Personality and Counterproductive Work Behavior
KeywordsIntegrity Test Personality Variable Emotional Stability Counterproductive Work Behavior Intentional Behavior
Counterproductive work behavior (CWB) comprises “any intentional behavior on the part of an organization member viewed by the organization as contrary to its legitimate interests” (Gruys and Sackett 2003, p. 30).
Personality characteristics have been found useful to describe, understand, and explain a broad array of work-related behavior and processes. One of the most important work-related constructs is employees’ job performance, defined as “the total expected value to the organization of the discrete behavioral episodes that an individual carries out over a standard period of time” (Motowidlo and Kell 2013, p. 82). Job performance is considered to encompass three broad domains, namely, task performance, organizational citizenship behavior (or contextual performance), and counterproductive work behavior (CWB; Rotundo and Sackett 2002). CWB can harm the well-being of others (e.g., co-workers) and can result in very considerable financial costs for the organization and society (Bennett and Robinson 2000). Correspondingly, researchers and practitioners have aimed to identify predictors or correlates of CWB in order to design interventions that can help reducing it. One approach in this regard has linked personality variables, which could be considered in staffing decision processes, to CWB. Current knowledge concerning the relations between personality and CWB is summarized herein.
Specifically, the chapter is structured as follows: After (i) CWB is introduced in more detail, the links are described between CWB and (ii) gender and age, (iii) cognitive abilities, (iv) basic personality traits (in terms of the Five-Factor Model of Personality and the HEXACO Model of Personality), and dark traits as well as (v) other personality characteristics. Thereafter, (vi) interaction effects between personality variables and context factors with regard to CWB are sketched, before finally (vii) a conclusion and outlook is given.
Counterproductive Work Behavior
Counterproductive work behavior (CWB) comprises “any intentional behavior on the part of an organization member viewed by the organization as contrary to its legitimate interests” (Gruys and Sackett 2003, p. 30). This definition implies several aspects. First, it relates CWB to behavior and not to the outcomes of behavior. For example, playing mean pranks on co-workers represents CWB whether or not it produces negative states in co-workers. Second, it relates CWB to intentional behavior. Behavior that is accidentally against the organizations’ interests does not represent CWB. For example, making a mistake due to a lack of knowledge does not represent CWB. Third, the behavior must be shown by a current member of the organization. Behavior shown by people outside of the organization such as clients, competitors, or former employees does not represent CWB. Fourth, the behavior must be viewed by the organization – if it knows or would know about the behavior – as contrary to its legitimate interests. This includes behavior that is directed toward the organization as an entity (e.g., stealing money from the organization’s budgets), as well as behavior that is directed toward people related to the organization (e.g., mobbing co-workers, insulting clients). Finally, CWB refers to all these kinds of behaviors, irrespective of their severity. For example, bullying a co-worker can be rather mild or intense, or taking too many breaks might be less detrimental for one’s organization as compared to stealing a lot of money from it, but all kinds of intentional behavior which are opposite to the organization’s interests are CWB (for further information on the severity of CWB, see, e.g., Robinson and Bennett 1995).
Researchers have described different types of CWB. Gruys and Sackett (2003), for instance, list the following 11 categories: theft and related behavior, destruction of property, misuse of information (e.g., falsifying documents), misuse of time and resources (e.g., doing personal business at work), unsafe behavior (e.g., not following safety instructions), poor attendance, poor quality work, alcohol use, drug use, inappropriate verbal actions, and inappropriate physical actions. A more condensed categorization has been introduced by Marcus et al. (2002), distinguishing between absenteeism/withdrawal, aggression, substance use, theft/property violations, and miscellaneous forms. Different categories of CWB are positively correlated with each other (e.g., .32 ≤ r ≤ .68 in Marcus et al. 2002).
Notwithstanding such categorizations describing different types of CWB, the most commonly used categorization – prominently introduced by Robinson and Bennett (1995) and Bennett and Robinson (2000) and also considered by Gruys and Sackett (2003) as well as Marcus et al. (2002) as a fundamental differentiation – distinguishes between the target of CWB, namely, either other people (interpersonal-directed CWB, CWB-I) or the organization (organization-directed CWB, CWB-O). CWB-I and CWB-O are highly correlated with each other; meta-analyses report associations of ρ = .62 (Berry et al. 2007) and ρ = .70 (Dalal 2005). Correspondingly, some have used one composite score of CWB only. On the other hand, as described in more detail in the following sections, CWB-I and CWB-O show some substantial differences with regard to their predictors and correlates, indicating support for their differentiation.
Summarizing thus far, CWB represents intentional behavior by organization members against the legitimate interests of the organization. CWB can have several forms and can vary in severity, but people who show some kind of CWB are more likely also to show other kinds of CWB. Most commonly, researchers and practitioners either use one composite score of CWB or differentiate between CWB-I and CWB-O.
Gender, Age, and CWB
According to meta-analyses by Berry et al. (2007) and Ng et al. (2016), there is a weak relation between gender and CWB, with women showing less CWB than men. This holds for CWB-I, CWB-O, and CWB overall. However, Ng et al. (2016) found variation in the strength of correlation based on the rating source of CWB. More precisely, they found sample-sized weighted corrected correlations between gender and CWB between .00 and |.19|, depending on the CWB category (overall, CWB-I, CWB-O) and rating source (self-rating, supervisor rating, peer rating, archival records).
Ng et al. (2016) explained these overall small gender differences with social role theory. Specifically, they assumed that women should show less CWB because they generally aim to foster social harmony and, in turn, refrain from behavior that threatens work-related cohesion (with regard to other people or the organization). This gender effect, however, might be – partly or largely – mitigated by another gender effect based on social role theory, namely, that men emphasize their careers more strongly and therefore refrain from behavior that threatens their careers (i.e., refraining from CWB because it threatens one’s career, if detected by others). Ng et al. (2016) speculated that these two opposite gender effects both derived from social role theory result in the weak association between gender and CWB overall, but they also emphasized the need for more research in this regard.
Concerning the link between age and CWB, the meta-analyses by Berry et al. (2012), Berry et al. (2007), and Ng and Feldman (2008) indicated that younger employees are slightly more likely to show CWB as compared to older employees. Again, in these meta-analyses, the strength of the relation differed due to both the CWB category and the CWB rating source, with correlations typically in between −.15 and −.05. Though there have been first attempts to explain this small linear relation (e.g., older employees might be better able to control and regulate their behavior), a thorough theorizing on the link between age and CWB is still lacking. Future researchers investigating this link might also consider the finding by Ng and Feldman (2008) that age does not have a negative linear but a curvilinear relation with CWB.
Cognitive Abilities and CWB
Compared to research on the link between cognitive abilities and task performance, there are only a few studies on the link between cognitive abilities and CWB. Recently, Gonzales-Mulé et al. (2014) summarized the studies linking general mental ability (GMA) to CWB, finding a mean true-score correlation between these constructs of −.02. While this implies that there is no link between GMA and CWB overall, Gonzales-Mulé et al. (2014) found that the strength of relation differed due to the rating source of CWB. Specifically, they found a very small positive relation when CWB was assessed via self-ratings (mean true-score correlation of .05), and a small negative relation when CWB was assessed via supervisor ratings or objective records (mean true-score correlation of −.11). This difference might be explained by the differential detection hypothesis, stating that people with a higher GMA do not show more or less CWB than those with a lower GMA but are better at concealing their CWB from supervisors and objective records.
Basic Personality Traits, Dark Traits, and CWB
Many studies have investigated the links between basic personality traits and CWB. Using the framework of the Five-Factor Model of Personality, Berry et al. (2007) found in their meta-analysis that especially agreeableness, conscientiousness, and emotional stability were negatively related to CWB. Again, they found some important differences based on the CWB category. Specifically, agreeableness was more strongly related to CWB-I (ρ = −.46) than to CWB-O (ρ = −.32), conscientiousness was more strongly related to CWB-O (ρ = −.42) than to CWB-I (ρ = −.23), while emotional stability was similarly related to both domains (ρ = −.23 for CWB-O and ρ = −.24 for CWB-I). No strong relations were found between CWB and openness to experience (ρ = −.04 for CWB-O and ρ = −.09 for CWB-I) and extraversion (ρ = −.09 for CWB-O and ρ = .02 for CWB-I), respectively.
Roughly speaking, researchers have explained these links as follows: Compared to their counterparts with low levels, employees high in agreeableness show less CWB because they are generally kind, sympathetic, warm, etc.; employees high in conscientiousness show less CWB because they are generally diligent, self-controlled, adhering to rules, etc.; and employees high in emotional stability show less CWB because they are generally calm, even-tempered, relaxed, etc.
In a recent meta-analysis, Berry et al. (2012) compared the links between the Big Five and CWB considering two rating sources of the latter: self-ratings and other ratings. They found differences concerning the strengths of correlations depending on the CWB rating source in the range between |.07| and |.18|. At this juncture, the pattern of correlations between, on the one hand, agreeableness, conscientiousness, and openness to experience and, on the other hand, CWB was quite similar across self-ratings and other ratings of CWB. Potentially negligible differences were found for extraversion as well, which was marginally negatively correlated with self-rated CWB and marginally positively correlated with other-rated CWB. Most strikingly, however, emotional stability was only marginally correlated with other-rated CWB (−.05) but more strongly with self-rated CWB (−.23).
As an adaptation and extension of the five-factor model, the HEXACO model of personality suggests that there are six basic personality traits, including honesty-humility (Ashton et al. 2014). People high in honesty-humility are characterized as being cooperative, honest, modest, sincere, etc. In the light of these attributes, it has been assumed that people high in honesty-humility show less CWB than their counterparts low in honesty-humility. Specifically, it has been assumed that those high in honesty-humility rather refrain from CWB because their cooperative and sincere nature may function as an internal inhibition capacity against showing CWB and because those high in honesty-humility are generally less interested in money, luxury goods, etc., so that they experience less motivation to obtain these via CWB (Zettler and Hilbig 2010). Corresponding studies have not only demonstrated a negative link between honesty-humility and CWB in general, but have also indicated that, besides conscientiousness, honesty-humility is the basic personality trait most strongly associated with CWB (e.g., Cohen et al. 2014; Marcus et al. 2007; Zettler and Hilbig 2010).
Given the negative connotation of CWB, researchers have also investigated its relations with dark personality traits, which are considered to lie somewhat outside of the personality space of the five-factor model. The most prominent dark traits are the components of the dark triad, i.e., Machiavellianism, narcissism, and psychopathy. Roughly speaking, Machiavellianism encompasses “a cynical view of human nature and a deceitful and calculating interpersonal style” (p. 169), narcissism encompasses “a sense of self-importance, exhibitionism, entitlement, and interpersonal exploitation” (pp. 169–170), and psychopathy encompasses “grandiosity, lack of empathy, a glib and manipulative interpersonal style, shallow affect, and a parasitic and antisocial lifestyle” (p. 169; all Lee et al. 2013).
O’Boyle et al. (2012) conducted a meta-analysis on the links between these dark traits and CWB and found that especially narcissism (corrected for unreliability, r = .43), but also Machiavellianism (corrected for unreliability, r = .25), and marginally psychopathy (corrected for unreliability, r = .07) were positively associated with CWB. This is well in line with the reported negative link between honesty-humility and CWB as low honesty-humility overlaps with the common core of the dark triad (e.g., Lee et al. 2013).
In addition to these linear links between basic or dark personality traits and CWB, Le et al. (2011) found support for a curvilinear relation between emotional stability and CWB. Specifically, intermediate levels of emotional stability were associated with the lowest levels of CWB. Le et al. (2011) speculated that people high in emotional stability are too overcontrolled to respond to work stressors in an appropriate manner, resulting in a U-shaped relation between emotional stability and CWB. Further more, Le et al. (2011) also found support in one out of two studies for a U-shaped relation between conscientiousness and CWB. That is, intermediate levels of conscientiousness were linked to the lowest levels of CWB. Given the increasing support for curvilinear relations in organizational life in general and these first indications, future researchers might more consistently investigate the possibility of such curvilinear relations between personality traits and CWB.
Researchers have also started to examine interaction effects between different traits when (statistically) predicting CWB. Jensen and Patel (2011), for example, investigated interaction effects between the three traits of the Big Five that were related to CWB most strongly in the meta-analysis by Berry et al. (2007), i.e., interaction effects between conscientiousness and emotional stability, between conscientiousness and agreeableness, as well as between agreeableness and emotional stability. For all three interactions and with regard to both CWB-I and CWB-O, they found that CWB was lowest when people had high levels in both traits (e.g., in both conscientiousness and emotional stability). Interestingly, there were virtually no differences between people having low levels in both traits (e.g., in both conscientiousness and emotional stability) or in one trait only (e.g., having low levels in conscientiousness but high levels in emotional stability). However, these interaction patterns do not correspond to the findings by Bowling et al. (2011) who investigated similar interaction effects. Specifically, Bowling et al. (2011) investigated interaction effects between, on the one hand, negative affectivity (a proxy for low emotional stability) and, on the other hand, conscientiousness and agreeableness. Bowling et al. (2011) found that CWB was highest when people had low levels in both emotional stability and conscientiousness/agreeableness but was virtually similarly low when people had either high levels in both emotional stability and conscientiousness/agreeableness or in one trait only. Besides these interaction effects, further studies have reported interaction effects between self- and observer ratings of the same trait or between honesty-humility and extraversion.
Overall, basic personality traits are related to CWB. Regarding the Big Five, links have been found between CWB and emotional stability (negatively linear and U-shaped), conscientiousness (negatively linear – though stronger with regard to CWB-O than to CWB-I – and, though in one study only, U-shaped), and agreeableness (negatively linear – though stronger with regard to CWB-I than to CWB-O). As compared to using the five-factor model, using the HEXACO model as a framework of basic personality traits seems advantageous for (statistically) predicting CWB. Specifically, from among all basic personality traits, honesty-humility, together with conscientiousness, has been found to be linked to CWB most strongly, namely, in a negative linear manner. This is further supported by positive relations between the dark triad components and CWB as the common core of the dark triad overlaps with low honesty-humility. Research has also indicated potential interaction effects between personality traits when (statistically) predicting CWB, although the exact patterns are unclear currently.
Other Personality Characteristics and CWB
In addition to gender, age, cognitive abilities, as well as basic and dark traits, many other personality characteristics have been linked to CWB. Among these, the most prominent correlates of CWB are arguably self-control and integrity tests. Capturing one’s tendency to repress present desires in order to avoid negative consequences and/or to reach long-term goals, self-control has been considered to play a major role in explaining crime in general and CWB in particular. And, indeed, several studies have found a relatively strong negative link between self-control and CWB (e.g., Marcus and Schuler 2004). This finding is well in line with the described importance of conscientiousness and honesty-humility for CWB as self-control has been found to overlap with these traits.
Integrity tests are a compound of items that are related to CWB. Generally, there are two types of integrity tests, namely, overt integrity tests and personality-based integrity tests. Overt integrity tests comprise items that are obviously related to counterproductive activities, e.g., attitudes toward theft. Personality-based integrity tests comprise items capturing personality characteristics that are related to counterproductive activities. Thus, while the purpose of overt integrity tests is relatively easy to read from the items, items from personality-based integrity tests are, for respondents, not that obviously linked to CWB.
Two meta-analyses summarized the associations of integrity tests with CWB (Sackett and Schmitt 2012). However, for predictive studies involving job applicants and using non-self-report measures of CWB, the meta-analyses differ quite strongly with regard to the found strength of relation. While one meta-analysis reported mean correlations between integrity tests and CWB for these kind of studies of .20 (personality-based integrity tests) and .27 (overt integrity tests), respectively, the other meta-analysis reported a mean correlation of .09 only (merging overt and personality-based integrity tests). Unfortunately, the meta-analyses differ with regard to some methodological aspects, and both meta-analyses have some specific strengths and limitations, so that it is rather difficult to say currently which meta-analysis reflects reality better (Sackett and Schmitt 2012).
Though the strength of relation between integrity tests and CWB is unclear, at least for studies involving job applicants and using non-self-report measures of CWB, the finding that integrity tests are correlated with CWB is further supported by studies linking integrity tests to basic personality traits. Specifically, corresponding studies have indicated that both kinds of integrity tests are correlated with conscientiousness and honesty-humility in particular (e.g., Marcus et al. 2007), which, as described above, are the two basic traits most strongly linked to CWB.
The wide array of further personality characteristics that have been linked to CWB can be summarized only roughly herein. In an impressive series of investigations, Cohen and her colleagues (e.g., Cohen et al. 2014), for instance, considered the role of more than 20 individual difference constructs – reflecting one’s abilities, identity, and motivation – for shaping a moral character, which, in turn, was negatively associated with CWB. Next to conscientiousness, honesty-humility, Machiavellianism, and self-control, Cohen et al. (2014) found support for the importance of such constructs as empathetic concern, guilt proneness, or an internalized moral identity for CWB. Further constructs that have shown substantial relations with CWB are circumplex traits (i.e., traits representing the intersection of two personality traits), core self-evaluations (i.e., a composite of emotional stability, generalized self-efficacy, internal locus of control, and self-esteem; negative relation), negative affectivity (positive relation), positive psychological capital (i.e., a composite of hope, optimism, resiliency, and self-efficacy; negative relation), and the RIASEC (realistic, investigative, artistic, social, enterprising, conventional) vocational interests when fitting to one’s job (negative relation).
Person-Situation Interactions and CWB
Several studies have investigated person-situation interactions when (statistically) predicting CWB. The basic reasoning underlying these studies is that both person-and situation-factors mutually shape behavior. Studies investigating person-situation interactions with regard to CWB are typically based on situation strength theory or on trait activation theory (for an overview of these theories, see, e.g., Judge and Zapata 2015).
In simplified terms, situation strength theory suggests that situations can be rather strong or weak. Strong situations are characterized by several aspects – e.g., consistency, guidelines, structures – which suggest a specific kind of behavior. As a consequence, there should be low variability in the behavior of different people. In other words, in strong situations, most people should behave similarly, while in weak situations individual differences should come into play. Whereas situation strength theory thus allows predictions about the interplay between different personality characteristics and situation strength at once (i.e., different personality characteristics are more important in weak situations), trait activation theory allows predictions about the specific interplay between one personality characteristic and the context of behavior. More precisely, according to trait activation theory, the relation between a personality characteristic and an outcome of interest is stronger when people are in situations which are trait relevant as compared to situations without trait-relevant aspects. So, trait activation theory focuses on the fit between personality, context, and the outcome of interest. Therefore, the two theories – situation strength theory and trait activation theory – are complementary not mutually contradictory.
Irrespective of the specific underlying theory, several studies have found support for person-situation interactions when (statistically) predicting CWB. Examples span across interaction effects between agreeableness and global situational strength (Meyer et al. 2014), between gender and organizational constraints (Spector and Zhou 2014) or between honesty-humility and perceptions of organizational politics (Zettler and Hilbig 2010). Often, the reported interactions indicate that CWB is highest when, on the one hand, employees have personality characteristics that are positively related to CWB and, on the other hand, contextual factors are present that are generally more likely to result in CWB. Interestingly, the combination of having personality characteristics that are negatively related to CWB and the presence of context factors that are negatively related to CWB often seems to result in the same low – but not lower – levels of CWB as the combination of either having personality characteristics that are negatively related to CWB or the presence of context factors that are negatively related to CWB. Note that some studies even found support for three-way interactions, e.g., between two personality traits and job stressors (Zhou et al. 2014).
As a dimension of employees’ job performance and potentially entailing very considerable financial and social costs, counterproductive work behavior is one of the most important constructs in organizational life. As it is the case with regard to many other important constructs, personality variables have been linked successfully to CWB. Whereas gender, age, and general mental ability show small to moderate correlations with CWB, especially the basic traits of conscientiousness and honesty-humility are negatively related to CWB. Agreeableness, Machiavellianism, narcissism, and self-control are further personality traits that could be associated with CWB substantially, all of which overlapping with conscientiousness and honesty-humility. The same is the case for – both overt and personality-based – integrity tests.
In addition to more basic personality characteristics, further personality variables have shown associations with CWB. Examples span across guilt proneness, negative affectivity, or vocational interests (when fitting to one’s job). Extending assumptions about simple linear relations, some curvilinear relations between personality variables and CWB have been reported, especially concerning emotional stability. In a similar vein, studies have found interaction effects between different personality variables as well as between personality variables and organizational context factors (e.g., global situational strength, perceptions of organizational politics) when (statistically) predicting CWB. However, corresponding results sometimes differ with regard to the exact patterns of interactions.
Across the studies investigating the role of personality for CWB, some differences based on the CWB category in question (most prominently, CWB composite, CWB-I, or CWB-O) or rating source (e.g., self-reports, other reports, objective records) were found. Future research might thus more strongly focus on diagnostic-related issues when investigating the role of personality for CWB, preferably assessing different forms of CWB, using different rating sources, and relying on generalizable samples. Indeed, studies have indicated that CWB research generally suffers from low response rates (and low base rates) of CWB (Grecko et al. 2015) and that employees who report CWB of others differ from employees who do not report CWB of others (Gruys et al. 2010).
From a conceptual perspective, future research might more strongly aim to clarify the specific contributions of overlapping personality characteristics for understanding CWB. Further, more research on mediation processes on the importance of trait and state differences of personality or on the importance of personality change for CWB seems needed. Currently, most research focuses on the role of personality for showing CWB. Extending this perspective, future studies might more thoroughly investigate potential effects of CWB on personality (change) or emphasize the role of personality for becoming a target – and not only an actor – of CWB. For instance, a study by Scott and Judge (2013) indicated important individual differences between employees becoming more vs. less often targets of CWB.
Definitely, linking personality to CWB has been a fruitful research area for many years and has significantly contributed to the acceptance of considering personality variables in research and practice. Indeed, many robust associations between personality and CWB have been identified. Hopefully, there will be corresponding efforts in the future to even further understand the relations between personality and CWB.
- Bowling, N. A., Burns, G. N., Stewart, S. M., & Gruys, M. L. (2011). Conscientiousness and agreeableness as moderators of the relationship between neuroticism and counterproductive work behaviors: A constructive replication. International Journal of Selection and Assessment, 19, 320–330. doi:10.1111/j.1468-2389.2011.00561.x.CrossRefGoogle Scholar
- Meyer, R. D., Dalal, R. S., José, I. J., Hermida, R., Chen, T. R., Vega, R. P., et al. (2014). Measuring job-related situational strength and assessing its interactive effects with personality on voluntary work behavior. Journal of Management, 40, 1010–1041. doi:10.1177/0149206311425613.CrossRefGoogle Scholar
- Motowidlo, S. J., & Kell, H. J. (2013). Job performance. In N. W. Schmitt & S. Highhouse (Eds.), Handbook of psychology, Industrial and organizational psychology (Vol. 12, 2nd ed., pp. 82–103). Hoboken: Wiley.Google Scholar
- Zettler, I., & Hilbig, B. E. (2010). Honesty-humility and a person-situation interaction at work. European Journal of Personality, 24, 569–582. doi:10.1002/per.757.Google Scholar