Maladaptive Personality at Work: Exploring the Darkness
Important changes in how personality is conceptualized and measured are occurring in clinical psychology. We focus on 1 aspect of this work that industrial psychologists have been slow to embrace, namely, a new trait model that can be viewed as a maladaptive counterpart to the Big 5. There is a conspicuous absence of work psychology research emerging on this trait model despite important implications for how we understand personality at work. We discuss objections to the trait model in a work context and offer rejoinders that might make researchers and practitioners consider applying this model in their work. We hope to stimulate discussion of this topic to avoid an unnecessary bifurcation in the conceptualization of maladaptive personality between industrial and clinical settings.
Personality researchers agree on a number of points. There are between three and seven broad primary dimensions of normal personality, the Big Five plus or minus one or two (Ashton & Lee, 2008; Eysenck, 1991; Goldberg, 1990; Hogan, 1986; McCrae & Costa, 1999). These dimensions can be represented by a higher-order structure of fewer dimensions. Digman (1997) showed covariation between the Big Five factors of Neuroticism, Agreeableness, and Conscientiousness is accounted for by a factor called alpha, whereas covariation among the remaining factors of Extraversion and Openness of the Big Five is accounted for by a factor called beta. DeYoung (2010) reviewed psychometric and neuropsychological evidence for a similar higher-order factor structure and labeled the factors stability and plasticity. Beneath the primary factors reside two aspects per dimension, and under these sit a yet-to-be-determined number of narrower facets (DeYoung, Quilty, & Peterson, 2007; Perugini & Gallucci, 1997). The phenotypic expression of the genetic basis of personality is moderated by environmental factors (Roberts & Jackson, 2008).
Industrial psychologists have focused their research on the implications of personality for the workplace. Meta-analysis has established that personality is a moderately effective predictor of how workers go about their jobs, known as contextual performance (Hough, 1992; Salgado, 2003). Meta-analysis has also demonstrated that certain personality dimensions, if well chosen, are moderate predictors of what workers will do in their jobs, or task performance (Barrick & Mount, 1991; Schmidt & Hunter, 1998). Motivational mechanisms mediate the relationship between personality traits and job performance (Barrick, Stewart, & Piotrowski, 2002). Some dimensions are important to all jobs and the importance of others varies by job, (Barrick, Mount, & Judge, 2001). For greatest predictive efficacy, the breadth of trait measures should match the breadth of outcome models (Paunonen, 1998). Despite a seemingly sophisticated conception of personality, since the 1990s our understanding of personality at work has only been marginally refined, and incrementally at that. It seems reasonable to say that no major developments have redirected the course of work related personality research as significantly as the integration of the Big Five taxonomy and the meta-analytic method in the 1990s (e.g., Barrick & Mount, 1991; Salgado, 1997).
In the interim, dramatic changes have occurred in the field of abnormal personality. The essence of the drama is whether personality disorders are categorical “types” or continuous “dimensions.” The preponderance of evidence supports a dimensional view (Krueger & Eaton, 2010). Now attention is on what the dimensions underlying personality disorder are and how they relate to models of normal personality (Harkness, Finn, McNulty, & Shields, 2012; Krueger et al., 2011). The American Psychiatric Association (APA) recently refrained from adopting a proposed revision to the Diagnostic and Statistical Manual of Mental Disorders (DSM) IV's categorical approach to disorders that would have seen personality disorders represented, in part, as distinct profiles on a pathological dimensional model (Skodol et al., 2011). The maladaptive trait model will instead be included in a separate section on the diagnosis of personality disorder.
We believe that the profiling approach considered during the DSM revision has important implications for understanding personality at work. However, the new trait model, a maladaptive equivalent of the Big Five, has not yet been embraced by industrial psychologists. Very few publications are emerging on this model in work contexts. In this article we suggest that the field of personality at work is now at a point reminiscent of the 1990s, where substantive developments in the field of personality are ready to be integrated to advance understanding of personality at work. We wish to stimulate discussion of maladaptive personality traits at work to quicken the introduction of these new ideas into work-related personality research, as we believe industrial psychologists need to pay more attention to this important topic.
Various labels are used in the psychological literature to describe abnormal personality. These include pathological (e.g., Wright et al., 2012), abnormal (e.g., Tromp & Koot, 2010), deviant (e.g., Howard & McMurran, 2012), and aberrant (e.g., Edmundson, Lynam, Miller, Gore, & Widiger 2011). However, rather than assessing personality disorder, we are interested in studying traits that predispose individuals to personality disorder among normal working populations. Here, we will avoid these terms in lieu of the term “maladaptive.” Although some authors have preferred other of these terms to maladaptive (e.g., Wille, De Fruyt, & De Clercq, 2013), we suggest that our use of this term is appropriate given its use to describe the trait model in clinical settings where the model originated. Although the proposed trait model emerged in the context of pathology research, all individuals can be profiled on the underlying model (e.g., O'Connor, 2002; Saulsman & Page, 2004; Trull & Durrett, 2005). The term maladaptive is consistent with the notion that a profile on the trait model by itself does not equate to a disorder.
Our use of the term maladaptive further serves to differentiate the new trait model from “dark side” personality research. In the academic literature the term “dark side” is sometimes used to refer to models of maladaptive personality based on dimensionalized DSM IV Axis II categories (e.g., Hogan & Hogan, 2001). Later we explicitly advocate reconceptualizing maladaptive personality under the recently proposed DSM-5 trait framework.
Finally, research on the Dark Triad (see Furnham, Richards, & Paulhus, 2013 for a review) refers to a cluster of maladaptive traits including narcissism, Machiavellianism, and psychoticism. All three traits fall under one factor of the new maladaptive trait model, that is Antagonism. We discuss the Dark Triad research and position it relative to maladaptive personality as operationalized under the DSM-5 trait structure in a later section. Here, we note that the Dark Triad represents a subset of facets of the Antagonism factor of maladaptive personality.
Previous Calls for Research Into Maladaptive Personality at Work
De Fruyt and Salgado (2003) said that the fields of individual differences and industrial psychology too often evolve separately, and we agree. They then made a call for more research into maladaptive personality in the workplace. Our prompt differs from theirs in two important ways. First, their article suggested that profiles of disorders on normal personality inventories be used to screen for symptomology for certain functions in specific jobs (e.g., police, firefighter). Their rationale was the low population base rate of disorders, which they reported the APA estimated at 3%. We are explicitly suggesting industrial psychologists consider the relevance of maladaptive personality at work across all jobs that job analysis suggests warrants its consideration. Our rationale is that we are not screening for disorder, we are measuring maladaptive personality traits with implications for job performance. Second, De Fruyt and Salgado (2003) did not advocate an overarching framework for the study of maladaptive personality. In this article, we suggest aligning all maladaptive personality research in industrial psychology under the trait framework considered in the recent DSM revision.
The remainder of this article is structured as follows. First, we offer a discussion of the main elements of the revision to the personality disorder diagnosis considered during the DSM-5 revision. Next, we present the maladaptive trait model itself in more detail, followed by potential objections to using inventories assessing this model at work. Following each potential objection, we discuss points that we believe might alleviate these concerns.
Changes Considered During the DSM Revision With Relevance to the Workplace
Certain diagnostic criteria for personality disorders under DSM IV, such as the age at which they begin and level of stability, do not adequately differentiate personality disorders from other mental illnesses (Krueger, Markon, Patrick, Benning, & Kramer, 2007). Dimensional measures of disorder can predict impairment better than their categorical equivalents (Skodol et al., 2005). There is also little evidence suggesting personality disorders are composed of latent classes that are categorical (Krueger & Eaton, 2010). Factors such as these promoted the APA's consideration of the new profiling approach discussed in this article.
Practical reasons for considering change also exist. Diagnosing personality disorders by the presence or absence of subsets of the 79 indicators in DSM IV might be more complex than required (Krueger & Eaton, 2010). Instead, the picture that has emerged of disordered personality is one where the symptomology is explainable in part by extreme standing on a core set of maladaptive personality traits (Widiger & Simonsen, 2005). In response, the APA considered profiling against a pathological trait model that predisposes individuals to personality problems. The trait model the APA considered is the focus of this article.
Evidence for the Relevance of Maladaptive Personality Traits at Work
Syntheses of Broad-Spectrum Work-Related Maladaptive Personality Research
For some time researchers have argued that personality measurement has considerably more to offer than the prediction of positive work-related outcomes. Judge and LePine (2007) summarized research showing that high standing on measures of aberrant personality could lead to problems at work. Narcissists, for example, overestimate their effectiveness as leaders, task performance, and contextual performance, while underestimating counterproductivity (Judge, Le Pine, & Rich, 2006). Impulsivity at work was understudied according to Judge et al., but they speculated impulsive individuals suffered impaired work-related reasoning. Trait hostility is linked with coronary heart disease (Miller, Smith, Turner, Guijarro, & Hallet, 1996), conflict (Newton & Kiecolt-Glaser, 1995), and interpersonal aggression (Archer & Webb, 2006). Finally, Judge et al. considered type-A personality, noting research that suggested it was linked with coronary heart disease (Booth-Kewley & Friedman, 1987; Matthews & Haynes, 1986), job dissatisfaction (Jiang, Yan, & Li, 2004), burnout (Alotaibi, 2003), and poor health (Kirkcaldy, Shephard, & Furnham, 2002).
Although indicative of the work relevance of the DSM trait model, the taxonomic framework Judge and LePine (2007) used to guide their review was rationally derived and diverges from the DSM-5 model. Given that research has shown that Big Five validities are generally higher when analyzed using a conceptual framework that is primary (Salgado, 2003), it is likely that syntheses of maladaptive personality research would benefit from being analyzed under the DSM maladaptive trait umbrella. In summary, this line of research suggests that the DSM maladaptive trait model is likely to have important work-related implications. We believe that conceptual clarity will come more quickly regarding the role of maladaptive personality at work if researchers nest their research questions under this organizing framework.
Dimensionalized Interpretations of DSM IV Axis II Categories
Much of what industrial psychologists know today about maladaptive personality in occupational settings stems from the work of Robert Hogan and his colleagues (e.g., Hogan, Hogan, & Kaiser, 2010). In an early and highly cited paper on maladaptive personality at work, Hogan and Hogan (2001) described the rationale for developing the Hogan Development Survey (HDS: Hogan & Hogan, 1997/2006). The HDS emerged from Hogan's desire to predict managerial incompetence. They offered three reasons for studying incompetence rather than the more common approach of studying effectiveness. First, although there is often disagreement on who is competent, they suggested that there was rarely disagreement on who in an organization was incompetent. Second, Hogan and Hogan argued that the base rate of incompetence is high. Finally, they suggested there was a moral imperative to mitigate managerial incompetence.
Hogan and Hogan (2001) reviewed early work by Bentz (1985, August); McCall, Lombardo, and Morrison (1988); and Leslie and Van Velsor (1996). Their review suggested an 11-dimension taxonomy of “dysfunctional dispositions” (p. 40). This taxonomy showed a very strong resemblance to the DSM IV Axis II personality disorders. They then presented psychometric properties based on analyses of a dataset of over 10,000 responses. These analyses suggested the 11 HDS dimensions can be grouped into three themes first identified by Horney (1950): moving toward people, moving away from people, and moving against people. Along with follower and situational characteristics, later work would position these 11 HDS dimensions as part of a “toxic triangle” that leads to destructive leadership (Padilla, Hogan, & Kaiser, 2007).
In summary, Hogan's pioneering work on maladaptive personality at work has led to a stream of research on dimensionalized DSM IV Axis II traits that predispose managers to incompetence. This work has established that some maladaptive constructs have negative implications for managerial derailment (Hogan, Raskin, & Fazzini, 1990), leadership (Hogan & Hogan, 2001), and productivity (Moscoso & Salgado, 2004). Some research in this vein has also supported positive outcomes for these constructs (Furnham, Hyde, & Trickey, 2013). But perhaps, the most common industrial psychology application of the HDS traits is in managerial coaching. For example, a recent paper by De Fruyt, Wille, and Furnham (2013) discussed the development of cut-off scores that would flag risks for various work contexts that might be addressed by coaching. They estimated that up to one quarter of all managers “qualified” as at risk for one problematic behavioral tendency.
Dimensional representations of DSM IV characteristics can be considered emerging or compound trait measures because they are underpinned by two or more personality traits that are more fundamental in nature. A consequence of assessing a compound trait is that we do not know whether a given individual's composition on the trait is due to equal standing on the contributing traits or different standing on the contributing traits. This is not a trivial issue because there may be situations where one of the traits is important and another is not. In other words, measuring the compound trait does not allow us to reduce a profile to its constituent elements. The research energy today is toward unpacking these tendencies to disorder into their primary attributes to better understand the phenomena (Widiger, Lynam, Miller, & Oltmanns, 2013). To summarize, the dimensionalized DSM IV research tradition suggests a broad taxonomic model of disordered personality traits could have application in occupational settings, but a model focusing on the primary traits seems preferable to one focusing on blends of primary traits.
Research Into Aberrant Profiles on Normal Inventories
Some researchers have investigated profiles indicative of aberrance based on normal personality inventories (e.g., De Fruyt et al., 2009). In one recent instance of this approach Wille et al. (2013) used the five-factor model (FFM) compound technique to derive disordered profiles from normal personality questionnaires. The five-factor compound technique essentially involves the computation of linear composites of five-factor facets that are related to specific personality disorders. Wille et al. noted that this approach performs as well as more complex prototype matching, that is, examining the similarity between assessee profiles and subject matter expert's views of disorders in terms of FFM scales.
The profiles Wille et al. (2013) considered corresponded to those expected to be represented in the DSM-5 (i.e., antisocial, narcissistic, borderline, schizotypal, obsessive–compulsive, and avoidant). Results suggested that these aberrant profiles predicted career success and that the aberrant compounds explained incremental variance over FFM scales. This line of work comes closest to our suggestion to adopt the maladaptive trait model. However, its basis in the Big Five skirts the issue of adopting the maladaptive trait model and all the benefits that a primary framework affords. Because the content of Big Five and maladaptive trait inventories are not the same, Big Five-based profiles will not be the same as maladaptive trait profiles.
Research Into Narrow Aspects of Maladaptive Personality
Another stream of research into maladaptive personality at work focuses on narrow aspects of the spectrum of problematic personality traits, for example, measures of the Dark Triad, that is psychoticism, narcissism, and Machiavellianism (Paulhus & Williams, 2002). Wu and LeBreton (2011) conducted a conceptual review of Dark Triad. They stated that despite an earlier hint from Rolland and De Fruyt (2003) that the Dark Triad traits might predict counterproductive behavior, considerably more was known about Big Five links with counterproductive behavior. Wu and LeBreton reviewed the literature before setting a series of research questions for further investigation.
Empirical studies of narrow aspects of the maladaptive model have shown that there is good reason for expecting the new trait model to be relevant in occupational settings. A meta-analysis by O'Boyle, Forsyth, Banks, and McDaniel (2011) found that all three Dark Triad components were related to counterproductivity and that Machiavellianism and Psychoticism were related to job performance. A case might be made that one or other of the Dark Triad traits are more relevant to the work place, and indeed some researchers have focused on the impact of a single element of the Dark Triad, for example psychoticism (Babiak & Hare, 2006) or narcissism (Campbell, Hoffman, Campbell, & Marchiso, 2011; Resick, Whitman, Weingarden, & Hiller, 2009). However, all aspects of the Dark Triad fall under the Antagonism factor of the new trait model. Studying narrow aspects of maladaptive personality does not offer the promise of a complete understanding of maladaptive personality at work. For this, broader taxonomic frameworks need to be adopted to frame our research endeavors.
In summary, we contend that researchers need to move beyond using DSM IV-based measures and studying narrow aspects of maladaptive personality. The problem with these approaches is that the former is imprecise and the latter is incomplete. Future quantitative and qualitative summaries should also be organized according to the maladaptive trait framework because it affords greater conceptual clarity regarding what is measured and the likelihood of stronger relations with important performance criteria. We suggest that a solution is to research maladaptive personality under the trait framework considered during the DSM-5 revision.
The Proposed DSM-5 Maladaptive Trait Model
The initial DSM-5 trait model proposed was rationally derived and comprised of six domain level traits (Skodol et al., 2011, p. 37): negative emotionality: “Experiences a wide range of negative emotions (anxiety, depression, guilt/shame, worry etc.), and the behavioral and interpersonal manifestations of those experiences.” Detachment: “Withdrawal from other people, ranging from intimate relationships to the world at large; restricted affective experience and expression; limited hedonic capacity.” Antagonism: Exhibits diverse manifestations of antipathy toward others, and a correspondingly exaggerated sense of self-importance. Disinhibition: Diverse manifestations of being present (vs. future- or past-) oriented so that behavior is driven by current internal and external stimuli rather than by past learning and consideration of future consequences. Compulsivity: “The tendency to think and act according to a narrowly defined and unchanging ideal, and the expectation that this ideal should be adhered to by everyone.” Psychoticism: “Exhibits a range of odd or unusual behaviors and cognitions, including both process (e.g., perception) and content (e.g., beliefs).” Subsequent research supported a five-trait domain structure where compulsivity is seen as the opposite pole of disinhibition (e.g., Krueger et al., 2011).
Provenance of the DSM Trait Model
The domain trait model bears similarities to measures of well-established traits predisposing individuals to disordered personality. These research programs reviewed in detail by Krueger, Derringer, Markon, Watson, and Skodol (2012) models include the Dimensional Assessment of Personality Pathology (DAPP: Livesley, 2001), the Schedule for Non-Adaptive and Adaptive Personality (SNAP: Clark, 1993), the Personality Psychopathology-Five (PSY-5: Harkness, 1992), the Dimensional Personality Symptom Item Pool (DIPSI: De Clercq, De Fruyt, Van Leeuwen, & Mervielde, 2006), the Millon Clinical Multiaxial Inventory III (MCMI-III: Millon & Davis, 1996), and the Shedler-Western Assessment Procedure (SWAP: Shedler & Westen, 2004).
Four of the trait domains in the new trait model, negative emotionality, detachment, antagonism, and disinhibition are widely regarded to be maladaptive variations of the Big Five in normal populations. The adaptive variants of these, respectively, are Neuroticism, Extraversion, Agreeableness, and Conscientiousness (Widiger & Simonsen, 2005). Clark and Krueger (2012) noted that the Openness in normal populations appears to have limited relevance for personality disorder. They also said compulsivity is included as it underlies obsessive–compulsive disorder and is not well captured by the maladaptive variants of the Big Five; and psychoticism because as it underpins schizotypal personality disorder, which is also not well captured by the maladaptive variants of the Big Five.
Less is known about the trait facet structure of pathological personality (Krueger et al., 2011), partly because factor analysis extracts factors accounting for the most shared variance among variables, and these are generally domains (Krueger et al., 2007). At the facet level, the rationally derived model comprised of 37 trait facets, whereas empirical analyses supported a 25 facet structure (Krueger et al., 2011; Wright et al., 2012).
Important Considerations in the Assessment of Maladaptive Personality at Work
Industrial psychologists have been slow to examine the taxonomic model considered during the DSM revision. They have even been slow to study the models on which the proposed maladaptive traits are based, despite a robust stream of evidence on their origins (Krueger, Derringer, et al., 2012; Widiger and Simonsen, 2005). There are likely to be a number of reasons for this state of affairs. Here we discuss what some of the barriers might be and present counterarguments that we believe might stimulate further consideration of the maladaptive trait model among academics and practitioners.
The barriers that we discuss here include (i) concerns that use of maladaptive inventories might infringe rights protected by law, (ii) social responsibility concerns regarding inadvertent and unnecessary exclusion of candidates with mental health problems from the work place, (iii) a belief that the new taxonomic model of personality pathology is redundant if measures of the Big Five are already used in assessment and would therefore have no incremental validity, (iv) concerns that personality tests show low validities generally and are not predictive of performance, and (v) concern that faking is too much of a concern with maladaptive inventories.
One reason for industrial psychologists' slow adoption of the trait model might be concern about whether assessing these traits infringes workers' rights covered by legislation, for example, the Americans with Disabilities Act (ADA) of 1990 in the United States or the Disability Discrimination Act of 1995 in the United Kingdom. Mental ill health is one form of disability typically covered by such legislation. For example, the ADA states that qualified individuals with disabilities cannot be discriminated against on the basis of their disability if they can perform the essential functions of the job either with or without accommodation.
Klimoski and Palmer (1993) stated that ADA has two major requirements in relation to testing, and these are relevant to the present discussion: “first, a test that screens out or tends to screen out an individual with a disability must be job related and consistent with business necessity” (p. 18), and second, “tests must reflect the skills and aptitudes of an individual with a disability rather than impaired sensory, manual, or speaking skills, unless those are job-related skills the test is designed to measure” (p. 18). The essence of their case is that purpose matters: Personality tests designed to be job related and predictive of performance are okay, diagnostic tests designed to indicate disorder and form the basis of treatment plans are not.
More recently, Colella and Bruyère (2010) noted that the Seventh Circuit Court of Appeals ruled in Karraker v. Rent-a-Center, Inc. (2005) that the Minnesota Multiphasic Personality Inventory (MMPI) could not be used in personnel selection unless the “job relatedness or business necessity” of the instrument could be demonstrated. This is because the MMPI can be considered a preemployment medical exam, and such exams are prohibited under ADA.
Disorder is more than a trait profile
A personality disorder diagnosis under DSM-5 trait model approach has more components than simply elevated standing on any subset of the contributing traits. The significance of this point is made very clearly by Wright (2011) who stated: “Rarely if ever are individuals with a certain trait or profile of traits found, and then subsequently diagnosed. What this leaves us with is the knowledge of what traits might be elevated if a person possesses a diagnosis, but not the reverse. It is not the case that in the population each individual with a given trait profile possesses the same PD diagnosis, or any diagnosis at all for that matter” (p. 374). We posit here that because a profile on the maladaptive trait model in and of itself is not typically sufficient to be disordered under the APA proposal, measures of maladaptive variants of the Big Five should not be considered medical tests in preemployment settings.
Maladaptive items differ from diagnostic test items
Items measuring the maladaptive trait model have high relevance to the workplace and can be contrasted with tests used in the diagnosis of disorder. A sample item for the irresponsibility facet of the Disinhibition factor, for example, might be “I break agreements.” Clearly its endorsement should give rise to concern in the mind of employers. This item and others like it might be contrasted with items in clinical tests that are not appropriate for the work place. Take, for example, the much-maligned Rorschach, which meta-analytic evidence from Mihura, Meyer, Dumitascu, and Bombel (2013), showed has validity for certain clinical applications. Mihura et al. noted that a possible response to a Rorschach inkblot indicative of psychosis might be “It's a Jesus head with smoke coming out of the eyes. The smoke is a sign that he's judging me. It's scary” (p. 6). We suggest that the content of the items in any inventory and the content of respondents be closely examined in making judgments of the work relatedness of maladaptive measures.
This could still leave industrial psychologists needing a way to determine whether maladaptive tests are job related. Industrial psychologists have in fact for many decades been operating in an environment where job relatedness must be shown before using cognitive ability tests, a selection technique known to result in impact against groups covered by the Civil Rights Act (1964). As with cognitive ability, job analysis is the primary way organizations can show whether tests are job related (Mitchell, Alliger, & Morfopolous, 1997). A key concern with the notion of job relatedness is the degree to which the definition of essential functions includes contextual performance, as contextual performance is arguably what maladaptive traits are most likely to predict. In their review of a sample of case law, Haimann, Gilmore, and Emmer (2013) considered over 200 cases and concluded that although the notion of essential functions certainly included task performance criteria, there was also evidence of essential functions that clearly resemble contextual performance. For example, a ruling from EEOC vs. Walmart (2007) indicated that the essential functions for a cashier included following company rules and procedures.
Finally, we note that our views are similar to Wu and LeBreton (2011), who said that their research agenda into the links between the Dark Triad and counterproductivity was very unlikely to violate the ADA because (i) assessments are designed explicitly for the work environment, (ii) clinical individuals are unlikely to be encountered because the base rate is just 1% in the population, and a large proportion of these individuals are institutionalized. We believe therefore that there is a good case to believe maladaptive inventories are work related and legally defensible.
Social Responsibility Concerns
Although some industrial psychologists' reluctance to use maladaptive personality traits is likely to be legally based, hesitance on the part of others may be due to concern of the social impact of the use of maladaptive personality as a prehire screen. At the same time that ADA legislation protects those with disability from discrimination on the basis of that disability if they are capable of performing the job, this legislation also affords provision for individuals with mental health concerns to request assistance from their employer for their disability. Support for workers with mental health issues is critical. Klimoski and Donahue (1997) have argued that perhaps the biggest threat to the spirit of ADA is not the access to employment opportunities but fair treatment once employed. Under ADA, this support is called accommodation. Employers must grant accommodation so long as the requests are reasonable. Similar provisions exist under the Disability Discrimination Act (1995), where the support is referred to as adjustments.
Perusal of the accommodations Colella and Bruyère (2010) describe indicated that they are primarily to assist with physical disabilities, but the personal assistance category, covering a job coach, might be a relevant accommodation that a employer might be asked to provide. Requiring additional time than typical employees receive from a manager is another form of accommodation that could be requested. We believe that effective accommodation and adjustment from employers are key to ameliorating concerns about social impact. In making this point, we acknowledge that requesting an accommodation is no guarantee that it will be granted, and sometimes even the process of requesting an accommodation will be extremely strenuous. Conceptual models of the likelihood of requesting and being granted accommodations under ADA exist and we refer readers to these sources for further discussion (e.g., Baldridge & Veiga, 2001, 2006, Florey & Harrison, 2000).
One of the criticisms of general personality as a selection methodology is that the validity coefficients that personality dimensions show for job performance criteria are generally small to moderate. We expect a possible reason practitioners are not examining maladaptive personality at work might be that they believe the validity gains will be small. Morgeson et al.'s (2007a) interpretation of the literature indicated that personality had low correlations with job performance, albeit slightly higher relations with contextual performance than with task performance. Ones, Dilchert, Viswesvaran, and Judge (2007) retorted, presenting validates for a variety of performance criteria, ranging from .11 to .49, and say the corrections they apply are conservative. Tett and Christiansen (2007) also presented a contrary view, arguing that the meta-analytic estimates are actually impressive and dependable. It seems reasonable to say that opinion is divided but also that validities likely to emerge for the maladaptive personality model would be considered useful by a substantial proportion of academic and practitioner readers. Based on the work reviewed in earlier sections of this article, we see good reason to investigate the efficacy of maladaptive personality to predict organizational relevant outcomes.
Construct Redundancy and Lack of Incremental Validity
Another reason that researchers and practitioners have been somewhat slow in researching the new maladaptive trait model might be concerns that such measures have low incremental validities. In fact, meta-analytic evidence supports the incremental validity of the Big Five (e.g., Salgado, 1998). Morgeson et al. (2007b) claimed, however, that the incremental validities are overestimates due to underestimating the intercorrelations of personality in meta-analyses. It could well be the case that researchers and practitioners feel that if the case for the incremental validity of the Big Five is marginal, adding another maladaptive Big Five is unlikely to be very helpful. Readers might be further pushed toward this conclusion because there is a move in clinical psychology to integrate the maladaptive trait model under the Big Five as an overarching framework (Thomas et al., 2013). Preliminary research, however, indicates that in fact dark side traits (i.e., dimensionalized scales measuring the DSM IV categorical outcomes) do show incremental validity for the prediction of important work outcomes (Furnham, Hyde, et al., 2013; Rolland & Du Fruyt, 2003). There is more to be said on this topic, but results so far appear promising.
Maladaptive Personality Inventories Are Too Easily Faked
Research has suggested that individuals do have the ability to fake, although the extent to which they do in practice is debated, as is the degree to which this erodes the psychometric properties of personality tests and incremental validity. We concur with a view expressed by one of the panelists from the Morgeson et al. (2007a) article: Faking should not be our biggest concern, rather we should be concerned by those who don't see the need to fake and a better measurement method than standard self-report is probably required. This is likely to be doubly so with maldadaptive personality inventories that ask people to rate their agreement with statements such as “I see how far I can push people,” a possible indicator of the callousness subfacet of the Antagonism factor. Few adroit test takers would rate this item highly when a job is at stake. Overall, we agree with these concerns over faking.
Short of using a more dependable source to report on applicant personalities there is little that can be done to deal with faking other than using social desirability scales, spotting the faking post hoc with statistical analyses or using forced choice response options in the measurement instrument. Each approach faces problems, but the news is best on the forced-choice front. Arguably among industrial psychologists greatest breakthroughs in personality research since the 1990s can be said to have been psychometric in nature, relating to improvements in measurement efficiency through advanced item response theory (e.g., Stark, Chernyshenko, Drasgow, & White, 2012) and structural equation modeling techniques (Brown & Maydeu-Olivares, 2012). This research has shown that normative scores can be recovered from forced choice designs, overcoming the biggest obstacle to their widespread adoption. Whether or not normative recovery of scores is possible, a recent meta-analysis by Salgado and Táuriz (2012) offers encouragement. This article showed that forced choice measures have greater validity than rating scale measures. It seems then that a solution to concern over faking on maladaptive inventories is a forced-choice design.
We contend that industrial psychologists need to be faster in their response to recent developments in clinical psychology to develop a full picture of personality at work. In particular, inadequate focus has been devoted to the maladaptive personality taxonomy considered during the recent DSM-5 revision. We suggested that this development is among the most exciting occurrences in personality research since meta-analysis and the Big Five. Up until now, however, industrial psychologists have been largely pursuing research into maladaptive personality without regard for the wider environment in which their research is taking place. Industrial psychologists need to carefully consider the developments occurring in the clinical field if an unnecessary separation between how clinical and industrial psychologists understand and assess maladaptive traits is to be avoided.