderated mediation between abusive supervision and outcomes. Because of the multilevel nature of the data, I used multilevel structural equation modeling (MSEM) as recommended by Preacher, Zyphur, and Zhang (2010). I performed these analyses in Mplus 8.2 (Muthén & Muthén, 2017), with the raw data entered. The output from Mplus included the path coefficient, indirect effects, and moderated mediation effects. I used the fit indexes described above to evaluate the model fit in this analysis. 57 CHAPTER 5 RESULTS In this chapter, I present the results of my dissertation. I first report the descriptive data and correlations among variables at both the employee and team level. Then I report the analytical results for antecedents and outcomes separately. 5.1 Data Aggregation, Descriptive Data, and Correlations Based on the analytical needs, I only aggregated one variable from the employee level to the leader level. Specifically, each leader’s abusive supervision rating was obtained by aggregating the ratings from employees supervised by that leader. The ICC of abusive supervision was .37, supporting the use of aggregated values of employee ratings as each leader’s abusive supervision behavior rating. I used the approach outlined by Croon and van Veldhoven (2007) to compute the adjusted group means of abusive supervision ratings as the team leader’s abusive supervision score based on employees’ ratings of their own leaders. As Croon and van Veldhoven (2007) have discussed, using the adjusted group means provides more accurate results and unbiased estimates than simply using the arithmetic means. I conducted the analysis using SPSS and present the results in Table 1. Along the diagonals, I report Cronbach alphas for each scale. 5.2 Antecedents to Abusive Supervision 5.2.1 Regression and Moderation Analysis Hypothesis 1 stated that leader personality variables are correlated with abusive supervision. Hypothesis 2 stated that ethical climate moderates the relationship between abusive supervision and leader personality, such that these relationships are stronger when ethical climate is low. Table 2 presents the results for the regression and moderation analysis. I tested the above hypotheses in a stepwise approach in SPSS. In Model 1, I included demographic variables as 58 control variables. In Model 2, I added personality variables and ethical climate as predictors to test the main effects of personality variables. In Model 3, I added the interaction terms between personality variables and ethical climate. To facilitate the interpretation of results, I centered the personality variables and ethical climate and created the interaction terms before I ran the analysis (Aiken et al., 1991). 59 Table 1. Means, SDs, reliabilities, and correlations 1 2 3 4 5 6 7 8 9 10 11 (.77) .47* .44* .32* -.16 .09 -.13* -.20* .18* .01 -.09* -.02 -.18* -.07* -.06* -.03 .14* (.81) .43* (.69) .39* -.27* -.24* -.17* (.71) -.25* .12 .10* -.30* -.04 -.18* -.14* -.11* -.13* .10* .21* -.18* .37* .07* -.12* -.05 .16* .04 -.11* -.07* .11* .09* -.06 -.07* -.07* .21* (.83) .20* -.02 .02 -.43* -.12* -.22* -.11* -.02 .01 .09* .13* .01 .11* .05 .00 .08* .13* .09* .20* .12* .15* .12* -.12* .47 .64* .72 -.07* .70 -.20* .84 -.17* .36 .25* -.88* .05 .03 .07* .26* -.29* .55* .14* -.19* .08* -.10* -.10* .24* -.23* .05 .03 .18* -.26* -.01 .20* .09* -.28* .11* -.07* -.08* 60 Mean 42.13 .36 2.15 SD 6.65 .48 .09 .52* .21* 1.15 -.70* -.33* .22* -.61* 129.38 133.76 (.70) -.26* .05 .50 .05 -.54* (.78) .26* -.13 .61 -.36* -.11 .51* -.54* .17 -.10 .58 -.11 -.06 .46* -.52* .44* -.14 .74 -.28* -.23* .23* .09 -.39* .00 .61 -.04 -.08 .34* -.21* -.22* .23* .98 -.24* -.15 -.20* .04 .58 .09 -.29* .54 .16* -.64* 8.69 .49 .34* -.31* .99 -.62* -.31* 4.03 2.21 2.24 1.68 1.54 2.70 3.90 1.45 37.44 .62 1.70 .12 .44* -.37* .45* .25* .45* -.13 .12 .17 .25* .52* .20* .16* -.21* .14* -.22* .29* .05 .00 .20* -.03 .06* -.09* .85* -.53* -.23* -.35* .01 -.20* .36* -.07* .05 57.48 68.09 .33* .01 .63 -.07* -.01 .05 .11* .73 4.05 2.82 .68 3.95 3.56 3.27 1.29 1. Leader Age 2. Leader Gender 3. Leader Education 4. Leader Tenure dept 5. Leader Agreeableness 6. Leader Neuroticism 7. Leader Aggressiveness 8. Leader Machiavellianism 9. Leader Psychopathy 10. Leader Narcissism 11. Leader Ethical climate 12. Leader Abusive supervision 13. Follower Age 14. Follower Gender 15. Follower Education 16. Follower Tenure dyadic 17. Follower Interactional justice 18. Follower Power distance orientation 19. Follower Job type 20. Follower Job performance 21. Follower OCB 22. Follower Creativity 23. Follower Deviance Table 1. (cont’d.) 12. Follower Abusive supervision 13. Follower Age 14. Follower Gender 15. Follower Education 16. Follower Tenure dyadic 17. Follower Interactional justice 18. Follower Power distance orientation 19. Follower Job type 20. Follower Job performance 21. Follower OCB 22. Follower Creativity 23. Follower Deviance 12 (.87) .19* -.04 -.18* .16* -.19* 13 14 15 .17* -.67* -.35* .44* -.05 .02 -.01 -.36* .06 17 16 -.10* (.90) .15* .20* .03 -.21* .14* -.12* (.77) 20 .19* -.20* -.13* -.04 .22* .36* -.05 -.01 .32* -.92* .67* .05 -.03 -.20* -.04 -.21* -.04 .12* -.05 .03 .20* -.08* .16* -.13* -.08* .24* .10* -.09* -.02 .10* -.14* -.19* .12* -.14* -.15* (.85) .56* (.92) .40* .60* (.89) .08* -.17* .09* .09* -.47* -.38* -.18* (.85) 18 19 21 22 23 Note. Gender: female = 1, male = 0. Job type: white-collar = 0, blue-collar = 1. The sample size for variable 1-11 is 136, the sample size for variable 12-23 is 1009. 61 Table 2. Regression analyses predicting abusive supervision Model 1 Model 2 Control variables Age Gender Education Tenure Main effects Agreeableness Neuroticism Aggressiveness Machiavellianism Psychopathy Narcissism Ethical climate (EC) Interaction effects Agreeableness*EC Neuroticism*EC Aggressiveness*EC Machiavellianism*EC Psychopathy*EC Narcissism*EC F R2 △R2 Note. Standardized coefficients are reported; n = 136. * p < .05 .00 -.02 -.04 .41* .06 -.02 -.21 .32* -.39* .04 -.15 .04 .06 .03 -.15 7.57* .40 .21* 7.70* .19 Model 3 Model 4 .08 .02 -.20 .33* -.34* .10 -.15 .07 .04 -.03 -.20* .07 .07 -.19 .10 .05 .19* 6.00* .46 .06* .06 .01 -.22 .33 -.35* .07 -.14 .07 .05 -.04 -.19* .21* 7.95* .44 .04* Results indicated in Table 2 partially supported Hypothesis 1. Leader agreeableness was the only personality predictor that was significantly associated with abusive supervision (b = -.39, p < .05). In other words, less agreeable leaders were more likely to be perceived as abusive supervisors. All other personality predictors were not significantly associated with abusive supervision. Results in Table 2 failed to support Hypothesis 2. As indicated in Model 3 in Table 2, ethical climate only moderated the relationship between narcissism and abusive supervision (b 62 = .19, p < .05). Because all other interaction terms were not significant, I ran a reduced model that only included the interaction term between leader narcissism and ethical climate. I present the results in Model 4. Consistent with Model 3, in Model 4 the interaction effect between narcissism and ethical climate was significant (b = .21, p < .05). To better understand the moderation effect, I plotted the interaction relationship between narcissism and ethical climate in Figure 2 following the approach outlined by Aiken et al. (1991). Specifically, I drew the two regression lines for the relationship between narcissism and abusive supervision at two levels of ethical climate, namely +1 SD above the mean of ethical climate and -1 SD below the mean of ethical climate. I also reported the simple slopes for these two regression lines. The results showed that when leaders perceived high ethical climates, the relationship between narcissism and abusive supervision was positive (b = .06, p < .05). Whereas when leaders perceived low ethical climates, the relationship between narcissism and abusive supervision was negative (b = -.08, p < .05). Because the interaction pattern was contrary to Hypothesis 2, Hypothesis 2 was not supported. Figure 2. The relationship between leader narcissism and abusive supervision with ethical climate as a moderator 63 5.3 Outcomes to Abusive Supervision Because employees were nested within teams, data at the individual level may not be independent. As a result, using HLM is more appropriate than OLS in this situation. To test the necessity of using HLM, I first analyzed the ICC for interactional justice, job performance, OCB, creativity, and CWB. Results indicated that the ICC for interactional justice was .25, the ICC for job performance was .53, the ICC for OCB was .73, the ICC for creativity was .65, and the ICC for deviance was .60. These ICC values were considered large (Bliese, 2000; James, 1982), indicating that the variances were significant at both the individual and the team level. Therefore, I decided to use HLM that considers the data dependence issue at the employee level (Raudenbush & Bryk, 2002). As a result, all the following analyses were based on HLM. 5.3.1 Moderation Analysis for Power Distance Orientation and Job Type Hypotheses 3 to 7 stated that power distance orientation moderated the relationship between abusive supervision and interactional justice, job performance, OCB, creativity, and deviance, such that these relationships were stronger for employees with low power distance orientations. Hypotheses 8 to 12 stated that job type moderated the relationship between abusive supervision and interactional justice, job performance, OCB, creativity, and deviance, such that these relationships were stronger for white-collar employees than blue-collar employees. To test the above hypotheses, I used HLM 7.03 to conduct the corresponding analyses. Specifically, I tested five models separately with interactional justice, job performance, OCB, creativity, and deviance as outcomes, and power distance orientation and job type as predictors simultaneously for each model. I tested the two moderators simultaneously because I was interested in their relative importance in impacting how abusive supervision was associated with each outcome. Because abusive supervision and the two moderators were all at the employee level, I group- 64 mean centered them and created the interaction terms before I conducted the analysis (Enders & Tofighi, 2007). Table 3. HLM analyses for power distance orientation and job nature as moderators Interactional Job OCB Creativity Deviance justice 3.87** performance 3.97** 3.56** 3.24** 1.13** .00 -.01 .02 .00 .13** .01 .09 -.02 -.11* .00 -.05 .08 .00 -.02 .00 -.01 .05 .00 -.09 .00 -.03 .00 .00 -.20* .00 .00 .04 .00 -.64** Intercept Control variables Age (γ10) Gender (γ20) Education (γ30) Tenure (γ40) Main effects Abusive supervision (AS) (γ50) Power distance orientation (PDO) (γ60) Job nature (γ70) Interaction effects AS*PDO (γ80) AS*Job nature (γ90) Note. Unstandardized coefficients are reported; N employee = 1009, N team = 136. ** p < .01; * p < .05. -.01 -.02 .58** -.02 .15* .10 -.02 .00 .01 .05 -.20 -.02 .03 -.14 .05 -.05 Table 3 presents the results for the HLM analyses for power distance orientation and job type as moderators. Among Hypotheses 3 to 7, only Hypothesis 4 was supported. Namely, power distance orientation moderated the relationship between abusive supervision and job performance (γ10 = .15, p < .05). To visualize the interaction effect, I plotted a figure in Figure 3. Following the steps outlined in Aiken et al. (1991) and Preacher et al. (2006), I drew two regression lines for the relationship between abusive supervision and job performance at two levels of power distance orientation, +1 SD above the mean and -1 SD below the mean of power 65 distance orientation. The simple slope calculation indicated that for employees with low power distance orientation, abusive supervision was negatively associated with job performance (γ = -.28, p < .05). For employees with high power distance orientation, abusive supervision was not associated with job performance (γ = -.11, n.s.). Figure 3. The relationship between abusive supervision and job performance with power distance orientation as a moderator Among Hypotheses 8 to 12, Hypotheses 8 and 12 were supported. Consistent with Hypothesis 8, the results indicated that job type moderated the relationship between abusive supervision and interactional justice (γ10 = .58, p < .05). To visualize this interaction effect, I plotted a figure in Figure 4. Similar to Figure 3, I drew two regression lines for the relationship between abusive supervision and interactional justice for white-collar employees and blue-collar employees separately. The simple slope calculation indicated that for white-collar employees, abusive supervision was negatively associated with interactional justice (γ = -.64, p < .05). For 66 blue-collar employees, abusive supervision was not associated with interactional justice (γ = -.06, n.s.). Figure 4. The relationship between abusive supervision and interactional justice with job type as a moderator Consistent with Hypothesis 12, the results showed that job type moderated the relationship between abusive supervision and deviance (γ10 = -.11, p < .05). To visualize this interaction effect, I plotted a figure in Figure 5. I drew two regression lines for the relationship between abusive supervision and deviance for white-collar employees and blue-collar employees separately. The simple slope analysis indicated that for white-collar employees, abusive supervision was positively associated with deviance (γ = .13, p < .05). For blue-collar employees, abusive supervision was not associated with deviance (γ = .02, n.s.). 67 Figure 5. The relationship between abusive supervision and deviance with job type as a moderator 5.3.2 Multilevel Confirmatory Factor Analysis Before I ran the path analysis model, I ran multilevel confirmatory factor analyses for my key variables, including abusive supervision, power distance orientation, interactional justice, job performance, OCB, creativity, and deviance (Mehta & Neale, 2005). In the first model, I included all the items used to measure these variables and modeled all the items to load on their corresponding latent constructs. To make sure the multilevel CFA model could converge, I used parcels to reduce the number of free parameters in the model (Little, Rhemtulla, Gibson, & Schoemann, 2013). The model showed an acceptable fit to the data: χ2(982) = 1562.665; CFI = 0.952; RMSEA = .024; SRMR (within) = .033, and SRMR (between) = .076, and all loadings were significant (p < .05). I compared this model with another model, in which I combined all the attitudinal constructs including abusive supervision, power distance orientation, and interactional justice as an overall attitudinal construct, as well as combined all the behavioral 68 variables including job performance, OCB, creativity, and deviance as an overall behavioral construct. The alternative model provided a poor fit to the data: χ2(1052) = 6932.828; CFI = .511; RMSEA = .478; SRMR (within) = .097 and SRMR (between) = .227. Thus, compared with the first model, the second model was significantly worse (Δ χ2 (Δdf = 70) = 5370.163, p > .05). These findings further supported the discriminant validity of the key variables in my model. 5.3.3 Multilevel Moderated Mediation Model Hypothesis 13 stated that interactional justice mediated the relationship between abusive supervision and employee outcomes. Hypothesis 14 and 15 stated that power distance orientation and job type moderated the above mediation effects, such that the mediation effects would be different at different levels of the moderators. To test the mediation and moderated mediation effects, I ran a path analysis model to test all these relationships simultaneously. Specifically, in the model abusive supervision was the predictor, interactional justice was the mediator, behaviors including job performance, creativity, OCB, and deviance were outcomes, power distance orientation, and job type were moderators. In addition, I also included age, gender, education, and dyadic tenure as control variables in this model. I conducted the analysis in Mplus 8.4. Figure 6 presents the structural model with paths among these variables. This is a full mediation model in which interactional justice carries the effects from abusive supervision and passes such effects to employee outcomes. Overall, the fit of this model was acceptable. χ2 (df = 20, n = 957) = 32.632, p < .05, CFI = .98, TLI = .94, RMSEA = .026, SRMRwithin = .047, SRMRbetween = .020. As indicated by this model, most of the paths were significant, except for the paths that were depicted using dashed lines. Figure 6 indicates that abusive supervision was 69 associated with interactional justice (β = -.64, p < .01). Interactional justice was associated with all behavioral outcomes, including job performance (β = .09, p < .01), OCB (β = .07, p < .01), creativity (β = .09, p < .01), and deviance (β = -.06, p < .01). In addition, job type moderated the relationship between abusive supervision and interactional justice (β = .58, p < .01). Unexpectedly, power distance orientation did not moderate this relationship as stated in Hypothesis 14 (β = -.02, n.s.). Power distance orientation -.02 Job performance -.02 .09** .09** Creativity Abusive supervision -.64** Interactional justice .58** -.01 Job type .07** OCB -.06** Deviance Note. N = 956. Model fit: χ2 (df = 20, n = 957) = 32.632, p < .05, CFI = .98, TLI = .94, RMSEA = .026, SRMRwithin = .047, SRMRbetween = .020. Gender, age, education, and dyadic tenure are modeled as control variables. Figure 6. A moderated mediation model In sum, Hypothesis 13 received full support. Table 4 presents the mediation results. As indicated in Table 4, the indirect effects from abusive supervision to job performance (indirect effect = -.06, p < .05), creativity (indirect effect = -.05, p < .05), OCB (indirect effect = -.05, p < .05), and deviance (indirect effect = .04, p < .05) via interactional justice were all significant. Therefore, interactional justice fully mediated the relationship between abusive supervision and all these behavioral outcomes. 70 Table 4. Indirect effects from abusive supervision to outcomes through interactional justice Dependent variable Job performance Creativity OCB Deviance Indirect effect -.06* -.05* -.05** .04** Hypothesis 14 was not supported, indicating that the mediation effects from abusive supervision to behavioral outcomes via interactional justice did not vary significantly among employees with different power distance orientations. In other words, the differences in the indirect effects were minimal for employees with different levels of power distance orientation. Table 5 presents the results associated with Hypothesis 14. 71 Table 5. Moderated indirect effects from abusive supervision to outcomes through interactional justice by power distance orientation Dependent variable Job performance Creativity OCB Deviance Job Type Low PD Orientation High PD Orientation Difference Low PD Orientation High PD Orientation Difference Low PD Orientation High PD Orientation Difference Low PD Orientation High PD Orientation Difference Indirect effect -.06* -.06** .00 -.05* -.06* .00 -.04** -.05* .00 .04** .04* .00 Hypothesis 15 received full support. As expected, the mediation effects from abusive supervision to behavioral outcomes via interactional justice were different for white-collar employees and blue-collar employees. Table 6 presents the results associated with Hypothesis 15. Specifically, the indirect effect from abusive supervision to job performance via interactional justice was -.06 (p < .05) for white-collar employees, whereas such indirect effect was not significant (indirect effect = -.01, n.s.) for blue-collar employees. This pattern also applied to other behavioral outcomes. The indirect effect from abusive supervision to creativity via interactional justice was -.05 (p < .05) for white-collar employees, whereas such indirect effect was not significant (indirect effect = -.01, n.s.) for blue-collar employees. The indirect effect from abusive supervision to creativity via interactional justice was -.05 (p < .05) for white-collar employees, whereas such indirect effect was not significant (indirect effect = .00, n.s.) for blue- collar employees. The indirect effect from abusive supervision to deviance via interactional justice was .04 (p < .05) for white-collar employees, whereas such indirect effect was not 72 significant (indirect effect = .00, n.s.) for blue-collar employees. In sum, these results indicated that interactional justice functioned as a mediator that explained the underlying mechanism of why abusive supervision had an influence on outcome variables only for white-collar employees, but not for blue-collar employees. In other words, for white-collar employees, abusive supervision influenced interactional justice which further influenced employee behaviors. However, this was not the case for blue-collar employees. For blue-collar employees, interactional justice did not function as a mediator that explained the effects of abusive supervision on outcomes. As a supplementary analysis. I also conducted analyses with interpersonal justice rather than interactional justice. The results with interpersonal analyses were similar to the analyses using interactional justice. Table 6. Moderated indirect effects from abusive supervision to outcomes through interactional justice by job type Dependent variable Job performance Creativity OCB Deviance Job Type White-collar workers Blue-collar workers Difference White-collar workers Blue-collar workers Difference White-collar workers Blue-collar workers Difference White-collar workers Blue-collar workers Difference Indirect effect -.06* -.01 .05* -.05* -.01 .05* -.05** .00 .04** .04** .00 -.03** 73 CHAPTER 6 DISCUSSION Overall, the results provide support for the proposed model and several of the hypotheses. The results indicate that leader personality is an important predictor of abusive supervision and that contextual factors moderate the relationship between leader personality and abusive supervision. Regarding the relationship between abusive supervision and employee outcomes, this study supports that interactional justice is an important mediator. In addition, the findings show that there are different indirect effects between white-collar and blue-collar employees, highlighting the importance of considering job type as a key boundary condition in future studies. In contrast to prior research, this study does not provide strong evidence for the moderating role of power distance orientation on the relationship between abusive supervision and employee outcomes. 6.1 The Role of Leader Personality in Abusive Supervision The results provide support for the role of leader personality traits as predictors of abusive supervision. As indicated in Table 1, among the six examined personality variables, only agreeableness (r = -.37, p < .05) and psychopathy (r = .20, p < .05) are correlated with abusive supervision. Based on standards on the magnitude of effect size (Cohen, Cohen, Aiken, & West, 1983), leader agreeableness has a moderate to strong association with abusive supervision, whereas psychopathy has a weak to moderate association with abusive supervision. Unlike Mackey et al. (2017) who found weak to moderate associations between subordinate personality traits and abusive supervision, this study indicates that certain leader personality traits (e.g., agreeableness) are possibly stronger predictors of abusive supervision. In the past, researchers have focused more on the extent to which subordinate personality traits color employee perceptions of abusive supervision (e.g., Brees et al., 2016; Tepper, 2007) and have largely 74 ignored the role of leader personality traits. In addition, the relationship between leader agreeableness and abusive supervision found in this study is generally higher than the relationship between personality traits (e.g., the big five) and other leadership behaviors such as transformational leadership and transactional leadership (e.g., Bono & Judge, 2004). The results of this study highlight the significant role of leader personality traits in abusive supervision behaviors. In addition to the correlation results, the regression analysis results further support the important role of agreeableness, which is a broadly defined personality trait. In the regression analysis, agreeableness was the only significant predictor among leader personality traits that were included. This study indicates that a less agreeable supervisor is more likely to be an abusive supervisor. This specific finding provides some evidence about the relative importance of broadly defined and narrowly defined personality variables, indicating that the broadly defined personality variables may have better predictive validity than narrowly defined personality traits in explaining abusive supervision. However, one study is not conclusive. In this study, I only examined agreeableness and neuroticism as broadly defined personality traits and trait aggressiveness and the dark triad as narrowly defined personality traits. Based on my review of the literature, no studies have compared the predictive validity of narrowly versus broadly defined leader personality variables in the context of abusive supervision. Most often, researchers only used either broadly or narrowly defined personality variables as predictors in their studies (e.g., Waldman et al., 2018; Wang et al., 2015). The idea of studying narrowly defined personality variables assumes that such variables may capture specific aspects of personality and thus serve as stronger predictors than widely defined variables in some cases (Hough et al., 2015). Therefore, I suggest additional 75 empirical research examine the relationship between leader personality and abusive supervision and compare the relative importance of broadly defined and narrowly defined personality traits with different operationalizations of personality traits. This study also sheds light on the role of leader narcissism. Leadership researchers in the past have paid some attention to narcissism, finding that it is an important antecedent of leadership behaviors (e.g., Grijalva & Harms, 2014; Judge, LePine, & Rich, 2006), including abusive supervision (Waldman et al., 2018). Based on trait activation theory, I hypothesized that a low ethical climate serves as an activator, whereas a high ethical climate serves as a buffer between leader narcissism and abusive supervision. Consistent with prior findings, this study also supports the importance of narcissism. Although narcissism was not correlated with abusive supervision, it interacted with ethical climate in predicting abusive supervision. Contrary to my hypothesis, the moderation results indicated that in a high organizational caring ethical climate, the relationship between narcissism and abusive supervision was positive, whereas in a low organizational caring ethical climate, the relationship between narcissism and abusive supervision was negative. These results, although unexpected, could be explained by Kristof’s (1996) idea of fit between leader personality and organizational ethical climate. Kristof (1996) defined the construct of person-organization fit and highlighted the importance of the supplementary fit between employee personality and organizational climate. It is possible that narcissistic leaders fit better in less caring ethical climates because of their tendency to behave unethically and overemphasis on their own interests (Blair, Hoffman, & Helland, 2008). In a low ethical context, followers perceive narcissistic leaders as less abusive because leaders’ behaviors fit better with the organizational climate. In comparison, in a high ethical context, followers perceive narcissistic leaders as more abusive because leaders’ narcissistic behaviors are more 76 salient to them as not caring. Interestingly, the results of this study are similar to the results in Hoffman, Strang, Kuhnert, Campbell, Kennedy, and LoPilato (2013). With ethical leadership as an outcome, Hoffman et al. (2013) found that leader narcissism was negatively related to ethical leadership in a high ethical context, and such a relationship was not significant in a low ethical context. Together, this study and Hoffman et al. (2013) support the fit perceptive of narcissism. Namely, narcissist leaders fit better in a less ethical context and are generally rated more negatively in a high ethical context. Another contribution of this study is examining trait aggressiveness as a predictor in the model. Research has supported that trait aggressiveness is an important antecedent to aggressive behaviors and has moderate to strong relationships with aggressive behaviors in general (Bettencourt et al., 2006). However, in the abusive supervision literature, trait aggressiveness is seldom studied as a predictor. This study filled the gap by treating leader trait aggressiveness as a predictor. The results indicated that trait aggressiveness was not correlated with abusive supervision nor it was a significant predictor in the regression model. This may indicate that the content domain of abusive supervision is significantly different from that of aggressive behaviors. Although researchers have conceptualized aggressive behaviors as both physical and verbal, a majority of the studies on aggressive behaviors have focused on physically aggressive behaviors (Bettencourt et al., 2006). Therefore, it is understandable that trait aggressiveness is a strong predictor for physical aggressive behaviors but not a strong predictor for non-physical hostile behaviors in the workplace, namely abusive supervision behaviors. 6.2 Ethical Climate and Trait Activation Theory In this study, I integrate the trait activation theory with the abusive supervision literature (Tett & Guterman, 2000) and propose that ethical climate moderates the relationship between 77 leader personality and abusive supervision. Although the hypothesis related to trait activation theory was not supported, results from this study still provided support for the relevance of situational factors in the abusive supervision context. Overall, little empirical attention has been devoted to understanding the boundary conditions of leader personality on abusive supervision from the perspectives of situational factors. Some studies examined the boundary conditions from the perspectives of individual differences, such as political skills (Waldman et al., 2018). One exception is a study conducted by Wisse and Sleebos (2016) who found that the perceived position power strengthened the relationship between leader Machiavellianism and abusive supervision. In the future, I encourage researchers to devote more attention to examining theory- based situational factors to provide a more thorough understanding of the relationship between leader/follower personality and abusive supervision. For example, researchers in the future could examine other dimensions of ethical climate including law and code, rules, instrumental, and independence (Victor & Cullen, 1988). The caring dimension of ethical climate primarily captures the interpersonal aspect of the ethical climate, and it is possible that other dimensions function differently from the caring dimension. In addition, as a related construct, ethical culture could also serve as a moderator based on the trait activation theory. According to Treviño, Butterfield, and McCabe (1998), “the term "climate" suggests meteorological climate and qualities such as temperature, humidity, precipitation, wind, and other atmospheric conditions that can affect individuals (e.g., feelings),” whereas “the notion of "culture" evokes notions of rules, codes, rewards, leadership, rituals, and stories-sensemaking devices that more explicitly guide and shape behavior.” It is possible that ethical culture has a stronger influence on the relationship between leader personality and abusive supervision than ethical climate because of ethical culture’s stronger emphasis on rules and codes rather than atmospheric conditions. 78 6.3 The Relative Importance of Power Distance Orientation and Job Type Different from the several published studies on power distance orientation, this study does not provide much support for power distance orientation as a moderator. In this study, power distance orientation was found to be only a significant moderator for the relationship between abusive supervision and job performance, but not for other outcomes such as interactional justice, OCB, creativity, and deviance. Nor was it a significant moderator in the moderated mediation model. This is unexpected, especially since past research tended to support that power distance orientation moderates the relationship between abusive supervision and outcomes such as interactional justices and deviance. For example, research in the past has found that for employees with low power distance orientation, their interactional justice perception is more strongly influenced by abusive supervision (Lian et al., 2012; Vogel et al., 2015; Wang et al., 2012). Findings in this study did not support power distance orientation as a moderator for the relationship between abusive supervision and deviance, providing different results from prior research (e.g., Lian et al., 2012). However, as a moderator job type outperformed power distance orientation in this study. Results indicate that job type strongly influences employees’ interactional justice and deviant behaviors when they experience abusive supervision. White-collar employees tend to react more strongly than blue-collar employees. When white-collar employees experience abusive supervision, they are more likely to have decreased interactional justice perceptions and more deviant behaviors. The moderated mediation model reveals that the mediation mechanisms for white-collar and blue-collar employees are different as well. Interactional justice is found only to mediate the relationship between abusive supervision and outcomes for white-collar employees but not for blue-collar employees. 79 The purpose of including power distance orientation and job type as moderators and testing them simultaneously is to test their relative importance. In the past, researchers typically have studied power distance orientation by itself, assuming that power distance orientation serves as a boundary condition and makes a difference regarding how employees respond to abusive supervision (e.g., Lian et al., 2012; Lin et al., 2013). However, in the workplace setting, how employees respond to their leaders is not solely determined by personal values. More likely, it is also influenced by other contextual factors, such as what type of job they have. In sum, future research should devote more attention to exploring whether power distance orientation and other cultural values are important boundary conditions. The current study contributed to the mixed findings in this area. 6.4 Generalizability Issues of Current Research with a Focus on White-Collar Employees Similar to other areas in the organizational behavior field, the abusive supervision literature has heavily relied on white-collar employees as the samples. The underlying assumption appears to be that the job type does not matter, and results obtained from white-collar employees are generalizable to other job types. However, the results of this study indicate that there are significant differences regarding how white-collar and blue-collar employees respond to abusive supervision. The results indicate that the extent to which abusive supervision influences interactional justice perceptions is different for white-collar and blue-collar employees. White-collar employees respond more intensively to abusive supervision than blue- collar employees in terms of interactional justice perceptions and deviant behaviors. Also, interactional justice only mediates the relationship between abusive supervision and employee outcomes for white-collar employees but not blue-collar employees. This reveals that job type is possibly an important boundary condition for abusive supervision. In the past, the role of job 80 type has been ignored, even though some studies have included blue-collar employees in their samples, their purpose was to increase the generalizability of their results rather than specifically examining differences between the groups (Bamberger & Bacharach, 2006; Harr et al., 2016; Lin et al., 2013; Kluemper et al., 2019). The results from this study indicate that the role of job type is possibly more important in influencing employee reactions to abusive supervision than researchers have considered. In addition, this study also challenges the generalizability of the current findings to non- white collar employees. For example, Tepper (2000) specifically highlighted the role of interactional justice and proposed a justice-based model for abusive supervision. Researchers have proposed and supported that interactional justice is an important mediator that explains why abusive supervision impacts employees (e.g. Aryee et al., 2007; Rafferty & Restubog, 2011). However, this study demonstrates that interactional justice is not a mediator for blue-collar employees. Therefore, I would suggest that future research examine the role of job type to test whether it is a moderator for the relationship between abusive supervision and other important constructs. 6.5 Strengths, Limitations, and Future Directions There are several strengths of this study. First, because I believe abusive supervision is a complex phenomenon, I used a complex model to study its antecedents and outcomes. At the same time, I integrated multiple theories into abusive supervision literature. For antecedents, I incorporated personality literature and the trait activation theory (Tett & Burnett, 2003) to the abusive supervision literature, and thereby found support for leader personality as a key predictor. This dissertation highlights the important role of leader personality, which represents an understudied group of variables in the abusive supervision. In the future, researchers could 81 continue examining the role of leader personality to better understand why personality traits have an impact on leaders’ abusive supervision behaviors. Regarding outcomes, I tested job type that represents a traditionally ignored boundary condition, in the relationship between abusive supervision and employees. The significant results of job type challenge the current dominant focus on white-collar employees as well as the generalizability of the current findings to other job types. At the same time, this study also tested the relative importance of power distance orientation. Contrary to prior research findings (e.g., Lian et al., 2012), this study indicates that power distance orientation is not as important as prior studies have indicated. Future research should continue to examine the role of power distance orientation with more contextual factors considered at the same time. Second, in this study, I was able to use appropriate statistical models to test the corresponding hypotheses. Especially for the relationship between abusive supervision and outcomes, I hypothesized that two moderators and one mediator influence the above-mentioned relationships. Specifically, I hypothesized that interactional justice serves as the mediator, and such mediated relationships are different for employees with different power distance orientation and different job types. In other words, this model also answers the questions of whether interactional justice serves as the mediator that could explain why abusive supervision interacts with power distance and job type to explain the employee outcomes. To test these effects simultaneously in a multilevel situation, I used a multilevel moderated mediation model to test these moderated mediation effects to examine how abusive supervision influences employee outcomes. By using this model, I was able to test all the related effects simultaneously rather than using a piece-meal approach. This model considers mutual influences among these variables and therefore provides more reliable results. 82 Third, I used a multiple-wave multilevel-source design to collect the data. I collected data in three waves with a month time interval, and two data sources, from both supervisors and subordinates. This data collection procedure helps mitigate the influence of common method bias (Podsakoff, MacKenzie, & Podsakoff, 2012). In addition, to guarantee the quality of the data, I also used all validated scales to measure the constructs and the obtained reliabilities are all acceptable. However, this dissertation is not free of limitations. Similar to all studies using cross- sectional data, this study cannot draw conclusions on the casual relationships among variables. Although in this study, I used a multiple-wave design to collect data, with the predictor, mediator, and outcomes collected sequentially, I could not confidently establish casual directions by ruling out other possibilities. In the future, researchers could use other approaches, such as experiments, to draw conclusions based on casual relationships. One limitation of this study is dichotomizing all jobs into white-collar vs. blue-collar employees. Research on job classification has indicated that although most people generally understand the distinction between “white-collar” and “blue-collar” and use these terms in their daily lives, the distinction is not so clear especially when it comes to “low-level” white-collar jobs (DeVault, 1990). “Low-level” white-collar employees, such as clerical and sales workers, are wage employees that have little personal control and may work in similar employment situations as employees traditionally labeled as blue collar. For the blue-collar employees in this study, they are all from a textile manufacturing company and work in different departments that are responsible for the different steps of producing textile products. They typically have six to nine years of education and work on irregular shifts with very low wages doing repetitive tasks. In contrast, in this study, white-collar employees include urban planning/designing employees 83 and employees working in key professional functional departments in companies such as human recourses and finance. All these employees have very decent salaries and high education, usually a bachelor’s degree or master’s degree. The distinction in my sample is very clear in terms of white-collar vs. blue-collar, or professional vs. non-professional workers. This sample provides a strong basis for revealing the potential differences among these two major job categories. In the future, researchers should consider examining the influence of job types in more refined categories and provide more detailed results. Another limitation of this study is the use of samples from a single country. Because power distance orientation is a key moderator in my model, ideally, I should collect data from multiple cultures. Collecting data from a single country, namely China in my study, may raise the concern that scores for power distance orientation have range restriction. Range restriction is a phenomenon where the variance of a variable in a particular sample is smaller than the variance of the variable in the large population (Cohen et al., 1983). Range restricted data may shrink the obtained effect size in the analysis. However, when I compared the power distance orientation data from this study to other studies that have used samples from multiple countries, I found that the standard deviation in my sample (SD = .73) is very similar to the standard deviations from those cross-cultural samples (e.g. SD = .70, Cavazotte, Hartman, & Bahiense, 2014; SD = .77, Vogel et al., 2015). This may indicate that the power distance orientation is not range-restricted in my sample. In the context of globalization, researchers have noted the convergence of cultural values among different countries (Sarala & Vaara, 2010; Ralston, Holt, Terpstra, & Kai-Cheng, 1997). For example, the GLOBE project has found that the power distance orientations of people from different cultures are quite similar to each other (GLOBE, 2004). With a 7-point Likert Scale, the 84 overall mean value across different cultures was 2.75. Specifically, the mean value in China was 3.1, the mean value in Thailand was 2.86, the mean value in the U.S. was 2.85, the mean value in France was 2.76. This indicates that the cultural differences are not as large as previously thought. Therefore, it is less likely that data from a single country in this study is associated with severe range restriction issues. On the other hand, I also suggest future research to study abusive supervision with data from multiple cultures to obtain a better understanding of the role of cultural values such as power distance orientation. 6.6 Conclusions This dissertation aims to understand the leader personality traits as antecedents and the mechanisms regarding how abusive supervision influences employee outcomes. The findings supported that leader personality, especially leader agreeableness is an important predictor. Leader narcissism also has an impact on abusive supervision, and its effects depend on the ethical climate perception. In addition, the results indicated that job type is an important boundary condition for the abusive supervision-employee outcome relationships. Interactional justice only mediates the abusive supervision-employee outcome relationships for white-collar employees but not blue-collar employees. These results challenge the generalizability of prior studies in the abusive supervision literature that has relied on white-collar employees as samples. Moreover, the results indicated that power distance orientation is not as important as prior research suggests, especially compared with other moderators such as job type examined in this study. This is unexpected, but it may suggest that in the workplace, how employees respond to abusive supervision is largely influenced by other factors and cultural values play a less significant role. 85 APPENDIX 86 APPENDIX. Measures and Survey Items Big five (John, Donahue, & Kentle, 1991) Agreeableness Is considerate and kind to almost everyone Likes to cooperate with others Is helpful and unselfish with others Has a forgiving nature Is generally trusting Tends to find fault with others (RC) Starts quarrels with others (RC) Can be cold and aloof (RC) Is sometimes rode to others (RC) Neuroticism Worries a lot Can be tense Gets nervous easily Is depressed, blue Can be moody Remains calm in tense situations (RC) Is emotionally stable, not easily upset (RC) Is relaxed, handles stress well (RC) How much do you agree with the following statements? I am generally… 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 Trait aggressiveness (Bryant & Smith, 2001) 1 2 3 4 5 6 7 8 9 10 11 12 Dark triad (Jonason & Webster, 2010) 1 2 3 4 5 How much do you agree with the following statements? Given enough provocation, I may hit another person. There are people who pushed me so far that we came to blows. I have threatened people I know. I often find myself disagreeing with people. I can't help getting into arguments when people disagree with me. My friends say that I'm somewhat argumentative. I flare up quickly but get over it quickly. Sometimes I fly off the handle for no good reason. I have trouble controlling my temper. At times I feel I have gotten a raw deal out of life. Other people always seem to get the breaks. I wonder why sometimes I feel so bitter about things. How much do you agree with the following statements? I tend to manipulate others to get my way. I have used deceit or lied to get my way. I have use flattery to get my way. I tend to exploit others towards my own end. I tend to lack remorse. 87 The most efficient way is always the right way in this company. In this company, each person is expected above all to work efficiently. I tend to not be too concerned with morality or the morality of my actions. I tend to be callous or insensitive. I tend to be cynical. I tend to want others to admire me. I tend to want others to pay attention to me. I tend to seek prestige or status. I tend to expect special favors from others. 6 7 8 9 10 11 12 1 2 3 4 5 public. 6 7 1 2 3 4 5 6 7 8 9 10 11 Makes negative comments about me to others 12 13 14 15 How much do you agree with the following statements? 1 subordinates 2 3 is done efficiently. 4 objective in dealing with subordinates. 5 Abusive supervision (Tepper, 2000) How often does your direct leader treat you in the following ways: Ridicules me Tells me my thoughts or feelings are stupid Gives me the silent treatment Puts me down in front of others Invades my privacy Reminds me of my past mistakes and failures Doesn't give me credit for jobs requiring a lot of effort Blames me to save himself/herself embarrassment Breaks promises he/she makes Expresses anger at me when he/she is mad for another reason Is rude to me Does not allow me to interact with my coworkers Tells me I'm incompetent Lies to me Power distance orientation (Dorfman & Howell, 1988) 88 Ethical climate (Victor & Cullen, 1988) What is best for everyone in the company is the major consideration here. The most important concern is the good of all the people in the company as a whole. Our major concern is always what is best for the other person. In this company, people look out for each other's good. In this company, it is expected that you will always do what is right for the customers and It is frequently necessary for a manager to use authority and power when dealing with Employees should not disagree with management decisions A supervisor's use of authority and power is often necessary in order to assure that work Social interaction with one's subordinates may decrease a manager's ability to be Managers should make most decisions without consulting subordinates Managers should not delegate important tasks to employees. The following items refer to (the authority figure who enacted the procedure). To what Rate the overall level of performance that you observe for this subordinate What is your personal view of your subordinate in terms of his or her overall Overall, to what extent do you feel your subordinate has been effectively fulfilling his or Overall, to what extent do you feel the subordinate is performing his job the way you Has (he/she) treated you in a polite manner? Has (he/she) treated you with dignity? Has (he/she) treated you with respect? Has (he/she) refrained from improper remarks or comments? Has (he/she) been candid in (his/her) communications with you? Has (he/she) explained the procedures thoroughly? Were (his/her) explanations regarding the procedures reasonable? Has (he/she) communicated details in a timely manner? Has (he/she) seemed to tailor (his/her) communications to individuals' specific needs? 6 Interactional justice (Colquitt, 2001) extent: 1 2 3 4 5 6 7 8 9 Job performance (Liden et al., 1993) How much do you agree with the following statements? This subordinate is… 1 2 effectiveness? 3 her roles and responsibilities? 4 would like it to be performed? How much do you agree with the following statements? This subordinate is… 1 2 3 4 5 business or personal situations. 6 7 8 9 10 11 12 13 14 15 16 Give up time to help others who have work or nonwork problems. Assist others with their duties. Share personal property with others to help their work. Attend functions that are not required but that help the organizational image. Keep up with developments in the organization. Defend the organization when other employees criticize it. Show pride when representing the organization in public. Offer ideas to improve the functioning of the organization. Express loyalty toward the organization. Take action to protect the organization from potential problems. Demonstrate concern about the image of the organization. Help others who have been absent. Willingly give your time to help others who have work-related problems. Adjust your work schedule to accommodate other employees’ requests for time off. Go out of the way to make newer employees feel welcome in the work group. Show genuine concern and courtesy toward coworkers, even under the most trying OCB (Lee & Allen, 2002) 89 Creativity (Zhou & George, 2001) Suggests new ways to achieve goals or objectives. Comes up with new and practical ideas to improve performance. Promotes and champions ideas to others. Exhibits creativity on the job when given the opportunity to. Comes up with creative solutions to problems. CWB (Bennett & Robinson, 2000) 1 2 3 4 5 How much do you agree with the following statements? This subordinate is… 1 2 3 4 5 6 7 8 9 10 11 12 Made fun of someone at work Said something hurtful to someone at work Cursed at someone at work Played a mean prank on someone at work Acted rudely toward someone at work Publicly embarrassed someone at work Spent too much time fantasizing or daydreaming instead of working Taken an additional or longer break than is acceptable at your workplace Come in late to work without permission Neglected to follow your boss's instructions Intentionally worked slower than you could have worked Put little effort into your work 90 REFERENCES 91 REFERENCES Ahmed, P. K. (1998). Culture and climate for innovation. European Journal of Innovation Management, 1(1), 30-43. Aiken, L. S., West, S. G., & Reno, R. R. (1991). Multiple regression: Testing and interpreting interactions. Sage. Aquino, K., Galperin, B. L., & Bennett, R. J. (2004). Social status and aggressiveness as moderators of the relationship between interactional justice and workplace deviance. Journal of Applied Social Psychology, 34(5), 1001-1029. Aryee, S., Chen, Z. X., Sun, L. Y., & Debrah, Y. A. (2007). Antecedents and outcomes of abusive supervision: test of a trickle-down model. Journal of Applied Psychology, 92(1), 191-201. Ashforth, B. (1994). Petty tyranny in organizations. Human Relations, 47(7), 755-778. Ashforth, B. (1997). Petty tyranny in organizations: A preliminary examination of antecedents and consequences. Canadian Journal of Administrative Sciences/Revue Canadienne des Sciences de l'Administration, 14(2), 126-140. Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182. Barrick, M. R., & Mount, M. K. (1991). The big five personality dimensions and job performance: a meta‐analysis. Personnel Psychology, 44(1), 1-26. Bass, B. M. (1985). Leadership and performance beyond expectations. Collier Macmillan. Baumgartner, H., & Homburg, C. (1996). Applications of structural equation modeling in marketing and consumer research: A review. International Journal of Research in Marketing, 13(2), 139-161. Bamberger, P. A., & Bacharach, S. B. (2006). Abusive supervision and subordinate problem drinking: Taking resistance, stress and subordinate personality into account. Human Relations, 59(6), 723-752. Benet-Martínez, V., Leu, J., Lee, F., & Morris, M. W. (2002). Negotiating biculturalism: Cultural frame switching in biculturals with oppositional versus compatible cultural identities. Journal of Cross-Cultural Psychology, 33(5), 492-516. Bennett, R. J., & Robinson, S. L. (2000). Development of a measure of workplace deviance. Journal of Applied Psychology, 85(3), 349-360. 92 Bennett, R. J., & Robinson, S. L. (2003). The past, present, and future of workplace deviance research. In J. Greenberg (Ed.), Organizational behavior: The state of the science (pp. 247- 281). Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers. Bettencourt, B., Talley, A., Benjamin, A. J., & Valentine, J. (2006). Personality and aggressive behavior under provoking and neutral conditions: a meta-analytic review. Psychological Bulletin, 132(5), 751-777. Bies, R. J. (1989). Managing conflict before it happens: The role of accounts. In M. A. Rahim (Ed.), Managing conflict: An interdisciplinary approach (pp. 83-91). New York: Praeger Bies, R. J., & Moag, J. S. (1986). Interactional communication criteria of fairness. Research in Organizational Behavior, 9, 289-319. Blair, C. A., Hoffman, B. J., & Helland, K. R. (2008). Narcissism in organizations: A multisource appraisal reflects different perspectives. Human Performance, 21(3), 254-276. Bliese, P. D. (2000). Within-group agreement, non-independence, and reliability: Implications for data aggregation and analysis. In K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions (pp. 349-381). San Francisco, CA, US: Jossey-Bass. Bono, J. E., & Judge, T. A. (2004). Personality and transformational and transactional leadership: a meta-analysis. Journal of Applied Psychology, 89(5), 901-910. Brees, J. R. (2012). Relationship Between Subordinates' Individual Differences and Their Perceptions of Abusive Supervision. Unpublished dissertation, The Florida State University. Brees, J., Martinko, M., & Harvey, P. (2016). Abusive supervision: subordinate personality or supervisor behavior?. Journal of Managerial Psychology, 31(2), 405-419. Breevaart, K., & de Vries, R. E. (2017). Supervisor's HEXACO personality traits and subordinate perceptions of abusive supervision. The Leadership Quarterly, 28(5), 691-700. Brown, M. E., Treviño, L. K., & Harrison, D. A. (2005). Ethical leadership: A social learning perspective for construct development and testing. Organizational Behavior and Human Decision Processes, 97(2), 117-134. Bryant, F. B., & Smith, B. D. (2001). Refining the architecture of aggression: A measurement model for the Buss–Perry Aggression Questionnaire. Journal of Research in Personality, 35(2), 138-167. Buffardi, L. E., & Campbell, W. K. (2008). Narcissism and social networking web sites. Personality and Social Psychology Bulletin, 34(10), 1303-1314. Bushman, B. J., Baumeister, R. F., Thomaes, S., Ryu, E., Begeer, S., & West, S. G. (2009). Looking again, and harder, for a link between low self‐esteem and aggression. Journal of Personality, 77(2), 427-446. 93 Buss, A. H., & Perry, M. (1992). The aggression questionnaire. Journal of Personality and Social Psychology, 63(3), 452-459. Campbell, W. K. (1999). Narcissism and romantic attraction. Journal of Personality and Social Psychology, 77(6), 1254-1270. Camps, J., Stouten, J., & Euwema, M. (2016). The relation between supervisors’ big five personality traits and employees’ experiences of abusive supervision. Frontiers in Psychology, 7, 1-11. Carlson, D., Ferguson, M., Hunter, E., & Whitten, D. (2012). Abusive supervision and work– family conflict: The path through emotional labor and burnout. The Leadership Quarterly, 23(5), 849-859. Carlson, D. S., Ferguson, M., Perrewé, P. L., & Whitten, D. (2011). The fallout from abusive supervision: An examination of subordinates and their partners. Personnel Psychology, 64(4), 937-961. Cavazotte, F., Hartman, N. S., & Bahiense, E. (2014). Charismatic leadership, citizenship behaviors, and power distance orientation: Comparing Brazilian and US workers. Cross- Cultural Research, 48(1), 3-31. Centers, R., & Bugental, D. E. (1966). Intrinsic and extrinsic job motivations among different segments of the working population. Journal of Applied Psychology, 50(3), 193-197. Cho, J., & Dansereau, F. (2010). Are transformational leaders fair? A multi-level study of transformational leadership, justice perceptions, and organizational citizenship behaviors. The Leadership Quarterly, 21(3), 409-421. Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (1983). Applied multiple regression/ Correlation analysis for the behavioral sciences. Cohen-Charash, Y., & Spector, P. E. (2001). The role of justice in organizations: A meta- analysis. Organizational Behavior and Human Decision Processes, 86(2), 278-321. Colquitt, J. A. (2001). On the dimensionality of organizational justice: a construct validation of a measure. Journal of Applied Psychology, 86(3), 386-400. Colquitt, J. A., Scott, B. A., Rodell, J. B., Long, D. M., Zapata, C. P., Conlon, D. E., & Wesson, M. J. (2013). Justice at the millennium, a decade later: A meta-analytic test of social exchange and affect-based perspectives. Journal of Applied Psychology, 98(2), 199-236. Costa Jr, P. T., & McCrae, R. R. (1992). Four ways five factors are basic. Personality and Individual Differences, 13(6), 653-665. Costa, P. T., McCrae, R. R., & Dembroski, T. M. (1989). Agreeableness versus antagonism: Explication of a potential risk factor for CHD. In A. W. Siegman & T. M. Dembroski (Eds.), In search of coronary prone behavior (pp. 41–63). Hillsdale, NJ: Erlbaum. 94 Croon, M. A., & van Veldhoven, M. J. (2007). Predicting group-level outcome variables from variables measured at the individual level: a latent variable multilevel model. Psychological Methods, 12(1), 45-57. Cropanzano, R., Byrne, Z. S., Bobocel, D. R., & Rupp, D. E. (2001). Moral virtues, fairness heuristics, social entities, and other denizens of organizational justice. Journal of Vocational Behavior, 58(2), 164-209. Cropanzano, R., Prehar, C. A., & Chen, P. Y. (2002). Using social exchange theory to distinguish procedural from interactional justice. Group & Organization Management, 27(3), 324-351. Cropanzano, R., Rupp, D. E., Mohler, C. J., & Schminke, M. (2001). Three roads to organizational justice. In G. R. Ferris (Ed.), Research in personnel and human resources management: Vol. 20. Research in personnel and human resources management, Vol. 20 (p. 1–123). Elsevier Science/JAI Press. Cullen, J. B., Parboteeah, K. P., & Victor, B. (2003). The effects of ethical climates on organizational commitment: A two-study analysis. Journal of Business Ethics, 46(2), 127- 141. Da Costa, S., Páez, D., Sánchez, F., Garaigordobil, M., & Gondim, S. (2015). Personal factors of creativity: A second order meta-analysis. Revista de Psicología del Trabajo y de las Organizaciones, 31(3), 165-173. Daniels, M. A. (2015). Shame as an alternate mechanism for the abusive supervision- performance relation and the role of power distance values. Unpublished doctoral dissertation, Bowling Green State University. Daniels, M. A., & Greguras, G. J. (2014). Exploring the nature of power distance: Implications for micro-and macro-level theories, processes, and outcomes. Journal of Management, 40(5), 1202-1229. De Hoogh, A. H., Den Hartog, D. N., & Koopman, P. L. (2005). Linking the Big Five‐Factors of personality to charismatic and transactional leadership; perceived dynamic work environment as a moderator. Journal of Organizational Behavior, 26(7), 839-865. Den Hartog, D. N., House, R. J., Hanges, P. J., Ruiz-Quintanilla, S. A., Dorfman, P. W., Abdalla, I. A., ... & Akande, B. E. (1999). Culture specific and cross-culturally generalizable implicit leadership theories: are attributes of charismatic/transformational leadership universally endorsed?. The Leadership Quarterly, 10(2), 219-256. Daniels, M. A., & Greguras, G. J. (2014). Exploring the nature of power distance: Implications for micro-and macro-level theories, processes, and outcomes. Journal of Management, 40(5), 1202-1229. 95 Detert, J. R., Treviño, L. K., Burris, E. R., & Andiappan, M. (2007). Managerial modes of influence and counterproductivity in organizations: A longitudinal business-unit-level investigation. Journal of Applied Psychology, 92(4), 993-1005. DeVault, I. A. (1990). White Collar/Blue Collar. Dheer, R., Lenartowicz, T., Peterson, M. F., & Petrescu, M. (2014). Cultural regions of Canada and United States: Implications for international management research. International Journal of Cross Cultural Management, 14(3), 343-384. Dorfman, P. W., & Howell, J. P. (1988). Dimensions of national culture and effective leadership patterns: Hofstede revisited. Advances in International Comparative Management, 3(1), 127-150. Duffy, M. K., Ganster, D. C., & Pagon, M. (2002). Social undermining in the workplace. Academy of Management Journal, 45(2), 331-351. Dulebohn, J. H., Bommer, W. H., Liden, R. C., Brouer, R. L., & Ferris, G. R. (2012). A meta- analysis of antecedents and consequences of leader-member exchange: Integrating the past with an eye toward the future. Journal of Management, 38(6), 1715-1759. Dulebohn, J., Wu, D., Liao, C., & Hoch, J. E. (2017). Transformational Leadership and National Culture: A Meta-analysis across 36 Countries. In Academy of Management Proceedings (Vol. 2017, No. 1, p. 17512). Briarcliff Manor, NY 10510: Academy of Management. Eden, D. and Leviathan, U. (1975), “Implicit leadership theory as a determinant of the factor structure underlying supervisory behavior scales”, Journal of Applied Psychology, 60 (6), 736-741. Edwards, J. R., & Lambert, L. S. (2007). Methods for integrating moderation and mediation: a general analytical framework using moderated path analysis. Psychological Methods, 12(1), 1-22. Eissa, G., & Lester, S. W. (2017). Supervisor role overload and frustration as antecedents of abusive supervision: The moderating role of supervisor personality. Journal of Organizational Behavior, 38(3), 307-326. Enders, C. K., & Tofighi, D. (2007). Centering predictor variables in cross-sectional multilevel models: a new look at an old issue. Psychological Methods, 12(2), 121-138. Epitropaki, O., & Martin, R. (2004). Implicit leadership theories in applied settings: Factor structure, generalizability, and stability over time. Journal of Applied Psychology, 89(2), 293-310. Farmer, S. M., Tierney, P., & Kung-Mcintyre, K. (2003). Employee creativity in Taiwan: An application of role identity theory. Academy of management Journal, 46(5), 618-630. 96 Fischer, R. (2009). Where is culture in cross cultural research? An outline of a multilevel research process for measuring culture as a shared meaning system. International Journal of Cross Cultural Management, 9(1), 25-49. Fisk, G. M., & Friesen, J. P. (2012). Perceptions of leader emotion regulation and LMX as predictors of followers' job satisfaction and organizational citizenship behaviors. The Leadership Quarterly, 23(1), 1-12. Fleishman, E. A. (1995). Consideration and structure: Another look at their role in leadership research. In F. Dansereau & F. J. Yammarino (Eds.), Leadership: The multiple-level approaches (pp. 51–60). Stamford, CT: JAI Press. Foti, R. J., & Luch, C. H. (1992). The influence of individual differences on the perception and categorization of leaders. The Leadership Quarterly, 3(1), 55-66. Furnham, A., Richards, S. C., & Paulhus, D. L. (2013). The Dark Triad of personality: A 10 year review. Social and Personality Psychology Compass, 7(3), 199-216. Garcia, P. R. J. M., Restubog, S. L. D., Kiewitz, C., Scott, K. L., & Tang, R. L. (2014). Roots run deep: Investigating psychological mechanisms between history of family aggression and abusive supervision. Journal of Applied Psychology, 99(5), 883-897. Gardner, H., & Hatch, T. (1989). Educational implications of the theory of multiple intelligences. Educational Researcher, 18(8), 4-10. Gelfand, M. J., Aycan, Z., Erez, M., & Leung, K. (2017). Cross-cultural industrial organizational psychology and organizational behavior: A hundred-year journey. Journal of Applied Psychology, 102(3), 514-529. Gelfand, M. J., Erez, M., & Aycan, Z. (2007). Cross-cultural organizational behavior. Annu. Rev. Psychol., 58, 479-514. George, J. M. (2007). Creativity in organizations. The Academy of Management Annals, 1(1), 439-477. Gleason, K. A., Jensen-Campbell, L. A., & Richardson, D. S. (2004). Agreeableness as a predictor of aggression in adolescence. Aggressive Behavior, 30, 43–61. GLOBE, 2004. Retrieved from https://globeproject.com/results?page_id=country#country. Gong, Y., Huang, J. C., & Farh, J. L. (2009). Employee learning orientation, transformational leadership, and employee creativity: The mediating role of employee creative self-efficacy. Academy of Management Journal, 52(4), 765-778. Graziano, W. G., Jensen-Campbell, L. A., & Hair, E. C. (1996). Perceiving interpersonal conflict and reacting to it: The case for Agreeableness. Journal of Personality and Social Psychology, 70, 820–835. 97 Greenbaum, R. L., Hill, A., Mawritz, M. B., & Quade, M. J. (2017). Employee Machiavellianism to unethical behavior: The role of abusive supervision as a trait activator. Journal of Management, 43(2), 585-609. Greenberg, J. (2001). Studying organizational justice cross-culturally: Fundamental challenges. International Journal of Conflict Management, 12(4), 365-375. Greenfield, P. M. (2014). Sociodemographic differences within countries produce variable cultural values. Journal of Cross-Cultural Psychology, 45(1), 37-41. Greenleaf, R. K. (2002). Servant leadership: A journey into the nature of legitimate power and greatness. Paulist Press. Grijalva, E., & Harms, P. D. (2014). Narcissism: An integrative synthesis and dominance complementarity model. Academy of Management Perspectives, 28(2), 108-127. Haar, J. M., de Fluiter, A., & Brougham, D. (2016). Abusive supervision and turnover intentions: The mediating role of perceived organisational support. Journal of Management & Organization, 22(2), 139-153. Hackman, J. R. (1992). Group influences on individuals in organizations. In M. D. Dunnette & L. M. Hough (Eds.), Handbook of industrial and organizational psychology (Vol. 3, pp. 199-267). Palo Alto, CA: Consulting Psychologists Press. Hale, J. R., & Fields, D. L. (2007). Exploring servant leadership across cultures: A study of followers in Ghana and the USA. Leadership, 3(4), 397-417. Hare, R. D., & Neumann, C. S. (2009). Psychopathy: Assessment and forensic implications. The Canadian Journal of Psychiatry, 54(12), 791-802. Harris, K. J., Kacmar, K. M., & Zivnuska, S. (2007). An investigation of abusive supervision as a predictor of performance and the meaning of work as a moderator of the relationship. The Leadership Quarterly, 18(3), 252-263. Harris, T. C., & Locke, E. A. (1974). Replication of white-collar-blue-collar differences in sources of satisfaction and dissatisfaction. Journal of Applied Psychology, 59(3), 369-370. Harvey, P., Stoner, J., Hochwarter, W., & Kacmar, C. (2007). Coping with abusive supervision: The neutralizing effects of ingratiation and positive affect on negative employee outcomes. The Leadership Quarterly, 18(3), 264-280. Herr, R. M., Bosch, J. A., Loerbroks, A., van Vianen, A. E., Jarczok, M. N., Fischer, J. E., & Schmidt, B. (2015). Three job stress models and their relationship with musculoskeletal pain in blue-and white-collar workers. Journal of Psychosomatic Research, 79(5), 340-347. Herr, R. M., Bosch, J. A., van Vianen, A. E., Jarczok, M. N., Thayer, J. F., Li, J., ... & Loerbroks, A. (2015). Organizational justice is related to heart rate variability in white- 98 collar workers, but not in blue-collar workers—findings from a cross-sectional study. Annals of Behavioral Medicine, 49(3), 434-448. Hershcovis, M. S., Turner, N., Barling, J., Arnold, K. A., Dupré, K. E., Inness, M., ... & Sivanathan, N. (2007). Predicting workplace aggression: a meta-analysis. Journal of Applied Psychology, 92(1), 228-238. Hoch, J. E., Bommer, W. H., Dulebohn, J. H., & Wu, D. (2018). Do ethical, authentic, and servant leadership explain variance above and beyond transformational leadership? A meta- analysis. Journal of Management, 44(2), 501-529. Hoel, H., Rayner, C., & Cooper, C. L. (1999). Workplace bullying. John Wiley & Sons Ltd. Hoffman, B. J., Strang, S. E., Kuhnert, K. W., Campbell, W. K., Kennedy, C. L., & LoPilato, A. C. (2013). Leader narcissism and ethical context: Effects on ethical leadership and leader effectiveness. Journal of Leadership & Organizational Studies, 20(1), 25-37. Hofmann, D. A. (1997). An overview of the logic and rationale of hierarchical linear models. Journal of Management, 23(6), 723-744. Hofmann, D. A., Griffin, M. A., & Gavin, M. B. (2000). The application of hierarchical linear modeling to organizational research. In K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions (pp. 467-511). San Francisco, CA, US: Jossey-Bass. Hofstede, G. (1980a). Motivation, leadership, and organizations: Do American theories apply abroad? Organizational Dynamics, 9,42-63. Hofstede, G. (1980b). Culture’s consequences. Beverly Hills, CA: Sage. Hofstede, G. (2001). Culture's consequences: Comparing values, behaviors, institutions and organizations across nations. Sage publications. Hon, A. H., & Lu, L. (2016). When will the trickle-down effect of abusive supervision be alleviated? The moderating roles of power distance and traditional cultures. Cornell Hospitality Quarterly, 57(4), 421-433. Hong, Y. Y., Morris, M. W., Chiu, C. Y., & Benet-Martinez, V. (2000). Multicultural minds: A dynamic constructivist approach to culture and cognition. American Psychologist, 55(7), 709-720. Hough, L. M., Oswald, F. L., & Ock, J. (2015). Beyond the Big Five: New directions for personality research and practice in organizations. Annu. Rev. Organ. Psychol. Organ. Behav., 2(1), 183-209. House, R., Javidan, M., Hanges, P., & Dorfman, P. (2002). Understanding cultures and implicit leadership theories across the globe: an introduction to project GLOBE. Journal of World Business, 37(1), 3-10. 99 Howell, J. P., Dorfman, P. W., & Kerr, S. (1986). Moderator variables in leadership research. Academy of Management Review, 11(1), 88-102. Hu, X., Kaplan, S., & Dalal, R. S. (2010). An examination of blue-versus white-collar workers’ conceptualizations of job satisfaction facets. Journal of Vocational Behavior, 76(2), 317- 325. Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1-55. Iacobucci, D. (2010). Structural equations modeling: Fit indices, sample size, and advanced topics. Journal of Consumer Psychology, 20(1), 90-98. Inoue, A., Kawakami, N., Tsutsumi, A., Shimazu, A., Tsuchiya, M., Ishizaki, M., ... & Kivimäki, M. (2009). Reliability and validity of the Japanese version of the Organizational Justice Questionnaire. Journal of Occupational Health, 51(1), 74-83. James, L. R. (1982). Aggregation bias in estimates of perceptual agreement. Journal of Applied Psychology, 67(2), 219-229. Javidan, M., Dorfman, P. W., De Luque, M. S., & House, R. J. (2006). In the eye of the beholder: Cross cultural lessons in leadership from project GLOBE. Academy of Management Perspectives, 20(1), 67-90. Jian, Z., Kwan, H. K., Qiu, Q., Liu, Z. Q., & Yim, F. H. K. (2012). Abusive supervision and frontline employees' service performance. The Service Industries Journal, 32(5), 683-698. John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The big five inventory—versions 4a and 54. Jonason, P. K., & Webster, G. D. (2010). The dirty dozen: A concise measure of the dark triad. Psychological Assessment, 22(2), 420-432. Jones, D. N., & Paulhus, D. L. (2009). Machiavellianism. In M. R. Leary & R. H. Hoyle (Eds.), Handbook of individual differences in social behavior (pp. 93-108). New York, NY, US: The Guilford Press. Judge, T. A., & Bono, J. E. (2000). Five-factor model of personality and transformational leadership. Journal of Applied Psychology, 85(5), 751-765. Judge, T. A., Bono, J. E., Ilies, R., & Gerhardt, M. W. (2002). Personality and leadership: a qualitative and quantitative review. Journal of Applied Psychology, 87(4), 765-780. Judge, T. A., Heller, D., & Mount, M. K. (2002). Five-factor model of personality and job satisfaction: A meta-analysis. Journal of Applied Psychology, 87(3), 530-541. Judge, T. A., LePine, J. A., & Rich, B. L. (2006). Loving yourself abundantly: relationship of the narcissistic personality to self-and other perceptions of workplace deviance, leadership, and task and contextual performance. Journal of Applied Psychology, 91(4), 762-776. 100 Judge, T. A., Rodell, J. B., Klinger, R. L., Simon, L. S., & Crawford, E. R. (2013). Hierarchical representations of the five-factor model of personality in predicting job performance: integrating three organizing frameworks with two theoretical perspectives. Journal of Applied Psychology, 98(6), 875-925. Judge, T. A., & Zapata, C. P. (2015). The person–situation debate revisited: Effect of situation strength and trait activation on the validity of the Big Five personality traits in predicting job performance. Academy of Management Journal, 58(4), 1149-1179. Jung, D. I., Bass, B. M., & Sosik, J. J. (1995). Bridging leadership and culture: A theoretical consideration of transformational leadership and collectivistic cultures. Journal of Leadership Studies, 2(4), 3-18. Kalshoven, K., Den Hartog, D. N., & De Hoogh, A. H. (2011). Ethical leader behavior and big five factors of personality. Journal of Business Ethics, 100(2), 349-366. Kant, L., Skogstad, A., Torsheim, T., & Einarsen, S. (2013). Beware the angry leader: Trait anger and trait anxiety as predictors of petty tyranny. The Leadership Quarterly, 24(1), 106- 124. Kanten, S., & Sadullah, O. (2012). An empirical research on relationship quality of work life and work engagement. Procedia-Social and Behavioral Sciences, 62, 360-366. Keller Hansbrough, T., & Jones, G. E. (2014). Inside the minds of narcissists: How narcissistic leaders’ cognitive processes contribute to abusive supervision. Zeitschrift für Psychologie, 222(4), 214-220. Kenrick, D. T., & Funder, D. C. (1988). Profiting from controversy: Lessons from the person- situation debate. American Psychologist, 43(1), 23-34. Kernan, M. C., Watson, S., Chen, F. F., & Kim, T. G. (2011). How cultural values affect the impact of abusive supervision on worker attitudes. Cross Cultural Management: An International Journal, 18(4), 464-484. Kiazad, K., Restubog, S. L. D., Zagenczyk, T. J., Kiewitz, C., & Tang, R. L. (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. Kiewitz, C., Restubog, S. L. D., Shoss, M. K., Garcia, P. R. J. M., & Tang, R. L. (2016). Suffering in silence: Investigating the role of fear in the relationship between abusive supervision and defensive silence. Journal of Applied Psychology, 101(5), 731-742. Kirkman, B. L., Chen, G., Farh, J. L., Chen, Z. X., & Lowe, K. B. (2009). Individual power distance orientation and follower reactions to transformational leaders: A cross-level, cross- cultural examination. Academy of Management Journal, 52(4), 744-764. 101 Kirkman, B. L., Lowe, K. B., & Gibson, C. B. (2006). A quarter century of culture's consequences: A review of empirical research incorporating Hofstede's cultural values framework. Journal of International Business Studies, 37(3), 285-320. Kirkman, B. L., & Shapiro, D. L. (2001). The impact of cultural values on job satisfaction and organizational commitment in self-managing work teams: The mediating role of employee resistance. Academy of Management Journal, 44(3), 557-569. Kish-Gephart, J. J., Detert, J. R., Treviño, L. K., & Edmondson, A. C. (2009). Silenced by fear: The nature, sources, and consequences of fear at work. Research in Organizational Behavior, 29, 163-193. Kish-Gephart, J. J., Harrison, D. A., & Treviño, L. K. (2010). Bad apples, bad cases, and bad barrels: meta-analytic evidence about sources of unethical decisions at work. Journal of Applied Psychology, 95(1), 1-31. Kluemper, D. H., Mossholder, K. W., Ispas, D., Bing, M. N., Iliescu, D., & Ilie, A. (2019). When core self-evaluations influence employees’ deviant reactions to abusive supervision: The moderating role of cognitive ability. Journal of Business Ethics, 159(2), 435-453. Kristof, A. L. (1996). Person‐organization fit: An integrative review of its conceptualizations, measurement, and implications. Personnel Psychology, 49(1), 1-49. Kuenzi, M., & Schminke, M. (2009). Assembling fragments into a lens: A review, critique, and proposed research agenda for the organizational work climate literature. Journal of Management, 35(3), 634-717. Lam, S. S., Hui, C., & Law, K. S. (1999). Organizational citizenship behavior: Comparing perspectives of supervisors and subordinates across four international samples. Journal of Applied Psychology, 84(4), 594-601. Lee, K., & Allen, N. J. (2002). Organizational citizenship behavior and workplace deviance: The role of affect and cognitions. Journal of Applied Psychology, 87(1), 131-142. Lee, S., Yun, S., & Srivastava, A. (2013). Evidence for a curvilinear relationship between abusive supervision and creativity in South Korea. The Leadership Quarterly, 24(5), 724- 731. Lian, H., Ferris, D. L., & Brown, D. J. (2012). Does power distance exacerbate or mitigate the effects of abusive supervision? It depends on the outcome. Journal of Applied Psychology, 97(1), 107-123. Liden, R. C., Wayne, S. J., & Stilwell, D. (1993). A longitudinal study on the early development of leader-member exchanges. Journal of Applied Psychology, 78(4), 662-674. Likert, R. (1967). The human organization. New York: McGraw-Hill. 102 Lin, W., Wang, L., & Chen, S. (2013). Abusive supervision and employee well‐being: The moderating effect of power distance orientation. Applied Psychology, 62(2), 308-329. Littek W, Heisig U. Work organisation under technological change: sources of differentiation and the reproduction of social inequality in processes of change. In: Clegg S, ed. Organisation Theory and Class Analysis. Berlin/New York: de Gruyter; 1989: 299-314. Little, T. D., Rhemtulla, M., Gibson, K., & Schoemann, A. M. (2013). Why the items versus parcels controversy needn’t be one. Psychological Methods, 18(3), 285-300. Liu, D., Liao, H., & Loi, R. (2012). The dark side of leadership: A three-level investigation of the cascading effect of abusive supervision on employee creativity. Academy of Management Journal, 55(5), 1187-1212. Liu, J., Kwong Kwan, H., Wu, L. Z., & Wu, W. (2010). Abusive supervision and subordinate supervisor‐directed deviance: The moderating role of traditional values and the mediating role of revenge cognitions. Journal of Occupational and Organizational Psychology, 83(4), 835-856. Locke, E. A. (1973). Satisfiers and dissatisfiers among white-collar and blue-collar employees. Journal of Applied Psychology, 58(1), 67-76. Mackey, J. D., Frieder, R. E., Brees, J. R., & Martinko, M. J. (2017). Abusive supervision: A meta-analysis and empirical review. Journal of Management, 43(6), 1940-1965. Malatesta, R. M., & Byrne, Z. S. (1997). The impact of formal and interactional procedures on organizational outcomes. In 12th annual conference of the Society for Industrial and Organizational Psychology, St. Louis, MO. Martin, K. D., & Cullen, J. B. (2006). Continuities and extensions of ethical climate theory: A meta-analytic review. Journal of Business Ethics, 69(2), 175-194. Martinko, M. J., Harvey, P., Brees, J. R., & Mackey, J. (2013). A review of abusive supervision research. Journal of Organizational Behavior, 34(S1), S120-S137. Masterson, S. S., Lewis, K., Goldman, B. M., & Taylor, M. S. (2000). Integrating justice and social exchange: The differing effects of fair procedures and treatment on work relationships. Academy of Management Journal, 43(4), 738-748. Mawritz, M. B., Dust, S. B., & Resick, C. J. (2014). Hostile climate, abusive supervision, and employee coping: Does conscientiousness matter?. Journal of Applied Psychology, 99(4), 737-747. Mayer, D. M., Bardes, M., & Piccolo, R. F. (2008). Do servant-leaders help satisfy follower needs? An organizational justice perspective. European Journal of Work and Organizational Psychology, 17(2), 180-197. 103 Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709-734. McGinnis, J. T. L. (2010). Leaders Behaving Badly: Antecedents and Consequences of Abuse. Unpublished doctoral dissertation, North Carolina State University. McGregor, D. (1967). The professional manager. New York: McGrawHill Mehta, P. D., & Neale, M. C. (2005). People are variables too: Multilevel structural equations modeling. Psychological Methods, 10(3), 259-284. Miller, J. D., Lynam, D., & Leukefeld, C. (2003). Examining antisocial behavior through the five-factor model of personality. Aggressive Behavior, 29, 497–514. 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. Morris, W. R., Conrad, K. M., Marcantonio, R. J., Marks, B. A., & Ribisl, K. M. (1999). Do blue-collar workers perceive the worksite health climate differently than white-collar workers?. American Journal of Health Promotion, 13(6), 319-324. Muthén, L., & Muthén, B. (2017). Mplus (Version 8)[computer software].(1998–2017). Los Angeles, CA: Muthén & Muthén. Nandkeolyar, A. K., Shaffer, J. A., Li, A., Ekkirala, S., & Bagger, J. (2014). Surviving an abusive supervisor: The joint roles of conscientiousness and coping strategies. Journal of Applied Psychology, 99(1), 138-150. Neuman, J. H., & Baron, R. A. (1998). Workplace violence and workplace aggression: Evidence concerning specific forms, potential causes, and preferred targets. Journal of Management, 24(3), 391-419. Ng, K. Y., Ang, S., & Chan, K. Y. (2008). Personality and leader effectiveness: A moderated mediation model of leadership self-efficacy, job demands, and job autonomy. Journal of Applied Psychology, 93(4), 733-743. O’Boyle, E. H., Forsyth, D. R., Banks, G. C., & McDaniel, M. A. (2012). A meta-analysis of the Dark Triad and work behavior: a social exchange perspective. Journal of Applied Psychology, 97(3), 557-579. Offermann, L. R., Kennedy Jr, J. K., & Wirtz, P. W. (1994). Implicit leadership theories: Content, structure, and generalizability. The Leadership Quarterly, 5(1), 43-58. Oldham, G. R., & Cummings, A. (1996). Employee creativity: Personal and contextual factors at work. Academy of Management Journal, 39(3), 607-634. 104 Organ, D. W. (1977). A reappraisal and reinterpretation of the satisfaction-causes-performance hypothesis. Academy of Management Review, 2(1), 46-53. Organ, D. W., & Ryan, K. (1995). A meta‐analytic review of attitudinal and dispositional predictors of organizational citizenship behavior. Personnel Psychology, 48(4), 775-802. Ostroff, C. (1993). The effects of climate and personal influences on individual behavior and attitudes in organizations. Organizational Behavior and Human Decision Processes, 56(1), 56-90. Pan, S. Y., & Lin, K. J. (2018). Who suffers when supervisors are unhappy? The roles of leader– member exchange and abusive supervision. Journal of Business Ethics, 151(3), 799-811. Park, H., Hoobler, J. M., Wu, J., Liden, R. C., Hu, J., & Wilson, M. S. (2017). Abusive supervision and employee deviance: A multifoci justice perspective. Journal of Business Ethics, 1-19. Paulhus, D. L., & Williams, K. M. (2002). The dark triad of personality: Narcissism, Machiavellianism, and psychopathy. Journal of Research in Personality, 36(6), 556-563. Peterson, M. F., & Hunt, J. G. J. (1997). International perspectives on international leadership. The Leadership Quarterly, 8(3), 203-231. Petty, M. M., McGee, G. W., & Cavender, J. W. (1984). A meta-analysis of the relationships between individual job satisfaction and individual performance. Academy of Management Review, 9(4), 712-721. Pillai, R., Schriesheim, C. A., & Williams, E. S. (1999). Fairness perceptions and trust as mediators for transformational and transactional leadership: A two-sample study. Journal of Management, 25(6), 897-933. Podsakoff, P. M., MacKenzie, S. B., & Podsakoff, N. P. (2012). Sources of method bias in social science research and recommendations on how to control it. Annual Review of Psychology, 63, 539-569. Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools for probing interactions in multiple linear regression, multilevel modeling, and latent curve analysis. Journal of Educational and Behavioral Statistics, 31(4), 437-448. Preacher, K. J., Zyphur, M. J., & Zhang, Z. (2010). A general multilevel SEM framework for assessing multilevel mediation. Psychological Methods, 15(3), 209-233. Rafferty, A. E., & Restubog, S. L. D. (2011). The influence of abusive supervisors on followers' organizational citizenship behaviours: The hidden costs of abusive supervision. British Journal of Management, 22(2), 270-285. 105 Ralston, D. A., Holt, D. H., Terpstra, R. H., & Kai-Cheng, Y. (1997). The impact of natural culture and economic ideology on managerial work values: a study of the United States, Russia, Japan, and China. Journal of International Business Studies, 28(1), 177-207. Randall, D. M. (1990), ‘The Consequences of Organizational Commitment: Methodological Investigation’, Journal of Organizational Behavior, 11, 361–78. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (Vol. 1). Sage. Raymond, B., & Bruschi, I. G. (1989). Psychological abuse among college women in dating relationships. Perceptual and Motor Skills, 69(3), 1283-1297. Resick, C. J., Hanges, P. J., Dickson, M. W., & Mitchelson, J. K. (2006). A cross-cultural examination of the endorsement of ethical leadership. Journal of Business Ethics, 63(4), 345-359. Resick, C. J., Whitman, D. S., Weingarden, S. M., & Hiller, N. J. (2009). The bright-side and the dark-side of CEO personality: examining core self-evaluations, narcissism, transformational leadership, and strategic influence. Journal of Applied Psychology, 94(6), 1365-1381. Rhodewalt, F., & Peterson, B. (2009). Narcissism. In M. R. Leary & R. H. Hoyle (Eds.), Handbook of individual differences in social behavior (pp. 547-560). New York, NY, US: The Guilford Press. Richard, O. C., Boncoeur, O. D., Chen, H., & Ford, D. L. (2018). Supervisor Abuse Effects on Subordinate Turnover Intentions and Subsequent Interpersonal Aggression: The Role of Power-Distance Orientation and Perceived Human Resource Support Climate. Journal of Business Ethics, 1-15. Riketta, M. (2002). Attitudinal organizational commitment and job performance: a meta‐ analysis. Journal of Organizational Behavior, 23(3), 257-266. Robinson, W. S. (1950). Ecological correlations and the behavior of individuals. American Sociological Review, 15, 351-357. Robinson, S. L., & Bennett, R. J. (1995). A typology of deviant workplace behaviors: A multidimensional scaling study. Academy of Management Journal, 38(2), 555-572. Rousseau, D. M., Sitkin, S. B., Burt, R. S., & Camerer, C. (1998). Not so different after all: A cross-discipline view of trust. Academy of Management Review, 23(3), 393-404. Rousseau, V., & Aubé, C. (2018). When leaders stifle innovation in work teams: The role of abusive supervision. Journal of Business Ethics, 151(3), 651-664. Salancik, G. R., & Pfeffer, J. (1978). A social information processing approach to job attitudes and task design. Administrative Science Quarterly, 224-253. 106 Sarala, R. M., & Vaara, E. (2010). Cultural differences, convergence, and crossvergence as explanations of knowledge transfer in international acquisitions. Journal of International Business Studies, 41(8), 1365-1390. Schat, A. C. H., Desmarais, S., & Kelloway, E. K. (2006). Exposure to workplace aggression from multiple sources: Validation of a measure and test of a model. Unpublished manuscript, McMaster University, Hamilton, Canada, 331-351. Schaubroeck, J., Lam, S. S., & Peng, A. C. (2011). Cognition-based and affect-based trust as mediators of leader behavior influences on team performance. Journal of Applied Psychology, 96(4), 863-871. Scherbaum, C. A., & Ferreter, J. M. (2009). Estimating statistical power and required sample sizes for organizational research using multilevel modeling. Organizational Research Methods, 12(2), 347-367. Schneider, B., & Reichers, A. E. (1983). On the etiology of climates. Personnel Psychology, 36(1), 19-39. Shepard, M. F., & Campbell, J. A. (1992). The Abusive Behavior Inventory: A measure of psychological and physical abuse. Journal of Interpersonal Violence, 7(3), 291-305. Shoss, M. K., Eisenberger, R., Restubog, S. L. D., & Zagenczyk, T. J. (2013). Blaming the organization for abusive supervision: The roles of perceived organizational support and supervisor's organizational embodiment. Journal of Applied Psychology, 98(1), 158-168. Smith, C. A., Organ, D. W., & Near, J. P. (1983). Organizational citizenship behavior: Its nature and antecedents. Journal of Applied Psychology, 68(4), 653-663. Smith, P. C., Kendall, L. M., & Hulin, C. (1969). The measurement of satisfaction in work and behavior. Chicago: Raud McNally. Spector, P. E., Cooper, C. L., Sanchez, J. I., O'Driscoll, M., Sparks, K., Bernin, P., ... & Miller, K. (2001). Do national levels of individualism and internal locus of control relate to well‐ being: an ecological level international study. Journal of Organizational Behavior, 22(8), 815-832. Suls, J., Martin, R., & David, J. P. (1998). Person-environment fit and its limits: Agreeableness, neuroticism, and emotional reactivity to interpersonal conflict. Personality and Social Psychology Bulletin, 24, 88–98. Taras, V., Kirkman, B. L., & Steel, P. (2010). Examining the impact of culture's consequences: A three-decade, multilevel, meta-analytic review of Hofstede's cultural value dimensions. Journal of Applied Psychology, 95(3), 405-439. Taras, V., Steel, P., & Kirkman, B. L. (2016). Does country equate with culture? Beyond geography in the search for cultural boundaries. Management International Review, 56(4), 455-487. 107 Taylor, M. (2004). The relationships between workplace violence, deviant workplace behavior, ethical climate, organizational justice, and abusive supervision. Unpublished dissertation, Aliiant International University. Taylor, S. G., Griffith, M. D., Vadera, A. K., Folger, R., & Letwin, C. R. (2019). Breaking the cycle of abusive supervision: How disidentification and moral identity help the trickle-down change course. Journal of Applied Psychology, 104(1), 164-182. Tedeschi, J. T., & Felson, R. B. (1994). Violence, aggression, and coercive actions. Washington, DC, US: American Psychological Association. Tepper, B. J. (2000). Consequences of abusive supervision. Academy of Management Journal, 43(2), 178-190. Tepper, B. J. (2007). Abusive supervision in work organizations: Review, synthesis, and research agenda. Journal of Management, 33(3), 261-289. Tepper, B. J., Carr, J. C., Breaux, D. M., Geider, S., Hu, C., & Hua, W. (2009). Abusive supervision, intentions to quit, and employees’ workplace deviance: A power/dependence analysis. Organizational Behavior and Human Decision Processes, 109(2), 156-167. Tepper, B. J., Duffy, M. K., Henle, C. A., & Lambert, L. S. (2006). Procedural injustice, victim precipitation, and abusive supervision. Personnel Psychology, 59(1), 101-123. Tepper, B. J., Henle, C. A., Lambert, L. S., Giacalone, R. A., & Duffy, M. K. (2008). Abusive supervision and subordinates' organization deviance. Journal of Applied Psychology, 93(4), 721-732. Tett, R. P., & Burnett, D. D. (2003). A personality trait-based interactionist model of job performance. Journal of Applied Psychology, 88(3), 500-517. Tett, R. P., & Guterman, H. A. (2000). Situation trait relevance, trait expression, and cross- situational consistency: Testing a principle of trait activation. Journal of Research in Personality, 34(4), 397-423. Toppinen-Tanner, S., Kalimo, R., & Mutanen, P. (2002). The process of burnout in white‐ collar and blue‐collar jobs: eight‐year prospective study of exhaustion. Journal of Organizational Behavior, 23(5), 555-570. Treviño, L. K., Butterfield, K. D., & McCabe, D. L. (1998). The ethical context in organizations: Influences on employee attitudes and behaviors. Business Ethics Quarterly, 8(3), 447-476. Tsui, A. S., Nifadkar, S. S., & Ou, A. Y. (2007). Cross-national, cross-cultural organizational behavior research: Advances, gaps, and recommendations. Journal of Management, 33(3), 426-478. 108 Tsui, A.S. and O’Reilly, C.A. (1989), “Beyond simple demographic effects: the importance of relational demography in supervisor-subordinate dyads”, Academy of Management Journal, 32 (2), 402-423. Tyler, T. R., & Bies, R. J. (1990). Beyond formal procedures: The interpersonal context of procedural justice. Applied social psychology and organizational settings, 77, 98. Van Dierendonck, D. (2011). Servant leadership: A review and synthesis. Journal of Management, 37(4), 1228-1261. Victor, B., & Cullen, J. B. (1988). The organizational bases of ethical work climates. Administrative Science Quarterly, 101-125. Vogel, R. M., Mitchell, M. S., Tepper, B. J., Restubog, S. L., Hu, C., Hua, W., & Huang, J. C. (2015). A cross‐cultural examination of subordinates' perceptions of and reactions to abusive supervision. Journal of Organizational Behavior, 36(5), 720-745. Volmer, J., Koch, I. K., & Göritz, A. S. (2016). The bright and dark sides of leaders' dark triad traits: Effects on subordinates' career success and well-being. Personality and Individual Differences, 101, 413-418. Waldman, D. A., Wang, D., Hannah, S. T., Owens, B. P., & Balthazard, P. A. (2018). Psychological and neurological predictors of abusive supervision. Personnel Psychology, 71(3), 399-421. Walter, F., Lam, C. K., Van Der Vegt, G. S., Huang, X., & Miao, Q. (2015). Abusive supervision and subordinate performance: Instrumentality considerations in the emergence and consequences of abusive supervision. Journal of Applied Psychology, 100(4), 1056-1072. Walumbwa, F. O., Hartnell, C. A., & Oke, A. (2010). Servant leadership, procedural justice climate, service climate, employee attitudes, and organizational citizenship behavior: a cross-level investigation. Journal of Applied Psychology, 95(3), 517-529. Walumbwa, F. O., Lawler, J. J., & Avolio, B. J. (2007). Leadership, individual differences, and work‐related attitudes: a cross‐culture investigation. Applied Psychology, 56(2), 212- 230. Walumbwa, F. O., & Schaubroeck, J. (2009). Leader personality traits and employee voice behavior: mediating roles of ethical leadership and work group psychological safety. Journal of Applied Psychology, 94(5), 1275-1286. Wang, G., Harms, P. D., & Mackey, J. D. (2015). Does it take two to tangle? Subordinates’ perceptions of and reactions to abusive supervision. Journal of Business Ethics, 131(2), 487-503. Wang, W., Mao, J., Wu, W., & Liu, J. (2012). Abusive supervision and workplace deviance: The mediating role of interactional justice and the moderating role of power distance. Asia Pacific Journal of Human Resources, 50(1), 43-60. 109 Wang, Y. D., & Hsieh, H. H. (2014). Employees' reactions to psychological contract breach: A moderated mediation analysis. Journal of Vocational Behavior, 85(1), 57-66. Washington, R. R., Sutton, C. D., & Feild, H. S. (2006). Individual differences in servant leadership: The roles of values and personality. Leadership & Organization Development Journal, 27(8), 700-716. Weaver, C. N. (1975). Job preferences of white collar and blue collar workers. Academy of Management Journal, 18(1), 167-175. Westerlaken, K. M., & Woods, P. R. (2013). The relationship between psychopathy and the Full Range Leadership Model. Personality and Individual Differences, 54(1), 41-46. 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. Wright, I., Bengtsson, C., & Frankenberg, K. (1994). Aspects of psychological work environment and health among male and female white-collar and blue-collar workers in a big Swedish industry. Journal of Organizational Behavior, 177-183. Wu, L. Z., Kwong Kwan, H., Liu, J., & Resick, C. J. (2012). Work-to-family spillover effects of abusive supervision. Journal of Managerial Psychology, 27(7), 714-731. Wu, L. Z., Zhang, H., Chiu, R. K., Kwan, H. K., & He, X. (2014). Hostile attribution bias and negative reciprocity beliefs exacerbate incivility’s effects on interpersonal deviance. Journal of Business Ethics, 120(2), 189-199. Wu, T. Y., & Hu, C. (2009). Abusive supervision and employee emotional exhaustion: Dispositional antecedents and boundaries. Group & Organization Management, 34(2), 143- 169. Xu, E., Huang, X., Lam, C. K., & Miao, Q. (2012). Abusive supervision and work behaviors: The mediating role of LMX. Journal of Organizational Behavior, 33(4), 531-543. Yukl, G. (1989). Managerial leadership: A review of theory and research. Journal of Management, 15(2), 251-289. Zehir, C., Akyuz, B., Eren, M. S., & Turhan, G. (2013). The indirect effects of servant leadership behavior on organizational citizenship behavior and job performance: Organizational justice as a mediator. International Journal of Research in Business and Social Science, 2(3), 1-13. Zellars, K. L., Tepper, B. J., & Duffy, M. K. (2002). Abusive supervision and subordinates' organizational citizenship behavior. Journal of Applied Psychology, 87(6), 1068-1076. Zhang, X., & Bartol, K. M. (2010). Linking empowering leadership and employee creativity: The influence of psychological empowerment, intrinsic motivation, and creative process engagement. Academy of Management Journal, 53(1), 107-128. 110 Zhang, Y., & Bednall, T. C. (2016). Antecedents of abusive supervision: A meta-analytic review. Journal of Business Ethics, 139(3), 455-471. Zhang, Y., & Liao, Z. (2015). Consequences of abusive supervision: A meta-analytic review. Asia Pacific Journal of Management, 32(4), 959-987. Zhang, H., Kwan, H. K., Zhang, X., & Wu, L. Z. (2014). High core self-evaluators maintain creativity: A motivational model of abusive supervision. Journal of Management, 40(4), 1151-1174. Zhao, H., & Seibert, S. E. (2006). The Big Five personality dimensions and entrepreneurial status: A meta-analytical review. Journal of Applied Psychology, 91(2), 259-271. Zhou, J., & George, J. M. (2001). When job dissatisfaction leads to creativity: Encouraging the expression of voice. Academy of Management Journal, 44(4), 682-696. 111