FLUID REASONING AS A PREDICTOR OF DEVIANT WORKPLACE BEHAVIORS By James Garrett Matusik A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Business Administration – Organizational Behavior – Human Resource Management – Doctor of Philosophy 2020 FLUID REASONING AS A PREDICTOR OF DEVIANT WORKPLACE BEHAVIORS ABSTRACT By James Garrett Matusik Relative to the sizable body of work investigating the relationship between intelligence and task performance, one facet of overall job performance, very little research has been conducted on the relationship between intelligence and counterproductive work behavior, a different but equally important facet of job performance. Furthermore, the little research that has been conducted on the intelligence-counterproductive work behavior relationship has yielded entirely inconsistent results – while some researchers have found a negative relationship between intelligence and counterproductive work behavior, others have found null or positive relationships. This, coupled with the reality that none of these studies have explicitly tested causal mechanisms, provides us with an entirely unclear understanding of this relationship. Thus, the goal of this dissertation is to more carefully examine the potential relationship(s) between intelligence and counterproductive work behavior by (a) capturing both overt and covert counterproductive work behaviors, (b) testing a previously identified, but thus far untested, mediating mechanism, and (c) incorporating moderators, selected based upon criminology research that has leveraged a psychological approach in the explanation of individual deviance. Copyright by JAMES GARRETT MATUSIK 2020 ACKNOWLEDGEMENTS I would like to acknowledge everyone who played a role in my academic upbringing. First, my family, who supported me every step of the way. Without them, I could have never made it to this point. Second, my committee members – Drs. Hollenbeck, Hays, Scott, and Ferris – whom have provided patient advice and guidance throughout the entire research process. Thank you all for your unwavering support. iv TABLE OF CONTENTS LIST OF TABLES ....................................................................................................................... viii LIST OF FIGURES ....................................................................................................................... ix INTRODUCTION ...........................................................................................................................1 LITERATURE REVIEW: HUMAN INTELLIGENCE .................................................................6 Human Intelligence – Conceptual Evolution and Current Status.................................................7 Gf-Gc Theory ..........................................................................................................................8 Cattell-Horn Gf-Gc Theory.....................................................................................................9 Thorndike’s Intelligence Theory...........................................................................................10 Gardner’s Theory of Multiple Intelligences .........................................................................10 Carroll’s Three-Stratum Theory ...........................................................................................11 Cattell-Horn-Caroll Theory of Intelligence ..........................................................................13 Conceptualization of Intelligence Leveraged Herein .................................................................14 Fluid Reasoning ...................................................................................................................16 LITERATURE REVIEW: COUNTERPRODUCTIVE WORK BEHAVIOR .............................18 Typologies & Taxonomies of Counterpoductive Work Behavior ............................................19 LITERATURE REVIEW: PREDICTING COUNTERPRODUCTIVE WORK BEHAVIOR WITH INTELLIGENCE ...............................................................................................................24 Intelligence and Counterproductive Work Behavior ................................................................24 Limitations Associated with these Studies .........................................................................28 HYPOTHESIS DEVELOPMENT.................................................................................................31 Episodic Foresight .....................................................................................................................31 Episodic Foresight – Inhibition ...........................................................................................33 Episodic Foresight – Facilitation ........................................................................................37 MODERATING ROLES OF MORAL IDENTITY AND POWER .............................................40 The Moderating Role of Moral Identity ....................................................................................41 Moral Reasoning .................................................................................................................41 Moral Identity .....................................................................................................................44 The Moderating Role of Power .................................................................................................51 Hierarchy.............................................................................................................................52 Power ..................................................................................................................................53 STUDY 1 METHOD .....................................................................................................................63 Sample and Procedure ..............................................................................................................63 Item Generation ........................................................................................................................64 v Analytic Approach and Results ................................................................................................65 Phase One.........................................................................................................................65 Phase Two ........................................................................................................................66 Phase Three ......................................................................................................................68 STUDY 2 METHOD .....................................................................................................................70 Sample ......................................................................................................................................70 Procedure ..................................................................................................................................70 Measures ...................................................................................................................................72 Fluid Reasoning ...............................................................................................................72 Moral Identity ..................................................................................................................73 Episodic Foresight ...........................................................................................................73 Power ...............................................................................................................................76 Counterproductive Work Behavior ..................................................................................76 Control Variables .............................................................................................................77 Analytic Approach ...................................................................................................................78 RESULTS ......................................................................................................................................80 Tests of Hypotheses .................................................................................................................80 Model 1 ............................................................................................................................81 Model 2 ............................................................................................................................82 Model 3 ............................................................................................................................83 SUPPLEMENTAL ANALYSES ..................................................................................................86 Supplemental Analyses – Alternative Cue Word Sets .............................................................87 Model 4 ...........................................................................................................................87 Model 5 ...........................................................................................................................89 Supplemental Analyses – Additional Interactions ...................................................................89 Power x Fluid Reasoning Predicting Episodic Foresight ...............................................90 Power x Fluid Reasoning x Moral Identity Predicting Episodic Foresight ....................90 Status x Fluid Reasoning Predicting Episodic Foresight ................................................90 Status x Fluid Reasoning x Moral Identity Predicting Episodic Foresight .....................91 Moral Identity x Episodic Foresight Predicting Covert CWB ........................................91 Moral Identity x Episodic Foresight Predicting Overt CWB ..........................................91 Moral Identity x Episodic Foresight x Power Predicting Covert CWB ..........................92 Moral Identity x Episodic Foresight x Power Predicting Overt CWB ...........................92 Status x Episodic Foresight Predicting CWB .................................................................93 Moral Identity x Status x Episodic Foresight Predicting Covert CWB ..........................93 Moral Identity x Status x Episodic Foresight Predicting Overt CWB ............................93 Supplemental Analyses – Alternative Calculations of CWB ...................................................94 Original Measure ............................................................................................................94 Original Measure – Dichotomized ..................................................................................95 Property and Production Deviance .................................................................................95 Property and Production Deviance – Dichotomized .......................................................95 vi Personal Aggression and Political Deviance ..................................................................96 Personal Aggression and Political Deviance – Dichotomized ........................................96 Organizational and Interpersonal Deviance ....................................................................96 Organizational and Interpersonal Deviance – Dichotomized .........................................97 Measure from Study 1 – Dichotomized ..........................................................................97 Supplemental Analyses – Recreation of CWB Measures ........................................................97 DISCUSSION ..............................................................................................................................100 Theoretical Implications .........................................................................................................101 Practical Implications .............................................................................................................103 Limitations .............................................................................................................................105 Future Directions ....................................................................................................................109 Future Research related to Fluid Reasoning ................................................................109 Future Research related to Emotional Intelligence ......................................................111 Future Research related to Counterproductive Work Behavior Antecedents ..............112 Profile Research related to Engagement in Counterproductive Work Behavior .........114 Future Upper Echelons Research .................................................................................116 Past Engagement in Counterproductive Work Behavior .............................................117 CONCLUSION ............................................................................................................................120 APPENDICES .............................................................................................................................121 APPENDIX A – Measures .....................................................................................................122 APPENDIX B – Figures and Tables ......................................................................................138 REFERENCES ............................................................................................................................167 vii LIST OF TABLES TABLE 1: Cattell-Horn-Carroll Theory of Intelligence ..............................................................154 TABLE 2: Results of Item-Sort Pretask ......................................................................................156 TABLE 3: Results from 12-item Confirmatory Factor Analysis.................................................158 TABLE 4: Descriptive Statistics and Correlations among all Variables .....................................159 TABLE 5: Exploratory Factor Analysis that includes all 50 Items: Two-Factor Solution .........161 TABLE 6: Exploratory Factor Analysis that includes all 50 Items: Three-Factor Solution .......163 TABLE 7: Exploratory Factor Analysis that includes all 50 Items: Four-Factor Solution .........165 viii LIST OF FIGURES FIGURE 1: Theoretical Model ....................................................................................................138 FIGURE 2: Scree Plot for Study 1 Exploratory Factor Analysis ................................................139 FIGURE 3: Path Estimates from Model 1 ...................................................................................140 FIGURE 4: Model 1 Interaction between Power and Episodic Foresight in the Prediction of Covert Counterproductive Work Behavior ..................................................................................141 FIGURE 5: Path Estimates from Model 2 ...................................................................................142 FIGURE 6: Path Estimates from Model 3 ...................................................................................143 FIGURE 7: Model 3 Interaction between Power and Episodic Foresight in the Prediction of Covert Counterproductive Work Behavior ..................................................................................144 FIGURE 8: Path Estimates from Model 4 ...................................................................................145 FIGURE 9: Model 4 Interaction between Power and Episodic Foresight in the Prediction of Covert Counterproductive Work Behavior ..................................................................................146 FIGURE 10: Path Estimates for Model 5 ....................................................................................147 FIGURE 11: Interaction between Power, Moral Identity Internalization, and Fluid Reasoning in the Prediction of Episodic Foresight ............................................................................................148 FIGURE 12: Interaction between Moral Identity Symbolization and Episodic Foresight in the Prediction of Covert Counterproductive Work Behavior ............................................................149 FIGURE 13: Interaction between Moral Identity Symbolization and Episodic Foresight in the Prediction of Covert Counterproductive Work Behavior ............................................................150 FIGURE 14: Interaction between Power, Moral Identity Internalization, and Episodic Foresight in the Prediction of Overt Counterproductive Work Behavior ....................................................151 FIGURE 15: Interaction between Power, Moral Identity Internalization, and Episodic Foresight in the Prediction of Overt Counterproductive Work Behavior ....................................................152 FIGURE 16: Scree Plot for Exploratory Factor Analysis that includes all 50 Items ..................153 ix INTRODUCTION There are few, if any, absolutes in the social sciences. Whereas the natural and physical sciences are characterized by laws that may be consistently applied in a wide range of contexts, universal principles are rare, and quite possibly nonexistent, in social science disciplines such as organizational behavior (and the closely related field of organizational psychology; Robbins & Judge, 2014). That said, perhaps the most robust, widely replicated finding in organizational behavior and organizational psychology (OB/OP) is the positive relationship between intelligence, or the ability to reason, solve, plan, learn, and think abstractly across domains (Gottfredson, 1997), and individual job performance (Hunter & Hunter, 1984; Ree, Earles, & Teachout, 1994; Scherbaum, Goldstein, Yusko, Ryan, & Hanges, 2012; Schmidt, 2002; Schmidt & Hunter, 1998). Indeed, this relationship is so well-established that the notion that one may leverage intelligence to predict individual job performance may arguably represent the closest thing that OB/OP has to a “universal principle.” Consistent with this claim, some researchers have gone so far as to suggest that any further debate on the existence and significance of the intelligence-performance relationship is illogical (e.g., Gottfredson, 2002; Schmidt, 2002). In fact, Schmidt (2002: 187) literally warns that the rejection of the overwhelming body of evidence substantiating this relationship threatens OB/OP’s legitimacy as a science-based field. Moreover, one survey-based study of over 700 organizational psychologists determined that a majority consider tests of intelligence to be highly relevant in personnel selection contexts given intelligence’s relationship with job performance (Murphy, Cronin, & Tam, 2003). Thus, anyone brazen enough to challenge the position that intelligence is predictive of desirable performance outcomes would not only be questioning an 1 extensive body of research, but would also be standing in direct opposition to the general consensus of the field. Hence the purpose of this dissertation. Although I acknowledge that intelligence is positively related to certain aspects of job performance, we cannot assume intelligence has desirable consequences across all dimensions of performance as there is simply no research to support such an assumption. More precisely, our current understanding of the intelligence- performance relationship is inherently deficient because researchers have largely focused on task-relevant facets of performance and ignored non-task facets such as deviant workplace behavior, or, more commonly, counterproductive work behavior (CWB) (Campbell & Wiernik, 2015; Robinson & Bennett, 1995). As such, it is impossible to say with confidence that intelligence is positively or negatively related to CWB. To be sure, some research on the potential relationship between intelligence and CWB has been conducted (e.g., Dilchert, Ones, Davis, & Rostow, 2007; Gonzalez-Mulé, Mount, & Oh, 2014; Roberts, Harms, Caspi, & Moffitt, 2007), but studies explicitly addressing this relationship are incredibly scarce (Mercado, Dilchert, Giordano, & Ones, 2018). This paucity of research is somewhat surprising as managers often consider CWB to be just as important, if not more important, than task-relevant aspects of performance in their evaluations of overall job performance (Rotundo & Sackett, 2002). At the same time, it is also somewhat unsurprising as personality traits may be considered more theoretically proximal predictors of CWB (Dilchert et al., 2007) than abilities (Sackett & Lievens, 2008). Regardless as to whether the scarcity of this research comes as a surprise or not, it represents a major void in the knowledge base considering the abundance of research on the intelligence-(task) performance relationship. 2 Worsening matters, the very limited research that has been conducted on the intelligence- CWB relationship is entirely inconclusive. Of the three major studies conducted on the intelligence-CWB relationship, one found a negative relationship (Dilchert et al., 2007), one found a null, or non-significant, relationship (Gonzalez-Mulé et al., 2014), and one found a positive relationship (Roberts et al., 2007), despite all three hypothesizing a negative relationship. Additionally, all three of these studies failed to explicitly test any explanatory mechanisms for this theoretically negative relationship, despite offering several (e.g., foresight). Taken together, there appears to be theoretical consensus as to the direction of this relationship, and this consensus aligns with the potential assumption that intelligence is predictive of desirable performance outcomes across all facets of performance, yet there is little empirical evidence to support or explain it. Thus, the primary goal of this dissertation is to more carefully examine the relationship(s) between intelligence and CWB. By doing so, I hope to create substantively-based (i.e., empirically-informed) consensus (Hollenbeck, 2008) regarding the relationship between intelligence and CWB, and shift the apparent theoretical consensus (Hollenbeck, 2008) that the relationship between intelligence and CWB is unambiguously negative. As it pertains to the latter objective (i.e., consensus shifting), I hope to show that intelligence is indeed negatively related to overt forms of CWB, but positively related to covert forms of CWB. If evidence were to suggest that intelligence is positively related to covert CWB, it would bound academics’ claims that intelligence is an excellent predictor of desirable performance outcomes. Such evidence would also prove incredibly insightful to practitioners given that (a) tests of intelligence are frequently employed in personnel selection contexts (Murphy et al., 2003), and (b) CWB is 3 often considered to be just as important, if not more important, than task performance when it comes to evaluating individuals’ overall job performance (Rotundo & Sackett, 2002). The second goal of this dissertation is to identify and test the mediating mechanism underlying the intelligence-CWB relationship. As noted, there are currently no investigations of the intelligence-CWB relationship that have explicitly tested explanatory, mediating mechanisms, despite speculating on them. Accordingly, we have no substantively-based explanation for “why” (Whetten, 1989) intelligence might be related to CWB – we are currently limited to purely theoretical explanations. Thus, I seek to identify one potential mechanism (episodic foresight) and, in doing so, demonstrate how this single mediator can have both positive and negative downstream consequences in terms of engagement in CWB. The final goal of this dissertation is to provide a more nuanced understanding of the relationship(s) between intelligence and CWB through the integration of two moderators: moral identity and power. More specifically, I will investigate how moral identity, or the extent to which being and behaving morally is central to one’s self-concept (Aquino & Reed, 2002), affects the relationship between intelligence (operationalized as fluid reasoning) and episodic foresight, and how power, or the extent to which an individual controls valuable resources (Magee & Galinsky, 2008), enables and incentivizes engagement in various CWBs. Figure 1 provides a visualization of my proposed conceptual model. With these objectives in mind, my dissertation proposal is laid out as follows. First, I will review the literature on intelligence, tracing its history up to its current and most widely-accepted conceptualization. I will then select and justify my operationalization of intelligence, namely fluid reasoning. Second, I will review the major frameworks of CWB offered by organizational scholars before selecting and justifying the framework I will ultimately use herein (Robinson and 4 Bennett's (1995) framework). Third, I will critically review the few investigations that have attempted to substantiate the theoretical relationship(s) between intelligence and CWB before hypothesizing what I believe to be the actual direction of, and causal mechanism (i.e., mediator) underlying, this relationship. Fourth, I will define moral identity and power in greater detail, justify why I selected these two constructs by drawing upon psychological theories of criminology, and explain how they may serve as critical boundary conditions of the relationship(s) between intelligence and CWB. Finally, I will provide tests of my theoretical model before ending with a discussion of the results. 5 LITERATURE REVIEW: HUMAN INTELLIGENCE The domain of human intelligence is extensive and complex, with initial conceptualizations of the construct potentially dating as far back as the fourth century B.C. (Ree & Carretta, 2007). In terms of more contemporary research, intelligence as a construct of academic study can be traced back to at least Francis Galton’s work in the late 1860s (Scherbaum et al., 2012). Furthermore, in the more than century-long history of approaching intelligence from a psychometric angle (Spearman, 1904), several prolific scholars have proposed varied, competing frameworks to conceptually organize it (e.g., hierarchical models of human intelligence; Carroll, 1993; McGrew, 1997), considerable debate has surrounded whether certain sets of abilities qualify as part of it (e.g., emotional intelligence; Locke, 2005; MacCann, Joseph, Newman, & Roberts, 2014), and individual tendencies and traits have contaminated attempts to measure it (DeNisi & Shaw, 1977). To provide some background on intelligence’s elaborate history, I will chronicle its development as a theoretical construct, from inception to current status in the literature, in the subsections that follow. In doing so, I detail the frameworks and theories widely considered to be the most influential in the field, and explain how successive, eminent frameworks built upon their predecessors. At the conclusion of this section, and before transitioning to the next section, I will explain and defend the conceptualization, or “type” (McGrew, 2009), of intelligence that will be the focus of my theorizing in the hypothesis building sections of this dissertation. Specifically, I will focus on fluid reasoning in my theory development section as fluid reasoning (a) most closely corresponds to the conceptualization of intelligence widely agreed upon by intelligence researchers (Gottfredson, 1997), (b) is not context-specific (and is therefore likely to be assessed across selection contexts), and (c) is most theoretically proximal to outcomes of 6 organizational interest. I believe it is necessary to emphasize in advance that I will be leveraging a specific conceptualization of intelligence as a wide variety of other forms of intelligence have been proposed in the literature (e.g., olfactory abilities), and therefore will be discussed throughout the following subsections. Human Intelligence – Conceptual Evolution and Current Status Although the study of intelligence predates the 1900s (Scherbaum et al., 2012), Charles Spearman (1904) is often accredited with inspiring academic research on, and introducing the psychometric approach to the study of, intelligence via the seminal article in which he proposed a unitary factor underlying various assessments of cognitive ability. In this article, Spearman leveraged factor analysis, demonstrated that scores on various tests of mental ability were positively related (i.e., “positive manifold”), and argued that individual performance on these tests may be a product of a single, biologically-based source of intelligence. That is, Spearman theorized and provided evidence that there was a single, intrinsic force driving performance across these varied mental tests, which he referred to as general intelligence. Many consider this work to be the catalyst to the study of the unitary construct still known as general intelligence today (e.g., McGrew, 2009; Schmidt & Hunter, 2004), but also frequently referred to as general mental ability (i.e., GMA), the g factor, or, simply, g. Though Spearman provided evidence that general intelligence may very well exist, a precise definition for this construct has proven difficult to ascertain. Indeed, numerous, varying definitions of general intelligence were offered by researchers subsequent to Spearman’s work (e.g., Neisser et al., 1996; Weschler, 1958). In recognition of these varying definitions and the confusion they may sow, an editorial with 52 signatories, all of whom were experts on the construct of human intelligence, formally defined intelligence in the 1990s (Gottfredson, 1997). 7 These experts collectively defined intelligence as the mental capability to “reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience,” such that “it is not merely book learning, a narrow academic skill, or test-taking smarts,” but rather “reflects a broader and deeper capability for comprehending our surroundings,” (Gottfredson, 1997: 13). Thus, the consensus is that intelligence does not reflect any single talent, area of expertise, or specialty. Rather, intelligence encompasses a range of interrelated mental abilities and, consequently, accounts for a significant portion of variance in between-individual performance differences across a variety of cognitive tests (Glutting, Watkins, & Youngstrom, 2003; Spearman, 1904). This definition and conceptual understanding of intelligence has been widely accepted by the field (e.g., Nisbett et al., 2012). However, given the extensive, and somewhat indeterminate, set of abilities that intelligence may encompass, several researchers tried to provide the construct with an organizing framework prior to the arrival of this definition (e.g., Carroll, 1993; Cattell, 1941; Das, Kirby, & Jarman, 1975; Gardner, 1983; Hebb, 1942; Horn, 1965; Sternberg, 1985; Thorndike, 1920). These various frameworks proved to be highly influential on the field and are therefore worthy of recognition by contemporary researchers. Thus, I will next review each of these theories, largely proceeding in chronological order but going out of chronological order when relevant to this narrative. Gf-Gc Theory One of the earliest and most pervasive frameworks for understanding intelligence is Cattell’s (1941) fluid and crystalized theory of intelligence, also known as Gf-Gc Theory (Alfonso, Flanagan, & Radwan, 2005). Raymond Cattell (Cattell, 1941, 1971, 1987), a student of Spearman’s (Schneider & McGrew, 2012), was heavily influenced by the work of Spearman as 8 well as that of his own contemporaries, Louis Thurstone and Donald Hebb (Alfonso et al., 2005). In fact, there is so much overlap between Cattell’s theorizing and Donald Hebb’s theory of two intelligences, which Hebb dubbed “Intelligence A” and “Intelligence B” (Hebb, 1942), that some argue Cattell simply adapted and renamed Hebb’s two types of intelligence (e.g., Brown, 2016). Regardless of the source of his inspiration, over the course of several years Cattell popularized what came to be known as Gf-Gc theory and argued for what is essentially a dichotomous conceptualization of human intelligence. Specifically, Cattell proposed that intelligence could be classified as fluid or crystalized. Fluid intelligence (the Gf in Gf-Gc theory), refers to the ability to reason, resolve novel problems, and quickly transform and manipulate different types of information, irrespective of the factual knowledge one has accumulated (Zaval, Li, Johnson, & Weber, 2015). Thus, fluid intelligence reflects the ability to think abstractly and perform successfully in the moment, independent of past experience, practice, or education. On the other hand, crystalized intelligence (the Gc in Gf-Gc theory), reflects experience-based, general knowledge that one has acquired over time (i.e., facts, know-how, and skills) (Zaval et al., 2015). That is, crystalized intelligence is a product of prior formal and informal learning. This early, dichotomous conceptualization of intelligence is still recognized and employed to this day (e.g., Checa & Fernández-Berrocal, 2015). However, researchers have expanded upon Gf-Gc theory considerably since its inception. Cattell-Horn Gf-Gc Theory In a series of investigations (Horn, 1965; Horn, 1968; Horn & Stankov, 1982), John Horn, a student of Cattell’s, built upon Gf-Gc theory so that it included several additional abilities (i.e., abilities beyond fluid and crystalized intelligence). The earliest additions to Gf-Gc theory included abilities that reflected visual perception/processing, short-term memory, long- 9 term storage and retrieval, and auditory processing. By the mid-1990s, Gf-Gc theory had been further expanded to include abilities that reflected reaction time and decision speed, as well as quantitative and broad reading and writing abilities (Alfonso et al., 2005). These developments of Gf-Gc theory eventually led to a multi-factor model of, and a modified name for, Gf-Gc theory: Cattell-Horn Gf-Gc theory (Horn, 1991; Horn & Blankson, 2005). Much like Gf-Gc theory, Cattell-Horn Gf-Gc theory became one of the most widely-accepted and popular theories in intelligence research (McGrew, 2009). Thorndike’s Intelligence Theory Although Cattell’s seminal Gf-Gc theory and Horn’s expansion of Cattell’s Gf-Gc theory (i.e., Cattell-Horn Gf-Gc theory) were highly influential in intelligence research, these researchers were not the only ones to propose a multifaceted model of intelligence. For example, Edward Thorndike, who worked closely with Cattell, proposed a tripartite view of intelligence nearly 20 years before Gf-Gc theory was formally introduced to the field. Specifically, Thorndike (1920) argued that human intellectual abilities (i.e., intelligence) include abstract intelligence, or the ability to process and comprehend concepts, mechanical intelligence, or the ability to control and manipulate one’s body as well as physical objects, and social intelligence, or the ability to effectively manage human interactions and perform in social situations. Though Thorndike’s theory of intelligence did not seem to gain acceptance as quickly or widely as Cattell’s, Thorndike’s theorizing did inspire subsequent work that attempted to introduce alternative “types” of intelligence (e.g., emotional intelligence; Salovey & Mayer, 1990). Gardner’s Theory of Multiple Intelligences Much more recently, Howard Gardner (1983) put forth a theory of intelligence reminiscent of Thorndike’s and Horn’s conceptualizations, such that Gardner similarly identified 10 unique “types” of intelligence. In a direct challenge to the concept of general intelligence, or the g factor (Spearman, 1904), Gardner argued that there are multiple, orthogonal types of intelligence. Gardner’s theory of multiple intelligences is based on the logic that intelligence reflects a “capacity to process a certain kind of information” and information can take various forms, such that it can be musical, bodily-kinesthetic, logical-mathematical, verbal-linguistic, spatial, interpersonal, intrapersonal, naturalistic, or, quite possibly, existential (Gardner, 1993). In other words, Gardner argued that humans differ in their capacity to understand and utilize various kinds of information, which reflects different types of individual intelligences rather than a general ability to process information, abstract, and reason. Although Gardner’s theory of multiple intelligences has been influential in some disciplines (e.g., education; Jones, 2015), it has not been without its critics (e.g., Visser, Ashton, & Vernon, 2006). Carroll’s Three-Stratum Theory These competing, diverse frameworks for understanding human intelligence, among others not detailed here (e.g., Sternberg, 1985), partially motivated John Carroll’s (1993) momentous, meta-factor analytic research on the structure of human intelligence. Recognizing that disparate frameworks of intelligence had been put forth since Spearman’s (1904; 1927) seminal work, Carroll (1993) ambitiously factor analyzed over 460 human intelligence datasets collected over the span of more than 5 decades and detailed the results in his treatise Human Cognitive Abilities: A Survey of Factor-Analytic Studies. In doing so, Carroll built upon, and, naturally, was influenced by, the work of prior intelligence researchers, especially those discussed herein (e.g., Cattell and Horn). Ultimately, Carroll (1993) provided evidence of, and proposed a three-stratum theory of, intelligence. According to Carroll’s three-stratum theory, various abilities are nested under 11 general intelligence and organized hierarchically, such that human abilities are (a) interrelated and (b) differ in terms of their complexity and discreteness (Alfonso et al., 2005). At the lowest level (i.e., stratum I) are the most basic, narrow, discrete abilities, such as the ability to understand the meaning of words. At the middle level (i.e., stratum II) are eight broader, more coherent groups of abilities that encompass the lower-level, stratum I abilities. An example of these broader, mid-level abilities include the ability to recognize and compare perceptual patterns (i.e., perceptual-organizational intelligence) (Mayer et al., 2008). At the highest level (i.e., stratum III) is general intelligence, or g, the conceptualization of which is consistent with Spearman’s (1904, 1927) own conceptualization of g, and involves the ability to reason abstractly across various domains (Mayer et al., 2008). Thus, all other lower-level abilities are theoretically nested under general intelligence, which may help to explain why performance across cognitive domains appears to be positively correlated (i.e., positive manifold; Burkart, Schubiger, & van Schaik, 2017). This three-stratum theory has received empirical support in subsequent research, such that a three-stratum model can be reproduced in hierarchical confirmatory factor analyses (e.g., Bickley, Keith, & Wolfle, 1995) and general intelligence emerges as an entity at an etiological level (e.g., Shikishima et al., 2009). Given its wide scope and arguable success in providing intelligence researchers with some level of consensus on the structure of general human intelligence, Carroll’s (1993) theory has been recognized as one of the most influential theories of human intelligence to date (e.g., Burns, 1994; Horn, 1998; Jensen, 2004). Indeed, Carroll’s (1993) contribution to the field of applied psychometrics has been lauded as “a work similar in stature to other principia publications in other fields,” such as Newton’s The Mathematical Principles of Natural Philosophy (McGrew, 2009: 2). Despite widespread acceptance, however, 12 intelligence researchers have attempted to, and largely succeeded in, further developing the three-stratum theory (e.g., McGrew, 1997). Cattell-Horn-Carroll Theory of Intelligence As alluded to, the work of prior intelligence researchers was highly influential on Carroll’s own work, particularly the work of Cattell and Horn. Accordingly, Carroll’s three- stratum theory shares several similarities with Cattell and Horn’s fluid and crystallized theory of intelligence (i.e., Cattell-Horn Gf-Gc theory), such as the recognition of several broad, conceptually similar sets of abilities (e.g., fluid and crystalized intelligence are recognized by both theories; McGrew, 1997, 2009). Although there are differences among these theories, with the most notable being the recognition of a g factor (i.e., general intelligence) by Carroll but not Cattell and Horn (McGrew, 1997), their considerable similarities gave rise to what is now known as the Cattell-Horn-Carroll (CHC) theory of intelligence (Alfonso et al., 2005). Originally proposed as an “integrated” theory by Kevin McGrew (1997), the CHC theory of intelligence has become “the most comprehensive and empirically supported psychometric theory of the structure of cognitive and academic abilities to date” (Alfonso et al., 2005: 85), and therefore has been leveraged extensively by researchers and proprietors of intelligence tests (e.g., Flanagan, McGrew, & Ortiz, 2000; Flanagan & Ortiz, 2001; McGrew & Flanagan, 1998). CHC theory encompasses over 70 narrow, first-stratum abilities that are subsumed by several broader, second-stratum abilities, which themselves are subsumed by g, the general ability factor. Thus, CHC theory is hierarchically organized and contains three stratums, like Carroll’s three-stratum theory, but contains a larger set of second-stratum abilities. While most treatments of CHC theory typically identify nine broad, second-stratum ability domains (i.e., fluid reasoning, comprehension-knowledge, short-term memory, visual 13 processing, long-term storage and retrieval, cognitive processing speed, decision and reaction speed, reading and writing, and quantitative knowledge), McGrew proposes the inclusion of additional domains that reflect general (but domain-specific) knowledge, auditory processing, tactile abilities, kinesthetic abilities, olfactory abilities, psychomotor abilities, and psychomotor speed (McGrew, 2009; Schneider & McGrew, 2012), for a total of 16 second-stratum, broad abilities. Considering its breadth, CHC theory is by far the most comprehensive, cumulative model of human intelligence to date. CHC theory is summarized in greater detail in Table 1. All information in this table is adapted or directly borrowed from Schneider and McGrew (2012). General intelligence (i.e., g) is omitted for the ease of presentation, as well as because it encompasses all lower-level abilities (i.e., it is at the apex, stratum III, and therefore all abilities are nested underneath it). The second column of this table provides the name of the second-level (stratum II), broad abilities recognized in CHC theory, while the first column provides information regarding the conceptual groupings among these broad abilities. The third column provides the verbatim definitions of these abilities, borrowed directly from Schneider and McGrew (2012). Lastly, the fourth column provides examples of the narrow, lowest-level (stratum I) abilities nested within the broader abilities (stratum II). Conceptualization of Intelligence Leveraged Herein There are a few important takeaways from Table 1, as well as CHC theory more broadly. First, there are several second-stratum, broad abilities recognized by intelligence researchers that are seemingly distal to organizational outcomes (e.g., counterproductive work behavior, the outcome examined in this dissertation). For example, there are many second-stratum abilities that can be conceptually grouped (column 1) into abilities that reflect the capacity to successfully (a) 14 use one’s senses (i.e., sensory abilities, such as olfactory processing) or (b) employ motor skills (i.e., motor abilities, such as kinesthetic abilities). Though individuals may indeed differ in terms of their prowess in these domains, neither of these ability sets appear to reflect a capacity to reason, plan, or think abstractly (i.e., definitional elements of intelligence; Gottfredson, 1997), particularly as it pertains to engagement in CWB. Similarly, several second-stratum abilities can be conceptually grouped into abilities that reflect the efficiency of one’s memory, the speed with which one can operate, or the knowledge one has accumulated (e.g., quantitative knowledge). While many of these second-stratum abilities may enhance (or moderate) one’s ability to reason about, or actually facilitate one’s engagement in, behavior such as CWB (e.g., leveraging quantitative knowledge to successfully “cook the books”), these abilities themselves do not directly speak to the capacity to think or reason about potential behavior. That is, these abilities may enable behavior, but they do not capture the extent to which one may be able to reason about it. Finally, I acknowledge that domain-specific knowledge could facilitate engagement in deviant behavior, but have elected not to examine this second-stratum ability in this dissertation. There are two reasons for this. The first, perhaps more practical, reason is because domain- specific knowledge, of which could include knowledge regarding an organization’s or industry’s functioning, is something acquired through experience in a particular organization or role. It is not an ability with universal applications – domain-specific knowledge is tied to particular industries, occupations, or organizations, and thus there is no single best method to capture it, it is not universally predictive of performance across organizations, and it is not likely to be measured in a consistent way across hiring contexts. The second, perhaps more theoretical, reason is that domain-specific knowledge is explicitly excluded in the agreed upon definition 15 provided by intelligence researchers (Gottfredson, 1997). Therefore, for both practical and theoretical reasons, I will not focus on domain-specific knowledge in this dissertation. All said, this leaves just 1 of these 16 second-stratum abilities, notably grouped with no other abilities in the CHC model: fluid reasoning. In what follows, I argue that fluid reasoning is (a) most proximal to outcomes of relevance to organizations (e.g., CWB) and (b) closely mirrors the widely-accepted definition of human intelligence previously provided. Thus, it will be my focus throughout the rest of this dissertation. Fluid Reasoning As illustrated in Table 1, fluid reasoning may be conceptualized as “the deliberate but flexible control of attention to solve novel ‘on-the-spot’ problems that cannot be performed by relying exclusively on previously learned habits, schemas, and scripts,” (Schneider & McGrew, 2012: 111). Consistent with this definition, higher levels of fluid reasoning should, theoretically speaking, facilitate the formation and recognition of concepts, inductive reasoning, the solving of unfamiliar problems (i.e., novel problem-solving), the accurate identification of relationships, and the mental reorganization of information (Otero, 2017). Furthermore, measures of fluid reasoning are considered to be excellent measures of general intelligence (Frick, Barry, & Kamphaus, 2010), such that fluid reasoning provides a “window into an individual’s overall intellectual ability” (Otero, 2017: 208). In other words, the type of thinking that measures of fluid reasoning assess is conceptually similar to the type of thinking associated with general notions of intelligence. Indeed, and as previously explained, broadly defined general intelligence may be described as the ability to “reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience,” (Gottfredson, 1997: 13). This definition of 16 general intelligence is quite similar to the definition of fluid reasoning just provided. Neither fluid reasoning, specifically, nor general intelligence, more broadly, reflect expertise in a specific area, both are independent of the past knowledge one has accumulated, and both capture the ability to use, manipulate, and comprehend information in a way that facilitates decision-making, problem-solving, and planning. Given the wide body of research that suggests that general intelligence (i.e., the g factor) may be one of the best individual differences for predicting proximal, organizational outcomes such as task performance (Hunter & Hunter, 1984; Scherbaum et al., 2012; Schmidt & Hunter, 1998), fluid reasoning, specifically, should also be strongly related to organizationally-relevant outcomes. Thus, in order to bound my theorizing as well as accurately measure the form of intelligence most likely to affect individual behavior in organizational contexts, I henceforth focus explicitly on fluid reasoning to the preclusion of the other 15 second-stratum abilities recognized in the CHC model. That is, fluid reasoning is my conceptualization of intelligence and I will therefore use a measure of fluid reasoning in assessing individual intelligence1. With this in mind, I next review the literature on my criterion variable: counterproductive work behavior. 1 Fluid reasoning is distinct from, but correlated with, cognitive flexibility. Cognitive flexibility has been defined various ways (e.g., one’s ability to alternate between and/or consider various concepts simultaneously, Scott, 1962; one’s ability to adapt cognitive processing strategies to face novel conditions, Moore & Malinowski, 2009), and research suggests that it is positively related to fluid reasoning/intelligence (as measured by Raven's Standard Progressive Matrices; Colzato, Van Wouwe, Lavender, & Hommel, 2006). Given fluid reasoning’s prominence in widely-accepted, comprehensive frameworks for organizing human intelligence (and cognitive flexibility’s lack thereof), as well as its extensive history in this domain, I will focus on fluid reasoning. 17 LITERATURE REVIEW: COUNTERPRODUCTIVE WORK BEHAVIOR Job performance may be defined as the actions and behaviors under the control of employees that contribute to or detract from the organization’s goals, regardless whether these actions are part of one’s formal job description (Campbell & Wiernik, 2015). The reason that “job performance” constitutes both beneficial and detrimental employee behaviors is because it consists of three broad components: task performance, counterproductive work behavior (CWB), and organizational citizenship behavior (OCB) (Rotundo & Sackett, 2002; Viswesvaran & Ones, 2000). Task performance refers to behaviors often stipulated in one’s formal job description and involves the transformation of raw materials into goods or services (i.e., “doing one’s job”). OCBs represent discretionary behaviors that contribute to the technical core and generally promote organizational effectiveness (Podsakoff, Whiting, Podsakoff, & Blume, 2009). Thus, both task performance and OCBs generally promote the welfare of the organization. CWB, on the other hand, has overwhelmingly negative implications for the organization. CWB is an umbrella term that encompasses a wide range of volitional, intentional actions and behaviors employees engage in that are harmful to the interests of the organization or its members, undermine the organization’s or its members’ goals, and detract from the organization’s or its members’ well-being (Bennett & Robinson, 2000; Dalal, 2005; Ones & Dilchert, 2013). That is, CWBs consist of scalable, deliberate employee behaviors that are directed at, and inflict harm on either, the organization or its constituents. Given the inclusive nature of this definition, researchers have studied a wide range of behaviors that may be classified as CWBs and have done so using a variety of labels. Indeed, prior researchers have studied narrowly defined behaviors (e.g., Greenberg, 1990) as well as broad sets of ostensibly distinct behaviors (e.g., Chen & Spector, 1992) that may be categorized as CWBs, and they have 18 studied CWBs under the guise of various labels, such as deviance (e.g., Berry, Ones, & Sackett, 2007), retaliation (e.g., Skarlicki & Folger, 1997), and workplace aggression (e.g., O’Leary, Griffin, & Glew, 1996), to name a few. Despite the various labels employed to capture and describe CWBs, investigations into the diverse measures used by researchers generally suggest that each consists of an overlapping group of behaviors (Spector & Fox, 2005). In other words, researchers often use different terminology when describing behaviors that can largely be categorized under the broader term “counterproductive work behaviors.” This is not to say that all CWBs are equivalent; rather, this suggests that CWB is a broad phenomenon that manifests in various manners, manners that may still be distinct and worthy of unique theoretical and empirical attention. Accordingly, several typologies and taxonomies of CWBs have been put forth in the literature. Typologies & Taxonomies of Counterproductive Work Behavior Given the nonspecific definition of CWB, as well as the numerous behaviors and labels examined and used (respectively) in the CWB literature, it is perhaps unsurprising that various typologies have been proposed by prior researchers in an attempt to provide a theoretical framework for understanding CWB (Marcus, Taylor, Hastings, Sturm, & Weigelt, 2016). For example, Gruys and Sackett (2003) identified eleven categories of CWB, some of which include destruction of property, inappropriate verbal and physical actions, and misuse of time and resources. On the other hand, Spector and colleagues (2006) argued for smaller set of CWBs. Specifically, these authors argued that there are only five specific, discernable facets of CWB (i.e., abuse towards others, production deviance, sabotage, theft, and withdrawal), and they provided some support for their typology by demonstrating that these five facets of CWB differ in terms of their antecedents (e.g., abuse and sabotage were strongly related to anger while theft 19 was unrelated to emotion). As another example, Martinko and colleagues (2002) broadly grouped individual CWBs into those that are primarily self-destructive (e.g., absenteeism, passivity) and those that are primarily retaliatory (e.g., aggression, stealing). Perhaps the most notable taxonomy of CWBs, however, was provided by Robinson and Bennett (1995) in their multidimensional scaling study. In this study, Robinson & Bennett (1995) determined that CWBs could generally be grouped into four categories, production deviance, property deviance, political deviance, and personal aggression, that vary along two dimensions, (1) minor versus serious and (2) interpersonally- versus organizationally-oriented. Production deviance consists of minor CWBs that affect the organization, with examples including leaving work early and taking excessive breaks, while property deviance consists of more serious CWBs that affect the organization, such as stealing from the company and sabotaging equipment. Political deviance consists of minor CWBs that affect organizational members, with examples including gossiping about and blaming one’s coworkers, while personal aggression consists of more serious CWBs that affect organizational members, such as sexual harassment and verbal abuse. This seminal research, as well as these researchers’ subsequent work (e.g., Bennett & Robinson, 2000), has been cited thousands of times and become one of the most widely-accepted CWB typologies in the literature (and perhaps for good reason). Indeed, considerable empirical evidence following Robinson and Bennett’s seminal work suggests that CWB may be targeted at either the organization or its members (i.e., one’s coworkers; Campbell & Wiernik, 2015), one dimension identified in Robinson and Bennett’s (1995) typology. Much like Robinson and Bennett (1995), Gruys and Sackett (2003) utilized a dimensional scaling approach and determined that there is an interpersonal-organizational dimension underlying CWB. Furthermore, Berry et al. (2007) provided meta-analytic evidence 20 that, while correlated, organizationally-directed and interpersonally-directed CWBs represent distinct categories that have different relationships with personality characteristics and organizational citizenship behaviors. Importantly, Berry et al. (2007) were not the only researchers to provide meta-analytic support for this dimension of CWB; Dalal (2005) similarly provided meta-analytic support for the distinction between interpersonally-oriented and organizationally-oriented CWBs. Naturally, support for this dimension is not reserved to just dimensional scaling and meta-analytic investigations. Support is also provided by a considerable number of factor analyses, including Bennett and Robinson's (2000) own widely-cited measure of CWB, that yield factors reflecting organizationally- and interpersonally-oriented CWBs (Campbell & Wiernik, 2015). There is also a body of research suggesting that organizationally-oriented CWBs and interpersonally-oriented CWBs differ in terms of their antecedents. For example, Fox and colleagues (2001) found that organizational stressors were more closely associated with organizationally-oriented CWBs while interpersonal conflict was more closely associated with interpersonally-oriented CWBs. Likewise, Bruk-Lee and Spector (2006) determined that interpersonal conflict was likely to coincide CWB directed at individuals, but not the organization. Thus, a large body of evidence supports the contention that CWB can vary in terms of its target. Investigations into, and replications of, the other dimension identified by Robinson and Bennett (1995) (the minor versus serious distinction), however, are far scarcer. Some have argued that such research has been hampered by the low base-rates associated with serious CWBs (e.g., sexual harassment; Spector et al., 2006), whereas others have argued that CWB measurement instruments themselves affect respondents’ attendance to the minor-serious 21 dimension (Gruys & Sackett, 2003). Regardless, most widely used measures of CWB include both minor and serious CWB content (Bowling & Gruys, 2010), and the minor-serious distinction has been recognized in, and incorporated into, empirical research. For example, in their investigation into the effects of perceived overqualification on CWB, Fine and Edward (2017) provided evidence that the minor-serious distinction mattered. Specifically, these researchers determined that employees who perceive themselves to be overqualified not only engage in more organizationally-directed CWB, in general, but also found that this CWB tended to be more minor than serious in nature. Furthermore, researchers have theoretically integrated the eleven categories of CWB identified by Gruys and Sackett (2003) into the Robinson and Bennett (1995) typology by leveraging the minor-serious distinction, as well as theorized how different individual dispositional traits relate to more or less severe categories of CWB (Wu & Lebreton, 2011). Given the well-accepted status of Robinson and Bennett’s (1995) typology and its underlying dimensions, I have chosen to focus on the four categories of CWBs identified by these researchers (but will ultimately group them into two categories; more information on this is provided below). To reiterate, the four categories of CWB identified by Robinson and Bennett (1995) include production deviance, property deviance, political deviance, and personal aggression, and the two dimensions along which CWBs vary reflect their target (i.e., the organization or its members) and its severity (i.e., minor versus serious). Furthermore, any one of several behaviors may be categorized in this typology. Production deviance includes relatively minor but organizationally harmful acts that violate formal norms reflecting the quality and quantity of work one must accomplish, with examples including leaving work early, taking excessive breaks, intentionally working slow, and 22 wasting resources. Property deviance includes more major, organizationally harmful acts, with examples including theft, embezzlement, and destruction of equipment. Political deviance includes relatively minor and interpersonally harmful social interactions that disadvantage others personally or politically, with examples including gossiping about one’s coworkers, showing favoritism, spreading rumors, and blaming coworkers. Finally, personal aggression includes more serious, interpersonally harmful acts that are directed at individuals and are hostile in nature, with examples including sexual harassment, verbal abuse, and stealing from or endangering coworkers. As noted, I will argue that these four categories of CWB may be grouped into two categories. In particular, I will argue that some are inherently overt (production deviance and personal aggression) and that some can be conducted more covertly (property deviance and political deviance). Together, these two (grouped) categories will serve as the dependent variables in my theoretical model. 23 LITERATURE REVIEW: PREDICTING COUNTERPRODUCTIVE WORK BEHAVIOR WITH INTELLIGENCE While the positive relationship between intelligence and task performance is robust and well-accepted (Hunter & Hunter, 1984; Scherbaum et al., 2012; Schmidt & Hunter, 1998), research on the relationship between intelligence and CWB is sparse (Mercado et al., 2018), and the nature of the relationship (i.e., positive or negative) is uncertain (Dilchert et al., 2007; Gonzalez-Mulé et al., 2014; Roberts et al., 2007). In the following subsections, I review the major findings regarding the intelligence-CWB relationship before arguing that intelligence (conceptualized and measured as fluid reasoning) may have both positive and negative implications for engagement in CWB. Specifically, I will argue that intelligence is indeed related to CWB, but that the direction of this relationship is contingent upon the specific type of CWB under consideration (i.e., covert or overt forms of CWB). Importantly, I want to preface this section by stating that I do not believe the direct effect of fluid reasoning on CWB will be significant once the mediating mechanism is accounted for (more information on this mediator is provided later). Intelligence and Counterproductive Work Behavior In the first study meant to explicitly test the relationship between intelligence and CWB, Dilchert and colleagues (2007) found that a standardized psychometric test of intelligence was indeed predictive of objectively measured CWB, as captured by formally recorded incidents. More specifically, these researchers determined that higher performance on the Shipley Institute of Living Scale (SILS), a two-part test frequently used in personnel selection as well as clinical settings, was negatively related to recorded incidents of CWB directed at both individuals and organizations in a sample of police officers. These researchers did not capture self-reported 24 CWB (i.e., possibly covert, undetected CWB) nor did they distinguish between minor and serious types of CWB. Furthermore, these researchers drew heavily from the criminology literature to inform their hypotheses and propose several potential explanatory mechanisms (e.g., foresight), but they did not explicitly test any of these mechanisms. In a direct, two-study response to Dilchert et al. (2007), Marcus and colleagues (2009) also examined the intelligence-CWB relationship. In their first study, a field study, these researchers tested the intelligence-CWB relationship using a slightly more heterogenous sample, a measure of CWB directed primarily at the organization, and a different measure of intelligence: the Wonderlic Personnel Test. Marcus et al. (2009) ultimately found no significant relation between intelligence and CWB. In their second study, a lab study, Marcus and colleagues (2009) measured intelligence with a different measure, the Personnel Assessment Form, and captured CWB with both a 50-item, self-report scale as well as several objective indicators (e.g., theft of pencils). Again, intelligence was unrelated to self-reported CWB. However, intelligence was negatively related to one objective indicator of CWB: participant’s late arrival to the study. Though this is arguably a weak operationalization of CWB given the lab context, it did provide at least some findings that were consistent with those of Dilchert et al. (2009). In later, meta-analytic research, Gonzalez-Mulé and colleagues (2014) identified and coded 35 independent samples that provided information on the intelligence-CWB relationship. These studies utilized various non-self-reported measures of intelligence, both self- and non-self- reported measures of CWB, and a meta-sample that was heavily skewed towards military and police personnel. Ultimately, these researchers determined that the overall true-score correlation between intelligence and CWB was essentially zero (consistent with Marcus et al.’s (2009) findings). However, they also found some evidence of a positive relationship between self-rated 25 CWB and intelligence and, interestingly, a negative relationship between non-self-rated CWB and intelligence. Importantly, the 95% confidence intervals of these two estimates did not overlap, suggesting that the intelligence-CWB correlation is indeed moderated by rating source. Yet, it should be noted, the 95% confidence interval for self-rated CWB and intelligence did contain zero (-.01, .11). Furthermore, and potentially of relevance in light of Dilchert et al.’s (2007) findings based upon a police sample, Gonzalez-Mulé et al. (2014) found a negative relationship between intelligence and CWB for police samples, compared to a non-significant relationship between intelligence and CWB for nonpolice samples (importantly, however, the confidence intervals for these two estimates did overlap). Much like Dilchert et al. (2007), Gonzalez-Mulé and colleagues (2014) neither distinguished between minor and severe CWBs nor did they test explanatory mechanisms of the intelligence-CWB relationship despite their discussion of such mechanisms, many of which were similar to those discussed by Dilchert et al. (2007) (e.g., foresight). While the studies discussed thus far generally found either null or negative relationships between intelligence and CWB, other studies have found positive relationships. In one longitudinal study utilizing a heterogenous sample, it was determined that childhood measures of intelligence, as measured by the Wechsler’s Intelligence Scale for Children-Revised, were positively related to engagement in CWB in adulthood (Roberts et al., 2007). Specifically, the results of this study suggested that children who scored higher on this test of intelligence were more likely to steal money and materials from their workplace, as well as use company equipment for personal reasons (e.g., the company car) as working adults. This effect remained even after controlling for adolescent conduct disorder, criminal convictions as a minor, several personality characteristics, and job characteristics. Roberts and colleagues (2007) speculated that 26 this may be due to a stronger sense of entitlement among, and different job opportunities available to, more intelligent individuals, or potentially to the fact that they used a more heterogenous sample than that used by Dilchert et al. (2007). This is not the only study to find a positive relationship between intelligence and CWB. Indeed, Marcus, Schuler, Quell, and Humpfner (2002) found a relationship of similar magnitude between intelligence and a broad, self-report measure of CWB (though this relationship was not significant at the p < .05 level), and research on overqualification, which is calculated using measures of intelligence, suggests a positive relationship between overqualification and CWB. Indeed, Fine and Edward (2017) provide evidence that perceptions of overqualification are positively associated with CWB. Specifically, objective overqualification, operationalized as the standardized difference between the required level of intelligence necessary to do a job and participants’ scores on a national scholastic aptitude college entrance test, was highly correlated with perceptions of overqualification (r = .73), which, in turn, were related to organizationally- directed CWB (r = .29). These findings are similar to those reported in other work on overqualification (Liu, Luksyte, Zhou, Shi, & Wang, 2015; Luksyte, Spitzmueller, & Maynard, 2011). Though overqualification is clearly not synonymous with raw intelligence, these findings may be insightful considering that the existing body of literature on the intelligence-CWB relationship is meager. Although other studies in organizational research have examined the relationship between intelligence and outcomes conceptually akin to CWB, such as domestic violence, fraud, and destruction of property, as well as other forms of criminal off-duty deviance (e.g., Lyons, Hoffman, Bommer, Kennedy, & Hetrick, 2016), the reviewed studies best represent our current understanding of the intelligence-CWB relationship (an understanding that is cloudy, at best) as 27 they explicitly capture deviant behaviors in the work context. Most other studies have included intelligence as a moderator of some independent variable-CWB relationship (e.g., Kluemper et al., 2018), or have controlled for intelligence in their assessment of some other independent variable-CWB relationship (e.g., some of the studies used by Gonzalez-Mulé et al., 2014). Thus, I have chosen to focus explicitly on these studies’ findings up to this point. In the next section, I will delineate their limitations. Limitations Associated with these Studies There are a couple of differences between, and limitations associated with, these studies that may help to explain their inconsistent findings, the first of which pertains to their measurement of CWB. Dilchert et al. (2007) did not capture self-reported CWB and only distinguished between interpersonally-directed and organizationally-directed CWB (i.e., they did not distinguish between covert and overt forms of CWB). Given the objective, documented nature of the reports used to operationalize CWB, it may be argued that only those CWBs that were overt were included in Dilchert et al.’s (2007) analyses. This operationalization biases analyses such that CWBs not explicitly documented were not included (e.g., people who frequently “get away with” stealing from the organization). In support of this possibility, Gonzalez-Mulé et al. (2014) determined that the relationship between CWB and intelligence was moderated by reporting source (i.e., self vs. non-self), such that the intelligence-CWB relationship was negative when CWB was measured with non-self-reports (possibly capturing overt, observable CWB) while it was positive when CWB was measured with self-reports (possibly capturing covert, undetected CWB). Echoing the suggestions made by Mercado et al. (2018), I believe that more nuanced measures of CWB would perhaps paint a more accurate picture of the intelligence-CWB relationship. 28 The second limitation associated with the studies reviewed here, and noted throughout, is that none of them tested causal mechanisms (i.e., mediators), despite proposing them. For example, Dilchert et al. (2007) argued that intelligence should be negatively related to CWBs because individuals higher in intelligence should have greater moral reasoning abilities as well as greater foresight to see the consequences of their actions (something these researchers call the inhibitory effect of intelligence). However, neither of these mechanisms were tested. These two potential explanatory mechanisms were echoed in later work by Gonzalez-Mulé et al. (2014), yet they still went untested. In addition to these two mechanisms, Gonzalez-Mulé et al. (2014) also offered something they called the “frustration-aggression cycle” as an explanation for the intelligence-CWB relationship, arguing that those lower in intelligence experience less success and, consequently, greater frustration, alienation, and rejection that, in turn, triggers aggression (i.e., a feedback loop). Finally, and as noted, Roberts et al. (2007) attempted to rationalize the positive relationship they found between intelligence and CWB by arguing that those higher in intelligence may have a greater sense of entitlement as well as greater access to resources they may steal (due to their positions in their organizations). Yet again, these proposed mechanisms went untested. Despite the inconsistent results and limitations associated with existing investigations into the intelligence-CWB relationship, I nevertheless argue that there is theoretical reason to believe there are systematic, significant relationships between intelligence and CWB. I say relationships, rather than relationship, because I argue that taking a nuanced view of CWB would allow researchers to detect the “true” relationships that exist between intelligence and CWB. That is, the inconsistent results obtained by researchers thus far may be a result of how CWB is measured (i.e., true relationships are obscured due to deficiencies in measurement; for 29 example, objective reports of CWB may only capture overt incidents). Furthermore, I believe that identifying and testing causal mechanisms may help to make these relationships clearer. Thus, I invoke research in criminology (much like Dilchert et al., 2007) to pinpoint a potential mediating mechanism of the intelligence-CWB relationship: episodic foresight. Rather than argue that episodic foresight acts solely as a deterrent of CWB, as the researchers of the studies just discussed would, I argue that episodic foresight deters overt, easily detectable CWBs but facilitates engagement in covert, less noticeable CWBs. 30 Episodic Foresight HYPOTHESIS DEVELOPMENT A common and well-supported argument in the criminology literature is that individuals higher in intelligence are less likely to engage in deviant behavior (Gottfredson & Hirschi, 1990; Herrnstein & Murray, 1994; Jensen, 1998; Mercado et al., 2018; White, Moffitt, & Silva, 1989; Wilson & Herrnstein, 1985). The belief is that the greater mental flexibility and abstraction skills associated with heightened intelligence enable individuals to (a) imagine future events and (b) foresee the negative consequences associated with their potential, deviant actions, thus inhibiting participation in illicit activities (Dilchert et al., 2007). That is, the general belief is that foresight, specifically, is the explanatory mechanism of the well-evidenced, negative relationship between intelligence and deviant criminal behavior. This proposition has come to be known as the inhibitory effect of intelligence, and it has received attention in scholarly research (Mercado et al., 2018), more generally, as well as in the limited OB/OP research that has sought to understand the nature of the intelligence-CWB relationship, specifically (e.g, Dilchert et al., 2007; Gonzalez-Mulé et al., 2014; Lyons et al., 2016; Marcus et al., 2009). Given the explanatory role theoretically fulfilled by foresight in the intelligence-CWB relationship, a detailed explanation of the construct is necessary. Although there are many different ways to conceptualize foresight (Amsteus, 2008), the conceptualization of foresight in research that leverages the inhibitory effect of intelligence is most similar to what may be called episodic foresight (Buckner & Carroll, 2007; Suddendorf, 2010). Whereas foresight, more generally, may be construed as a behavior or an ability (i.e., a noun; Amsteus, 2008), episodic foresight (EpF) is the mental process of imagining future events and projecting oneself into these possible futures (i.e., an action or process; a verb), such that one 31 can “pre-experience” an event before it occurs (Buckner & Carroll, 2007; Lyons et al., 2014; Vierra, 2016). Given that a process view of foresight most closely aligns with the conceptualization of foresight employed in theorizing on the inhibitory effect of intelligence, I similarly adopt a process view (i.e., EpF) and definition of foresight. Specifically, I borrow the definition of foresight provided by Slaughter (1996: 156): foresight is “a directed process which broadens the boundaries of perception through careful scanning of possible futures and the clarification of emerging situations.” Seemingly unique to primates (Suddendorf, 2006), EpF has been partially accredited with humans’ success as a species (Suddendorf & Corballis, 2007). This is because foresight allows humans to anticipate and plan for the future (i.e., it has survival value), something long recognized by both foresight researchers and management scholars (e.g., Dufty, 1961). Indeed, EpF is thought to enhance individual judgement, decision-making, and planning, and facilitate goal attainment as a vision of the future can serve as a useful guide for one’s behavior in the present (Atance & O’Neill, 2001; Suddendorf & Corballis, 2007). EpF is also thought to facilitate the assessment of potential consequences associated with particular actions, the proactive formulation of strategies, the detection of warning signs and avoidance of dangerous situations, and the visualization of desired future states (as well as paths for achieving them; Slaughter, 1996). Thus, EpF is useful to humans because it may (a) facilitate advantageous, strategic behavior, while also (b) inhibiting disadvantageous, risky behavior. Rephrasing that last sentence, EpF could act as both an inhibitor and a facilitator of behavior. Taking CWB as an example of such behavior, EpF could inhibit or suppress certain deviant actions (e.g., overt, easily observable CWB such as production deviance or personal aggression) given that EpF elicits the recognition of the consequences associated with these 32 potential actions. This logic aligns with the arguments made by proponents of the inhibitory effect of intelligence: intelligence facilitates foresight, which allows individuals to anticipate the consequences of their actions and, thus, deters participation in behaviors that would bring about these unfavorable consequences (i.e., punishment, public shame, etc.). However, I argue that EpF could also facilitate other deviant behavior (e.g., covert, less observable CWB such as property and political deviance) given that EpF enables strategy formulation and the visualization of paths to desired outcomes. Indeed, planning has been referred to as “the exercise of foresight” (Dufty, 1961: 51), such that it allows individuals to modify their future environment to meet their desires (Suddendorf & Corballis, 2007). This possibility directly challenges the logic underlying the inhibitory effect of intelligence. Given that foresight may have both inhibitory and faciliatory effects on individuals’ deviant behavior, I elaborate on each of these effects in the following subsections and formulate formal hypotheses. Episodic Foresight – Inhibition As explained, the inhibitory effect is based on the belief that intelligent individuals are better able to engage in EpF (i.e., foresight is the explanatory mechanism underlying the negative intelligence-criminal behavior relationship in criminology research). However, all OB/OP research that has leveraged the inhibitory effect to theorize a negative intelligence-CWB relationship has failed to test the mediating role of foresight, EpF (specifically) or otherwise. Additionally, some OB/OP researchers leveraging the inhibitory effect have mistakenly synonymized the concept of “foresight” with the concept of “self-control.” For example, Lyons et al. (2016) do not test mediation yet cite an assortment of criminology literature to argue that (a) there is positive relationship between intelligence and self-control, (b) self-control is negatively related to deviant behavior, and, thus, (c) there is evidence in support of the inhibitory 33 effect. Similarly, Gonzalez-Mulé and colleagues (2014) cite research on self-control (i.e., delayed gratification) in their defense of the inhibitory effect. This is problematic because these arguments overlook the fact that the inhibitory effect specifies foresight as the causal mechanism between intelligence and deviant behavior (criminal behavior or CWB), not self-control. Indeed, literature on the inhibitory effect explicitly states: “Intelligence has been suggested to possess an inhibitory effect with regard to deviant behaviors, in that individuals of above average and high intelligence possess enough foresight to consider possible consequences of their actions and subsequently choose those most beneficial to them (White, Moffitt, & Silva, 1989)…It has been argued that intelligence influences the ability to anticipate and evaluate possible consequences of one’s actions (Terman, 1916). Hence, the argument has often been connected with the observation that individuals low on cognitive ability seem to have limited time horizons (Lubinski, 2000). Such limited foresight makes it difficult for some individuals to grasp the serious implications of certain offenses (e.g., losing one’s job, facing a jail sentence), and thus they display poor decision making (i.e., engage in deviant behaviors)...” (Dilchert et al., 2007: 618-619) Foresight as the explanatory mechanism has been echoed in subsequent investigations of the intelligence-CWB relationship (e.g., Marcus et al., 2009), including in investigations by those individuals who have committed the fallacy of leveraging research on self-control as support for the existence of the inhibitory effect (i.e., Gonzalez-Mulé et al., 2014; Lyons et al., 2016). Though it is indeed possible that self-control curbs deviant behavior, and may even micro- mediate the relationship between EpF and deviant behavior, I argue that self-control itself is not the explanatory mechanism because it does not address the fundamental question a mediator is 34 meant to address: why a given relationship exists (Whetten, 1989). That is, if one were to use self-control as the explanatory mechanism in the intelligence-CWB relationship, the question as to why intelligence leads to self-control would remain. This question does not remain when EpF is used as an explanatory mechanism: intelligence enables EpF as those who are more intelligent have the abstraction skills and cognitive flexibility necessary to imagine alternative futures. As noted, however, I have chosen to focus explicitly on fluid reasoning, rather than general intelligence or one of the other many facets of intelligence, in this dissertation. As a quick reminder, fluid reasoning (a) reflects the ability to deliberately and flexibly control one’s attention in order to solve problems, think abstractly, see relationships, and reason (Schneider & McGrew, 2012), (b) embodies the core concepts underlying general intelligence (Frick et al., 2010; Otero, 2017), such that it shares several definitional similarities with well-accepted conceptualizations of general intelligence (Gottfredson, 1997), and (c) is more proximal to organizationally relevant outcomes than ability sets that are based on motor skills, sensory abilities, memory, speed, or crystalized, domain-specific knowledge. Thus, I argue that examining the causal relationship between fluid reasoning and EpF is likely to be the most fruitful and appropriate approach in understanding the inhibitory effect of intelligence. Accordingly, I hypothesize the following: Hypothesis 1: Fluid reasoning is positively related to episodic foresight. Although no studies in OB/OP have explicitly examined the link between EpF (or foresight, more generally) and behavioral inhibition in the context of CWB (i.e., the inhibitory effect), there is quite a bit of research that evidences a negative relationship between EpF and deleterious, impulsive behavior (i.e., acting without foresight) outside of OB/OP. Impulsive behavior “encompasses a range of actions which are poorly conceived, prematurely expressed, unduly risky or inappropriate to the situation and that often result in undesirable consequences” 35 (Daruna & Barnes, 1993). More succinctly, however, impulsive behavior can be defined as acting without foresight (Dalley, Everitt, & Robbins, 2011; Winstanley, Eagle, & Robbins, 2006), and research suggests that engagement in EpF can reduce undesirable behavior in a variety of contexts. For example, imagining the future may help reduce the impulsive use of alcohol (e.g., Bulley & Gullo, 2017), consumption of densely caloric food (e.g., Daniel, Said, Stanton, & Epstein, 2015; Daniel, Stanton, & Epstein, 2013), and participation in immoral behavior (Vierra, 2016). With this in mind, I argue that EpF inhibits engagement in certain forms of CWB, particularly overt forms as these CWBs tend to be more (a) impulsive and (b) likely to result in negative consequences. Although Robinson and Bennett (1995) specify two dimensions underlying their taxonomy of CWBs (minor-serious and organizational-interpersonal), I argue that CWBs also differ in terms of their overtness, such that some deviant acts are more easily observable and, hence, punishable than others (i.e., consequences are more of a “sure thing”). Importantly, I am not the first to make this claim – other researchers have similarly argued that CWBs may differ in terms of their conspicuousness (i.e., Spector & Fox, 2002; Zhang, Wan, Zhao, & Bashir, 2011). For example, leaving work early, taking excessive breaks, or showing up late (i.e., behaviors encompassed by production deviance), and sexual harassment, verbal abuse, or endangerment of coworkers (i.e., behaviors encompassed by personal aggression) are more predisposed to impulsivity and overt than other forms of deviance. Therefore, they are more likely to be observed and punished. Individuals of higher intelligence may be acutely aware of the consequences associated with these more compulsive, overt behaviors, and thus avoid engaging in them. 36 This prediction is consistent with Dilchert et al.’s (2007) finding that intelligence shares a negative relationship with formally recorded incidents of CWB, as the recording of CWB implies its detection by others. This prediction is also consistent with Marcus et al.’s (2009) finding that intelligence shares a negative relationship with tardiness, and Gonzalez-Mulé et al.'s (2014) finding that the intelligence-CWB relationship is moderated by rating source, such that the relationship is negative when using non-self-report measures of CWB. Thus, I formally hypothesize that EpF is negatively related to overt, observable CWB because overt CWB tends to be more impulsive and detectable by others (i.e., consequences are more of a certainty): Hypothesis 2: Episodic foresight is negatively related to engagement in overt CWB. Episodic Foresight - Facilitation As noted, EpF may not only have inhibitory effects on behavior. In fact, EpF may also facilitate behavior. Indeed, an abundant body of research, particularly in the discipline of developmental psychology (Suddendorf & Redshaw, 2013), suggests that EpF influences predetermined, goal-directed behavior, such that, “the tendency to anticipate and prepare for the potential future,” or EpF, “is central to both routine and adaptive planning on a daily basis” (Lyons, Henry, Rendell, Corballis, & Suddendorf, 2014: 873). Indeed, it flows naturally that EpF, or the process of imagining oneself in future scenarios, should facilitate one’s imminent and subsequent actions in order to control one’s future (Suddendorf, 2017). Thus, EpF enables individuals to shape their activities in a way that allows them to secure future benefits while also avoiding disastrous situations (Suddendorf & Moore, 2011). Accordingly, I argue that EpF may facilitate engagement in certain forms of CWB, particularly those that are inclined to being pre-planned and conducted covertly. While tardiness, absentness, sexual assault, and verbal harassment are fairly impulsive and observable, theft and misuse of company equipment (i.e., behaviors encompassed by property deviance), as well as 37 blame-shifting and gossiping (i.e., behaviors encompassed by political deviance) can be done in a much more secretive fashion. Given that they may be undetected and, thus, go unpunished, those who are better-equipped to engage in EpF (i.e., those with heightened levels of fluid reasoning), may see ways to avoid consequences and capitalize on opportunities to engage in them. This prediction is consistent with Roberts et al.'s (2007) findings suggesting that intelligence is positively related to theft of resources and misuse of company equipment. This prediction is also consistent with the evidence provided by Marcus et al. (2002), who found a positive relationship between intelligence and CWB, as well as the evidence Gonzalez-Mulé et al. (2014) provide regarding rating source (i.e., a positive relationship between intelligence and CWB when using self-reported measures, rather than non-self-reported measures). Thus, I formally hypothesize that EpF is positively related to covert, less observable CWB because EpF enables the pre-planning of such behavior (and in fact requires it if these behaviors are to be done with any sort of frequency): Hypothesis 3: Episodic foresight is positively related to engagement in covert CWB. Finally, I argue that episodic foresight mediates the negative relationship between fluid reasoning and overt CWB as well as the positive relationship between fluid reasoning and covert CWB. More formally: Hypothesis 4: Episodic foresight mediates the negative relationship between fluid reasoning and overt CWB. Hypothesis 5: Episodic foresight mediates the positive relationship between fluid reasoning and covert CWB. Thus far I have focused explicitly on an ability (i.e., fluid reasoning), the mediating cognitive process (i.e., episodic foresight), and behavioral outcomes (i.e., CWB) without 38 consideration of potential boundary conditions. As such, I next draw upon psychological theories of criminal behavior, from which the inhibitory effect of intelligence was derived (Marsh, 2011), to identify two potential moderating mechanisms relevant to the study of CWB: moral identity and power. 39 MODERATING ROLES OF MORAL IDENTITY AND POWER Although there is reason to believe that heightened levels of fluid reasoning provide individuals with the ability to (a) anticipate the consequences associated with overt deviant behavior and (b) pre-plan covert deviant behavior, fluid reasoning, or even intelligence more broadly, does not speak to one’s dispositional tendency to actually consider or plan deviant acts. Thus, in the following sections I review the literature on moral identity, and argue that the two dimensions of moral identity, symbolization and internalization (Aquino & Reed, 2002), moderate the path between fluid reasoning and EpF. I have elected to focus on moral identity, specifically, for three reasons. First, it has been identified as an individual difference that is consistently predictive of unethical behavior (Hertz & Krettenauer, 2016; Shao, Aquino, & Freeman, 2008; Trevino, den Nieuwenboer, & Kish- Gephart, 2014). Second, it consists of both a “public” (i.e., symbolization) and a “private” (i.e., internalization) dimension (Aquino & Reed, 2002), which, I argue, are particularly relevant given the arguments I have developed thus far regarding overt (public, more discernable) and covert (private, less discernable) forms of deviant behavior. Finally, morality (and, specifically, moral reasoning) has historically been positioned as a major psychological explanation for criminal behavior, much like intelligence (Marsh, 2011). Given that OB researchers have leveraged the criminology literature to justify the theoretically negative relationship between intelligence and CWB (i.e., Dilchert et al., 2007), and that intelligence has been positioned as a psychological explanation for criminal behavior (Marsh, 2011), I have elected to draw my moderating mechanisms from this same realm of thought. 40 The Moderating Role of Moral Identity Prior to the advent of moral identity as a theoretical construct, moral reasoning was presumed to be one of the driving forces behind individuals’ ethical behavior (Rest, Narvaez, Thoma, & Bebeau, 2000; Shao et al., 2008). Moral reasoning, or the process by which individuals differentiate right from wrong through the use of logic, is considered to be the key construct underlying the cognitive developmental model of morality first proposed by Kohlberg in 1969 (Shao et al., 2008; Trevino et al., 2014). According to this model of morality, which has largely been applied in the context of child development (Marsh, 2011), there is a “culturally universal” and “invariant” sequence of stages of moral development individuals undergo (Kohlberg, 1973: 630) that depend, at least to some degree, on one’s cognitive and perspective- taking capacities (Aquino & Reed, 2002). Moral Reasoning Per Kohlberg (1973), there are three levels of moral development, each of which are characterized by two stages (for a total of six stages). These stages of moral development are as follows. First, individuals are at the “preconventional level” of moral development. In the first stage of the preconventional level (i.e., the punishment-and-obedience stage), individuals determine “right” actions from “wrong” actions by the consequences associated with those actions. That is, individuals defer to the power of authority at this stage of moral development, presuming that punishment and severity of punishment by authority figures are indicative of the morality of the behavior that led to the punishment. In the second stage of the preconventional level (i.e., the instrumentalist-relativist stage), individuals determine “right” from “wrong” by presuming that that which best serves their personal needs is “right.” In other words, “moral” behavior consists of behavior that benefits the individual actor. 41 The next two stages of moral development are at the conventional level. At the first stage of the conventional level (i.e., the interpersonal concordance, or “good boy – nice girl,” stage), individuals infer “right” from “wrong” by assuming that that behavior which is approved by and pleases others must be “right.” Accordingly, individuals at this stage of moral development actively conform and, thus, exhibit behavior that reflects those engaged in by the majority. At the second stage of the conventional level (i.e., the “law and order” stage), “right” behavior involves respecting authority, maintaining social order, and following fixed rules. That is, moral behavior involves blind adherence to laws, regulations, and so forth. The final two stages of moral development are at the postconventional level. At the first stage of the postconventional level (i.e., the social-contract legalistic stage), individuals begin to think more independently about “right” and “wrong,” and are more open to considering different ideas and viewpoints regarding morality. Individuals believe that they should engage in behavior that conforms to society’s consensual interpretation of what is “right” or “wrong,” and understand that rational changes to laws should not be off the table. At the final stage of moral development (i.e., the universal-ethical-principle stage), ideas regarding universal moral principles have formed and will be considered “right” no matter what the laws of a society may be. These universal principles include concepts such as justice, reciprocity, equality of rights, and respect for human dignity, and laws are only seen as valid insofar as they are grounded in these ostensibly universal principles. Though highly influential, Kohlberg’s cognitive developmental model of morality was (and still is) not without its criticisms. For example, a major criticism of this theory is the assumption that there is consensus regarding deontic principles when, in reality, there is no such consensus (Rest at al., 2000). As another example, there are concerns that the stages proposed by 42 Kohlberg are somewhat arbitrary, such that individuals do not necessarily need to pass through them sequentially (Marsh, 2011). Most pertinent to the present line of inquiry, however, is that neither Kohlberg’s (1973) cognitive developmental model of morality nor the refinement of this model by Rest and colleagues (i.e., the "Neo-Kohlbergian Approach"; Rest, Narvaez, Bebeau, & Thoma, 1999; Rest et al., 2000) are strongly associated with actual ethical tendencies (Shao et al., 2008). That is, although it is possible that individuals differ in terms of their propensity to engage in moral reasoning, evidence largely suggests that measures of this ability are not predictive of behavioral outcomes. Given moral reasoning’s lack of predictive validity, scholars eventually turned their attention towards other theories of moral functioning. One of the most influential understandings of moral motivation and behavior to be borne by this trend was the social-cognitive perspective associated with moral identity (Aquino & Reed, 2002; Jennings, Mitchell, & Hannah, 2015). Moral identity, in essence, reflects the extent to which morality is central to an individual’s self-concept, or identity (Hardy & Carlo, 2011), and it is considered to be a sort of self-regulatory mechanism that motivates moral action (i.e., behavior) (Aquino & Reed, 2002). Recognizing that moral identity shares conceptual similarities with moral reasoning, Aquino and Reed (2002) differentiate the two on the basis that moral reasoning is developmentally-oriented (and, perhaps, partially a product of cognitive ability) while moral identity is an aspect of self- concept. Whereas the moral reasoning literature implies that the motivation to engage in moral behavior is at least somewhat contingent upon one’s cognitive development or capacity (Rest, 1979), the motivational driver underlying behavior implied by the moral identity literature is the extent to which individuals consider morality as important to their sense of “self.” That is, cognitive development presumably influences one’s ability to engage in moral reasoning, but the 43 extent to which one considers moral behavior as essential to one’s self-concept (i.e., his or her moral identity) is independent of his or her cognitive capacities. Given that (a) moral reasoning is less predictive of behavior than moral identity, and that (b) measures of moral reasoning may be confounded with intelligence, my theorizing henceforth will focus on moral identity. Moral Identity As noted, moral identity reflects the extent to which being a moral person is important to an individual’s self-concept or identity (Hardy & Carlo, 2011). Accordingly, moral identity has roots in both social identity theory (Ashforth & Mael, 1989) and social cognitive theory (Bandura, 1986; Trevino et al., 2014), which, taken together, gave rise to what may be called the “social-cognitive perspective” of morality (Shao et al., 2008). Consistent with the definition previously provided, the social-cognitive perspective conceptualizes moral identity as a cognitive schema of values, goals, traits, and behavioral scripts (Aquino, Freeman, Reed, Lim, & Felps, 2009). Put quite simply, one’s moral identity is an aspect of one’s self-definition. Relative to the theorized relationship between moral reasoning and behavior (Shao et al., 2008), empirical evidence more strongly supports the notion that higher levels of moral identity are associated with greater engagement in behaviors that may be considered moral and ethical. Indeed, meta-analytic and, naturally, primary research suggests that those who have higher levels of moral identity are more likely to engage in prosocial and ethical behaviors, and avoid antisocial and deviant behaviors (Hertz & Krettenauer, 2016; Shao et al., 2008). That said, some empirical research that has examined moral identity’s relationship with various outcomes has found that teasing apart the two dimensions of moral identity, symbolization and internalization (Aquino & Reed, 2002), may prove insightful as these two dimensions may have different relationships with outcomes of interest (e.g., Mayer, Aquino, Greenbaum, & Kuenzi, 2012; 44 Skarlicki, Jaarsveld, & Walker, 2008; Vitell et al., 2009; Winterich, Aquino, Mittal, & Swartz, 2013). Symbolization refers to the more “public” aspect of moral identity, namely the extent to which the importance an individual attaches to his or her perceived morality manifests, overtly, in social settings (Aquino & Reed, 2002; Gotowiec & van Mastrigt, 2018). It has previously been argued that individuals who score high on this dimension of moral identity are motivated to engage in moral actions out of a desire for self-verification (i.e., they attempt to confirm their moral identity through the recognition and acknowledgement of their moral behavior by others; Winterich et al., 2013), rather than simply because behaving morally is important to their self- concept. As such, those that score high in moral identity symbolization should engage in behaviors that indicate their moral character to others and avoid behaviors that would leave others with poor impressions. Consistent with this line of theorizing, Winterich and colleagues (2013) provided evidence, across two studies, suggesting that higher levels of moral identity symbolization coincide higher levels of prosocial behavior, provided that that behavior may be recognized by others. In the first study, these researchers determined that individuals high in moral identity symbolization (i.e., +1 SD) were up to five times more likely than those low in moral identity symbolization (i.e., -1 SD) to complete a voluntary survey when they were told that they would be recognized for their actions (i.e., their names would be listed on a public website). However, moral identity symbolization was not predictive of this behavior when there was no recognition for the completion of this survey. These researchers replicated the results of their first study in their second study. In this second study, prosocial behavior was operationalized as the voluntary completion of various 45 anagrams, with each successfully completed anagram resulting in a $0.05 donation (on behalf of the participant) to Feeding America (a system of food banks). Winterich et al. (2013) determined that heightened levels of moral identity symbolization were associated with the completion of a larger number of anagrams and, consequently, greater donations when participants were told they would be recognized for their actions (i.e., their names would be listed on a public website, much like the first study). However, moral identity symbolization was unrelated to anagram completion and, thus, donations when participants were not told they would be recognized for their actions. Thus, evidence from two studies suggests that higher levels of moral identity symbolization may result in desirable behaviors when this behavior is observable by others, but not when this behavior is unobservable. In a somewhat inversed extension of this logic, I argue that moral identity symbolization moderates the positive relationship between fluid reasoning and EpF (in terms of identifying potential consequences of CWB; the inhibitory function), strengthening it, as “traits can increase the accessibility of cognitive concepts” and thus affect how individuals process information (DeCelles, DeRue, Margolis, & Ceranic, 2012: 682). Indeed, DeCelles and colleagues (2012) argued and provided evidence that individuals with heightened moral identities may be particularly conscious of the moral implications associated with a given action as moral concepts are readily accessible to these individuals. As it pertains to my theorizing, those higher in moral identity symbolization should be especially conscious of the consequences associated with overt deviant behavior because they are keenly aware that their actions may be observed and evaluated by others. To reiterate previous theorizing, the negative relationship between overt CWB and fluid reasoning is mediated by EpF, such that individuals higher in fluid reasoning are more likely to 46 foresee the negative consequences associated with engagement in easily observable behaviors. I argue that the first path of this indirect effect, the positive relationship between fluid reasoning and EpF, is strengthened by moral identity symbolization because individuals with above average levels of symbolization are (a) very much aware that others monitor their moral/immoral behavior, and (b) are concerned with how others’ interpretation of this behavior reflects on them as an individual. Put differently, those high in moral identity symbolization are cognizant of the fact that their behavior may be observed by others, and are explicitly concerned with being perceived as a moral person. Therefore, the negative consequences associated with overt, deviant actions should be much more salient to these individuals. Thus, I formally hypothesize that moral identity symbolization moderates the path between fluid reasoning and EpF, and the indirect relationship between fluid reasoning and overt CWB, given the greater salience of potential consequences to those with heightened levels of moral identity symbolization: Hypothesis 6: Moral identity symbolization moderates the positive relationship between fluid reasoning and episodic foresight, such that the relationship is stronger when moral identity symbolization is high and weaker when moral identity symbolization is low. Hypothesis 7: Moral identity symbolization moderates the negative, indirect relationship between fluid reasoning and overt CWB, via episodic foresight, such that the indirect relationship is stronger when moral identity symbolization is high and weaker when moral identity symbolization is low. In contrast to symbolization, internalization refers to the more “private” aspect of moral identity, namely the extent to which morality, and thus engaging in moral behavior, is considered foundational to an individual’s sense of self (Aquino & Reed, 2002; Gotowiec & van Mastrigt, 2018). It has been argued by prior researchers that individuals who score high on this dimension 47 of moral identity are motivated to engage in moral behavior because engagement in immoral behavior would be inconsistent with their self-concept, and thus result in psychological distress (Winterich et al., 2013). As such, those that score high in moral identity internalization should engage in moral behaviors not because they are concerned with the impression they give to others (i.e., the interpretation of their moral character by others), but because moral behavior is consistent with their self-definitions. Given the importance of morality to the self, moral identity internalization is associated with moral behavior simply because immorality is against the individual’s nature. Consistent with this line of reasoning, internalization may be a more reliable predictor of desirable behaviors than symbolization, which may be contingent upon the behavior’s overtness. Indeed, Winterich and colleagues (2013), discussed in the prior section on moral identity symbolization, provided evidence that individuals who score high on internalization tend to engage in voluntary prosocial behavior (i.e., the completion of surveys and anagrams) regardless whether they were told that their actions will be recognized by others. Relatedly, some research has elected to focus exclusively on internalization given its predictive validity. For example, Aquino et al. (2009) operationalized “moral identity centrality” using the internalization dimension of moral identity and demonstrated that, across four studies, this measure was predictive of various moral behaviors (i.e., willingness to initiate a cause-related marketing program, avoid lying to a job candidate, and contribute to a public good), particularly when moral self-schemas were activated (i.e., when individuals’ moral identities were primed). In another (somewhat inversed) extension of this logic, I argue that moral identity internalization moderates the positive relationship between fluid reasoning and EpF (in terms of pre-planning CWB; the faciliatory function), weakening it, as those higher in moral identity 48 internalization are simply less adept at planning deviant acts. To reiterate previous theorizing, the positive relationship between covert CWB and fluid reasoning is mediated by EpF, such that individuals higher in fluid reasoning are better able to anticipate the future, imagine themselves in scenarios, and, consequently, plan deviant behavior. I argue that the first path of this indirect effect, the positive relationship between fluid reasoning and EpF, is weakened by internalization because immoral concepts are not as readily accessible to individuals with above average levels of internalization (DeCelles et al., 2012). If these individuals were prompted to consider engagement in covert deviant behavior, the scenarios they envision would not be particularly vivid or well-conceived as these individuals are not predisposed to even consider such behavior – it is simply not in their nature. In a sense, moral identity internalization acts as a hindrance to EpF (as it pertains to the planning of CWB). Thus, I hypothesize that moral identity internalization moderates the path between fluid reasoning and EpF, and therefore the indirect relationship between fluid reasoning and covert CWB, such that internalization acts as a hindrance factor. More formally: Hypothesis 8: Moral identity internalization moderates the positive relationship between fluid reasoning and episodic foresight, such that the relationship is weaker when moral identity internalization is high and stronger when moral identity internalization is low. Hypothesis 9: Moral identity internalization moderates the positive indirect relationship between fluid reasoning and covert CWB, via episodic foresight, such that the indirect relationship is weaker when moral identity internalization is high and stronger when moral identity internalization is low. I do not predict a moderating effect of moral identity internalization on the indirect effect of fluid reasoning on overt CWB because the theory underlying that indirect effect involves the 49 foresight of consequences (the “inhibitory effect”). That is, there is no theoretical reason to believe that moral identity internalization should strengthen or weaken this relationship because internalization reflects one’s “true” moral character – it should not increase or decrease one’s cognizance of (or the salience of) potential punishments. That said, I acknowledge that there is likely to be a negative direct effect of moral identity internalization on overt CWB given the abundant literature on the relationship between moral identity and moral behavior (Aquino et al., 2009; Hardy & Carlo, 2011; Hertz & Krettenauer, 2016; Jennings et al., 2015; Shao et al., 2008) Similarly, I do not predict a moderating effect of moral identity symbolization on the indirect effect of fluid reasoning on covert CWB because the theory underlying that indirect effect is based upon arguments that these behaviors can be done inconspicuously. That is, there is no theoretical reason to believe that moral identity symbolization should strengthen or weaken this relationship because symbolization involves concerns regarding public image and observed behavior (not private, unobserved behavior). Once again, however, I acknowledge that moral identity symbolization may have a negative direct effect on covert CWB per the literature on the relationship between moral identity and moral behavior. In sum, I argue that the negative indirect effect between fluid reasoning and overt CWB is exclusively moderated by moral identity symbolization, strengthening it, while the positive indirect effect between fluid reasoning and covert CWB is exclusively moderated by moral identity internalization, weakening it. In the following section, I draw upon research on power to argue that differences in autonomy, access to resources and information, and feelings of invulnerability associated with various hierarchical positions may attenuate or exacerbate the previously hypothesized, direct relationships between EpF and CWB. Succinctly, I will argue that power enables individuals to engage in a variety of CWBs. This proposition is aligned with the previously noted, post hoc 50 claims made by Roberts et al. (2007), who argued that the positive relationship they identified between intelligence and theft may be partly due to the tendency for more intelligent individuals to accrue positions of power in organizations. The Moderating Role of Power Although fluid reasoning may provide individuals with the ability to pre-plan covert deviant behavior and foresee the consequences associated with overt deviant behavior, and moral identity internalization and symbolization may affect an individual’s proclivity to engage in the associated cognitive processes (i.e., EpF), I argue that none of these constructs act as enabling factors. For example, it is possible that individuals with above average levels of fluid reasoning and below average levels of moral identity internalization may be both willing and able to steal from the company, but they may not be enabled to do. Thus, I next review the literature on power (a foundational base of hierarchy; Magee & Galinsky, 2008) and argue that it moderates the relationships between EpF and CWB, strengthening the positive relationship (i.e., covert CWB) and weakening the negative relationship (i.e., overt CWB). I have elected to focus on power, specifically, for two reasons. First, the psychological experience of power motivates action (Keltner, Gruenfeld, & Anderson, 2003), including both prosocial and self-interested behavior (Galinsky, Gruenfeld, & Magee, 2003). That is, power does not necessarily lead to “good” or “bad” behavior so much as it elicits action, more generally. Second, and relatedly, Keltner and colleague’s (2003) power-approach theory explains that power elicits disinhibited behavior, and psychological explanations for criminal behavior often point to impulsivity, a concept previously discussed, as an antecedent of such behavior (Marsh, 2011). Given that the experience of power may lead individuals to engage in disinhibited, agentic behavior (particularly in the presence of a weak moral identity; DeCelles et 51 al., 2012), it may allow for more impulsive tendencies to emerge. Thus, in the following sections I draw liberally from power-approach theory to argue that power enables and incentivizes CWB, supplementing this theory with additional logic and research where appropriate. Hierarchy Hierarchy, which may be defined as the differentiation of individuals along some socially valued dimension (Anderson & Brown, 2010), is a pervasive social construct. Indeed, hierarchical social structures are evident in various species (Diamond, 1999), tend to emerge quickly and naturally in most groups (Bales, 1958), and have historically played a major role in shaping human societies (Watts, Sheehan, Atkinson, Bulbulia, & Gray, 2016). As a result, hierarchy has been studied extensively across social science disciplines, including psychology, sociology, economics, and, importantly, organizational behavior (Anderson & Brown, 2010). The study of hierarchy is particularly relevant to organizational behavior as work organizations, which are themselves social constructions, are inherently hierarchical. There are so frequently rank orderings among organizational actors in terms of some social dimension (e.g., power; Gruenfeld & Tiedens, 2010), that some individuals have gone so far as to suggest that hierarchies may be universal features of organizational life (Anderson & Brown, 2010). Indeed, a simple observation of organizational charts, which literally define the structure of organizations, provide evidence to support this contention – organizational charts are traditionally organized hierarchically, with those individuals in positions of greater power situated higher on the chart than those of lesser power. Undoubtedly, power, or the extent to which an individual has asymmetric control over resources, is a very salient, foundational base of hierarchy made apparent by these charts (Magee & Galinsky, 2008). 52 Given the relevance of hierarchies based on power (i.e, resource) differences to organizations, I will discuss power, explicitly. Specifying which base of hierarchy will be the focus of my attention and theorizing is necessary as there are variety of social dimensions that may serve as bases of hierarchy, many of which are neither based upon formal positions of authority nor the structure of the organization like power (Magee & Galinsky, 2008). Hence, in the following section I will reiterate the conceptualization of power that will be the focus of my theorizing herein, highlight research that speaks to the enabling capacity of power (i.e., power- approach theory), and, finally, explain how power may moderate the paths between EpF and CWB. Power Power, which has also been defined as a function of resource dependency (i.e., Actor A’s power over B is a function of Actor B’s dependence on Actor A; Emerson, 1962), is considered to be property of the actor and one of the most foundational bases of hierarchy in organizational contexts (Magee & Galinsky, 2008). Given that power is a property of the actor (i.e., it does not have to be socially conferred by others), it is less mutable (Hays & Bendersky, 2015) and may motivate less selfless, prosocial behavior (Anderson & Brown, 2010; Blader & Chen, 2012) than other bases of hierarchy (e.g., status, which is a property of observers; Magee & Galinsky, 2008). That is, because one’s power (a) involves direct control over resources and (b) is less easily taken away than other hierarchical bases (e.g., status) it allows for more agentic behavior (Abele & Wojiciske, 2014). Relatedly, research suggests that the experience of power motivates and enables individuals to take action (e.g., Galinsky et al., 2003; Keltner et al., 2003). Partially motivated to address inconsistent findings regarding the relationship between power and various behaviors 53 (i.e., behaviors that would be considered “immoral” versus “moral”), Galinsky and colleagues (2003) conducted a series of experiments and consistently demonstrated that power motivates action, in a broad sense. In their first experiment, the researchers primed power by randomly placing participants into one of two roles, a “manager” or a “subordinate” role ostensibly based upon their responses to a “Leadership Questionnaire,” in the context of a Lego-building task. After priming participants, but before beginning the task, participants took part in a simulated blackjack scenario (which participants were told was part of another, separate experiment). The researchers ultimately found that those participants given the role of “manager” were significantly more likely, χ2 (1, N = 32) = 4.07, p = .04, than those assigned the role of “subordinates” to take a hit (i.e., an additional card) when given two cards that total to an amount that presents a particularly “vexing” situation. Specifically, 92% of the “managers” took the hit while only 58% of “subordinates” took the hit. Galinsky and colleagues (2003) replicated these findings in their second study. In the second study, the researchers primed power via a recall task – those in the high power condition were instructed to recall a particular incident in which they had power over others and write about it, while those in the low power condition were instructed to recall a particular incident in which someone had power over them and write about it. After this prime, participants (a) began the completion of a resource allocation task that reinforced these manipulations, and (b) were situated in front of an “annoying” fan that was blowing directly on them. Researchers determined that those participants in the high power condition were more likely to turn off the fan than those who were in the low power condition, χ2 (1, N = 59) = 4.21, p = .04, such that they were more than twice as likely to turn off the fan than ignore it (69% versus 31%). 54 In their final study, Galinsky et al. (2003) once again replicated the finding that participants primed with power were more likely to take action. The researchers primed high and low power using the same task as they did in the second study (i.e., narrative essays) and examined power’s relationship with both prosocial and antisocial behavior (i.e., contributing to versus taking from a resource pool). The researchers determined that those primed with high power were more likely to contribute to and take from a shared resource pool than those participants who were primed with low power, t(152) = 2.69, p = .01, or those situated in the non-primed, control condition, t(152) = 2.04, p = .04. Furthermore, these researchers found no evidence implying that power interacted with social dilemma type (i.e., prosocial or antisocial), suggesting that power motivates action, more generally, rather than prosocial or antisocial behavior, specifically. Additionally, and perhaps anecdotally, these researchers did not find evidence suggesting that those in the low power condition differed significantly from those in the control condition in terms of action orientation. Thus, across three studies Galinsky et al. (2003) empirically demonstrated that power motivates action. Incidentally, Keltner and colleagues (2003) published their power-approach theory that same year. Much like Galinsky et al. (2003), Keltner and colleagues reviewed a large body of literature evidencing the relationships power has with various positive and negative behaviors, and ultimately argued that the experience of power does not necessarily lead to “good” or “bad” behavior, but rather motivates action, broadly. More precisely, these researchers argued that power can transform individuals’ psychological states, such that elevated power (i.e., greater resources and freedom/non-dependence on others) coincides greater attention to rewards, positive emotions, automatic cognitions, and, importantly, disinhibited behiavior. On the other hand, reduced power (fewer resources and less freedom/more dependence on others) coincides 55 greater attention to threats, negative emotions, systematic and controlled cognition, and inhibited behavior. Put simply, high-power individuals tend to exhibit approach-oriented tendencies while low-power individuals tend to exhibit avoidance-oriented tendencies because power is liberating. In sum, there is reason to believe that power enables action. Although I have discussed only two publications in detail in this section, one empirical (i.e., Galinsky et al., 2003) and one theoretical (i.e., Keltner et al., 2003), both were motivated and informed by a slew of primary research evidencing relationships between power and various positive (e.g., Chen, Lee-Chai, & Bargh, 2001), negative (e.g., Georgesen & Harris, 1998, 2000), and neutral (e.g., Overbeck & Park, 2001) behaviors. Given that power elicits approach-oriented tendencies and, therefore, action, I argue that power moderates the previously hypothesized paths between episodic foresight (EpF) and counterproductive work behavior (CWB). Throughout the following sections I invoke power-approach theory, but, as noted, I supplement this theoretical rationale with additional evidence and logic. First, I argue that power moderates the positive relationship between EpF and covert CWB. As a reminder, covert CWB includes behaviors that would be generally be categorized as property deviance (e.g., theft, misuse of company property) and political deviance (e.g., undermining, blame-shifting) in the Robinson and Bennett (1995) typology. I argue that power strengthens this relationship not only because power coincides greater action (per power- approach theory), but also because power enables and incentivizes such behavior. Focusing first on those behaviors typically nested under property deviance, power, by definition, provides individuals with greater access to organizational resources. As noted, power may be succinctly defined as control over resources (Magee & Galinsky, 2008). Thus, in organizational contexts, individuals in positions of power have greater command over, and 56 opportunity to misuse, capital, raw materials, equipment, and other resources of material value. Though the exact nature of these “resources” varies by organization and role, they could include anything from company property (e.g., equipment and merchandise) to liquid assets (e.g., physical cash). Considering that positions of power provide individuals with greater, and perhaps direct, access to resources that can be abused, stolen, or deliberately damaged, it naturally follows that positions of power enable deviant behaviors that positions lacking power may not – it is easier to steal, destroy, and misuse resources when one can actually access them. In a sense, power acts as a faciliatory factor; one can engage in greater deviance related to company resources in positions of power because power puts these resources at the individual’s fingertips. This, coupled with the goal-directed actions that the experience of power typically elicits (Keltner et al., 2003), leads me to the argument that power enables covert deviance related to company assets. Shifting the focus to political deviance, power provides individuals with an incentive to undermine or take credit for the accomplishments of others, gossip about others, and engage in other political behavior. Indeed, there are a considerable number of benefits associated with high-power positions, including disproportionate credit and rewards, satisfaction of core needs (e.g., control and autonomy), and, often, status (Magee & Galinsky, 2008). Furthermore, because high-power individuals often pay greater attention to rewards (Keltner et al., 2003) they may be more conscious of the benefits that their positions entail (relative to those that are low in power, who tend to be characterized by greater attention towards threats; Keltner et al., 2003). Thus, they may be acutely aware of the benefits they enjoy and have an incentive to maintain their positions. 57 Consequently, high-power individuals may have greater reason to engage in a variety of behaviors that would usually be categorized as political deviance, a covert form of deviance. In a sense, power incentivizes those that have it to socially undermine others (e.g., gossiping) and refocus attention off their own faults by blaming others for their mistakes, whereas those without it have less reason to do so. In support of this line of reasoning, research suggests that high- power individuals not only seek to manipulate others, particularly the less powerful, but also devalue the performance of low-power individuals (Kipnis, 1972). This, coupled with the goal- directed actions that the experience of power typically elicits (Keltner et al., 2003), leads me to the argument that power acts as an incentive to engage in covert deviance related to more “political” behaviors. Taken altogether, this leads me to my hypothesis that power moderates the positive relationship between episodic foresight and covert CWB, such that the relationship is stronger when power is high and weaker when power is low. More formally: Hypothesis 10: Power moderates the positive relationship between episodic foresight and covert CWB, such that the relationship is stronger when power is high and weaker when power is low. I also argue that power moderates the negative relationship between EpF and overt CWB. As a reminder, overt CWB includes behaviors that would generally be categorized as production deviance (e.g., leaving work early, coming in late) and personal aggression (e.g., verbal abuse, harassment) in the Robinson and Bennett (1995) typology. I argue that power weakens this relationship not only because power coincides greater action (per power-approach theory), but also because high-power individuals (a) have greater independence, (b) exhibit more confidence, and (c) have a reduced aversion to loss. 58 Focusing first on production deviance, it has long been argued that elevated power signifies independence from other parties while a lack of power signifies dependence on other parties; an individual’s power over another is a function of the latter’s reliance on the former (Emerson, 1962). As a natural result, power is liberating – powerful individuals are not reliant upon the less powerful to dole out rewards or punishments, hence why elevated power is associated with freedom, disinhibition, and, relatedly, action (Galinsky et al., 2003; Keltner et al., 2003). In organizational contexts, individuals such as managers have greater power over their subordinates because they control the positively (i.e., rewards) and negatively (i.e., punishments) valued “resources” that employees receive (Magee & Galinsky, 2008). However, subordinates generally do not have direct power over managers. Accordingly, those in power have greater discretion in terms of the behavior they engage in, and are therefore enabled to act in ways that those of lesser power may face consequences for. In a sense, power is a countervailing force to the inhibitory effect. This is not to say that egregious acts by the powerful bear no consequences, but rather that less serious, overt CWBs, such as leaving early or coming in late, are more likely to go unpunished as there are fewer control mechanisms in place to prevent such behavior. While these arguments align with power-approach theory, let us assume for a moment that such behaviors are easily punishable. Even when production deviance is easily punishable across all levels, I argue that high-power individuals may still be more likely to engage in such behavior because power often begets heightened confidence and risk tolerance. Indeed, research suggests that the experience of power can lead to overconfidence (Fast, Sivanathan, Mayer, & Galinsky, 2012; See, Morrison, Rothman, & Soll, 2011) and reduced loss aversion (Inesi, 2010), such that high-power individuals are more self-assured in their actions and are less concerned about the negative consequences of these actions. That is, the powerful may believe that they can 59 “get away” with mild, overt deviant behavior and that, if “caught,” the consequences associated with their actions (i.e., their “losses”) will be tolerable. Therefore, they are more likely to enact these behaviors – the inhibitory function of EpF is undermined. Shifting the focus to personal aggression, power may enable such behavior because those in power are less likely to face retaliation from their victims. As noted in the explanation for the moderating role of power on the EpF-production deviance relationship, high-power individuals are less dependent on others, particularly those individuals of less power than themselves (Emerson, 1964). Therefore, they may feel enabled to engage in aggressive behavior because rewards are less likely to be withheld (by their victims), punishments are less likely to be doled out (by their victims), and high-power perpetrators may wield resources of value over their victims. That is, they may be less concerned about potential retaliation because their victims (a) cannot directly punish them and (b) do not want resources withheld from themselves as a product or their retaliation. That said, this logic may be more descriptive of aggressive behavior directed at low- power individuals given that these are the individuals that high-power individuals are presumably less dependent on and have greater control over. Hence the need for another explanation as for why high-power individuals may be more inclined to engage in personal aggression: high-power individuals are less concerned with saving face. Indeed, research suggests that high-power individuals tend to be less preoccupied with the “face-threatening” potential of their actions (Brown & Levinson, 1987). In other words, high-power individuals are less concerned with how others perceive them (i.e., their public self-image, or “face”; Goffman, 1959) than their low-power counterparts, and therefore may not behave as kindly towards others. 60 In support of this line of theorizing, a conceptual review of the literature suggested that high-power individuals are more likely to tease others than low-power individuals, and may do so in a more hostile manner (Keltner, Capps, Kring, Young, & Heerey, 2001). Relatedly, recent research suggests that power begets aggressiveness towards and exploitation of others (Cislak, Cichocka, Wojcik, & Frankowska, 2018), and the sense of confidence and competence that coincides power (Fast et al., 2012; Tost, Gino, & Larrick, 2012, 2013) arguably leads individuals to act on their aggressive impulses with less hesitation (Hershcovis et al., 2017). Thus, power once again acts as a countervailing force to the inhibitory effect of EpF (which curbs overt deviance). Taken together, this leads me to my hypothesis that power moderates the negative relationship between episodic foresight and overt CWB, such that the relationship is weaker when power is high and stronger when power is low. More formally: Hypothesis 11: Power moderates the negative relationship between episodic foresight and overt CWB, such that the relationship is weaker when power is high and stronger when power is low. In light of my theorizing regarding the moderating roles of power, I conclude my theory section by proposing two moderated mediation hypotheses. Specifically, I argue that the indirect effects of fluid reasoning on covert and overt counterproductive work behaviors (CWBs), via episodic foresight (EpF), are moderated by one’s power. Hypothesis 12: Power moderates the positive, indirect effect of fluid reasoning on covert CWB, via episodic foresight, such that the indirect effect is stronger when power is high and weaker when power is low. 61 Hypothesis 13: Power moderates the negative, indirect effect of fluid reasoning on overt CWB, via episodic foresight, such that the indirect effect is weaker when power is high and stronger when power is low. 62 STUDY 1 METHOD The purpose of this first study is to determine the factor structure of a revised version of a self-report measure of counterproductive work behavior (CWB). The primary reason I chose a self-report measure was because I previously developed theory regarding covert, and thus perhaps less discernable, deviant behaviors. As a result, focal participants (i.e., the “behavers”) are the most appropriate source of this data as these behaviors may go undetected by others. The secondary reason I chose a revised self-report measure of CWB is that existing measures of CWB (e.g., Bennett & Robinson, 2000; Stewart, Bing, Davison, Woehr, & McIntyre, 2009) do not explicitly tap into the into the overt-covert distinction I make herein. Sample and Procedure This first study consisted of three phrases after item generation (more detail on item generation is provided below), and thus I collected data for this study from three samples of TurkPrime workers who were also full-time employees. The first sample of TurkPrime employees (n = 104; average age of 44.8; 62.5% female; 72.1% Caucasian) engaged in an item- sort pretest task (Anderson & Gerbing, 1991; Hinkin, 1998), and the second (n = 105; average age of 39.4; 61.0% female; 74.3% Caucasian) and third (n = 118; average age of 37.4; 54.6% female; 74.8% Caucasian) samples of TurkPrime employees responded to a series of items reflecting their engagement in CWB over the past year. Participants were paid approximately $1.00 each for their participation. I used the data from the first sample to pretest an item set and establish substantive validity (Anderson & Gerbing, 1991), which informed item retention in subsequent analyses. More information on this item-sort pretest is provided in the analytic approach and results section of this study. I used the data from the second sample to conduct an exploratory factor 63 analysis and the data from the third sample to conduct a confirmatory factor analysis, with the results of the former informing item retention for the latter. As with the item-sort task, more information on these factor analyses is provided in the section pertaining to my analytic approach and results. Item Generation As noted throughout, my conceptualization of CWB was inspired by Robinson and Bennett (1995), who argued for four categories of CWB (though I make distinctions regarding overtness, which these researchers do not). Thus, I used all 28 items/behaviors originally piloted by Bennett and Robinson (2000), and supplemented them with additional items meant to capture behaviors that would be considered political deviance, property deviance, and personal aggression by these researchers for the sake of scope (the Bennett and Robison measure is heavily loaded with items reflecting production deviance). All of these supplemental items were derived from or inspired by Robinson and Bennett’s (1995) seminal typology. Although the behaviors captured by these supplemental items appear in Robinson and Bennett’s (1995) original typology of deviant workplace behaviors, it is not clear if they were originally represented in the early stages of development of the authors’ measure of deviance (i.e., Bennett & Robinson, 2000) and dropped as these researchers worked to refine that measure. The supplemental items include those meant to capture blaming, nonbeneficial competition, favoritism, taking credit for others’ work, manipulation, asking one to work beyond his or her job description, theft of cash, sabotage of merchandise or equipment, coverups, verbal and physical abuse, endangerment of coworkers, and sexual harassment, to name a few behaviors. This item generation process resulted in a total 50 items/behaviors (see Appendix A). 64 Analytic Approach and Results Phase One 104 TurkPrime workers participated in phase one of this study. All 104 participants passed quality checks, and therefore all 104 participants were retained. As noted, these participants completed an item-sort pretest task (Anderson & Gerbing, 1991; Hinkin, 1998; Hinkin & Tracey, 1999), which required them to read each of the 50 items/behaviors just discussed and assign them to the category that they believe each belonged to. Each behavior could be assigned to one, and only one, of two categories: deviant behaviors that cannot be done in secret (i.e., behaviors that are inherently observable, or overt) and deviant behaviors that can be done in secret (i.e., behaviors that are relatively unobservable, or covert). After data collection, and following Anderson and Gerbing (1991), two indices were calculated. These indices include the proportion of substantive agreement (i.e., the proportion of respondents who assigned an item to its intended category) and the substantive-validity coefficient (which reflects the extent to which respondents assigned an item to its posited category more than the one other available category). The latter of these two indices represents a more accurate estimate of substantive validity, and therefore can and should be used to determine whether or not an item (or behavior, in this case) was assigned to its intended category at a greater frequency than what would be expected by chance. To determine the critical value for the substantive-validity coefficient (or the value for this index that suggests that participants assigned an item to its intended category with greater frequency than what might be expected by chance at the p < .05 level), I assumed that, if the item/behavior under consideration did not clearly reflect one of the two categories in which it could be categorized, there was a 50% chance it could be assigned to either. As noted by 65 Anderson and Gerbing (1991: 734), a probability value of 0.5 is a reasonable choice when an item may tap into its intended construct or categorization “equally well” as it would one other. Given a sample size of 104 (and a desired p-value of .05), assuming random categorization, and using the formulas provided by Anderson and Gerbing (1991), I determined that my critical number of assignments (i.e., the minimum number of assignments to the intended category) was 61 out of 104 and, relatedly, that the critical value for the substantive-validity coefficient was 0.173. Of the 50 items/behaviors piloted, 31 were categorized in their intended categories and 8 were categorized in their non-intended categories (p < .05). The remaining 11 items were not clearly categorized as either overt or covert. The results of these analyses are summarized in Table 2. The 31 items that were correctly categorized were utilized in the exploratory factor analysis. Phase Two 105 TurkPrime workers participated in phase two of this study. 13 participants failed quality checks, and therefore 92 participants were retained. As noted, participants from the second sample responded to a series of items reflecting their engagement in CWB over the course of the past year. Although all 50 items/behaviors in Appendix A were used, only the 31 that were correctly categorized in Phase one were included in the exploratory factor analysis. Analyses were conducted in Mplus version 7.11 (Muthén & Muthén, 2012) using the default oblique rotation (geomin) as I anticipated that the factors would be correlated. Results supported a two-factor solution for the 31 items, explaining 63.85% of the variance (Factor 1 had an eigenvalue of 16.44, and Factor 2 had an eigenvalue of 3.36). Although Factors 3, 4, and 5 all had eigenvalues greater than 1.0 (1.76, 1.40, and 1.19, respectively), and 66 thus met the Kaiser criterion, examination of a scree plot (Gorsuch, 2003) suggested that a break occurred after Factor 2. See Figure 2 for this scree plot. Moreover, a two-factor solution made greater theoretical sense, as covert behaviors and overt behaviors largely grouped into their individual, respective categories (the correlation between factors was .49, p < .05). More specifically, those items meant to reflect covert organizationally-directed deviance (those classified as production deviance by Robinson and Bennett) and those mean to reflect overt organizationally-directed deviance (those classified as property deviance by Robinson and Bennett) generally loaded onto two distinct factors. However, not all factor loadings reached the recommended minimum criterion level of .40 (Ford, MacCallum, & Tait, 1986). Additionally, some items exhibited cross-loadings. Thus, I next dropped one item that did not have a loading on either factor equal to or greater than .40, and three items that exhibited cross-loadings. This left 27 items. Of the 27 remaining items, 6 items reflected what would be considered production deviance, 9 items reflected what would be considered property deviance, 11 items reflected what would be considered personal aggression, and 1 item reflected what would be considered political deviance. Whereas the organizationally-directed forms of deviance loaded onto the anticipated factors, as noted, the interpersonally-directed forms of deviance were somewhat inconsistent. The single item reflecting political deviance loaded onto the overt factor (contrary to expectations), while 8 of the 11 items reflecting personal aggression loaded onto the (anticipated) overt factor (3 of the 11 items loaded on the covert factor). Given that the interpersonal deviance items did not cleanly load onto one factor or the other, I decided to drop those 12 items. This left me with 15 items: 6 reflecting overt forms of deviance and 9 reflecting covert forms of deviance. 67 Although there are no set guidelines for determining the total number of items that should appear on a given scale, shorter measures are considered more desirable than longer ones as they help to prevent participant boredom and fatigue (Kraut, Wolfson, & Rothenberg, 1975). Indeed, it is recommended that researchers aim for four to six items per measure, as this should allow researchers to sufficiently tap into the content domain of the construct under examination while also ensuring acceptable homogeneity among retained items (Clark & Watson, 1995; Hinkin, 1998). Thus, following prior research (e.g., Yu, Hays, & Zhao, 2019), I dropped the 3 of the 9 items reflecting covert deviance that had the lowest factor loadings, leaving me with 12 items total (6 for each form of deviance). However, going into my CFA, I examined both the 15- and 12-item measure. Phase Three 118 TurkPrime workers participated in phase two of this study. 11 participants failed quality checks, and therefore 107 participants were retained. As noted, participants from the third sample responded to a series of items reflecting their engagement in CWB over the course of the past year. Although all 50 items/behaviors in Appendix A were used, only the 15 that were retained after the EFA in Phase two were included in the first set of confirmatory factor analyses. In these analyses, I contrasted a two-factor model of deviance (with items reflecting covert and overt forms of deviance loading on separate factors) with a one-factor model (with all items loading on a single factor) for these 15 items. Neither the one-factor model (χ2 (90) = 330.60, RMSEA = .16, CFI = .83, SRMR = .10) nor the two-factor model (χ2 (89) = 238.44, RMSEA = .13, CFI = .90, SRMR = .07) provided an acceptable fit to the data (Bentler, 1990; Hu & Bentler, 1999; Kline, 2011), although the two- factor model did provide a significant improvement in fit over the one-factor model (χ2 diff (1) = 68 92.16, p < .001). Considering that fit was not acceptable, I then engaged in a second set of confirmatory factor analyses using the 12 items that were retained once I had dropped the 3 of the 9 items reflecting covert deviance that had the lowest factor loadings in the EFA (leaving me with 6 items for each form of deviance). The one-factor model did not provide an acceptable fit to the data (χ2 (54) = 227.07, RMSEA = .17, CFI = .85, SRMR = .10). However, the two-factor model did provide an acceptable fit (χ2 (53) = 130.13, RMSEA = .08, CFI = .93, SRMR = .06), as well as a significant improvement in fit over the one-factor model (χ2 diff (1) = 96.94, p < .001). Thus, the 12-item two- factor model was retained. The items and factor loadings for this two-factor model are provided in Table 3. 69 Sample STUDY 2 METHOD Data for the second study, used to test my theoretical model, was collected from a sample of 158 full-time employees who were also native English speakers. Participants were recruited via ResearchMatch, a non-profit organization that connects researchers to individuals who are interested in participating in studies (Michigan State University has an account and liaison with ResearchMatch). Approximately 77.85% of the sample identified as female (n = 123), 20.89% as male (n = 33), and 1.27% as non-binary (n = 2). Participants were allowed to report multiple ethnicities. Approximately 85.81% (n = 133) identified as partially or entirely Caucasian, 3.80% (n = 6) as Asian or Pacific Islander, 6.96% (n = 11) as African American, 3.16% (n = 5) as Native American, 7.59% (n = 12) as Hispanic/Latino, and 5.70% (n = 9) as some other ethnicity. The average age of participants was 40.28 years. Additionally, 72.15% (n = 114) reported holding a Bachelor’s degree or higher, and 51.27% (n = 81) reported an annual income of $50,000 or higher. Procedure Upon consent, participants reported their demographics and completed survey items capturing their moral identity, power, and engagement in CWB over the course of the last year (as well as various control variables, e.g., entitlement). Participants also completed the Autobiographical Interview task at this time (more information on this task is provided below). One week after completing these survey items and tasks, participants were sent a link to a fluid reasoning assessment, specifically Raven’s Advanced Progressive Matrices (more information on this task is also provided below). Participants were compensated $30 each for their participation, given the intensity of their participation. 70 Although capturing all data from a single source could raise concerns regarding same- source bias, I argue that (a) focal employees are most qualified to report on their personality and covert CWB (and, naturally, must complete their own ability assessments and tasks designed to assess internal, cognitive processes) and (b) the relationships between fluid reasoning and EpF, and EpF and CWB should not be inflated by common method variance as each construct was indeed captured with a different method – an ability assessment, a coded task, and a self-report (as noted, more information on each of these is provided below). In other words, three different methods were utilized to collect data on focal variables. Objective measures (e.g., Raven’s Advanced Progressive Matrices and the Autobiographical Interview task) are resistant to many of the biases that can inflate the correlations between variables collected from the same source (Spector, 2006) Another concern might arise from the fact that I capture fluid reasoning (my independent variable) after participants had reported on CWB. However, it has long been recognized that the predictive and concurrent validities between measures of ability (e.g., the General Aptitude Test Battery, or GATB) and criteria of interest are similar in magnitude (Bemis, 1968). Furthermore, a review of the literature on personnel selection tests argues that the differences in estimates of concurrent and predictive validities are minimal at best, such that “concurrent validities provide useful estimates of predictive validities” when it comes to cognitive assessments (Barrett, Phillips, & Alexander, 1981: 5). Finally, CWB is a low base rate phenomenon, and therefore I asked participants the frequency with which they engaged in CWB over the course of the past year. Considering that abilities, my independent variable, are relatively stable and CWB, my dependent variable, is a low base rate phenomenon and thus was reported retrospectively, the argument that I must capture my independent variable before my dependent variable is moot. 71 Measures Although variables did not vary in terms of their source, their operationalization did vary in terms of structure. While CWB (the dependent variable), moral identity (the first stage moderator), and power (the second stage moderator) were captured using self-report survey measures, fluid reasoning (the independent variable) was captured with an ability measure and EpF (the mediator) was captured using a coded task. Fluid Reasoning The Raven’s Progressive Matrices (RPM) test is the most widely used measure of fluid reasoning (Ferrer, O’Hare, & Bunge, 2009). RPM is a nonverbal test that captures participants’ abilities to engage in abstract reasoning (Bilker et al., 2012), such that it requires participants “to identify relevant features based on the spatial organization of an array of objects, and then select the object that matches one or more of the identified features” (Ferrer et al., 2009: 46). Evidence of the validity and reliability of the RPM test is favorable (Burke, 1972; NCS Pearson Inc., 2007), and it has been found to be applicable to both child (e.g., Pind, Gunnardsdóttir, & Jóhannesson, 2003) and adult (e.g., Brown & Day, 2006) samples. Moreover, research that has used RPM with adult samples has provided evidence that performance on the RPM is positively related to decision-making performance (Gonzalez, Thomas, & Vanyukov, 2005), ratings of initiative and creativity (Chan, 1996), and one’s ability to obtain and maintain positions that require a high level of cognitive functioning (Raven, Raven, & Court, 2000). There are various forms that the RPM may take; RPM tests include the Standard Progressive Matrices, Colored Progressive Matrices, and Advanced Progressive Matrices tests. I employed the Advanced form of the RPM as it has been deemed appropriate for adult samples (Domino & Domino, 2006) and found to be both unidimensional and measurement invariant 72 across sex (Waschl, Nettelbeck, Jackson, & Burns, 2016). Unidimensionality and measurement invariance are of key importance given potential criticisms that the RPM taps into visuospatial ability and, thus, provides males with an advantage given research suggesting that men generally outperform women on tasks that require mental rotation (Waschl et al., 2016). Lastly, the most recent version of the Advance Progressive Matrices test is Raven’s Advanced Progressive Matrices-III (Item-banked), provided by Pearson. This test can be administered online and in a non-proctored setting, and therefore it was the version of the test used in this study. Moral Identity Both dimensions of moral identity were captured using the 10-item self-report measure developed by Aquino and Reed (2002). This measure contains five items for each dimension (i.e., symbolization and internalization). Additionally, this measure (a) prompts individuals with a list of nine traits, and then asks them to (b) visualize a person who has these traits (themselves or someone else), (c) imagine how that person would think, feel, and act, and (d) answer questions about themselves. This measure is provided in Appendix A. Coefficient alpha was .66 for the items reflecting the internalization dimension of moral identity, .73 for the items reflecting the symbolization dimension of moral identity, and .76 for the entire scale. Episodic Foresight Relative to fluid reasoning and moral identity, there appears to be no “gold standard” assessment or survey measure available for capturing EpF; each group of EpF researchers appears to favor a different measure. For example, Lyons and colleagues (Lyons et al., 2014; Lyons, Henry, Rendell, Robinson, & Suddendorf, 2016) utilize a version of the “two-rooms paradigm” that they modified to make appropriate for adult participants (the original is meant for children). This is a behavioral measure in which participants are presented “everyday,” virtual 73 problems that they must resolve by acquiring and using items without any form of cueing. For instance, one of these “everyday” problems is to pick up food to feed one’s cat. A distinctly different method for measuring EpF, and the one that was used in this dissertation, is the modified Autobiographical Interview (AI) task (Addis, Wong, & Schacter, 2008; Levine, Svoboda, Hay, Winocur, & Moscovitch, 2002). The original AI was developed by Levine and colleagues (2002: 678) and is meant to measure episodic memory, or the “recollection of an event from a specific time and place,” such that it involves, “the reexperiencing of contextual details and awareness of the self as a continuous entity across time.” This task was later adapted by Addis and colleagues (Addis et al., 2008) to measure episodic foresight (EpF). Therefore, the AI may be used to parse episodic and non-episodic content in either or both of two temporal phase conditions (past or future), providing an index of either episodic memory (recall) or episodic foresight (projection). Similar tasks have been used by other researchers (e.g., Hassabis, Kumaran, Vann, & Maguire, 2007; Rendell et al., 2012). Researchers who employ the future-oriented version of the AI provide participants with cue words (person, locations, and objects) regarding future events and require participants to provide as much detail as possible about that event within a prespecified timeframe. Researchers may choose either specific cue words (e.g., Mercuri et al., 2015) or entire stimulus sets (comprised of a person, place, and object; e.g., Addis, Musicaro, Pan, & Schacter, 2010) when prompting participants. I prompted participants with four different stimulus sets, two related to my theorizing regarding CWB and two unrelated to CWB. The purpose of these four different stimulus sets is threefold. First, I want to examine whether my model holds with non-context-specific measures of EpF. Second, I want to avoid raising suspicion regarding the purpose of my investigation (i.e., 74 the prediction of CWB). Finally, and most importantly, I developed theorizing above regarding the two different functions of EpF: planning (facilitation) and the avoidance of negative consequences (inhibition). Thus, each of the two CWB-relevant stimulus sets reflects one of these functions, allowing me to test my theoretical model with these measures left separate and/or combined as one variable. These four stimulus sets are provided in Appendix A. The imagined event had to be described from the participant’s subjective perspective, rather than that of an observer. That is, the participant was required to be personally involved in the event he or she described. Three minutes were allocated for each stimulus set, as this is the standard amount of time given (Addis et al., 2010). Timing started once the participant (a) was prompted with the stimulus set, and (b) decided on an event. Importantly, participants were told that imagined future events had to be plausible – they could not be totally farfetched. The standardized AI scoring procedure (Levine et al., 2002) was then used by a trained rater, blind to the hypotheses (I was given possession of this material by Dr. Donna Addis herself; see Appendix A). Each event was scored in the following manner. First, the central event was identified. If more than one event was mentioned, the event discussed in most detail would be selected as the central event. Second, the event was segmented into distinct details (e.g., a unique occurrence or thought), and these details were then categorized as internal (episodic information relating to the central event) or external (non-episodic information, including semantic details, extended events, and repetitions). Internal details involved the “core” of the imagined episode, including who, what, where, and when. External details involved unrelated facts, elaborations, or references to other events. For each event, the number of internal and external details were tallied. The internal details of the two events related to CWB were then standardized. 75 I also coded a subset (n = 48) of the sample’s events myself to calculate interrater agreement/reliability. Thus, there were 384 targets (48 participants x 4 cue word sets x 2 for categorization as internal or external) rated by 2 judges (myself and the blind coder). ICCs were calculated using two-way random effects models because each of the targets were rated by the same set of raters (myself and the blind coder), whom were assumed to be representative of a greater population of potential raters (Shrout & Fleiss, 1979). Specifically, both ICC(2,1) and ICC(2,k) were calculated as I was interested in both the reliability of a single rater (the blind coder) as well as the mean reliability of ratings between coders (Shrout & Fleiss, 1979). Lastly, I estimated both absolute levels of agreement as well as consistency in agreement (i.e., do ratings move in the same direction, regardless of agreement in terms of absolute levels?). All ICCs were calculated in Stata 13.1. In terms of absolute agreement, ICC(2,1) was .75 [.52, .85] and ICC(2,k) was .86 [.69, 92]. In terms of consistency, ICC(2,1) was .80 [.76, .83] and ICC(2,k) was .89 [.87, 91]. All ICCs were considered acceptable (Cicchetti, 1994), and therefore I utilized the blind coder’s ratings when testing my theoretical model. Power Power, as well as status, was measured using the scaled developed by Yu et al. (2019). This measure is provided in Appendix A. Coefficient alpha was .87 for power and .87 for status. Counterproductive Work Behavior The measure developed in the first study was used in this study, though all 50 items/behaviors originally piloted were captured. Given that there are many CWBs that may be specific to occupations, organizations, or industries (see Bowling & Gruys, 2010), participants were provided with the response option “not applicable to my job.” When this option was 76 selected, data was coded as missing. Coefficient alpha was .95 for overt CWB and .92 for covert CWB. Control Variables Several confounding variables have been put forth that may potentially explain any relationship(s) that exist between fluid reasoning (or intelligence, more generally) and CWB. These include educational level, socioeconomic status, ethnicity, power, feelings of entitlement, the ability to delay gratification, and feelings of frustration due to the consistent experience of failure (Dilchert et al., 2007; Gonzalez-Mulé et al., 2014; Roberts et al., 2007). Each of these variables (as well as additional demographics, e.g., age) were therefore captured in the self- report survey (see Appendix A). Entitlement (α = .81) was captured using the measure developed by Campbell and colleagues (2004), with an example item being, “I honestly feel I’m just more deserving than others.” The ability to delay gratification was captured using twelve of the items in the measure developed by Hoerger, Quirk, and Weed (2011), specifically those items that reflect the ability to delay gratification in the domains of achievement (α = .71), social life (α = .81), and money (α = .83) as they appeared most relevant, theoretically (the other two domains reflect the ability to delay gratification in terms of food and physical comfort). An example from each domain includes, “I would rather take the easy road in life than get ahead,” (reverse-coded), “I try to spend my money wisely,” and, “Usually I try to consider how my actions affect others.” I utilized four (rather than all seven) items for each of these domains to avoid participant fatigue. Frustration (α = .89) was measured using the brief irritability test developed by Holtzman, O’Connor, Barata, and Stewart (2015), adapted slightly to fit my context. An example item includes, “I often feel like I might snap.” 77 I also captured social desirability (α = .65) (Crowne & Marlowe, 1960) and the Big Five personality traits: extraversion (α = .66), openness to experience (α = .77), conscientiousness (α = .77), agreeableness (α = .71), and neuroticism (α = .81) (McCrae & John, 1992). The purpose for capturing social desirability is that those with heightened levels of social desirability may be less likely to divulge their engagement in CWB. The purpose for collecting the Big Five is to elucidate the nomological net for moral identity, as several of the adjectives in Aquino and Reed’s (2002) measure seem to tap agreeableness (e.g., kind, caring) and conscientiousness (e.g., hardworking). Social desirability was captured using a 10-item short form of the Marlowe- Crowne Social Desirability scale, developed by Strahan and Gerbasi (1972), as it has demonstrated acceptable reliability (α = .88) and been found to be highly correlated (r = .96) with the full 33-item scale in prior research (Fischer & Fick, 1993). The Big Five was captured using the short form of the Big Five Inventory-2 (BFI-2-S), which was developed to assess the Big Five personality traits as well as narrower (facet) traits (Soto & John, 2017). Both of these measures are provided in Appendix A. Analytic Approach The analyses testing my theoretical model took place at the individual level of analysis, and therefore it was appropriate to test the hypothesized model with structural equation modeling in Mplus version 7.11 (Muthén & Muthén, 2012). This method allowed for hypotheses to be tested and coefficients to be estimated simultaneously, while also providing a variety of fit indices (Bentler, 1990; Hu & Bentler, 1999; Kline, 2011). Furthermore, before creating product terms to test for moderation, episodic foresight (EpF), moral identity, and power were standardized. Finally, mediation and moderated mediation were tested with bootstrapping 78 (10,000 resamples) and Hayes’ index of moderated mediation (Hayes, 2015; Preacher, Rucker, & Hayes, 2007). 79 RESULTS Table 4 provides the descriptive statistics of, and pairwise correlations among, focal and control variables (control variables are italicized). The models tested included only 154 of the 158 participants as 4 participants did not complete the fluid reasoning assessment. As noted, my mediating variable, EpF, was captured with various configurations of the Autobiographical Interview task. These cue word sets include those geared towards capturing the facilitating effect of intelligence (cue word set #1) and the inhibiting effect of intelligence (cue word set #3), as well as two cue word sets that were not specific to engagement in, or punishment resulting from, CWB (cue word sets #2 and #4). My hypotheses were tested with a series of models that utilized the data derived from the CWB-specific cue word sets, given my theorizing. The data derived from the non-CWB-specific cue word sets were utilized in supplemental analyses. Tests of Hypotheses Hypotheses were tested with three models: Model 1, in which only the data from the first cue word set were utilized (capturing the facilitating effect of intelligence), Model 2, in which only the data from the third cue word set were utilized (capturing the inhibiting effect of intelligence), and Model 3, in which data from both cue word sets were utilized (with the two different functions teased apart). Only significant control variables were included in the model to avoid the interpretation of potentially spurious relationships (Becker, 2005). Significant controls included gender (females were less likely to engage in CWB) and irritability for covert CWB, and moral identity (internalization), ethnicity (Latinos reported lower levels of CWB), and conscientiousness for overt CWB. 80 Model 1 Model 1 provided acceptable fit to the data (χ2 (17) = 25.97 (p = .075), RMSEA = .06, CFI = .91, SRMR = .03). In support of Hypothesis 1, fluid reasoning was positively related to EpF (B = .027, p = .001). In support of Hypothesis 2, EpF was negatively related to engagement in overt CWB (B = -.114, p = .004). However, EpF was not positively relate to engagement in covert CWB (B = .034, p = .327), thus failing to support Hypothesis 3. In support of Hypothesis 4, EpF mediated the negative relationship between fluid reasoning and overt CWB (B = -.003, p = .047) as the 95% bias-corrected bootstrap confidence interval excluded zero (-.007, -.001). However, Hypothesis 5 was not supported as EpF did not mediate the positive relationship between fluid reasoning and covert CWB (B = .001, p = .422) as the 95% bias-corrected bootstrap confidence interval did include zero (-.001, .004). Transitioning focus to moderation, neither Hypothesis 6 nor Hypothesis 7 were supported as moral identity symbolization did not moderate the positive relationship between fluid reasoning and EpF (B = .004, p = .967) or the negative, indirect relationship between fluid reasoning and overt CWB (index = -.015 [-.041, .003]). Similarly, neither Hypothesis 8 nor Hypothesis 9 were supported as moral identity internalization did not moderate the positive relationship between fluid reasoning and EpF (B = .098, p = .313) or the positive, indirect relationship between fluid reasoning and covert CWB (index = -.006 [-.031, .004]). Hypothesis 10 was partially supported, as power moderated the relationship between EpF and covert CWB (B = .079, p = .044). Hypothesis 11 was not supported, as power did not moderate the relationship between EpF and overt CWB (B = -.024, p = .527). To better understand the nature of the EpF-covert CWB relationship, I conducted simple slopes computations at “high” (1 SD above the mean) and “low” (1 SD below the mean) levels of 81 power (Aiken & West, 1991). Results suggested that EpF was related to covert CWB at high levels of power (t-value = 2.527, p = .013) but not low levels of power (t-value = -1.006, p = .316). Finally, Hypothesis 12 was not supported as power did not moderate the indirect effect between fluid reasoning and covert CWB (index = .002 [.000, .005]) at the p < .05 level. Likewise, Hypothesis 13 was not supported as power did not moderate the indirect effect between fluid reasoning and overt CWB (index = -.001 [-.003, .001]). This said, it should be noted that Hypothesis 12 was supported at the p < .10 level (p = .095) as the 90% confidence interval for the index of moderated mediation excluded zero (index = .002 [.001, .005]). It has previously been suggested that a 90% confidence interval is acceptable when examining mediation (Preacher, Zyphur, & Zhang, 2010). The reason that Hypothesis 10 was partially supported was because I had formed arguments that power strengthened the positive relationship between EpF and covert CWB, yet there was no support for a positive relationship between EpF and covert CWB. Rather than a strengthening effect of power, I found evidence of a conditional effect. Figure 3 provides visualized path estimates and Figure 4 provides the interaction between EpF and power in the prediction of covert CWB. Model 2 Model 2 provided acceptable fit to the data (χ2 (17) = 22.94 (p = .151), RMSEA = .05, CFI = .93, SRMR = .03). In support of Hypothesis 1, fluid reasoning was positively related to EpF (B = .018, p = .025). In support of Hypothesis 2, EpF was negatively related to engagement in overt CWB (B = -.113, p = .010). However, EpF was not positively related to engagement in covert CWB (B = .044, p = .261), thus failing to support Hypothesis 3. Hypothesis 4 was not supported as EpF did not mediate the negative relationship between fluid reasoning and overt 82 CWB (B = -.002, p = .143) as the 95% bias-corrected bootstrap confidence interval included zero (-.005, .000). Likewise, Hypothesis 5 was not supported as EpF did not mediate the positive relationship between fluid reasoning and covert CWB (B = .001, p = .366) as the 95% bias- corrected bootstrap confidence interval included zero (.000, .003). Transitioning focus to moderation, neither Hypothesis 6 nor Hypothesis 7 were supported as moral identity symbolization did not moderate the positive relationship between fluid reasoning and EpF (B = .005, p = .963) or the negative, indirect relationship between fluid reasoning and overt CWB (index = -.015 [-.045, .003]). Similarly, neither Hypothesis 8 nor Hypothesis 9 were supported as moral identity internalization did not moderate the positive relationship between fluid reasoning and EpF (B = .080, p = .466) or the positive, indirect relationship between fluid reasoning and covert CWB (index = -.006 [-.034, .003]). Hypothesis 10 was not supported, as power did not moderate the relationship between EpF and covert CWB (B = .047, p = .237). Hypothesis 11 was also not supported, as power did not moderate the relationship between EpF and overt CWB (B = .001, p = .988). Finally, Hypothesis 12 was not supported as power did not moderate the indirect effect between fluid reasoning and covert CWB (index = .001 [.000, .003]). Likewise, Hypothesis 13 was not supported as power did not moderate the indirect effect between fluid reasoning and overt CWB (index = .000 [-.002, .001]). Figure 5 provides visualized path estimates. Model 3 Model fit was initially poor (χ2 (34) = 109.73, RMSEA = .12, CFI = .54, SRMR = .06). In an effort to improve model fit, the two different functions of foresight, as captured by cue word set #1 and cue word set #3, were allowed to covary. Upon this specification, Model 3 provided acceptable fit to the data (χ2 (33) = 49.63 (p = .032), RMSEA = .06, CFI = .90, SRMR = .04), 83 and a significant improvement in fit over the model in which the two functions of foresight were not allowed to covary (χ2 diff (1) = 60.10 (p < .001)). In support of Hypothesis 1, fluid reasoning was positively related to both operationalizations of EpF: that which was meant to capture the facilitating effect (B = .026, p = .001) and that which was meant to capture the inhibiting effect (B = .018, p = .015). In support of Hypothesis 2, EpF was negatively related to engagement in overt CWB (B = -.120, p = .006). However, EpF was not positively related to engagement in covert CWB (B = .044, p = .260), thus failing to support Hypothesis 3. Hypothesis 4 was not supported as EpF did not mediate the negative relationship between fluid reasoning and overt CWB (B = -.002, p = .104) as the 95% bias-corrected bootstrap confidence interval included zero (-.006, .000). Likewise, Hypothesis 5 was not supported as EpF did not mediate the positive relationship between fluid reasoning and covert CWB (B = .001, p = .311) as the 95% bias-corrected bootstrap confidence interval included zero (-.001, .004). Transitioning focus to moderation, neither Hypothesis 6 nor Hypothesis 7 were supported as moral identity symbolization did not moderate the positive relationship between fluid reasoning and EpF (B = .021, p = .768) or the negative, indirect relationship between fluid reasoning and overt CWB (index = -.004 [-.025, .012]). Similarly, neither Hypothesis 8 nor Hypothesis 9 were supported as moral identity internalization did not moderate the positive relationship between fluid reasoning and EpF (B = .050, p = .474) or the positive, indirect relationship between fluid reasoning and covert CWB (index = -.003 [-.019, .002]). Hypothesis 10 was supported, as power moderated the relationship between EpF and covert CWB (B = .083, p = .035). To better understand the nature of the EpF-covert CWB relationship, I conducted simple slopes computations at “high” (1 SD above the mean) and “low” 84 (1 SD below the mean) levels of power (Aiken & West, 1991). Results suggested that EpF was related to covert CWB at high levels of power (t-value = 2.840, p = .005) but not low levels of power (t-value = -.872, p = .385). Hypothesis 11 was not supported, as power did not moderate the relationship between EpF and overt CWB (B = .002, p = .956). Finally, Hypothesis 12 was not supported as power did not moderate the indirect effect between fluid reasoning and covert CWB (index = .002 [.000, .006]) at the p < .05 level (though it was supported at the p < .10 level; p = .077). Likewise, Hypothesis 13 was not supported as power did not moderate the indirect effect between fluid reasoning and overt CWB (index = .000 [-.002, .001]). Figure 6 provides visualized path estimates and Figure 7 provides the interaction between EpF and power in the prediction of covert CWB. 85 SUPPLEMENTAL ANALYSES To test the robustness of my results, I engaged in a series of supplemental analyses. In the first set of supplemental analyses I sought to determine whether my theoretical model held by using the non-CWB specific cue word sets collected in Study 2. These two cue word sets were not explicitly geared towards capturing the facilitating and inhibiting effect of intelligence (as it pertains to engagement in CWB). Cue word set 2 was used to test what I refer to as Model 4, and cue word set 4 was used to test what I refer to as Model 5. In the second set of supplemental analyses I tested several additional interactions. Specifically, I tested various interactions between power, fluid reasoning, and moral identity in the prediction of episodic foresight because power can transform one’s psychological state (Keltner et al., 2003), and thus may affect mental processes such as foresight. I also tested various interactions between status, fluid reasoning, and moral identity in the prediction of episodic foresight for the sake of comprehensiveness as status, like power, represents a foundational base of social hierarchy (Magee & Galinsky, 2008). Finally, I tested various interactions between moral identity, foresight, power, and status in the prediction of CWB because, in theory, those low in moral identity may be more disposed to engage in deviant actions (that is, moral identity may be proximal to behavior). In the third set of supplemental analyses I re-calculated my dependent variable (CWB) several ways and retested my theoretical model. The purpose in doing this was to determine whether episodic foresight was predictive of CWB, more generally (rather than just the “clean” measures created in Study 1). I tested my model using (a) the original Bennett and Robinson (2000) measure, (b) all 50 items I collected, separated into production deviance, property 86 deviance, interpersonal aggression, political deviance, interpersonal deviance, and organizational deviance, and (c) using dichotomized version of all CWB calculations. Finally, in the fourth set of supplemental analyses I re-conducted Study 1 and re- calculated my dependent variable. Given that the participants who completed the item sort task in Study 1 were not subject matter experts on CWB, I re-ran all 50 items in an EFA to determine what factor structure emerges. Additionally, and considering the number of items there were relative to the number of participants, I aggregated responses to these 50 survey items across all three samples of participants that responded to them – the EFA sample and the CFA sample from Study 1, as well as the sample whose data was used to test my theoretical model in Study 2. Although I realize that the data used to conduct an EFA and construct a measure should not then be used to test a theoretical model, this was purely for exploratory purposes. Supplemental Analyses – Alternative Cue Word Sets Model 4 Model 4 provided acceptable fit to the data (χ2 (17) = 19.83 (p = .283), RMSEA = .03, CFI = .96, SRMR = .03). In partial support of Hypothesis 1, fluid reasoning was marginally, but positively, related to EpF (B = .014, p = .087). In partial support of Hypothesis 2, EpF was marginally, but negatively, related to engagement in overt CWB (B = -.081, p = .078). However, EpF was not positively related to engagement in covert CWB (B = .029, p = .485), thus failing to support Hypothesis 3. Hypothesis 4 was not supported as EpF did not mediate the negative relationship between fluid reasoning and overt CWB (B = -.001, p = .289) as the 95% bias- corrected bootstrap confidence interval included zero (-.004, .000). Likewise, Hypothesis 5 was not supported as EpF did not mediate the positive relationship between fluid reasoning and 87 covert CWB (B = .000, p = .574) as the 95% bias-corrected bootstrap confidence interval included zero (-.001, .002). Transitioning focus to moderation, neither Hypothesis 6 nor Hypothesis 7 were supported as moral identity symbolization did not moderate the positive relationship between fluid reasoning and EpF (B = -.025, p = .780) or the negative, indirect relationship between fluid reasoning and overt CWB (index = -.007 [-.032, .005]). Similarly, neither Hypothesis 8 nor Hypothesis 9 were supported as moral identity internalization did not moderate the positive relationship between fluid reasoning and EpF (B = .001, p = .996) or the positive, indirect relationship between fluid reasoning and covert CWB (index = -.003 [-.027, .004]). Hypothesis 11 was supported, as power moderated the relationship between EpF and covert CWB (B = .085, p = .038). However, Hypothesis 11 was not supported, as power did not moderate the relationship between EpF and overt CWB (B = -.009, p = .839). To better understand the nature of the EpF-covert CWB relationship, I conducted simple slopes computations at “high” (1 SD above the mean) and “low” (1 SD below the mean) levels of power (Aiken & West, 1991). Results suggested that EpF was related to covert CWB at high levels of power (t-value = 2.430, p = .016) but not low levels of power (t-value = -1.32, p = .189). Finally, Hypothesis 12 was not supported as power did not moderate the indirect effect between fluid reasoning and covert CWB (index = .001 [.000, .004]). Likewise, Hypothesis 13 was not supported as power did not moderate the indirect effect between fluid reasoning and overt CWB (index = .000 [-.002, .001]). Figure 8 provides visualized path estimates and Figure 9 provides the interaction between EpF and power in the prediction of covert CWB. 88 Model 5 Model 5 did not provide an acceptable fit to the data (χ2 (17) = 28.96 (p = .035), RMSEA = .07, CFI = .86, SRMR = .04), based on the CFI estimate. Thus, Model 5 could not be interpreted. For the sake of exploration, however, the results of this model suggested that fluid reasoning was positively related to EpF (B = .023, p = .009), consistent with Hypothesis 1. This model also suggested that EpF was marginally, but negatively, related to engagement in overt CWB (B = -.085, p = .054), consistent with Hypothesis 2, and that power (marginally) moderated the relationship between EpF and covert CWB (B = .078, p = .099). No other hypotheses would have been supported given the estimates of this model, if they were acceptable to interpret. Figure 10 provides a visualization of this failed model. Supplemental Analyses – Additional Interactions As noted, I also explored a variety of other interactions. I focused first on first-stage moderation. Initially, I examined interactions between power and fluid reasoning in the prediction of EpF, given that power can transform psychological states (Keltner et al., 2003). After examining these two-way interactions, I explored three-way interactions between power, fluid reasoning, and moral identity. Then, I examined interactions between status and fluid reasoning in the prediction of EpF, as status represents a major basis of hierarchy like power (Magee & Galinsky, 2008), and therefore might also affect one’s psychological state. Finally, I examined interactions between status, fluid reasoning, and moral identity. As it pertains to second-stage moderation, I first explored interactions between moral identity and EpF in the prediction of covert and overt CWB. Then, I explored three-way interactions between moral identity, EpF, and power in the prediction of covert and overt CWB. After that, I replaced power with status in my model and explored interactions between status 89 and EpF in the prediction of covert and overt CWB. Finally, I explored three-way interactions between moral identity, EpF, and status in the prediction of covert and overt CWB. Power x Fluid Reasoning Predicting Episodic Foresight Power did not moderate the relationship between fluid reasoning and episodic foresight, either using the cue word set geared towards capturing the facilitating effect of foresight (B = - .021, p = .371) or the inhibiting effect of foresight (B = .014, p = .547). Power x Fluid Reasoning x Moral Identity Predicting Episodic Foresight The coefficient for the product term representing the interaction between power, moral identity symbolization, and fluid reasoning was not significant in the prediction of episodic foresight, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .037, p = .207) or the inhibiting effect of foresight (B = -.023, p = .345). The coefficient for the product term representing the interaction between power, moral identity internalization, and fluid reasoning was not significant in the prediction of episodic foresight when using the cue word set geared towards capturing the facilitating effect of foresight (B = - .029, p = .294). However, it was significant when using the cue word set geared towards capturing the inhibiting effect of foresight (B = -.068, p = .013). A visualization of this interaction is provide by Figure 11. Status x Fluid Reasoning Predicting Episodic Foresight Status did not moderate the relationship between fluid reasoning and episodic foresight, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .007, p = .771) or the inhibiting effect of foresight (B = .023, p = .372). 90 Status x Fluid Reasoning x Moral Identity Predicting Episodic Foresight The coefficient for the product term representing the interaction between status, moral identity symbolization, and fluid reasoning was not significant in the prediction of episodic foresight, either using the cue word set geared towards capturing the facilitating effect of foresight (B = -.002, p = .941) or the inhibiting effect of foresight (B = .005, p = .863). Likewise, the coefficient for the product term representing the interaction between status, moral identity internalization, and fluid reasoning was not significant in the prediction of episodic foresight, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .028, p = .309) or the inhibiting effect of foresight (B = .021, p = .363). Moral Identity x Episodic Foresight Predicting Covert CWB Moral identity internalization did not moderate the relationship between episodic foresight and covert CWB, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .007, p = .771) or the inhibiting effect of foresight (B = .023, p = .372). However, moral identity symbolization did moderate the relationship between episodic foresight and covert CWB, using both the cue word set geared towards capturing the facilitating effect of foresight (B = .049, p = .006) and the inhibiting effect of foresight (B = .037, p = .021). Visualizations of these interactions are provided in Figures 12 and 13, respectively. Moral Identity x Episodic Foresight Predicting Overt CWB Moral identity internalization did not moderate the relationship between episodic foresight and overt CWB, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .036, p = .269) or the inhibiting effect of foresight (B = .001, p = .978). Likewise, moral identity symbolization did not moderate the relationship between episodic 91 foresight and overt CWB, using either the cue word set geared towards capturing the facilitating effect of foresight (B = -.006, p = .775) or the inhibiting effect of foresight (B = .013, p = .497). Moral Identity x Episodic Foresight x Power Predicting Covert CWB The coefficient for the product term representing the interaction between power, moral identity symbolization, and episodic foresight was not significant in the prediction of covert CWB, either using the cue word set geared towards capturing the facilitating effect of foresight (B = -.019, p = .281) or the inhibiting effect of foresight (B = -.005, p = .753). Likewise, the coefficient for the product term representing the interaction between power, moral identity internalization, and episodic foresight was not significant in the prediction of covert CWB, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .013, p = .721) or the inhibiting effect of foresight (B = .033, p = .269). Moral Identity x Episodic Foresight x Power Predicting Overt CWB The coefficient for the product term representing the interaction between power, moral identity symbolization, and episodic foresight was not significant in the prediction of overt CWB, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .033, p = .127) or the inhibiting effect of foresight (B = .034, p = .131). However, the coefficient for the product term representing the interaction between power, moral identity internalization, and episodic foresight was significant in the prediction of overt CWB, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .087, p = .049) or the inhibiting effect of foresight (B = .104, p = .003). Visualizations of these interactions are provided in Figures 14 and 15, respectively. 92 Status x Episodic Foresight Predicting CWB Status did not moderate the relationship between episodic foresight and covert CWB, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .011, p = .586) or the inhibiting effect of foresight (B = .012, p = .568). Likewise, status did not moderate the relationship between episodic foresight and overt CWB, using either the cue word set geared towards capturing the facilitating effect of foresight (B = .005, p = .828) or the inhibiting effect of foresight (B = .012, p = .648). Moral Identity x Status x Episodic Foresight Predicting Covert CWB The coefficient for the product term representing the interaction between status, moral identity symbolization, and episodic foresight was not significant in the prediction of covert CWB, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .008, p = .801) or the inhibiting effect of foresight (B = .017, p = .601). Likewise, the coefficient for the product term representing the interaction between status, moral identity internalization, and episodic foresight was not significant in the prediction of covert CWB, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .025, p = .675) or the inhibiting effect of foresight (B = .006, p = .901). Moral Identity x Status x Episodic Foresight Predicting Overt CWB The coefficient for the product term representing the interaction between status, moral identity symbolization, and episodic foresight was not significant in the prediction of overt CWB, either using the cue word set geared towards capturing the facilitating effect of foresight (B = .000, p = .999) or the inhibiting effect of foresight (B = .023, p = .544). Likewise, the coefficient for the product term representing the interaction between status, moral identity internalization, and episodic foresight was not significant in the prediction of overt CWB, either 93 using the cue word set geared towards capturing the facilitating effect of foresight (B = .029, p = .678) or the inhibiting effect of foresight (B = .027, p = .652). Supplemental Analyses – Alternative Calculations of CWB I collected all 50 CWB items that I piloted in Study 1 in Study 2. Therefore, for exploratory purposes, I decided to test my theoretical model using alternative calculations of CWB to see if my results changed. First, I tested my theoretical model using the items in the original measure developed by Bennett and Robinson (2000): 12 items reflecting organizational deviance and 7 items reflecting interpersonal deviance. I then tested my theoretical model using measures that utilized all 50 items: production deviance (14 items), property deviance (15 items), and organizational deviance (14 items + 15 items = 29 items), as well as personal aggression (13 items), political deviance (8 items), and interpersonal deviance (13 items + 8 items = 21 items). Given the low base rate of these behaviors, I also created dichotomized versions of my CWB variables that reflected whether Study 2 participants reported that they had or had not engaged in the focal CWB at least once over the course of the past year. Finally, in reporting the results of these additional models I will focus on the operationalization of EpF derived from my first cue word set as this cue word set returned the most favorable results in my focal analyses. This said, the interpretation of results does not differ notably when using the other measures of EpF. Original Measure Fluid reasoning was positively related to EpF (B = .027, p = .001). Neither moral identity internalization (B = .098, p = .274) nor moral identity symbolization (B = .004, p = .965) moderated this relationship. EpF was not related to interpersonal deviance (B = -.023, p = .386) or organizational deviance (B = -.028, p = .365). Power did not moderate the relationship 94 between EpF and interpersonal deviance (B = .004, p = .877) or the relationship between EpF and organizational deviance (B = -.005, p = .849). Original Measure – Dichotomized Fluid reasoning was positively related to EpF (B = .025, p = .004). Neither moral identity internalization (B = .089, p = .345) nor moral identity symbolization (B = .002, p = .980) moderated this relationship. EpF was not related to interpersonal deviance (B = .004, p = .970) or organizational deviance (B = .015, p = .904). Power did not moderate the relationship between EpF and interpersonal deviance (B = .070, p = .542) or the relationship between EpF and organizational deviance (B = -.185, p = .672). Property and Production Deviance Fluid reasoning was positively related to EpF (B = .027, p = .001). Neither moral identity internalization (B = .098, p = .274) nor moral identity symbolization (B = .004, p = .965) moderated this relationship. EpF was not related to production deviance (B = -.009, p = .810) or property deviance (B = .013, p = .565). Power did not moderate the relationship between EpF and production deviance (B = -.013, p = .677) or the relationship between EpF and property deviance (B = .032, p = .111). Property and Production Deviance – Dichotomized All participants reported in engaging in at least some form of production deviance over the course of the past year, based on the 14 items I used to operationalize this construct. As a result, a dichotomized variable could be created for this construct but there would be no meaningful results as there would be no variance. Therefore, I ran a model with property deviance alone as there was variance in the dichotomized version of this variable. 95 Fluid reasoning was positively related to EpF (B = .025, p = .004). Neither moral identity internalization (B = .079, p = .383) nor moral identity symbolization (B = .015, p = .874) moderated this relationship. Surprisingly, EpF was positively related to property deviance (B = .209, p = .043). However, power did not moderate this relationship (B = .011, p = .918). Personal Aggression and Political Deviance Fluid reasoning was positively related to EpF (B = .027, p = .001). Neither moral identity internalization (B = .098, p = .274) nor moral identity symbolization (B = .004, p = .965) moderated this relationship. EpF was not related to personal aggression (B = -.008, p = .703) or political deviance (B = -.002, p = .945). Power did not moderate the relationship between EpF and personal aggression (B = -.003, p = .869) or the relationship between EpF and political deviance (B = .016, p = .444). Personal Aggression and Political Deviance – Dichotomized Fluid reasoning was positively related to EpF (B = .025, p = .004). Neither moral identity internalization (B = .088, p = .356) nor moral identity symbolization (B = .008, p = .932) moderated this relationship. EpF was not related to personal aggression (B = .137, p = .203) or political deviance (B = .011, p = .918). Power did not moderate the relationship between EpF and personal aggression (B = .076, p = .560) or the relationship between EpF and political deviance (B = .100, p = .366). Organizational and Interpersonal Deviance Fluid reasoning was positively related to EpF (B = .027, p = .001). Neither moral identity internalization (B = .098, p = .274) nor moral identity symbolization (B = .004, p = .965) moderated this relationship. EpF was not related to interpersonal deviance (B = -.005, p = .804) or organizational deviance (B = .002, p = .935). Power did not moderate the relationship between 96 EpF and interpersonal deviance (B = .006, p = .732) or the relationship between EpF and organizational deviance (B = .009, p = .687). Organizational and Interpersonal Deviance – Dichotomized All participants reported in engaging in at least some form of organizational deviance over the course of the past year, based on the 29 items I used to operationalize this construct. As a result, a dichotomized variable could be created for this construct but there would be no meaningful results as there would be no variance. Therefore, I ran a model with interpersonal deviance alone as there was variance in the dichotomized version of this variable. Fluid reasoning was positively related to EpF (B = .025, p = .001). Neither moral identity internalization (B = .092, p = .335) nor moral identity symbolization (B = .002, p = .979) moderated this relationship. EpF was not related to interpersonal deviance (B = .071, p = .490). Finally, power did not moderate this relationship (B = .019, p = .916). Measures from Study 1 – Dichotomized Fluid reasoning was positively related to EpF (B = .025, p = .004). Neither moral identity internalization (B = .085, p = .351) nor moral identity symbolization (B = .007, p = .937) moderated this relationship. EpF was not related to overt deviance (B = -.033, p = .740) or covert deviance (B = .160, p = .119). Power did not moderate the relationship between EpF and overt deviance (B = -.080, p = .411) or the relationship between EpF and covert deviance (B = .066, p = .476). Supplemental Analyses – Recreation of CWB Measures Finally, and given that the participants who completed the item sort task in Study 1 were not subject matter experts on CWB, I decided to run all 50 items in an EFA to determine what factor structure emerges. Moreover, and considering the number of items there were relative to 97 the number of participants, I aggregated responses to these 50 survey items across all three samples of participants that responded to them – the EFA sample and the CFA sample from Study 1, as well as the sample whose data was used to test my theoretical model in Study 2. Although I realize that the data used to conduct an EFA and construct a measure should not then be used to test a theoretical model, this was purely for exploratory purposes. The eigenvalues and scree plot for this EFA are presented in Figure 16. Results generally supported a two-factor solution for the 50 items, explaining 51.46% of the variance (Factor 1 had an eigenvalue of 22.29, and Factor 2 had an eigenvalue of 3.44). Factors 3 through 7 all had eigenvalues greater than 1.0, and thus met the Kaiser criterion, but examination of a scree plot suggested that a break occurred after Factor 2. Additionally, factor loadings did not change notably between a two-factor (Table 5) and a three-factor (Table 6) solution (though there were cross-loadings in the three-factor solution), and the factor structure of a four-factor solution was not theoretically interpretable (Table 7). Factor loadings under .40 were suppressed for ease of presentation in Tables 5, 6, and 7. The two-factor solution reveals that the property deviance items (reflecting serious, organizationally-directed behaviors) load on the first factor and the that the production deviance items (reflecting minor, organizationally-directed behaviors) load on the second factor. More than half of the personal aggression items loaded on the same factor as did the production deviance items, while four personal aggression items loaded on the same factor as did the property deviance items. Conversely, more than half of the political deviance items loaded on the same factor as did the property deviance items, while two political deviance items loaded on the same factor as did the production deviance items. Although personal aggression is considered the more serious form of interpersonal deviance, the seemingly less egregious of these items (e.g., 98 “Acted rudely toward someone at work”) loaded on the same factor as the production deviance items while the more egregious of these items (e.g., “Sexually harassed a coworker”) loaded on the same factor as property deviance. The same is ostensibly true of political deviance. Thus, I considered Factor 1 to reflect serious forms of deviance and Factor 2 to reflect minor forms of deviance. Given this two-factor solution, I then created a measure and tested my theoretical model. Fluid reasoning was positively related to EpF (B = .027, p = .001). Neither moral identity internalization (B = .098, p = .274) nor moral identity symbolization (B = .004, p = .965) moderated this relationship. EpF was not related to serious forms of deviance (B = .009, p = .634) or minor forms of deviance (B = -.029, p = .305). Power did not moderate the relationship between EpF and serious forms of deviance (B = .019, p = .249) or the relationship between EpF and minor forms of deviance (B = -.005, p = .852). 99 DISCUSSION Individual differences in cognitive ability are undeniably relevant to hiring contexts (Murphy et al., 2003). Indeed, scholars have repeatedly found evidence of a strong, positive relationship between performance on cognitive ability assessments and performance in one’s job (Gottfredson, 2002), and there is some research suggesting that intelligence predicts future job performance better than any other ability, characteristic, or personality trait recruiters are capable of measuring (Schmidt & Hunter, 2004). Thus, it should be unsurprising that researchers frequently recommend that practitioners make hiring decisions with applicant cognitive ability in mind (e.g., Schmidt & Hunter, 1998). Although I do not entirely challenge this recommendation, as I provide evidence that cognitive ability (or, more specifically, fluid reasoning) shares a negative relationship with certain forms of CWB (e.g., tardiness), the results of Study 2 somewhat suggest that researchers should not tout cognitive ability as an unambiguous predictor of positive workplace outcomes. I say this because the second study of this dissertation provides some evidence that fluid reasoning might share a positive, indirect relationship with covert forms of CWB when the focal employee is in a position of power (p = .095). In particular, the results of this study suggest that greater levels of fluid reasoning might enable individuals to better imagine future deviant episodes, which, in turn, may differentially predict actual engagement in deviant behavior. Whereas the capability to vividly imagine these scenarios appears to deter overt forms of workplace deviance (or CWB; e.g., loafing), which is (a) consistent with predictions derived from the inhibitory effect of intelligence and (b) desirable for organizations, this same mental process may be associated with greater engagement in covert forms of workplace deviance when the focal employee is in a position that enables him or her to enact such behavior (i.e., when s/he 100 is in a position of power). This latter finding (c) presents a challenge to the inhibitory effect of intelligence and (d) naturally has undesirable implications for organizations that integrate cognitive ability into the hiring process. Thus, these results have both theoretical and practical implications, and they prompt several questions for future research. Theoretical Implications Perhaps the most important theoretical contribution of this research is that it provides tentative empirical support for the inhibitory effect of intelligence by demonstrating that foresight (specifically episodic foresight) indeed mediates a negative relationship between intelligence and (overt) CWB. Although proponents of the inhibitory effect of intelligence position foresight as the explanatory mechanism behind the theoretically negative relationship between intelligence and workplace deviance, these proponents have thus far failed to capture this mechanism (or even provide consistent evidence of a negative relationship, for that matter). By identifying and leveraging an operationalization of episodic foresight that is primarily used in developmental and cognitive psychology, I was able to provide some support for this line of theorizing. That is, this investigation is the first to find some support for the argument that intelligence shares a negative relationship with certain forms of CWB via foresight. This said, this research also provides evidence that challenges the inhibitory effect of intelligence, and it does so in two distinct ways. First, I demonstrate that intelligence only deters overt forms of deviance (e.g., tardiness, leaving one’s work unfinished), at least via foresight. In my review of the literature, proponents of the inhibitory effect have not distinguished between different forms of CWB (e.g., minor versus serious, interpersonal versus organizational, overt versus covert). Instead, they imply that intelligence should curb all forms of deviance. Yet, the evidence provided herein does not demonstrate that intelligence deters more covert forms of 101 deviance (e.g., theft, sabotage) whatsoever, either directly or indirectly (via episodic foresight). Thus, claims regarding the inhibitory effect of intelligence might need to be bounded as the inhibitory effect appears to apply solely to observable forms of deviance, or those forms that are more likely to result in negative consequences for the perpetrator. Second, and perhaps presenting an even greater challenge to the inhibitory effect of intelligence, is that Study 2 revealed that intelligence might share a positive relationship with more egregious, covert forms of deviance (at least when the focal employee is in a position of power). Admittedly, the conditional indirect effect (per the index of moderated mediation) was only significant at the p = .095 level, but I would nevertheless argue that these findings add additional nuance to the theory underlying the intelligence-CWB relationship (even more so considering that some have argued that a mediated effect only warrants a 90% confidence interval; Preacher et al., 2010). Indeed, future research employing similar designs with larger sample sizes (and, thus, greater statistical power) might finding additional (or even stronger) evidence for a conditional, positive, indirect relationship between intelligence and certain forms of CWB. Perhaps there is not only an inhibitory effect of intelligence but also a facilitating effect of intelligence. When both are taken together, they might help to explain prior, inconsistent findings. Importantly, the theoretical implications of this investigation are not limited to just the field of OB/HR. Indeed, the criminology literature might gain insight from the theoretical framing and empirical results presented herein. I say this because the criminology literature would categorize intelligence, moral identity, and power as “psychological explanations” of deviance (as noted throughout), and therefore would position all three of them as independent variables within theoretical models. However, the model presented herein positioned moral 102 identity and power as moderators. To be sure, only power moderated one of the hypothesized paths (and indeed shared a positive relationship with covert CWB, as would be expected per the criminology literature – see Table 4), but I would nevertheless argue that this moderating effect demonstrates the value in taking an alternative approach to the study of frequently examined constructs in criminology. Relatedly, criminologists might find value in drawing upon theories from our field, such as integrative self-control theory (Kotabe & Hofmann, 2015). The creators of integrative self- control theory, like myself, would consider a lack of power to be an enactment constraint, or an environmental factor “not under the person’s immediate control that constrain[s] the range of available behavioral options in a given situation” (Kotabe & Hofmann, 2015: 628), and, consequently, a moderator rather than an independent variable. In fact, in their theoretical model Kotabe and Hofmann (2015) explicitly position enactment constraints as moderating mechanisms, and specifically identify inaccessibility to resources as one such potential enactment constraint. Perhaps future research in criminology would benefit from drawing upon theories from other disciplines and/or construing other commonly-studied constructs as moderators instead of independent variables, as done herein. Practical Implications The most salient practical implication of these findings is that intelligent employees who occupy positions of authority may be inclined to steal from and sabotage their organization, as well as exploit other opportunities that their positions afford (e.g., accept kickbacks). This is a finding of key importance as intelligence and occupational attainment tend to be positively correlated, meaning that more intelligent individuals frequently secure positions higher up the corporate ladder (Schmidt & Hunter, 2004). Indeed, one post hoc explanation offered by Roberts 103 and colleagues (2007) for the unanticipated, positive relationship they found between cognitive ability and CWB was that higher intelligence leads individuals to occupy powerful positions that allow for engagement in deviant behaviors such as theft (and, as noted, I indeed find a positive relationship between power and covert CWB in my own data – see Table 4). Thus, these findings underscore the importance of having behavioral control mechanisms in place, particularly for employees in the upper echelons. Fortunately, and reflecting back on integrative self-control theory (Kotabe & Hofmann, 2015), work environments can be proactively designed to deter deviant behaviors, such that physical or social enactment constraints (or external factors that inhibit behavior) can be implemented in advance to avoid managerial CWB. For example, and particularly relevant to deterring deviant behavior among employees holding upper echelon (and, hence, powerful) positions, there is an entire literature on corporate governance (or the practices and processes by which corporations are directed and controlled; Misangyi & Acharya, 2014). Indeed, this literature has identified stock options, CEO duality (O’Connor, Priem, Coombs, & Gilley, 2006), and board composition (Cumming, Leung, & Rui, 2015) as just some corporate governance mechanisms that can be strategically managed to deter engagement in deviant behavior by members of the top management team. Another practical implication of this work is that it adds to the chorus of scholars recommending that recruiters hire employees based upon their performance on assessments of cognitive ability (or, in this case, fluid reasoning). Although I do present evidence that intelligence might be associated with covert forms of CWB, the reality is that organizations are unlikely to start deliberately hiring employees of below-average intelligence (particularly for high-level positions), and covert CWB can likely be curbed through the use of appropriate 104 enactment constraints (as just noted). Overt forms of CWB, which are perhaps less contingent on organizational enactment constraints (e.g., tardiness, leaving one’s work unfinished), appear to share a negative indirect relationship with intelligence, and therefore hiring based on intelligence may very well provide another, though non-task-related, benefit to the organization. Limitations The first major limitation of this investigation is that it does not provide conclusive support for causality. Per John Stuart Mill, one cannot definitively declare causality without establishing temporal precedence, or demonstrating that the supposed cause (i.e., a change in, or occurrence of, some independent variable) did in fact occur before the effect (i.e., a change in, or occurrence of, some dependent variable) (Wagner & Hollenbeck, 2015). This said, it was pragmatically impossible to capture change in the independent variable, fluid reasoning, in this investigation as (a) the research question necessitated a sample of working adults and (b) fluid reasoning stops increasing around mid- to late-adolescence (Ferrer et al., 2009) and declines at only a very gradual pace thereafter (McArdle, Ferrer-Caja, Hamagami, & Woodcock, 2002). The small, steady decline typically observed in fluid reasoning passed young adulthood is perhaps why differences in concurrent and predictive validities are minimal when it comes to cognitive assessments (Barrett, 1981). Nevertheless, future researchers might utilize research designs that allow them to more reliably claim a causal relationship between fluid reasoning and CWB, via episodic foresight. Relatedly, a second major limitation of this work is that the design did not allow me to capture specific instances of episodic foresight before actual engagement in CWB. On the one hand, the design of such a study seems impractical, if not impossible. To expect participants to carry around a physical notebook and record the mental episode they envision about a particular 105 deviant behavior before they engage in that behavior is unrealistic in more ways than one. First, participants are highly unlikely to willingly walk around with what are essentially recorded confessions of premeditated CWB. Second, even if a willing sample were obtained, such a design would likely result in a large amount of missing data. If we assume that the theorizing developed in this dissertation is “truth,” then we would expect that participants would often fail to report on their engagement in episodic foresight before their engagement in overt forms of CWB because, presumably, very little (if any) forethought would occur anyway. Finally, CWB is a relatively low base rate phenomenon, hence why participants in Study 2 were asked to report the frequency with which they engaged in specific deviant behaviors over the course of the prior year. Therefore, participants would have to be followed for an extended period of time to complete data collection. On the other hand, more enterprising researchers might find study designs that allow for a more “episodic” approach to the study of workplace deviance. As an example, Liao, Liu, Li, and Song (2019) utilized an event-contingent experience sampling methodology to examine outcomes of leader-member exchange. Specifically, these researchers utilized a mobile survey platform to capture employees’ momentary assessments of face-to-face interactions with their leaders, with the hopes of demonstrating that these interaction episodes had downstream consequences for work engagement and individual contributions at later points in time. Future researchers interested in taking a more episodic, or event-contingent, approach to the study of CWB might be able to adapt the study design utilized by Liao and colleagues (though I anticipate that some of the previously identified problems will nevertheless occur). A third major limitation of this investigation is that the results of Study 1, the study designed around measure development, did not yield measures that captured interpersonal forms 106 of CWB. Contrary to expectations, participants did not appear to consistently categorize behaviors that would be considered “minor” forms of interpersonal CWB (i.e., political deviance) as behavior that (a) was inherently overt or (b) could be done in secret/covertly. In fact, only one behavior that would be considered political deviance per the Robinson and Bennett typology survived the item-sort phase of measure development, which was the very first phase of this study. Although a greater number of behaviors that would be considered more serious forms of interpersonal CWB (i.e., personal aggression) did survive this first phase of measure development, they did not survive the second and third phases. Surprisingly, several of these behaviors exhibited cross-loadings on the two factors that emerged from the exploratory factor analysis (Phase Two), and even when these cross-loaded behaviors/items were removed several of the remaining items/behaviors loaded on the factor reflecting covert CWB (which was, again, contrary to expectations). Given that behaviors reflecting this interpersonal CWB were not clearly loading onto either factor, they were removed from subsequent analyses for the sake of ensuring clean measures that explicitly, and exclusively, included items that were appropriately categorized during Phase 1. Future researchers armed with a new, and perhaps larger, set of items/behaviors reflecting interpersonal CWB might want to explore this further, in the hopes that they find an overt-covert distinction in interpersonally-directed CWB that corresponds to what I found regarding organizationally-directed CWB. A fourth major limitation of this investigation is that I solely captured one aspect of intelligence (or cognitive ability): fluid reasoning. That is, I did not investigate alternative components of intelligence, of which there are at least 15 per the Cattell-Horn-Carroll theory of intelligence. Although this was intentional, as performance on assessments of fluid reasoning 107 theoretically provide insight into overall intellectual capability (Frick et al., 2010; Otero, 2017), investigations capturing other aspects of intelligence (e.g., short-term memory, processing speed, domain-specific knowledge; Schneider & McGrew, 2012), or using alternative operationalizations of intelligence, might provide interesting insights. For example, future research that captures domain-specific knowledge in addition to fluid reasoning might find that the relationship between fluid reasoning and EpF (specifically related to engagement in CWB) is moderated by domain-specific knowledge about the organization’s functioning. In particular, these researchers might find that domain-specific knowledge strengthens this relationship as individuals with greater levels of domain-specific knowledge about their organization(s) should have a greater reservoir of information that can be used to inform the deviant episodes they envision. A final limitation of this work is that I was constrained in terms of the number of cue word sets I could employ, for the sake of avoiding participant fatigue. Moreover, there was no precedent in terms of which cue word sets were most appropriate in addressing my research question. Relative to the literature on related constructs (e.g., episodic memory, self-control), the literature on episodic foresight is fairly limited. For example, an online library search for the term “episodic foresight” yields a mere 207 research articles while similar searches for the terms “episodic memory” and “self-control” yield 62,087 and 170,258 research articles, respectively (at least as of October 2019). Additionally, the little primary research on episodic foresight that does exist is generally in the fields of developmental, cognitive, and clinical psychology. As a result, there are currently no recommendations regarding which specific cue word sets to use, how these cue word sets should be composed, and so forth, for research questions in OB/HR. Although I was able to utilize task instructions and coding procedures from prior research, the 108 selected cue word sets were ultimately based upon my own judgement and, to some extent, intuition. Future research using different cue word sets might come to alternative conclusions regarding the relationship(s) between intelligence, episodic foresight, and CWB (or other behaviors of interest). Future Directions Future Research related to Fluid Reasoning The theorizing, methods, and results presented herein give rise to several streams of future research related to intelligence/fluid reasoning, CWB, and other workplace behaviors. As it pertains to future research on fluid reasoning, this is the first investigation, to my knowledge, that has provided empirical evidence that fluid reasoning indeed facilitates episodic foresight (both related and unrelated to engagement in CWB), and, as alluded to in the last section, measures of episodic foresight can be customized to capture any specific behavior or adapted to fit any particular context. In other words, and despite the fact that this investigation focused on episodic foresight as it pertained to engagement in CWB, episodic foresight does not have to be specific to any particular environment or action. As an example of how flexible this task can be, Mercuri and colleagues (2015) chose six cue words from the Affective Norms for English Words list that differed in their valence (i.e., positive, negative, or neutral) in their study on the long-term effects of opiate use on episodic foresight. Specifically, these researchers used the terms birthday and vacation (two positively valenced words), nightmare and accident (two negatively valenced words), and taxi and bench (two neutral words), though they could have chosen any cue words they would have liked. What this means is that future researchers interested in examining the relationship between intelligence 109 and some particular outcome, via foresight, could simply adapt the AI task to fit their specific research questions. For instance, future researchers might be interested in fleshing out the potential relationship(s) between intelligence, episodic foresight, and organizational citizenship behaviors (or OCB), another non-task component of overall job performance (Rotundo & Sackett, 2002). Indeed, this may provide a fruitful avenue of future research given the results of a relatively recent meta-analysis, which revealed that intelligence shared a positive, albeit modest, correlation with OCB (even though this meta-analysis was unable to test mediators; Gonzalez- Mulé et al., 2014). In fact, these researchers explicitly lament that there is a dearth of research capturing the mediating mechanisms between intelligence and non-task aspects of performance (i.e., CWB and OCB), and ultimately call for future research that captures these mechanisms. In response to these calls, future researchers might want to adapt the AI task and capture episodic foresight (related to engagement in OCB) to further explore the positive relationship Gonzalez- Mulé and colleagues (2014) found between intelligence and OCB. Another potentially interesting research question, arising from a perusal of Study 2’s correlation table, pertains to the relationship(s) between intelligence and perceptions of one’s own power and/or status. Although intelligence often begets positions of power (as noted), Table 4 reveals negative correlations between objectively-measured fluid reasoning and self-reported power and status. Although these were not focal relationships in my theoretical model, it might be interesting to explore why more intelligent individuals rate themselves lower in power and status than less intelligent individuals (particularly given that more intelligent individuals had higher levels of education on average in this sample, which should thus lead to positions of greater power/status). Of course this is purely speculative at this point, but perhaps more 110 intelligent individuals simply have a more accurate understanding of their place in the organizational hierarchy. Future Research related to Emotional Intelligence Another avenue of future research pertains to the relationship between emotional intelligence and CWB (perhaps via episodic foresight). Emotional intelligence reflects the “ability to perceive emotions, to access and generate emotions so as to assist thought, to understand emotions and emotional knowledge, and to reflectively regulate emotions so as to promote emotional and intellectual growth” (Mayer & Salovey, 1997: 5), and some empirical research suggests that the Cattell-Horn-Carroll theory of intelligence should be expanded to include emotional intelligence as another second-stratum factor (e.g., Evans, Hughes, & Steptoe- warren, 2019; MacCann et al., 2014), ultimately giving it the same standing as fluid reasoning in this model. Although the debate is still out as to whether emotional intelligence should be integrated into the Cattell-Horn-Carroll theory of intelligence (or should even be considered a valid construct whatsoever; Locke, 2005), research on emotional intelligence arguably appears promising (Côté, 2014). For example, and particularly relevant to the theorizing presented herein, emerging research suggests that emotional intelligence might be related to emotional manipulation (Nagler, Reiter, Furtner, & Rauthmann, 2014), as well as other forms of interpersonal deviance (Côté, DeCelles, McCarthy, van Kleef, & Hideg, 2011). Although interpersonal forms of CWB were not examined herein (due to the results of Study 1, which suggested that interpersonal deviance cannot be cleanly categorized as either overt or covert forms of CWB), future researchers might find that emotional intelligence shapes episodic foresight related to interpersonal CWB (political deviance or personal aggression; Robinson & Bennett, 1995), and, consequently, actual 111 engagement in interpersonal CWB. Extending the theorizing presented herein, perhaps emotional intelligence is associated with lower levels of personal aggression (e.g., verbal abuse) yet higher levels of political deviance (e.g., gossip), via episodic foresight. Future researchers interested in studying emotional intelligence and non-task facets of performance might also find value in examining emotional intelligence’s relationship with OCB. One study revealed that a major motive for employee engagement in OCB is impression management (Rioux & Penner, 2001), or the deliberate process that involves attempts to influence the perceptions others have of oneself (Jones & Pittman, 1982). Given that it involves the manipulation of others’ perceptions, it is arguably a highly interpersonal behavior (and, as noted, emotional intelligence has previously been linked to manipulation). Perhaps individuals with heightened levels of emotional intelligence are better at impression management, and thus are more strategic in terms of the OCB they engage in, due to their ability to more accurately and vividly foresee the reactions others will have to potential extra-role behaviors. Future Research related to Counterproductive Work Behavior Antecedents As it pertains future research focused on the differential prediction of various forms of CWB, I presented evidence herein that employees indeed distinguish between overt and covert deviance. Although some researchers have theorized that the different behaviors nested under the umbrella term counterproductive work behavior might differ from one another in terms of their conspicuousness (e.g., Spector & Fox, 2002; Zhang et al., 2011), this is the first empirical investigation, to my knowledge, that has actually disentangled overt and covert forms of CWB. Naturally, evidence of this distinction (coupled with corresponding scales, developed in Study 1) opens up entirely new lines of research that preexisting distinctions (e.g., organizationally- 112 versus interpersonally-directed deviance) and scales (e.g., Bennett & Robinson, 2000) have not made possible. First, the distinction between overt and covert CWB allows for the identification of personality characteristics that are more or less predictive of these two different forms of deviance. Indeed, meta-analytic evidence suggests that personality characteristics are more powerful than abilities in the prediction of CWB (Gonzalez-Mulé et al., 2014), and therefore investigations pinpointing which traits are more strongly associated with one form of CWB over the other would prove highly informative. For example, different facets of the Dark Triad (which includes Machiavellianism, narcissism, and psychopathy; O’Boyle, Forsyth, Banks, & McDaniel, 2012) might have differential relationships with the two unique forms of CWB identified herein. In fact, there is already some evidence suggesting that Machiavellianism is related to engagement in theft and sabotage (O’Boyle et al., 2012), two behaviors captured in the Study 1 scale designed to measure covert deviance. Moreover, and given that Machiavellians are typically concerned with maintaining their organizational standing, they tend to be somewhat “conscientious,” and thus might avoid engagement in other, more observable forms of CWB (O’Boyle et al., 2012: 557). In other words, these findings suggest that future researchers employing the scales developed herein might find a positive relationship between Machiavellianism and covert CWB, but a negative relationship between Machiavellianism and overt CWB. Given the only moderate correlation I found between these two forms of deviance (r = .25 in Study 2), it is quite possible that they have divergent relationships with this trait. Conversely, individuals with heightened levels of psychopathy tend to lack self- regulatory mechanisms, and thus are inclined to act more impulsively (O’Boyle et al., 2012). As 113 previously noted, impulsive behavior has literally been defined as acting without foresight (Dalley et al., 2011; Winstanley et al., 2006), and thus future researchers might find that psychopathy shares a positive relationship with overt CWB that is equally strong, if not stronger, in magnitude as its relationship with covert CWB. Aside from personality characteristics, other antecedents of CWB include job attitudes (e.g., job satisfaction and organizational commitment) and justice (Dalal, 2005), both of which are multifaceted (Colquitt, 2001; Judge & Kammeyer-Mueller, 2012). For example, organizational commitment, or one’s psychological attachment to his or her organization, can be affective, instrumental, or normative in orientation (Judge & Kammeyer-Mueller, 2012). Perhaps all three forms of commitment are negatively related to engagement in overt forms of CWB (as committed employees, presumably, do not want to lose their positions), but more instrumental forms of commitment (e.g., continuance commitment) are positively related to covert forms of CWB as employees that report high levels of continuance commitment often stay with their jobs out of financial necessity. Those individuals who are in greater need of financial resources may very well engage in covert forms of deviance, such as theft. Indeed, the distinction between overt and covert forms of deviance allows for an entirely new line of research focused on the identification of divergent antecedents. Profile Research related to Engagement in Counterproductive Work Behavior The separation of CWB into overt and covert forms also opens up the possibility for profile research. Profile analyses represent data-driven, person-centered approaches (as opposed to a variable-centered approaches) that result in the determination of how two or more constructs simultaneously exist, relate, and operate within-individuals (Gabriel, Campbell, Djurdjevic, Johnson, & Rosen, 2018). That is, profile research differs from variable-centered (or variance- 114 oriented; Morgeson, Mitchell, & Liu, 2015) research in that the focus is on how groups of individuals cluster together in terms of their relative endorsement of various constructs, rather than on how different constructs simply relate to one another. Once clusters of individuals with similar profiles are discerned, researchers typically then attempt to identify antecedents and outcomes unique to the profiles discovered. As an example, recent research on emotional labor (or “the management of emotions as part of the work role”) identified five different emotional labor profiles (Gabriel, Daniels, Diefendorff, & Greguras, 2015: 863). Moreover, upon identification of these profiles, the researchers were able to discern different antecedents and outcomes associated with the various profiles. Whereas individuals who primarily engaged in one form of emotional labor (“deep acting”) exhibited relatively high levels of job satisfaction and relatively low levels of emotional exhaustion, the opposite was true for those individuals who primarily engaged in the alternative form of emotional labor (“surface acting”). As noted, the delineation of overt and covert forms of CWB might give rise to similar profile research, though related to engagement in workplace deviance. That is, by separating covert CWB from overt CWB and engaging in some form of profile analysis, future researchers might find that some employees have a tendency to engage in one form of CWB significantly more than the other, that some employees have a tendency to engage in both forms of CWB at a high (though statistically similar) rate, and that one final group of employees has a tendency to engage in both forms of CWB at a low (though statistically similar) rate. Discovery of these different groups naturally opens questions regarding different antecedents and outcomes. 115 Future Upper Echelons Research As previously noted, intelligence and occupational attainment tend to be positively correlated (Schmidt & Hunter, 2004), meaning that top management team positions are frequently occupied by individuals with heightened levels of cognitive functioning/fluid reasoning. This might create the perfect environment for immoral behavior to occur at the organization level (that is, beyond individual-level CWB directed at the organization itself, which constituted the outcomes examined herein). Indeed, top management teams have a great deal of discretion in setting the course for their entire organization, and thus they have a heavy hand in determining the firm’s approach to activities such as corporate social responsibility. Corporate social responsibility “refers to firm behavior that goes beyond compliance and legal requirements to provide some social good,” and is reflected in the firm’s approach to community relations, employee relations, human rights, and the environment, to name a few domains (Ormiston & Wong, 2013: 862). Focusing on human rights, organizations in a variety of industries, including those in the tech (Ethiraj, Kale, Krishnan, & Singh, 2005) and apparel (Siggelkow, 2001) industries, frequently outsource their labor-intensive work to developing nations. As an example, consider the shipbreaking industry in countries such as Bangladesh, India, and Pakistan. Shipbreaking is the process by which ships are cheaply dismantled for parts, scrap, or raw materials. Oftentimes, shipowners can sell their decommissioned vessels to Asian shipyards for more cash than they would receive if they were to sell to recycling yards with higher labor standards (The Guardian, 2019). Importantly, the details of such transactions (as well as employee accident records) are frequently kept hidden (Heidegger, 2017). The reasons that these records are typically kept hidden is because, to date, thousands of laborers in Bangladesh, India, 116 and Pakistan have been seriously, if not fatally, injured as a result of shipbreaking practices (The Guardian, 2019). Although outsourcing decommissioned vessels to these yards is not illegal on an international level, top management teams of organizations that participate in this practice (e.g., McDermott International) do realize that (arguable) human rights violations are in fact occurring and thus are quick to deny responsibility. For example, a spokesman from McDermott International is quoted in saying that, although his organization does not “condone unsafe, unethical or noncompliant business practices,” they “are not in a position to comment” on the shipbreaking practices of supposedly “unrelated third parties.” (Bengali, 2016). The point of the preceding discussion is that, (a) organizations realize that sending their ships to less-than-ethical shipbreaking yards could result in the injury or death of low-paid, unskilled labor, yet (b) have a financial incentive to do so. Assuming that highly competent individuals occupy executive suite positions (a fairly safe assumption), I would argue that highly powerful top management teams are likely to send their ships to less-than-ethical shipbreaking yards, and do so without garnering media or watchdog attention, more frequently than top management teams that have far less power (e.g., those who are not constrained by external factors). Of course, this argument is not necessarily reserved for just shipbreaking – other forms of corporate social irresponsibility may be a function of competence and power among top management teams. Past Engagement in Counterproductive Work Behavior In building my arguments for the moderating role played by moral identity on the relationship between fluid reasoning and episodic foresight, I drew upon some of the logic provided by DeCelles and colleagues (2012), specifically their notion that moral identity increases the cognitive accessibility of moral concepts. In a sense, I argued that individuals with 117 below average levels of moral identity were more likely to have immoral concepts closer to the forefront of their minds, and thus would have an easier time envisioning deviant scenarios. Although this theorizing was not supported, it occurred to me that past experiences, much like personality traits, might affect the accessibility of various cognitive concepts. Indeed, prior engagement in CWB might facilitate the mental process of imagining future scenarios as it, in a sense, provides a preexisting foundation to build off of. In fact, emerging neurological research suggests that there are notable similarities between memory and imagination, and that memory and foresight/imagination might very well share a common brain network (Schacter et al., 2012). Indeed, there is a growing body of research that suggests that (a) memories from prior experiences inform imagination of future scenarios, (b) individuals that suffer from memory losses (e.g., amnesic patients) frequently struggle to imagine novel future events (suggesting a connection between the past and the future in the brain), and (c) similar regions of the brain become activated when individuals are asked to recall past events as when they are asked to imagine future ones (Schacter et al., 2012). Thus, memory/recall and imagination/foresight appear to be two very closely intertwined mental processes. Thus, future researchers might find that prior engagement in CWB could allow individuals to better envision future engagement in CWB. It would be particularly interesting to see if engagement in one form of CWB (e.g., overt CWB) facilitates episodic foresight related to another form of CWB (e.g., covert CWB), or if prior experiences are domain specific. Similarly, it would be interesting to see how successful versus unsuccessful experiences engaging in CWB affect episodic foresight. Perhaps unsuccessful experiences facilitate foresight vividness because it makes potential consequences more salient. Finally, it would be interesting to see how prior 118 deviance outside the workplace affects mental processes pertaining to deviance in the workplace. This again gets at the question: are prior experiences domain specific? 119 CONCLUSION The yearly estimated economic cost associated with employee deviance is somewhere in the range of billions of dollars (Dilchert et al., 2007). Accordingly, it is in the organization’s best interest to curb such behavior well before it occurs, perhaps even as early as in the hiring stage. Herein, I provided evidence that intelligence is indeed negatively associated with overt forms of CWB, consistent with the inhibitory effect of intelligence, and therefore it might be in recruiters’ best interest to follow the advice of prior scholars to hire based upon performance on intelligence assessments. However, I also heed organizations to have appropriate enactment constraints in place, as the preliminary evidence presented here suggests that intelligent individuals in positions of power might have a tendency to engage in covert forms of deviance, such as theft. In sum, I endorse the use of intelligence assessments in hiring contexts but also emphasize the importance of checks-and-balances in the workplace. 120 APPENDICES 121 APPENDIX A – Measures Moral Identity (Aquino & Reed, 2002) Listed below are some characteristics that may describe a person:  Caring  Compassionate  Fair  Friendly  Generous  Helpful  Hardworking  Honest  Kind The person with these characteristics could be you or it could be someone else. For a moment, visualize in your mind the kind of person who has these characteristics. Imagine how that person would think, feel, and act. When you have a clear image of what this person would be like, answer the following questions. 5-point Likert scale (1 = strongly disagree, 5 = strongly agree). Internalization 1. It would make me feel good to be a person who has these characteristics. 2. Being someone who has these characteristics is an important part of who I am. 3. I would be ashamed to be a person who has these characteristics. (R) 4. Having these characteristics is not really important to me. (R) 5. I strongly desire to have these characteristics. Symbolization 1. I often wear clothes that identify me as having these characteristics. 2. The types of things I do in my spare time (e.g., hobbies) clearly identify me as having these characteristics. 3. The kinds of books and magazines that I read identify me as having these characteristics. 4. The fact that I have these characteristics is communicated to others by my membership in certain organizations. 5. I am actively involved in activities that communicate to others that I have these characteristics. 122 Adapted Autobiographical Interview - (Addis et al., 2008, adapted from Levine et al., 2002); provided by Dr. Donna Addis (I also have the materials from Levine et al., 2002, if desired). Paradigm used in Addis et al. (2008) The paradigm in Addis et al. (2008) involved 4 temporal conditions: past few weeks, past few years, next few weeks, and next few years. In each condition, participants generated 8 events in response to cue words. Within each condition, cues were randomly shown. Both time periods (weeks and years) for one temporal direction (past or future) were completed before the conditions in the other temporal direction began. Conditions were blocked in this manner to reduce load and facilitate older adults’ understanding of the instructions for each condition. Order of presentation of temporal direction and time periods was counterbalanced. Moreover, the above lists cycled through conditions, such that the design was fully counterbalanced. Administration Manual This adapted version of the Autobiographical Interview involves showing subjects cue words in order to elicit either recall of events from the subject’s past or the generation of imagined events which may occur in the subject’s future. Subjects are given 3 minutes (from when the cue is first shown) to recall/generate and describe the event in as much detail as possible. The events are required to have episodic specificity. In other words, the event should be one that is a few hours in duration, specific in time and place, and not one that was experienced or could be experienced with high frequency (e.g., routine events). These requirements are explained to subjects in the detailed instructions below. Given the purpose of this dissertation, participants will be asked only to imagine future events. They will not be asked recall past events. General Administration Instructions (The following instructions are provided to subjects prior to commencing the task. They have been modified given that I am interested in future events only): I want to begin by telling you a bit about this task. I am looking at how people imagine events which might happen in the future. To help with scoring, I will be recording your responses to the following cue word sets but, as with all studies, your responses will be kept completely confidential and stored in a secure place. Over the course of the next twenty (or so) minutes, you will be seeing cue word sets on the screen. Each cue word set consists of a person, place, and action. For each cue word set, you will be asked to imagine an event which may occur in your future. The imagined event must involve all of the words in the cue word set. For each set, I ask you to imagine an event that could happen to you in “the next few weeks” (up to about 5 months from now). Importantly, I want you to describe a single event for each set of cue words that is specific to a time and place. For example, if I showed you the cue word set “best friend, vacation, monument,” you might tell me about a vacation to France. However, I 123 would want you to focus in on one particular thing that you will do with your friend on that trip that stands out in your mind, or that would happen on one particular day. So, for instance, you might tell me about the day on which you and your best friend will visit the Eiffel Tower. I want you to tell me everything you can visualize about that event, such as what you will do, what you would be thinking or feeling, your senses, etc. The goal is to determine how well you can project yourself into the future. Thus, when creating these future scenarios, you can be creative, but you cannot be totally unrealistic. For example, you can’t tell me about going to the moon with your “best friend” Neil Armstrong for a vacation. I want you to think about scenarios that are plausible given your plans and thoughts about the future. For all of these events, you will be given three minutes to tell me as much detail as you can about them. So, remember, you can tell me everything you can imagine about the event, that particular day, what you were doing, thinking, or feeling, and so forth. I am not as interested in which events you choose to describe so much as I am interested in how much detail you can imagine when you describe them. Therefore, it doesn’t matter whether the event is important or trivial, as long as it is an event which is specific in time and place, and is one that you are personally involved in. There are two ways to imagine events; either through our own eyes, as if you are in our own shoes seeing the event unfolding around you, or from a bird’s eye view, watching yourself. Regardless of which approach you take, you must be involved in this event. So, in sum, you will be seeing a set of cue words on the screen. Using all of these cue words, you will have to tell me about a future scenario you have made up. Finally, I ask that you tell me as much detail about the event as possible. Cue Word Sets: 1. You (person), at work (place), stealing company property (action) 2. Your best friend (person), in Europe (place), sightseeing (action) 3. Your supervisor (person), in his/her office (place), disciplining you for deviant behavior (action) 4. One of your relatives (person), at a party (place), trying new food (action) Adapted Autobiographical Interview Scoring Manual Overview The Adapted Autobiographical Interview quantifies elements of descriptions of specific events which may occur in the subject’s personal future. In each trial, a set of cue words is shown. The subject must imagine a specific event in the future that the cue words make him or her think of, and describe as much detail as possible within 3 minutes. The events must be specific to a particular time and place. The interview is recorded digitally (via Qualtrics). For each event, the scorer isolates or defines the main event, then divides the entire response (including information external to the main 124 event) into small segments (details). These details are categorized as either “internal” or “external” to the main event. This will be explained in more detail below. Isolating and defining the event Although the test instructions request specific events, many subjects give more than one event or events that are difficult to define (i.e., non-specific events). It is therefore necessary to be clear what the event is before any scoring takes place. This will come into play when categorizing segments, as segments that are not part of the event (external details) are tallied separately from those that are part of the event (internal details). Subjects are instructed to describe an event in which they are personally involved and that is singular (not repeated) and specific to a time and place. The event should be restricted in time, no more than a few hours in duration. If an event extends over days or weeks (e.g., a vacation), the scorer must restrict scoring to the best time-restricted event available. If more than one exists, choose the time-restricted event which is described in most detail. One of the most difficult scoring situations is when the event is very impoverished or non- existent (e.g., only factual information is given). In such cases, it may be possible to select some details as probably specific to an event and to score them accordingly, but qualitative ratings cannot be assigned. Text segmentation and categorization A segment, or detail, is an information bit; it is a unique occurrence, observation, fact, statement, or thought. This will usually be a grammatical clause -- a sentence or part of a sentence that independently conveys information (i.e., a subject and a predicate), although a single clause may contain more than one detail. For each clause, consider whether its constituent parts convey additional information. If so, the parts can be separated and scored as separate segments. For example, the statement “he had an old, brown fedora” would be segmented into three details: a “fedora” is different from a “brown fedora,” which in turn is different from an “old brown fedora.” Each of these details adds information that significantly alters the meaning of “fedora”, which on its own would receive one detail. The main categorical distinction for details is internal or external to the event. To be categorized as Internal, a detail must pertain directly to the main event, isolated as defined above. Internal details can include the following: 1. Event details: Overall, event details describe the unfolding of the story. They are usually happenings (e.g., "I fell down"), but also include who was there (1 point per name/person up to a maximum of 5), reactions/emotions in others, the weather, one’s clothing if it is part of the action, physical occurrences and actions of others, and temporal sequence or information about the sequence of events (“Mary came/later than Sam”, where “Mary came” is an event detail, and “later than Sam” is another event detail). If an item qualifies 125 to be in another category (e.g., perceptual richness), then priority is given to that more specific category. An item cannot be scored as an event detail if it is in another category. 2. Place details: Any information that involves localization in space, including countries, bodies of water, provinces, cities, streets, buildings, rooms, and locations within a room. Note that one's own orientation in space ("I was to the right of Edgar") is considered a perceptual detail. 3. Time details: Life epoch ("My twenties"), year, season, month, date, day of week, time of day, or clock time. Information about sequences of events (“Mary came later than Sam”) are scored as event, not time details. Note: It has been argued that one cannot directly encode or retrieve temporal information (i.e., when an event occurred), but only infer it from other information. That is, it is not possible to re-experience a given point in time without reference to some related episodic thought, feeling, or other detail. Therefore, when scoring time information, people should not be penalized for making inferences (which are usually scored as "other" details), because this is the normal way to figure out when something occurred. 4. Perceptual details: Perceptual details include auditory, olfactory, tactile/pain, taste, visual (including object details [see below], colors, clothes), spatial (i.e., details about positions, distances, and orientations in allocentric/egocentric space, e.g. one's own orientation in space: "I was to the right of Edgar"), temporal (e.g., duration, “We were there for 20 minutes,” but not temporal sequence). In the case of objects, it can be difficult to distinguish between a perceptual and an event detail. Objects that are directly involved in the unfolding of an event are considered event details ("We lit the candles") whereas objects that are part of the visual landscape are considered visual details ("There were lit candles everywhere"). 5. Emotion/Thought details: Any detail that pertains to the mental state of the subject at the time of the event. These include feeling states, thoughts, opinions, expectations, or beliefs. Thoughts expressed in retrospect (either at the time of the interview or at any time after the event occurred - "I found out later I was wrong") are tallied as external. Beliefs or opinions that are long-standing (not specific to the event - "I never believed in ghosts") are also external and are scored as semantic details. Inferences about other people's mental state ("She was sad") are considered event details, unless these inferences reflect the subjects' own mental state at the time ("I thought he was angry with me"), in which case they are internal thought details. NOTE: For a coding scheme that separates internal details into event, place, spatial orientation, time, duration/sequence, visual (non- spatial), perceptual (non-visual), emotion/thought details, see Martin (2013) Memory for the Future: The Encoding and Phenomenology of Episodic Simulations. PhD Thesis, University of Auckland. External details events that are not part of the main event or factual (semantic) information that is not specific to the main event. These can include the following: 126 1. Semantic details: Semantic details involve general knowledge or facts. They can represent general knowledge ("Paris is the capital of France") or be specific to the person ("I always hated yams." "I worked as an engineer"). The distinction between semantic and other kinds of details can depend on the context. For example, "Paris fell to the Germans" would be semantic if it is described as a historical fact ("We couldn't go to Paris because it was in German hands") or an event detail ("We watched in disbelief as Paris fell to the Germans."). In general, details that reflect a long-standing state of being or without a clear beginning or end are considered semantic. Semantic information can be "brought in" to episodic recollection (and scored as an internal detail) if it becomes an integral aspect of the episode: "Arizona is hot" is semantic, but "Arizona was hot when we went there" is episodic. Note that the richness of the description is independent from the episodic/semantic distinction; very richly described factual information is still semantic, and impoverished, minimal details can still be episodic. If relevant, semantic details can be further broken down into one of the following two types: Personal semantic details – semantic details as per the above description which refer specifically to the person. This includes semantic information which relates to the subject’s family or friends. General semantic details – semantic details as per the above description which refer to general knowledge or facts. Note that some participants may give semantic details which appear to have been learned for the first time as part of the event described. These should be coded as semantic unless there are clear signifiers or ‘tags’ which indicate that the information was learned for the first time as part of this event (e.g. “I remember when the tour guide said …”). 2. Repetitions: A detail is a repetition if it is an unsolicited repetition of a prior information- containing detail. It does not have to be a verbatim repetition, but it should not add any new information to the prior detail ("I hoped for the best. I kept my fingers crossed" -- second sentence is a repetition). Score all repetitions, even if they are part of normal discourse, except for repetitions that are clearly prompted by the examiner, which may occur if the examiner queries a detail that was given earlier. Repetitions must convey information (as opposed to just words that are repeated). In the example below, “… and stuff” is repeated, but there is no information in this utterance, so it is not considered a repetition. As well, only score repetitions when they convey the same information as in an earlier detail. In the example below, “They really really liked me” is not a repetition of “They were happy with my work.” Similarly, “I was a carpenter’s helper”, “I helped them”, and “They could depend on me” are all different. “They liked what I did” however is the same as “They liked my work.” Then he repeats this repetition straight away. 3. Other details: This category is for details that do not reflect recollection and do not fit into other categories. It includes meta-cognitive statements ("Let me see if I can remember that"), editorializing ("That doesn't matter." "That's amazing."), inferences ("I must have been wearing a coat because it was winter"), or other statements that convey verbosity but are not related to the main event. Replies to a query that are clauses (E: "Do you remember any more about that day? S: No, right now I don't. I don't remember anymore") are also scored as "other", although simple reflexive replies such as "No" are not scored. Do not score an "other" detail for any utterance - only those that contain 127 information. Generally, an "other" detail will be a clause of some sort. Fragments such as “um” are not scored. 4. External episodic details: Episodic events secondary to the main episodic event, e.g. if the person is imagining the birth of their child (main event), and talks about the ultrasound a few months before. Code as Ext. Event 2, Ext. Event 3, etc. These events can be further sub-categorized according to internal detail categories (event, place, time, perceptual, thought/emotion) if so wished. 5. External generic events: Details that refer to repeated or routine events (but not general knowledge), e.g. “I always go to the dairy down the road.” (“Always” is a good indication that the event is repeated) or extended events (e.g., “I went on holiday to Fiji for 3 weeks”). These can be scored as Generic Events, or broken down further into Generic Repeated or Generic Extended, as appropriate. Events which are both repeated and extended (e.g. a week-long yacht race which the subject participated in every year) should be coded as Generic Repeated. The sums of internal and external details are important measures of a subject’s performance. With experience, a scorer will be able to simultaneously segment and categorize a response. In some cases, it can be difficult to distinguish internal from external details. The rule of thumb in these cases (the “benefit of the doubt” rule) is that if a detail could reasonably be internal, it is scored as such. This rule, however, should not be applied to all details that could possibly be internal; only those that could reasonably be internal. Scoring example Categorization: All details are classed as external as there is no specific, time-limited event described. The subject is describing the company he worked for and his role. However, this is somewhat open to interpretation. Another scorer might decide that the description of another company coming in (i.e., “another company came in and did the finish work but they were all happy with my work and saw I listened”) is a single episode rather than a matter of due course on every job. This is an example of a judgement call. Many scoring decisions are judgement calls. Scorers will be somewhat influenced by their own knowledge and experience with the subject matter. Score according to your knowledge. If two people could reasonably score a detail more than one way, simply score it the way that seems best rather than agonize over it. 128 Segmentation: The clause “It was a company out of New Bedford that was building” contains three details, a company, from New Bedford, that builds. Thus, “company” can stand alone (i.e., he works for a company, and not, for instance, himself) but the subject tells us something about what type of company it is (i.e., they build). The second detail is a place detail, telling us that the company was based in New Bedford. This clause illustrates that one cannot always find the dividing line between details. The dividing of segments can be somewhat arbitrary. Where one places the dividing lines is not as important as the number of information bits one scores. The “shelves … and rough carpenter work” can be segmented into three details: “doing shelves” (a type of building), “carpenter work” (another description of the type of work), and then further refining the carpenter work (as “rough”). Next, another company comes in. This instance of “company” is not a repetition of the first, as it is a different company. The “coming in” was scored as a separate detail because it implies a happening, something this other company did. Their being happy is a state of being/emotion; the cause of the happiness (i.e., the subject’s work) is a further detail. Likewise, the subject imparts a number of details about his role: a “carpenter’s helper”; the task was to help; but not just whenever, but “when something was needed”; he was dependable (“they could depend on me”); the company liked him; and they also liked the work he did (“what I did” and repeated in “the work I did”) We have come up with some other segmentation rules, as a result of scoring dilemmas that have arisen in the Addis lab: 1. Time details: The location of the event in time (e.g., “next few weeks”, “in a couple years”, “yesterday”) should not be segmented as this usually reflects the time period given as part of the cue. 2. Relationship details: The relationship of the subject to someone else (e.g., “boyfriend”, “last boyfriend”, “uncle”, “great uncle”, “friend”, “best friend”, “Donna’s friend”). Often, as the subject will consistently refer to someone as “my best friend.” However, if they have used the name and are using the phrase to describe the relationship, then it can be segmented accordingly (e.g., “she was my best friend;” she’s not just a friend, but a best friend.) 3. Senses: “I saw the tower”, “I heard a noise” are internal details as the sense description is part of the experience of the content (i.e., you can’t see a noise or hear a tower). Also, the sense verbs cannot stand alone (e.g., “I saw.”) 4. Dialogue: Whether the dialogue is external (speech) or internal (thoughts), each statement/thought represents one detail and so it is not segmented (e.g., “I thought, blah blah blah” or “She said, ‘blah blah blah’” are both internal). If there are masses of dialogue, then divide it up reasonably, by phrases. 5. Emotions: If a feeling is followed by the cause or target of the feeling (e.g., “I was happy that he came over”), then it is a SCORE=2 (Internal) phrase. This is because “I was happy” can stand alone, and more information is provided by describing the reason. 129 6. Metacognitive statements (e.g., “I remember”, “let me see if I remember”, “I can envisage” SCORE=1 (External). 7. Quantities: “There were skins” SCORE = 1 (internal); “there were 500 skins” SCORE = 2 (internal) Other segmenting and scoring tips – "Negative" events, or the absence or failure of something to occur ("Bob wasn't there") are still scoreable. External events can include both external episodes and semantic details. In cases where the two are difficult to distinguish, apply the benefit of the doubt rule. Scoring of fragmented sentences should allow for natural speech patterns even when they do not appear fluent in the transcription. The scorer should attempt to interpret fragmented sentences in a way that would be transparent to others. Repetitions should be segmented as finely as internal and other external details Do not give credit for information that is not there. "We went to a place where we could swim with the dolphins" contains one descriptive event detail, but the actual location is not mentioned, so it is not scored under place details. The place is implied, but is not scored until it is mentioned. Remember: Segmentation of details should be consistent regardless of whether the details are internal or external 130 Power and Status (Yu, Hays, & Zhao, 2019) Please indicate the extent to which you agree with the following statements: (1 = strongly disagree, 5 = strongly agree) Power 1. I supervise a large number of subordinates. 2. I formally manage many other people. 3. I can provide rewards to others at my own discretion. 4. I have a great deal of power at work. 5. I have authority to discipline others when needed. 6. My designated role allows me to control a lot of resources. Status 7. Others often seek my opinion because they respect me. 8. I have a good reputation among those I work with. 9. I am highly respected by others at work. 10. People look up to me because I am good at my job. 11. I am admired by others at work because I am seen as competent in my work. 12. Coworkers come to see me because they trust my judgment. 131 Counterproductive Work Behavior (Bennett & Robinson, 2000; Robinson & Bennett, 1995; Stewart et al., 2009) Please indicate the frequency with which you have engaged in the following behaviors at work over the last year. As a reminder, all information is completely anonymous. No one outside of myself (the researcher) will ever have direct access to this data. (1 = never, 2 = infrequently, 3 = somewhat frequently, 4 = frequently, 5 = very frequently, 6 = not applicable to my job) Original 28 Items Over the course of the past year, I have… Production Deviance 1. Worked on a personal matter instead of work for the company 2. Spent too much time fantasizing or daydreaming instead of working 3. Taken an additional or a longer break than is acceptable at the workplace 4. Came in late to work without permission 5. Intentionally worked slower than I could have worked 6. Put little effort into my work 7. Left my work for someone else to finish 8. Called in sick when I was not 9. Littered or dirtied my work environment 10. Repeated a rumor or gossip about the company 11. Neglected to follow my boss’s instructions 12. Discussed confidential company information with an unauthorized person 13. Told someone about the lousy place where I work 14. Left work early without permission Property Deviance 15. Took property from work without permission 16. Used an illegal drug or consumed alcohol on the job 17. Falsified a receipt to get reimbursed for more money than I spent on business expenses 18. Dragged out work in order to get overtime Personal Aggression 19. Said something hurtful to someone at work 20. Acted rudely toward someone at work 21. Lost my temper while at work 22. Made fun of someone at work 23. Cursed at someone at work 24. Publicly embarrassed someone at work 25. Played a mean prank on someone at work 26. Made an obscene comment at work 132 Political Deviance 27. Repeated a rumor or gossip about a coworker or manager at work 28. Made an ethnic, religious, or racial remark or joke at work Supplemental 3 Items Meant to Capture Property Deviance 29. Stole money from the company 30. Sabotaged merchandise 31. Sabotaged equipment 32. Intentionally made errors 33. Misused an expense account 34. Overcharged for services for own profit 35. Covered up mistakes 36. Stole a customer’s possessions 37. Accepted kickbacks 38. Lied about hours worked 39. Misused a discount privilege Supplemental Items Meant to Capture Personal Aggression 40. Verbally abused a coworker 41. Physically abused a coworker 42. Refused to give earned benefits or pay 43. Endangered coworkers by engaging in reckless behavior 44. Sexually harassed a coworker Supplemental Items Meant to Capture Political Deviance 45. Blamed others for mistakes 46. Competed in a way that was not beneficial 47. Showed favoritism 48. Asked another to work beyond their job description 49. Manipulated coworkers to get my way 50. Took credit for another person’s work In the space provided below, please describe a time when you engaged in one of these activities at work. Your response must be in English and at least 150 words in length. 133 Demographics & Control Variables Age (Fill in the blank) What is your current age in years (e.g., 30, 35, 40)? Male Female Other Prefer not to answer Caucasian/White African American/Black Hispanic/Latino Asian/Pacific Islander Native American/American Indian Prefer not to answer Gender (Multiple choice) Ethnicity (Multiple choice, select all that apply) Highest Level of Education Attained (Multiple choice) Annual Income Did not complete high school High school or GED Some college Associate’s degree Bachelor’s degree Master’s degree Doctoral degree or other advanced graduate work <$10,000 $10,000-$19,999 $20,000-$29,999 $30,000-$39,999 $40,000-$49,999 $50,000-$59,999 $60,000-$69,999 $70,000-$79,999 $80,000-$89,999 $90,000-$99,999 $100,000+ 134 Psychological Entitlement (Campbell et al., 2004) Please indicate the extent to which you agree with the following statements: (1 = strongly disagree, 5 = strongly agree) 1. I honestly feel I’m just more deserving than others 2. Great things should come to me 3. If I were on the Titanic, I would deserve to be on the first lifeboat 4. I demand the best because I’m worth it 5. I do not necessarily deserve special treatment (R) 6. I deserve more things in my life 7. People like me deserve an extra break now and then 8. Things should go my way 9. I feel entitled to more of everything Ability to Delay Gratification (Hoerger et al., 2011) Please indicate the extent to which you agree the following statements: (1 = strongly disagree, 5 = strongly agree) Social 1. Usually I try to consider how my actions affect others 2. I try to consider how my actions will affect other people in the long-term 3. I do not consider how my behavior affects other people (R) 4. There is no point in considering how my decisions affect other people (R) Money 5. It is hard for me to resist buying things I cannot afford (R) 6. I try to spend my money wisely 7. I cannot be trusted with money (R) 8. I manage my money well Achievement 9. I am capable of working hard to get ahead in life 10. I cannot motivate myself to accomplish long-term goals (R) 11. I have always felt like my hard work would pay off in the end. 12. I would rather take the easy road in life than get ahead (R) Frustration/Irritability (Holtzman et al., 2015) Please indicate the extent to which you agree with the following statements: 135 (1 = strongly disagree, 5 = strongly agree) 1. I often feel grumpy 2. I often feel like I might snap 3. Other people often get on my nerves 4. Things bother me more than they do the usual person 5. I often feel irritable Social Desirability (Strahan & Gerbasi, 1972; Marlowe & Crown, 1960) Listed below are a number of statements concerning personal attitudes and traits. Read each item and decide whether the statement is true or false as it pertains to you personally. 1. I’m always willing to admit it when I make a mistake. 2. I always try to practice what I preach. 3. I never resent being asked to return a favor. 4. I have never been irked when people expressed ideas very different from my own. 5. I have never deliberately said something that hurt someone’s feelings. 6. I like to gossip at times (R). 7. There have been occasions when I took advantage of someone (R). 8. I sometimes try to get even rather than forgive and forget (R). 9. At times I have really insisted on having things my own way (R). 10. There have been occasions when I felt like smashing things (R). BFI-2-S (Soto & John, 2017) Please indicate the extent to which you agree with the following statements: (1 = strongly disagree, 5 = strongly agree) I am someone who… 1. Tends to be quiet. 2. Is compassionate, has a soft heart. 3. Tends to be disorganized. 4. Worries a lot. 5. Is fascinated by art, music, and literature. 6. Is dominant, acts as a leader. 7. Is sometimes rude to others. 8. Has difficulty getting started on tasks. 9. Tends to feel depressed, blue. 10. Has little interest in abstract ideas. 11. Is full of energy. 12. Assumes the best about people. 136 13. Is reliable, can always be counted on. 14. Is emotionally stable, not easily upset. 15. Is original, comes up with new ideas. 16. Is outgoing, sociable. 17. Can be cold and uncaring. 18. Keeps things neat and tidy. 19. Is relaxed, handles stress well. 20. Has few artistic interests. 21. Prefers to have others take charge. 22. Is respectful, treats others with respect. 23. Is persistent, works until the task is finished. 24. Feels secure, comfortable with self. 25. Is complex, a deep thinker. 26. Is less active than other people. 27. Tends to find fault with others. 28. Can be somewhat careless. 29. Is temperamental, gets emotional easily. 30. Has little creativity. 137 APPENDIX B – Figures and Tables FIGURE 1: Theoretical Model 138 FIGURE 2: Scree Plot for Study 1 Exploratory Factor Analysis 139 FIGURE 3: Path Estimates from Model 1 140 FIGURE 4: Model 1 Interaction between Power and Episodic Foresight in the Prediction of Covert Counterproductive Work Behavior 141 FIGURE 5: Path Estimates from Model 2 142 FIGURE 6: Path Estimates from Model 3 143 FIGURE 7: Model 3 Interaction between Power and Episodic Foresight in the Prediction of Covert Counterproductive Work Behavior 144 FIGURE 8: Path Estimates from Model 4 145 FIGURE 9: Model 4 Interaction between Power and Episodic Foresight in the Prediction of Covert Counterproductive Work Behavior 146 FIGURE 10: Path Estimates from Model 5 147 FIGURE 11: Interaction between Power, Moral Identity Internalization, and Fluid Reasoning in the Prediction of Episodic Foresight 148 FIGURE 12: Interaction between Moral Identity Symbolization and Episodic Foresight in the Prediction of Covert Counterproductive Work Behavior 149 FIGURE 13: Interaction between Moral Identity Symbolization and Episodic Foresight in the Prediction of Covert Counterproductive Work Behavior 150 FIGURE 14: Interaction between Power, Moral Identity Internalization, and Episodic Foresight in the Prediction of Overt Counterproductive Work Behavior 151 FIGURE 15: Interaction between Power, Moral Identity Internalization, and Episodic Foresight in the Prediction of Overt Counterproductive Work Behavior 152 FIGURE 16: Scree Plot for Exploratory Factor Analysis that Includes all 50 Items 153 TABLE 1: Cattell-Horn-Carroll Theory of Intelligence CHC Abilities Definition Narrow abilities within Conceptual Groupings Reasoning Memory and efficiency Memory and efficiency Fluid Reasoning (Gf) “the deliberate but flexible control of attention to solve novel ‘on-the-spot’ problems that cannot be performed by relying exclusively on previously learned habits, schemas, and scripts” (p. 111) Short-term “the ability to encode, maintain, and memory (Gsm) Long-term memory and retrieval (Glr) manipulate information in one’s immediate awareness” (p. 114) “the ability to store, consolidate, and retrieve information over periods of time measured in minutes, hours, days, and years” (p. 116) Induction   General sequential reasoning  Quantitative reasoning  Memory span  Working memory capacity  Associative memory  Meaningful memory  Naming facility  Perceptual speed  Reading speed  Writing speed  Semantic processing speed  Mental comparison speed  Simple reaction time  Speed of limb  Writing speed  Speed of articulation  General verbal information  Language development  Listening ability  Foreign-language proficiency  Skill in lip reading  Geography achievement  Reading decoding  Reading comprehension  Reading speed  Mathematical knowledge  Mathematical achievement  Visualization  Speeded rotation  Length estimation  Phonetic coding  Memory for sound patterns Speed and efficiency Processing speed (Gs) “the ability to perform simple, repetitive cognitive tasks quickly and fluently” Speed and efficiency Reaction and decision speed (Gt) (p. 119) “the speed of making very simple decisions or judgments when items are presented one at a time” (p. 120) Speed and efficiency speed (Gps) Psychomotor “the speed and fluidity with which physical body movements can be made” movement (p. 121) Acquired knowledge Comprehension- “the depth and breadth of knowledge and knowledge skills that are valued by one’s culture” (Gc) (p. 122) Acquired knowledge Domain-specific “the depth, breadth, and mastery of knowledge (Gkn) specialized knowledge (knowledge that not all members of a society are expected Acquired knowledge Acquired knowledge Sensory Sensory to have)” (p. 123) Reading and “the depth and breadth of knowledge and writing (Grw) skills related to written language” (p. 125) Quantitative knowledge (Gq) “the depth and breadth of knowledge related to mathematics” (p. 127) Visual processing “the ability to make use of simulated mental imagery (often in conjunction (Gv) with currently perceived images) to solve Auditory processing (Ga) problems” (p. 129) “the ability to detect and process meaningful nonverbal information in sound” (p. 131) 154 TABLE 1 (cont’d) 155 TABLE 2: Results of Item-Sort Pretask Item/Behavior Taken an additional or a longer break than is acceptable at the workplace Robinson & Bennett (1995) Categorization Production Deviance Anticipated Categorization Results Overt Correctly Categorized Came in late to work without Production Deviance Overt Correctly Categorized permission Left my work for someone else to Production Deviance Overt Correctly Categorized finish Littered or dirtied my work Production Deviance Overt Correctly Categorized environment Repeated a rumor or gossip about the Production Deviance Overt Correctly Categorized company Neglected to follow my boss’s Production Deviance Overt Correctly Categorized instructions Left work early without permission Used an illegal drug or consumed Production Deviance Property Deviance Overt Covert Correctly Categorized Correctly Categorized alcohol on the job Dragged out work in order to get Property Deviance Covert Correctly Categorized overtime Sabotaged merchandise Sabotaged equipment Intentionally made errors Overcharged for services for own profit Covered up mistakes Accepted kickbacks Lied about hours worked Took property from work without permission Property Deviance Property Deviance Property Deviance Property Deviance Property Deviance Property Deviance Property Deviance Property Deviance Manipulated coworkers to get my way Said something hurtful to someone at Political Deviance Personal Aggression work Acted rudely toward someone at work Lost my temper while at work Made fun of someone at work Cursed at someone at work Publicly embarrassed someone at Personal Aggression Personal Aggression Personal Aggression Personal Aggression Personal Aggression work Covert Covert Covert Covert Covert Covert Covert Covert Covert Overt Overt Overt Overt Overt Overt Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Played a mean prank on someone at Personal Aggression Overt Correctly Categorized work Made an obscene comment at work Verbally abused a coworker Physically abused a coworker Refused to give earned benefits or pay Endangered coworkers by engaging in Personal Aggression Personal Aggression Personal Aggression Personal Aggression Personal Aggression reckless behavior Sexually harassed a coworker Spent too much time fantasizing or daydreaming instead of working Personal Aggression Production Deviance Overt Overt Overt Overt Overt Overt Overt Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Correctly Categorized Ambiguous Put little effort into my work Production Deviance Overt Ambiguous 156 TABLE 2 (cont’d) 157 TABLE 3: Results from 12-item Confirmatory Factor Analysis Item Overt Deviance 1. Take an additional or longer break than is acceptable at the workplace 2. Came in late to work without permission 3. Left my work for someone else to finish 4. Littered or dirtied my work environment 5. Repeated a rumor or gossip about the company 6. Neglected to follow my boss’s instructions Covert Deviance 7. Sabotaged merchandise 8. Sabotaged equipment 9. Overcharged for services for own profit 10. Accepted kickbacks 11. Lied about hours worked 12. Took property from work without permission Factor 1 Factor 2 .64 .57 .56 .70 .62 .66 - - - - - - - - - - - - .60 .52 .48 .48 .46 .62 158 TABLE 4: Descriptive Statistics and Correlations among all Variables 159 TABLE 4 (cont’d) 160 TABLE 5: Exploratory Factor Analysis that includes all 50 Items: Two-Factor Solution Item Production Deviance 1. Worked on a personal matter instead of work for the company. 2. Spent too much time fantasizing or daydreaming instead of working. 3. Taken an additional or longer break than it acceptable at the workplace. .547* 4. Came in late to work without permission. 5. Intentionally worked slower than I could have worked. 6. Put little effort into my work. 7. Left my work for someone else to finish. 8. Called in sick when I was not. 9. Littered or dirtied my work environment. 10. Repeated a rumor or gossip about the company. 11. Neglected to follow my boss’s instructions. 12. Discussed confidential company information with an unauthorized person. 13. Told someone about the lousy place where I work. 14. Left work early without permission. Property Deviance 15. Took property from work without permission. 16. Used an illegal drug or consumed alcohol on the job. 17. Falsified a receipt to get reimbursed for more money than I spent on business expenses. 18. Dragged out work in order to get overtime. 19. Stole money from the company. 20. Sabotaged merchandise. 21. Sabotaged equipment. 22. Intentionally made errors. 23. Misused an expense account. 24. Overcharged for services for own profit. 25. Covered up mistakes. 161 .630* .462* .505* .620* .642* .706* .581* Factor 1 Factor 2 .505* .846* .859* .669* .579* .694* .559* TABLE 5 (cont’d) 26. Stole a customer’s possessions. 27. Accepted kickbacks. 28. Lied about hours worked. 29. Misused a discount privilege. Personal Aggression 30. Said something hurtful to someone at work. 31. Acted rudely toward someone at work. 32. Lost my temper while at work. 33. Made fun of someone at work. 34. Cursed at someone at work. 35. Publicly embarrassed someone at work. 36. Played a mean prank on someone at work. 37. Made an obscene comment at work. 38. Verbally abused a coworker. 39. Physically abused a coworker. 40. Refused to give earned benefits or pay. 41. Endangered coworkers by engaging in reckless behavior. 42. Sexually harassed a coworker. Political Deviance 43. Repeated a rumor or gossip about a coworker or manager at work. 44. Made an ethnic, religious, or racial remark or joke at work. 45. Blamed others for mistakes. 46. Competed in a way that was not beneficial. 47. Showed favoritism. 48. Asked another to work beyond their job description. 49. Manipulated coworkers to get my way. 50. Took credit for another person’s work. 162 .960* .565* .469* .672* .606* .886* .747* .915* .502* .602* .587* .433* .840* .653* .804* .672* .763* .714* .424* .477* .761* .584* .543* TABLE 6: Exploratory Factor Analysis that includes all 50 Items: Three-Factor Solution Factor 1 Factor 2 Factor 3 .532* .851* .888* .506* .566* .430* .579* .678* .653* Item Production Deviance 1. Worked on a personal matter instead of work for the company. 2. Spent too much time fantasizing or daydreaming instead of working. 3. Taken an additional or longer break than it acceptable at the workplace. 4. Came in late to work without permission. 5. Intentionally worked slower than I could have worked. 6. Put little effort into my work. 7. Left my work for someone else to finish. 8. Called in sick when I was not. 9. Littered or dirtied my work environment. 10. Repeated a rumor or gossip about the company. 11. Neglected to follow my boss’s instructions. .571* .634* .484* .500* .616* .640* .709* 12. Discussed confidential company information with an unauthorized person. 13. Told someone about the lousy place where I work. .579* 14. Left work early without permission. Property Deviance 15. Took property from work without permission. 16. Used an illegal drug or consumed alcohol on the job. 17. Falsified a receipt to get reimbursed for more money than I spent on business expenses. 18. Dragged out work in order to get overtime. 19. Stole money from the company. 20. Sabotaged merchandise. 21. Sabotaged equipment. 22. Intentionally made errors. 23. Misused an expense account. 24. Overcharged for services for own profit. 163 TABLE 6 (cont’d) 25. Covered up mistakes. 26. Stole a customer’s possessions. 27. Accepted kickbacks. 28. Lied about hours worked. 29. Misused a discount privilege. Personal Aggression 30. Said something hurtful to someone at work. 31. Acted rudely toward someone at work. 32. Lost my temper while at work. 33. Made fun of someone at work. 34. Cursed at someone at work. 35. Publicly embarrassed someone at work. 36. Played a mean prank on someone at work. 37. Made an obscene comment at work. 38. Verbally abused a coworker. 39. Physically abused a coworker. 40. Refused to give earned benefits or pay. 41. Endangered coworkers by engaging in reckless behavior. 42. Sexually harassed a coworker. Political Deviance .622* .765* .621* .738* .720* .411* .479* .411* .750* .947* .445* .494* .488* .534* .566* .610* .910* .726* .909* .450* .574* 43. Repeated a rumor or gossip about a coworker or manager at work. .534* 44. Made an ethnic, religious, or racial remark or joke at work. .546* 45. Blamed others for mistakes. 46. Competed in a way that was not beneficial. 47. Showed favoritism. 48. Asked another to work beyond their job description. 49. Manipulated coworkers to get my way. 50. Took credit for another person’s work. 164 .468* .476* .815* TABLE 7: Exploratory Factor Analysis that includes all 50 Items: Four-Factor Solution Item Production Deviance Factor 1 Factor 2 Factor 3 Factor 4 1. Worked on a personal matter instead of work for the .519* company. 2. Spent too much time fantasizing or daydreaming .685* instead of working. 3. Taken an additional or longer break than it acceptable at the workplace. 4. Came in late to work without permission. 5. Intentionally worked slower than I could have worked. .713* 6. Put little effort into my work. 7. Left my work for someone else to finish. 8. Called in sick when I was not. 9. Littered or dirtied my work environment. 10. Repeated a rumor or gossip about the company. 11. Neglected to follow my boss’s instructions. 12. Discussed confidential company information with an unauthorized person. 13. Told someone about the lousy place where I work. .706* .473* 14. Left work early without permission. .473* Property Deviance 15. Took property from work without permission. 16. Used an illegal drug or consumed alcohol on the job. 17. Falsified a receipt to get reimbursed for more money than I spent on business expenses. 18. Dragged out work in order to get overtime. 19. Stole money from the company. 20. Sabotaged merchandise. 21. Sabotaged equipment. 22. Intentionally made errors. 23. Misused an expense account. .445* .540* .515* .575* .425* .542* .772* .875* .735* .756* .588* .951* 165 TABLE 7 (cont’d) 24. Overcharged for services for own profit. 25. Stole a customer’s possessions. 26. Accepted kickbacks. 27. Lied about hours worked. 28. Misused a discount privilege. Personal Aggression 29. Said something hurtful to someone at work. 30. Acted rudely toward someone at work. 31. Lost my temper while at work. 32. Made fun of someone at work. 33. Cursed at someone at work. 34. Publicly embarrassed someone at work. 35. Played a mean prank on someone at work. 36. Made an obscene comment at work. 37. Verbally abused a coworker. 38. Physically abused a coworker. 39. Refused to give earned benefits or pay. 40. Endangered coworkers by engaging in reckless behavior. 41. Sexually harassed a coworker. Political Deviance 42. Repeated a rumor/gossip about a coworker or manager at work. 43. Made an ethnic, religious, or racial remark or joke at work. 44. Blamed others for mistakes. 45. Competed in a way that was not beneficial. 46. Showed favoritism. 47. Asked another to work beyond their job description. 48. Manipulated coworkers to get my way. 49. Took credit for another person’s work. 166 .853* .874* .651* .783* .624* .713* .725* .476* .653* .684* .740* .654* .490* .441* .731* .511* .562* .600* .876* .983* .673* .844* .471* .708* REFERENCES 167 REFERENCES Abele, A. E., & Wojciszke, B. (2014). Communal and agentic content in social cognition: A dual perspective model. In J. M. Olson & M. P. Zanna (Eds.), Advances in Experimental and Social Psychology, 50, 195-255. Addis, D. R., Musicaro, R., Pan, L., & Schacter, D. L. (2010). Episodic simulation of past and future events in older adults: Evidence from an experimental recombination task. Psychology and Aging, 25, 369–376. Addis, D. R., Wong, A. T., & Schacter, D. L. (2008). Age-related changes in the episodic simulation of past and future events. Psychological Science, 19, 33–41. Alfonso, V. C., Flanagan, D. P., & Radwan, S. (2005). The impact of the Cattell-Horn-Carroll theory on test development and interpretation of cognitive and academic abilities. In D. P. Flanagan & Harrison, P. L. (Eds.), Contemporary intellectual assessment: Theories, tests and issues (185–202). New York, NY: Guilford Publications. Amsteus, M. (2008). Managerial foresight: concept and measurement. Foresight, 10, 53–66. Anderson, C., & Brown, C. E. (2010). The functions and dysfunctions of hierarchy. Research in Organizational Behavior, 30, 55–89. Anderson, J. C., & Gerbing, D. W. (1991). Predicting the performance of measures in a confirmatory factor analysis with a pretest assessment of their substantive validities. Journal of Applied Psychology, 76, 732-740. Aquino, K., Freeman, D., Reed, A., Lim, V. K. G., & Felps, W. (2009). Testing a social- cognitive model of moral behavior: The interactive influence of situations and moral identity centrality. Journal of Personality and Social Psychology, 97, 123–141. Aquino, K., & Reed, A. (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83, 1423–1440. Ashforth, B. E., & Mael, F. (1989). Social identity theory and the Ooganization. Academy of Management Review, 14, 20–39. Atance, C. M., & O’Neill, D. K. (2001). Episodic future thinking. Trends in Cognitive Sciences, 5, 533–539. Bales, R. F. (1958). Task roles and social roles in problem-solving groups. In E. Macoby, T. Newcomb, & E. Hartley (Eds.), Readings in social psychology (437-447). New York, NY: Holt. 168 Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall. Barrett, G. V, Phillips, J. S., & Alexander, R. A. (1981). Concurrent and predictive validity designs: A critical reanalysis. Journal of Applied Psychology, 66, 1–6. Bemis, S. E. (1968). Occupational validity of the General Aptitude Test Battery. Journal of Applied Psychology, 52, 240–244. Bengali, S. (2016, March 9). Adult and underage workers risk their lives in Bangladesh's rising ship-breaking industry. Retrieved from https://www.latimes.com/world/asia/la-fg- bangladesh-ships-20160309-story.html Bennett, R. J., & Robinson, S. L. (2000). Development of a measure of workplace deviance. Journal of Applied Psychology, 85, 349–360. Bentler, P. (1990). Comparative fit indices in structural models. Psychological Bulletin, 107, 238-246. Berry, C. M., Carpenter, N. C., & Barratt, C. L. (2012). Do other-reports of counterproductive work behavior provide an incremental contribution over self-reports? A meta-analytic comparison. Journal of Applied Psychology, 97, 613–636. Berry, C. M., Ones, D. S., & Sackett, P. R. (2007). Interpersonal deviance, organizational deviance, and their common correlates: A review and meta-analysis. Journal of Applied Psychology, 92, 410–424. Bickley, P. G., Keith, T. Z., & Wolfle, L. M. (1995). The three-stratum theory of cognitive abilities: Test of the structure of intelligence across the life span. Intelligence, 20, 309–328. Bilker, W. B., Hansen, J. A., Brensinger, C. M., Richard, J., Gur, R. E., & Gur, R. C. (2012). Development of abbreviated nine-item forms of the Raven’s Standard Progressive Matrices Test. Assessment, 19, 354–369. Blader, S. L., & Chen, Y. R. (2012). Differentiating the effects of status and power: A justice perspective. Journal of Personality and Social Psychology, 102, 994-1014. Bowling, N. A., & Gruys, M. L. (2010). Overlooked issues in the conceptualization and measurement of counterproductive work behavior. Human Resource Management Review, 20, 54–61. Boye, M. W., & Slora, K. B. (1993). The severity and prevalence of deviant employee activity within supermarkets. Journal of Business and Psychology, 8, 245–253. Brown, R. E. (2016). Hebb and Cattell: The genesis of the theory of fluid and crystallized intelligence. Frontiers in Human Neuroscience, 10. 169 Brown, R. P., & Day, E. A. (2006). The difference isn’t black and white: Stereotype threat and the race gap on Raven’s Advanced Progressive Matrices. Journal of Applied Psychology, 91, 979–985. Brown, P., & Levinson, S. C. (1987). Politeness: Some universals in language usage. Cambridge, UK: Cambridge University Press. Bruk-Lee, V., & Spector, P. E. (2006). The social stressors-counterproductive work behaviors link: Are conflicts with supervisors and coworkers the same? Journal of Occupational Health Psychology, 11, 145–156. Buckner, R. L., & Carroll, D. C. (2007). Self-projection and the brain. Trends in Cognitive Sciences, 11, 49–57. Bulley, A., & Gullo, M. J. (2017). The influence of episodic foresight on delay discounting and demand for alcohol. Addictive Behaviors, 66, 1–6. Burkart, J. M., Schubiger, M. N., & van Schaik, C. P. (2017). The evolution of general intelligence. Behavioral and Brain Sciences, 1–65. Burke, H. R. (1972). Raven’s Progressive Matrices: Validity, reliability, and norms. Journal of Psychology: Interdisciplinary and Applied, 82, 253–257. Burns, R. B. (1994). Surveying the cognitive terrain. Educational Researcher, 23, 35−37. Campbell, J. P., & Wiernik, B. M. (2015). The modeling and assessment of work performance. Annual Review of Organizational Psychology and Organizational Behavior, 2, 47–74. Campbell, W. K., Bonacci, A. M., Shelton, J., Exline, J. J., & Bushman, B. J. (2004). Psychological entitlement: Interpersonal consequences and validation of a self-report measure. Journal of Personality Assessment, 8, 29–45. Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. New York, NY: Cambridge University Press. Cattell, R. B. (1941). Some theoretical issues in adult intelligence testing. Psychological Bulletin, 38, 592. Cattell, R. B. (1971). Abilities: Their structure, growth, and action. Boston, MA: Houghton- Mifflin. Cattell, R. B. (1987). Intelligence: Its structure, growth and action. Amsterdam, Netherlands: Elsevier Chan, D. (1996). Criterion and construct validation of an assessment centre. Journal of Occupational and Organizational Psychology, 69, 167–181. 170 Checa, P., & Fernández-Berrocal, P. (2015). The role of intelligence quotient and emotional intelligence in cognitive control processes. Frontiers in Psychology, 6, 1–8. Chen, S., Lee-Chai, A. Y., & Bargh, J. A. (2001). Relationship orientation as a moderator of the effects of social power. Journal of Personality and Social Psychology, 80, 173–187. Chen, P. Y., & Spector, P. E. (1992). Relationships of work stressors with aggression, withdrawal, theft and substance abuse: An exploratory study. Journal of Occupational and Organizational Psychology, 65, 177-184. Cislak, A., Cichocka, A., Wojcik, A. D., & Frankowska, N. (2018). Power corrupts, but control does not: What stands behind the effects of holding high positions. Personality and Social Psychology Bulletin, 44, 944–957. Colzato, L. S., Van Wouwe, N. C., Lavender, T. J., & Hommel, B. (2006). Intelligence and cognitive flexibility: Fluid intelligence correlates with feature "unbinding" across perception and action. Psychonomic Bulletin & Review, 13, 1043-1048. Crowne, D., & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349-354. Dalal, R. S. (2005). A meta-analysis of the relationship between organizational citizenship behavior and counterproductive work behavior. Journal of Applied Psychology, 90, 1241– 1255. Dalley, J. W., Everitt, B. J., & Robbins, T. W. (2011). Impulsivity, compulsivity, and top-down cognitive control. Neuron, 69, 680–694. Daniel, T. O., Said, M., Stanton, C. M., & Epstein, L. H. (2015). Episodic future thinking reduces delay discounting and energy intake in children. Eating Behaviors, 18, 20-24. Daniel, T. O., Stanton, C. M., & Epstein, L. H. (2013). The future is now: Reducing impulsivity and energy intake using episodic future thinking. Psychological Science, 24, 2339–2342. Daruna, J. H., & Barnes, P. A. (1993). A neurodevelopmental view of impulsivity. In W. G. McCown, J. L. Johnson, & M. B. Shure (Eds.), The impulsive client: Theory, research and treatment (23). Washington, D.C.: American Psychological Association. Das, J. P., Kirby, J., & Jarman, R. F. (1975). Simultaneous and successive synthesis: An alternative model for cognitive abilities. Psychological Bulletin, 82, 87-103. DeCelles, K. A., DeRue, D. S., Margolis, J. D., & Ceranic, T. L. (2012). Does power corrupt or enable? When and why power facilitates self-interested behavior. Journal of Applied Psychology, 97, 681–689. DeNisi, A. S., & Shaw, J. B. (1977). Investigation of the uses of self-reports of abilities. Journal 171 of Applied Psychology, 62, 641–644. Diamond, J. (1999). Guns, germs, and steel: The fates of human societies. New York, NY: Norton. Dilchert, S., Ones, D. S., Davis, R. D., & Rostow, C. D. (2007). Cognitive ability predicts objectively measured counterproductive work behaviors. Journal of Applied Psychology, 92, 616–627. Domino, G., & Domino, M. L. (2006). Psychological testing: An introduction. Cambridge, UK: Cambridge University. Dufty, N. F. (1961). The planning function in the business enterprise. Academy of Management Journal, 4, 51–58. Emerson, R. M. (1962). Power-dependence relations. American Sociological Review, 27, 31–41. Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4, 272– 299. Fast, N. J., Sivanathan, N., Mayer, N. D., & Galinsky, A. D. (2012). Power and overconfident decision-making. Organizational Behavior and Human Decision Processes, 117, 249–260. Ferrer, E., O’Hare, E. D., & Bunge, S. A. (2009). Fluid reasoning and the developing brain. Frontiers in Neuroscience, 3, 46–51. Fine, S., & Edward, M. (2017). Breaking the rules, not the law: The potential risks of counterproductive work behaviors among overqualified employees. International Journal of Selection and Assessment, 25, 401–405. Fischer, D. G., & Fick, C. (1993). Measuring social desirability: Short forms of the Marlowe- Crowne social desirability scale. Educational and Psychological Measurement, 53, 417- 424. Flanagan, D. P., McGrew, K. S., & Ortiz, S. O. (2000). The Wechsler intelligence scales and Gf- Gc theory: A contemporary approach to interpretation. Needham Heights, MA: Allyn & Bacon. Flanagan, D. P., & Ortiz, S. O. (2001). Essentials of cross-battery assessment. New York, NY: Wiley. Fox, S., Spector, P. E., & Miles, D. (2001). Counterproductive work behavior (CWB) in response to job stressors and organizational justice: Some mediator and moderator tests for autonomy and emotions. Journal of Vocational Behavior, 59, 291–309. 172 Frick, P. J., Barry, C., & Kamphaus, R. W. (2010). Clinical assessment of child and adolescent personality and behavior. New York, NY: Springer Science & Business Media Galinsky, A. D., Gruenfeld, D. H., & Magee, J. C. (2003). From power to action. Journal of Personality and Social Psychology, 85, 453–466. Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York, NY: Basic Books. Gardner, H. (1993). In a nutshell. In H. Gardner (Ed.), Multiple intelligences: New horizons in theory and practice (5–31). New York, NY: Basic Books. Georgesen, J. C., & Harris, M. J. (1998). Why’s my boss always holding me down? A meta- analysis of power effects on performance evaluations. Personality and Social Psychology Review, 2, 184–195. Georgesen, J. C., & Harris, M. J. (2000). The balance of power: Interpersonal consequences of differential power and expectancies. Personality and Social Psychology Bulletin, 26, 1239– 1257. Glutting, J. J., Watkins, M. W., & Youngstrom, E. A. (2003). Multifactored and cross-battery ability assessments: Are they worth the effort? In C. R. Reynolds & R. W. Kamphaus (Eds.), Handbook of psychological and educational assessment of children: Intelligence, aptitude, and achievement (343–374). New York, NY: Guilford Press. Goffman, E. (1959). The presentation of the self in everyday life. New York, NY: Doubleday. Gonzalez-Mulé, E., Mount, M., & Oh, I.-S. (2014). A meta-analysis of the relationship between general mental ability and non-task performance. Journal of Applied Psychology, 99, 1222– 1243. Gonzalez, C., Thomas, R. P., & Vanyukov, P. (2005). The relationships between cognitive ability and dynamic decision making. Intelligence, 33, 169–186. Gotowiec, S., & van Mastrigt, S. (2018). Having versus doing: The roles of moral identity internalization and symbolization for prosocial behaviors and symbolization for prosocial behaviors. The Journal of Social Psychology, 1–17. Gottfredson, L. S. (1997). Mainstream science on intelligence: An editorial with 52 signatories, history, and bibliography. Intelligence, 24, 13–23. Gottfredson, L. S. (2002). Where and why g matters: Not a mystery. Human Performance, 15, 25–46. Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stanford University Press. 173 Greenberg, J. (1990). Employee theft as a reaction to underpayment inequity: The hidden cost of pay cuts. Journal of Applied Psychology, 75, 561–568. Gruenfeld, D. H., & Tiedens, L. Z. (2010). Organizational preferences and their consequences. In S. T. Fiske, D. T. Gilbert, L. Gardner (Eds.), Handbook of Social Psychology (1252-1287). Hoboken, NJ: John Wiley & Sons. Gruys, M. L., & Sackett, P. R. (2003). Investigating the dimensionality of counterproductive work behavior. International Journal of Selection and Assessment, 11, 30–42. Hardy, S. A., & Carlo, G. (2011). Moral identity: What is it, how does it develop, and is it linked to moral action?, 5, 212–218. Hassabis, D., Kumaran, D., Vann, S. D., & Maguire, E. A. (2007). Patients with hippocampal amnesia cannot imagine new experiences. Proceedings of the National Academy of Sciences, 104, 1726–1731. Hauser, M., Cushman, F., Young, L., Kang-Xing, J., & Mikhail (2007). A dissociation between moral judgments and justifications. Mind & Language, 22, 1-21. Hays, A. F. (2015). An index and test of linear moderated mediation. Multivariate Behavioral Research, 50, 1-22. Hays, N. A., & Bendersky, C. (2015). Not all inequality is created equal: Effects of status versus power hierarchies on competition for upward mobility. Journal of Personality and Social Psychology, 108, 867–882. Hebb, D. O. (1942). The effect of early and late brain injury upon test scores, and the nature of normal adult intelligence. Proceedings of the American Philosophical Society, 85, 275-292. Heidegger, P. (2017, February 1). Press release - Platform publishes list of ships dismantled worldwide in 2016. Retrieved from https://www.shipbreakingplatform.org/press-release- platform-publishes-list-of-ships-dismantled-worldwide-in-2016/ Hershcovis, M. S., Neville, L., Reich, T. C., Christie, A. M., Cortina, L. M., & Shan, J. V. (2017). Witnessing wrongdoing: The effects of observer power on incivility intervention in the workplace. Organizational Behavior and Human Decision Processes, 142, 45–57. Herrnstein, R. J., & Murray, C. (1994). The bell curve: Intelligence and class structure in American life. New York, NY: Free Press. Hinkin, T. (1998). A brief tutorial on the development of measures for use in survey questionnaires. Organizational Research Methods, 1, 104-121. Hertz, S. G., & Krettenauer, T. (2016). Does moral identity effectively predict moral behavior?: A meta-analysis. Review of General Psychology, 20, 129–140. 174 Hoerger, M., Quirk, S. W., & Weed, N. C. (2011). Development and validation of the delaying gratification inventory. Psychological Assessment, 23, 725–738. Hollenbeck, J. R. (2008). The role of editing and knowledge development: Consensus shifting and consensus creation. In Y. Baruch, A. M. Konrad, H. Aguinis, & W. H. Starbuck (Eds.), Opening up the black box of editorship (16-26). London: Palgrave Macmillan. Holtzman, S., O’Connor, B. P., Barata, P. C., & Stewart, D. E. (2015). The brief irritability test (BITe): A measure of irritability for use among men and women. Assessment, 22, 101–115. Horn, J. L. (1965). Fluid and crystalized intelligence: A factor analytic study of the structure among primary mental abilities. Horn, J. L. (1968). Organization of abilities and the development of intelligence. Psychological Review, 75, 242–259. Horn, J. L. (1991). Measurement of intellectual capabilities: A review of theory. In K. S. McGrew, J. K. Werder, & R. W. Woodcock (Eds.), WJ-R technical manual (197–232). Chicago, IL: Riverside. Horn, J. (1998). A basis for research on age differences in cognitive abilities. In J. J. McArdle & R. W. Woodcock (Eds.), Human cognitive abilities in theory and practice (57−92). Mahwah, NJ: Lawrence Erlbaum. Horn, J. L., & Blankson, N. (2005). Foundations for better understanding of cognitive abilities. In D. P. Flanagan, & P. L. Harrison (Eds.), Contemporary intellectual assessment: Theories, tests, and issues (41–68). New York, NY: Guilford Press. Horn, J. L., & Stankov, L. (1982). Auditory and visual factors of intelligence. Intelligence, 6, 165–185. Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6, 1–55. Hunter, J. E., & Hunter, R. F. (1984). Validity and utility of alternative predictors of job performance. Psychological Bulletin, 96, 72–98. Inesi, M. E. (2010). Power and loss aversion. Organizational Behavior and Human Decision Processes, 112, 58–69. Jennings, P. L., Mitchell, M. S., & Hannah, S. T. (2015). The moral self: A review and integration of the literature. Journal of Organizational Behavior, 36, 104–169. Jensen, A. R. (1998). The g factor: The science of mental ability. Westport, CT: Praeger. 175 Jensen, A. R. (2004). Obituary—John Bissell Carroll. Intelligence, 32, 1−5. Jones, E. (2015). One size fits all? Multiple intelligences and legal education. The Law Teacher, 9400, 1–13. Keltner, D., Capps, L., Kring, A. M., Young, R. C., & Heerey, E. A. (2001). Just teasing: A conceptual analysis and empirical review. Psychological Bulletin, 127, 229–248. Keltner, D., Gruenfeld, D. H., & Anderson, C. (2003). Power, approach, and inhibition. Psychological Review, 110, 265–284. Kipnis, D. (1972). Does power corrupt? Journal of Personality and Social Psychology, 24, 33– 41. Kline, R. B. (2011). Principles and practice of structural equation modeling. New York, NY: Guilford Press. Kluemper, D. H., Mossholder, K. W., Ispas, D., Bing, M. N., Iliescu, D., & Ilie, A. (2018). When core self-evaluations influence employees’ deviant reactions to abusive supervision: The moderating role of cognitive ability. Journal of Business Ethics, 1–19. Kohlberg, L. (1973). The claim to moral adequacy of a highest stage of moral judgment. Journal of Philosophy, 70, 630–646. Levine, B., Svoboda, E., Hay, J. F., Winocur, G., & Moscovitch, M. (2002). Aging and autobiographical memory: Dissociating episodic from semantic retrieval. Psychology and Aging, 17, 677–689. Litman, L., Robinson, J., & Abberbock, T. (2017). TurkPrime.com: A versatile crowdsourcing data acquisition platform for the behavioral sciences. Behavior Research Methods, 49, 433– 442. Liu, S., Luksyte, A., Zhou, L., Shi, J., & Wang, M. (2015). Overqualification and counterproductive work behaviors: Examining a moderated mediation model. Journal of Organizational Behavior, 36, 250–271. Locke, E. A. (2005). Why emotional intelligence is an invalid concept. Journal of Organizational Behavior, 26, 425–431. Luksyte, A., Spitzmueller, C., & Maynard, D. C. (2011). Why do overqualified incumbents deviate? Examining multiple mediators. Journal of Occupational Health Psychology, 16, 279–296. Lyons, A. D., Henry, J. D., Rendell, P. G., Corballis, M. C., & Suddendorf, T. (2014). Episodic foresight and aging. Psychology and Aging, 29, 873–884. 176 Lyons, A. D., Henry, J. D., Rendell, P. G., Robinson, G., & Suddendorf, T. (2016). Episodic foresight and schizophrenia. British Journal of Clinical Psychology, 55, 107–122. Lyons, B. D., Hoffman, B. J., Bommer, W. H., Kennedy, C. L., & Hetrick, A. L. (2016). Off- duty deviance: Organizational policies and evidence for two prevention strategies. Journal of Applied Psychology, 101, 463–483. MacCann, C., Joseph, D. L., Newman, D. A., & Roberts, R. D. (2014). Emotional intelligence is a second-stratum factor of intelligence: Evidence from hierarchical and bifactor models. Emotion, 14, 358–374. Magee, J. C., & Galinsky, A. D. (2008). Social hierarchy: The self‐ reinforcing nature of power and status. Academy of Management Annals, 2, 351–398. Marcus, B., Schuler, H., Quell, P., & Humpfner, G. (2002). Measuring counterproductivity: Development and initial validation of a German self-report questionnaire. International Journal of Selection and Assessment, 10, 18–35. Marcus, B., Taylor, O. A., Hastings, S. E., Sturm, A., & Weigelt, O. (2016). The structure of counterproductive work behavior: A review, a structural meta-analysis, and a primary study. Journal of Management, 42, 203–233. Marcus, B., Wagner, U., Poole, A., Powell, D. M., & Carswell, J. (2009). The relationship of GMA to counterproductive work behavior revisited. European Journal of Personality, 23, 489–507. Marsh, I. (2011). Crime and criminal justice. London, UK: Routledge. Martinko, M. J., Gundlach, M. J., & Douglas, S. C. (2002). Toward an integrative theory of counterproductive workplace behavior: A causal reasoning perspective. International Journal of Selection and Assessment, 10, 36–50. Mayer, D. M., Aquino, K., Greenbaum, R. L., & Kuenzi, M. (2012). Who displays ethical leadership, and why does it matter? An examination of antecedents and consequences of ethical leadership. Academy of Management Journal, 55, 151–171. Mayer, J. D., Roberts, R. D., & Barsade, S. G. (2008). Human abilities: Emotional intelligence. Annual Review of Psychology, 59, 507–536. McCrae, R. R., & John, O. P. (1992). An introduction to the Five-Factor Model and its applications. Journal of Personality, 60, 172-215. McGrew, K. S. (1997). Analysis of the major intelligence batteries according to a proposed comprehensive Gf-Gc framework. In D. P. Flanagan & Harrison, P. L. (Eds.), Contemporary intellectual assessment: Theories, tests and issues (119–151). New York, NY: Guilford Publications. 177 McGrew, K. S., & Flanagan, D. P. (1998). The inteligence test desk reference (ITDR): Gf-Gc cross-battery assessment. Boston, MA: Allyn & Bacon. McGrew, K. S. (2009). CHC theory and the human cognitive abilities project: Standing on the shoulders of the giants of psychometric intelligence research. Intelligence, 37, 1–10. Mercado, B. K., Dilchert, S., Giordano, C., & Ones, D. S. (2018). Counterproductive work behaviors. In D. S. Ones, C. Viswesvaran, & H. K. Sinangil (Eds.), The SAGE handbook of industrial, work, and organizational psychology: Personnel psychology and employee performance (109-211). Thousand Oaks, CA: Sage. Mercuri, K., Terrett, G., Henry, J. D., Bailey, P. E., Curran, H. V., & Rendell, P. G. (2015). Episodic foresight deficits in long-term opiate users. Psychopharmacology, 232, 1337– 1345. Moore, A., & Malinowski, P. (2009). Meditation, mindfulness and cognitive flexibility. Consciousness and Cognition, 18, 176-186. Murphy, K. R., Cronin, B. E., & Tam, A. P. (2003). Controversy and consensus regarding the use of cognitive ability testing in organizations. Journal of Applied Psychology, 88, 660– 671. Muthén, L., & Muthén, B. (1998-2012). Mplus User’s Guide. Los Angeles, CA: Muthén & Muthén. NCS Pearson Inc. (2007). Raven’s Standard Progressive Matrices ( SPM ). Neisser, U., Boodoo, G., Bouchard, T. J., Boykin, A. W., Brody, N., Ceci, S. J., Halpern, D. F., Loehlin, J. C., Perloff, R. Sternberg, R. J., & Urbina, S. (1996). Intelligence: Knowns and unknowns. American Psychologist, 51, 77–101. Nisbett, R. E., Aronson, J., Blair, C., Dickens, W., Flynn, J., Halpern, D. F., & Turkheimer, E. (2012). Intelligence: New findings and theoretical developments. American Psychologist, 67, 130–159. O’Leary, A. M., Griffin, R. W., & Glew, D. J. (1996). Organization-motivated aggression: A research framework. Academy of Management Review, 21, 225–253. Ones, D. S., & Dilchert S. (2013). Counterproductive work behaviors: Concepts, measurement, and nomological network. In K. F. Geisinger, B. A. Bracken, J. F. Carlson, J.-I. C. Hansen, N. R. Kuncel, S. P. Reise, & M. C. Rodriguez (Eds.), APA handbooks in psychology. APA handbook of testing and assessment in psychology, Vol. 1: Test theory and testing and assessment in industrial and organizational psychology (643–659). Washington, DC, US: American Psychological Association. Otero, T. M. (2017). Brief review of fluid reasoning: Conceptualization, neurobasis, and 178 applications. Applied Neuropsychology: Child, 6, 204–211. Overbeck, J. R., & Park, B. (2001). When power does not corrupt: Superior individuation processes among powerful perceivers. Journal of Personality and Social Psychology, 81, 549–564. Pind, J., Gunnardsdóttir, E. K., & Jóhannesson, H. S. (2003). Raven’s Standard Progressive Matrices: New school age norms and a study of the test’s validity. Personality and Individual Differences, 34, 375–386. Podsakoff, N. P., Whiting, S. W., Podsakoff, P. M., & Blume, B. D. (2009). Individual- and organizational-level consequences of organizational citizenship behaviors: A meta-analysis. Journal of Applied Psychology, 94, 122-141. Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research, 42, 185–227. Raven, J., Raven, J. C., & Court, J. H. (2000). Raven manual: Section 3. Standard progressive matrices. Ree, M. J., & Carretta, T. R. (2007). Tests of cognitive ability. In D. L. Whetzel & G. R. Wheaton (Eds.), Applied measurement: Industrial psychology in human resources management (131–159). New York, NY: Taylor & Francis/Erlbaum. Ree, M. J., Earles, J. A., & Teachout, M. S. (1994). Predicting job performance: Not much more than g. Journal of Applied Psychology, 79, 518–524. Rendell, P. G., Bailey, P. E., Henry, J. D., Phillips, L. H., Gaskin, S., & Kliegel, M. (2012). Older adults have greater difficulty imagining future rather than atemporal experiences. Psychology and Aging, 27, 1089–1098. Rest, J., Narvaez, D., Bebeau, M., & Thoma, S. (1999). A neo-Kohlbergian approach: The DIT and schema theory. Educational Psychology Review, 11, 291–324. Rest, J. R., Narvaez, D., Thoma, S. J., & Bebeau, M. J. (2000). A neo-Kohlbergian approach to morality research. Journal of Moral Education, 29, 381–395. Robbins, S. P., & Judge, T. A. (2014). Essentials of organizational behavior. Upper Saddle River, NJ: Pearson Education, Inc. Roberts, B. W., Harms, P. D., Caspi, A., & Moffitt, T. E. (2007). Predicting the counterproductive employee in a child-to-adult prospective study. Journal of Applied Psychology, 92, 1427–1436. Robinson, S. L., & Bennett, R. J. (1995). A typology of deviant workplace behaviors: A 179 multidimensional scaling study. Academy of Management Journal, 38, 555–572. Rotundo, M., & Sackett, P. (2002). The relative importance of task, citizenship, and counterproductive performance to global ratings of job performance: A policy-capturing approach. Journal of Applied Psychology, 87, 66–80. Sackett, P. R., & Lievens, F. (2008). Personnel selection. Annual Review of Psychology, 59, 419– 450. Salovey, P., & Mayer, J. D. (1990). Emotional intelligence. Imagination, Cognition and Personality, 9, 185–211. Scherbaum, C. A., Goldstein, H. W., Yusko, K. P., Ryan, R., & Hanges, P. J. (2012). Intelligence 2.0: Reestablishing a research program on g in I-O psychology. Industrial and Organizational Psychology, 5, 128–148. Schmidt, F. L. (2002). The role of general cognitive ability and job performance: Why there cannot be a debate. Human Performance, 15, 187–210. Schmidt, F. L., & Hunter, J. (2004). General mental ability in the world of work: Occupational attainment and job performance. Journal of Personality and Social Psychology, 86, 162– 173. Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124, 262–274. Schneider, W. J., & McGrew, K. S. (2012). The Cattell-Horn-Carroll model of intelligence. In D. P. Flanagan & Harrison, P. L. (Eds.), Contemporary intellectual assessment: Theories, tests and issues (99–144). New York, NY: Guilford Publications. Scott, W. A. (1962). Cognitive complexity and cognitive flexibility. Sociometry, 25, 405-414. See, K. E., Morrison, E. W., Rothman, N. B., & Soll, J. B. (2011). The detrimental effects of power on confidence, advice taking, and accuracy. Organizational Behavior and Human Decision Processes, 116, 272–285. Shao, R., Aquino, K., & Freeman, D. (2008). Beyond moral reasoning: A review of moral identity research and its implications for business ethics. Business Ethics Quarterly, 18, 513–540. Shikishima, C., Hiraishi, K., Yamagata, S., Sugimoto, Y., Takemura, R., Ozaki, K., Okada, M., Toda, T., & Ando, J. (2009). Is g an entity? A Japanese twin study using syllogisms and intelligence tests. Intelligence, 37, 256–267. Skarlicki, D. P., & Folger, R. (1997). Retaliation in the workplace: The roles of distributive, 180 procedural, and interactional justice. Journal of Applied Psychology, 82, 434–443. Skarlicki, D. P., Jaarsveld, D. D. Van, & Walker, D. D. (2008). Getting even for customer mistreatment: The role of moral identity in the relationship between customer interpersonal injustice and employee sabotage. Journal of Applied Psychology, 93, 1335–1347. Slora, K. B. (1989). An empirical approach to determining employee deviance base rates. Journal of Business and Psychology, 4, 199–219. Soto, C. J., & John, O. P. (2017). Short and extra-short forms of the Big Five Inventory-2: The BFI-2-S and BFI-2-XS. Journal of Research in Personality, 68, 69-81. Spearman, C. E. (1904). "General intelligence," objectively determined and measured. American Journal of Psychology, 15, 201-292. Spearman, C. E. (1927). The abilities of man. London, UK: Macmillan. Spector, P. E. (2006). Method variance in organizational research. Organizational Research Methdods, 9, 221–232. Spector, P. E., & Fox, S. (2002). An emotion-centered model of voluntary work behavior. Human Resource Management Review, 12, 269–292. Spector, P. E., & Fox, S. (2005). The stressor-emotion model of counterproductive work behavior. In S. Fox & P. E. Spector (Eds.), Counterproductive work behavior: Investigations of actors and targets (151-174). Washington, DC, US: American Psychological Association. Spector, P. E., Fox, S., Penney, L. M., Bruursema, K., Goh, A., & Kessler, S. (2006). The dimensionality of counterproductivity: Are all counterproductive behaviors created equal? Journal of Vocational Behavior, 68, 446–460. Sternberg, R. J. (1985). Beyond IQ: A triarchic theory of human intelligence. Cambridge, UK: Cambridge University Press. Stewart, S. M., Bing, M. N., Davison, H. K., Woehr, D. J., & McIntyre, M. D. (2009). In the eyes of the beholder: A non-self-report measure of workplace deviance. Journal of Applied Psychology, 94, 207–215. Strahan, R., & Gerbasi, K. C. (1972). Short, homogeneous versions of the Marlow-Crowne social desirability scale. Journal of Clinical Psychology, 28, 191-193. Suddendorf, T. (2006). Foresight and evolution of the human mind. Science, 312, 1006–1007. Suddendorf, T. (2010). Episodic memory versus episodic foresight: Similarities and differences. Wiley Interdisciplinary Reviews: Cognitive Science, 1, 99–107. 181 Suddendorf, T. (2017). The emergence of episodic foresight and its consequences. Child Development Perspectives, 11, 191–195. Suddendorf, T., & Corballis, M. C. (2007). The evolution of foresight: What is mental time travel and is it unique to humans. Behavioral and Brain Sciences, 30, 299–351. Suddendorf, T., & Moore, C. (2011). Introduction to the special issue: The development of episodic foresight. Cognitive Development, 26, 295–298. Suddendorf, T., & Redshaw, J. (2013). The development of mental scenario building and episodic foresight. Annals of the New York Academy of Sciences, 1296, 135–153. The Guardian (2017, December 2). 'This is the world's cheapest place to scrap ships' - but in Chittagong, it's people who pay the price. Retrieved from https://www.theguardian.com/global-development/2017/dec/02/chittagong-shipbreaking- yards-legal-fight Thorndike, E. L. (1920). Intelligence and its use. Harper's Magazine, 140, 227-235. Tost, L. P., Gino, F., & Larrick, R. P. (2012). Power, competitiveness, and advice taking: Why the powerful don’t listen. Organizational Behavior and Human Decision Processes, 117, 53–65. Tost, L. P., Gino, F., & Larrick, R. P. (2013). When power makes others speechless: The negative impact of leader power on team performance. Academy of Management Journal, 56, 1465–1486. Trevino, L. K., den Nieuwenboer, N. A., & Kish-Gephart, J. J. (2014). (Un)ethical behavior in organizations. Annual Review of Psychology, 65, 635–660. Vierra, A. (2016). Psychopathy, mental time travel, and legal responsibility. Neuroethics, 9, 129– 136. Visser, B. A., Ashton, M. C., & Vernon, P. A. (2006). g and the measurement of multiple intelligences: A response to Gardner. Intelligence, 34, 507–510. Viswesvaran, C., & Ones, D. S. (2000). Perspectives on models of job performance. International Journal of Selection and Assessment, 8, 216–226. Vitell, S. J., Bing, M. N., Davison, H. K., Ammeter, A. P., Garner, B. L., & Novicevic, M. M. (2009). Religiosity and moral Iidentity: The mediating role of self-control. Journal of Business Ethics, 88, 601–613. Waschl, N. A., Nettelbeck, T., Jackson, S. A., & Burns, N. R. (2016). Dimensionality of the Raven’s Advanced Progressive Matrices: Sex differences and visuospatial ability. Personality and Individual Differences, 100, 157–166. 182 Watts, J., O. Sheehan, Q. D. Atkinson, J. Bulbulia, & Gray, R. D. (2016). Ritual human sacrifice promoted and sustained the evolution of stratified societies. Nature, 532, 228-231. Wechsler, D. (1958). The measurement and appraisal of adult intelligence. Baltimore, MD: Williams & Wilkins. Whetten, D. A. (1989). What Constitutes a theoretical contribution? Academy of Management Review, 14, 490–495. White, J. L., Moffitt, T. E., & Silva, P. (1989). A prospective replication of the protective effects of IQ in subjects at high risk for juvenile delinquency. Journal of Consulting and Clinical Psychology, 57, 719–724. Wilson, J. Q., & Herrnstein, R. J. (1985). Crime and human nature. New York, NY: Simon & Schuster. Winstanley, C. A., Eagle, D. M., & Robbins, T. W. (2006). Behavioral models of impulsivity in relation to ADHD: Translation between clinical and preclinical studies. Clinical Psychology Review, 26, 379–395. Winterich, K. P., Aquino, K., Mittal, V., & Swartz, R. (2013). When moral identity symbolization motivates prosocial behavior: The role of recognition and moral identity internalization. Journal of Applied Psychology, 98, 759–770. Wu, J., & Lebreton, J. M. (2011). Reconsidering the dispositional basis of counterproductive work behavior: The role of aberrant personality. Personnel Psychology, 64, 593–626. Zaval, L., Li, Y., Johnson, E. J., & Weber, E. U. (2015). Complementary contributions of fluid and crystallized intelligence to decision making across the life span. In T. Hess, J. Strough, & C. Lockenhoff (Eds.), Aging and Decision Making: Empirical and Applied Perspectives (149–168). Cambridge, MA: Academic Press. Zhang, Y.-J., Wan, B., Zhao, J., & Bashir, M. (2011). A Studying on Proactive and Reactive Counterproductive Work Behavior. 2011 International Conference on Information Management, Innovation Management and Industrial Engineering, (2000), 150–153. 183