The growing support for the merging of “normal” and “abnormal” personality conceptualizations within the field of applied psychology has culminated in the adoption of the five-factor model (FFM) of personality within the DSM-V as part of a new approach to diagnosing personality disorders (PD). The focal article (Melson-Silimon, Harris, Shoenfelt, Miller, & Carter, Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) discusses the possibility that the shift in PD diagnostic procedures may lead to a rise in legal challenges in the I-O psychology field as the line becomes blurred between normal personality and pathological, disordered personality. However, we argue that the line between normal and abnormal personality model structures should not have been drawn in the first place, and we do not feel this shift is cause for alarm for those using personality testing in employee recruitment and selection. It is possible to employ the same over-arching theoretical framework to meet different goals. The fact that comprehensive trait-based, work-related personality measures and similar trait-based measures in clinical diagnostics are related to the same model structure does not mean the resulting measures can be or are being used interchangeably.
Although we do not necessarily agree with the authors’ level of concern that the link of PD to the FFM will result in increased noncompliance with ADA, we welcome their introduction into the more broad discussion of mental health (including PD). This is in line with a considered drive to increase awareness of mental health globally in employment settings in recent years. However, by inciting “cause for concern for legal action” we feel the topic is approached from the wrong angle. After all, the reason for strict legislation like the ADA is to eliminate any discrimination in the workplace and to remove the divide, not to emphasize the divide. Work-related personality measures, common within the field of industrial and organizational (I-O) psychology, are not designed to diagnose PD. In our argument, we pose that a shared theoretical model does aid in bridging the gap between the two fields of I-O psychology and clinical psychology, and we clarify the complexity of PD diagnostics, of which a personality measure is only one aspect, albeit a very valuable one. Despite the shift in the DSM-V to use dimensional, trait-based personality assessments, we argue that trait-based personality measures within the workplace remain non-clinical and highly predictive of job performance, and as such are valuable tools for employers.
Is there a line at all?
Prior to DSM-V (2013), a categorical medical model of personality disorders had been used for diagnosing personality disorders—a model that was criticized as early as the 1980s (Eysenck, Wakefield, & Friedman Reference Eysenck, Wakefield and Friedman1983; Kato, Reference Kato, Mezzich and von Cranach1988). The fact that the DSM adopted a trait theory approach in 2013 was therefore a long time coming. Dimensional approaches have been found as less arbitrary, more informative, and more consistent with empirical data on the nature of personality disorders (Trull, Widiger, & Guthrie, Reference Trull, Widiger and Guthrie1990; Widiger & Frances, Reference Widiger, Frances, Costa and Widiger1994). The DSM-V move toward a dimensional approach emphasizes the continuity with normal variations in personality traits and facilitates an integration between “normal” and “abnormal” personality, fields that have traditionally been quite separate. The consensus that has been reached over the years regarding the relationship between normal and abnormal personality structure is that abnormal personality generally represents extremes on continua in common with normal personality, as summarized by Markon, Krueger, and Watson Reference Markon, Krueger and Watson(2005). The most important historical hurdle in combining the two fields however was the lack of a universally accepted model of personality. This hurdle has long been overcome. Work by authors such as Costa and McCrae (Costa & McCrae, Reference Costa and McCrae1992; McCrae & Costa, Reference McCrae, Costa, John, Robins and Pervin2008) over the past 20 to 30 years has resulted in a consensus on personality structure. In particular, the FFM has been found to account for most of the variance from a wide variety of personality theories.
The FFM can be viewed as an overarching personality “umbrella,” which describes the full breadth of personality. However, it is an overview of the functioning of a person across their lifespan and omits many specifics that a complete theory of personality would include (McCrae & Costa, Reference McCrae, Costa, John, Robins and Pervin2008). Although normal and abnormal personality are on a continuum, there remains an adaptive and maladaptive variant of each factor within the FFM (Widiger, Costa, Gore, & Crego, Reference Widiger, Costa, Gore, Crego, Widiger and Costa2002). These adaptive and maladaptive variants are well illustrated in the Five Factor Form (Widiger, Reference Widiger, Leary and Hoyle2009). To give an example of the difference between adaptive and maladaptive variants, take for example “Activity,” a subfacet of Extraversion. It is generally adaptive and beneficial to be energetic (high in activity) but not to the point where a person would become frantic (maladaptive in high activity). On the other end of the scale it can be adaptive to be slow paced (low in activity) but when this moves into lethargy and sedentary behavior, it slides into a maladaptive variant of being low in activity.
Thus it follows that specific personality measures for either maladaptive or normal personality, including more specific job-related personality traits, are different branches of the same tree. Even though both are measuring at dimensional trait level and most if not all of the variance will be accounted for by the FFM, we have to consider the direction of the relationship between psychopathological and normal/work-related personality. A measure of psychopathology is looking for a different outcome to a workplace personality measure and therefore assesses different characteristic adaptations and dynamic processes than what a work-related measure would assess. Although the theoretical lines are blurring as measures belong to a common continuum, we cannot draw the conclusion that typical workplace personality testing could thus be regarded as a suitable alternative for maladaptive personality diagnostics and vice versa.
This links directly to the ADA, which states very clearly that an employer cannot use a medical examination as a test to screen applicants before a conditional offer of a job has been made. Thus, it is unsuitable and potentially illegal to use a dimensional trait-based measure designed for assessing pathological maladaptive personality in a work-related setting. It is not surprising that some of the few case-law studies mentioned by the authors refer to the MMPI, which was not designed to directly measure job-related behavior but actually to assess people for mental health (Butcher et al., Reference Butcher, Graham, Ben-Porath, Tellegen, Dahlstrom and Kaemmer2001). This highlights the need to continue to draw a line between the I-O psychology and clinical fields in the sense of making sure the appropriate measure is used for the given situation. Moreover, international guidance on employment and hiring practices such as the EEOC’s Guidelines (1978), the SIOP Principles (2018), the APA Standards (AERA, 2014), and ITC Guidelines (2014) all emphasize that measures used for selection and hiring decisions must be job relevant and predictive of successful performance of the job in question. We will return to this point later on, but first let us consider the more specific premise the authors make regarding the relationship of normative personality testing and the diagnosis of personality disorders in particular.
Using personality measures and the diagnostic procedure in the DSM-V
We propose that a lack of understanding on how the novel dimensional DSM-V diagnostic procedure would be applied may have led to unwarranted concern among I-O psychologists. Within this new approach, there are two criteria that a clinician must use to diagnose a PD. One is a maladaptive personality traits section, where clinicians are able to rate people on the occurrence of maladaptive traits—largely based on the FFM. People must exhibit a specified number of such traits for any given PD diagnoses. However, evidence of maladaptive traits must be combined with the second diagnostic criterion, which is evidence of adaptive functioning impairment; persons with PD must indicate that they are at least moderately impaired in at least one area of personality functioning: identity, self-direction, empathy, or intimacy. Furthermore, the person must show that his or her impairment is inflexible; that is, it occurs persistently and across most contexts.
A psychometric assessment measure that collects data on personality traits could in principle be used as part of a clinical diagnosis. However, what differentiates using this information for selection and recruitment versus clinical diagnoses is that the presence of maladaptive personality traits needs to be combined with evidence of adaptive functioning impairment. Assessment of such personality functioning would definitely not be included in any legally defensible pre-hire assessment, nor be administered by an I-O psychologist. Given these additional criteria, it is safe to conclude that an I-O psychologist or HR professional would not be able to label someone with a PD based on his/her personality trait assessment results alone. It is also important to remember that, from an ADA and legal perspective, a plaintiffs can only win judgments if they can prove that they have been classified within a group that has protected status and that this classification has been used to discriminate against them. Within the current discussion, plaintiffs would need to prove that they have been labeled as having a PD under the mental health umbrella and that their assessment scores were used to determine this classification. This seems an unlikely scenario based on the criteria we have outlined above.
Occurrence of maladaptive traits and extreme scores
In addition to postulating that nonclinical personality measures could be used for clinical classification and lead to direct discrimination, the authors of the focal article argue that screening out individuals with extreme trait levels may produce disparate treatment against individuals with PD—a form of indirect discrimination. They pose that typically for selection decisions, candidates with extreme trait levels are sifted out either overtly (via PD classification) or indirectly; someone with a PD, who would show more extreme trait scores, is more likely to be sifted out and excluded from the hiring process. Hence, their recommendation is to avoid assessing extreme traits (such as dark side traits, which could be seen as subclinical measures of PDs), as this would avoid this type of discrimination. This is an interesting position when considering research has found that aspects of dark-side personality in certain contexts can be associated with negative workplace outcomes, such as deviant and counterproductive work behavior (Judge & LePine, Reference Judge, LePine, Langan-Fox, Cooper and Klimoski2007, Penney & Spector, Reference Penney and Spector2005).
The maladaptive personality traits stated in the DSM-V that form part of PD diagnoses are indeed at the extreme end of the FFM factors. We are not refuting the possibility that, using either clinical or nonclinical personality measures, a person with a PD would probably score at the extreme ends on some of the traits measured. We are refuting that this would lead to discrimination because an I-O psychologist following strict standards and guidelines for assessment design (as mentioned before, i.e., the APA standards/ITC guidelines) should not sift out candidates on any trait unless this is directly job relevant and predictive of future job performance. For certain jobs this could result in people with PDs being screened out as they exhibit extreme scores on domains that were deemed job relevant, but this is legally defensible. Following I-O best practice guidelines, it would be highly unlikely that someone could accidentally discriminate against a protected group (including candidates with PD) by using personality measures.
The future of personality measures used for selection and hiring decisions
Using personality measures for selection and hiring decisions remains an appropriate and valid tool, and we should not be fearful of litigation when using personality measures in the workplace. However, using the right personality tools and using them properly for selection is critical. The original authors have provided a platform for beginning the discussion of how to approach PD and workplace personality measures, and they cite some practical recommendations. Below we summarize our own recommendations, some in line with the original authors’ but provided against the perspective we have shared above.
Recommendation 1: Conduct a job analysis
This step should be taken not simply due to an attempt to avoid ADA challenges but because a job analysis is a fundamental start to any recruitment, selection, or development trajectory design. Inclusion of a personality assessment for every job should not be customary but based on the outcome of the job analysis. This is especially relevant when considering research findings such as those from Barrick, Mount, and Judge Reference Barrick, Mount and Judge(2001) that have shown that the FFM does not predict overall job performance across all occupations, and prediction of job performance using personality traits has been found to be context-dependent. However, substantial research also promotes the use of personality measures in predicting specific aspects of job performance (e.g., risk). Studies have found low agreeableness and conscientiousness to be strongly associated with work accidents for instance (Clarke & Robertson, Reference Clarke and Robertson2008). Thus, when appropriately administered, personality assessments have been shown to be significant predictors of job related performance and thus should not be routinely discounted.
Recommendation 2: Use a tool specifically designed for work-related personality measurement
Ensure you select measure(s) that are supported by evidence of reliability and validity for their intended purpose (employee selection/hiring), as required by the standards and guidelines (i.e., SIOP, APA, ITC). Although the authors recommend that practitioners should avoid measuring constructs that are closely related to PDs, we have argued that there is no divide between normal and abnormal personality structure that makes avoiding measuring all constructs associated with PD challenging. However, what can be avoided is using an inappropriate measure of personality in the wrong context. Thus the only division that should remain is the application of clinical and nonclinical personality measures.
Recommendation 3: Do not avoid the use of dark-sided personality traits due to fear of ADA challenges
The authors have taken a stance against the use of dark-side personality measures, as dark-side traits are often subclinical measures of PD and could be used for discrimination. However, we believe that the growing body of research and findings suggests that the dark side of personality in organizations is worth paying attention to. Moreover, it is legally defensible and appropriate for some jobs to specify extreme personality traits (that may form part of these dark-side personality constructs) that would result in hiring exclusion (e.g., risk taking and being a surgeon, hostility and being a customer service representative). Again, this highlights the need to document the appropriate personality traits required for the job and the correct, nonclinical assessment(s) to measure them.
Recommendation 4: Consider reporting personality test results in a behavioral-based competency framework
Prehire assessment of potential job performance through personality constructs tends to be done by collapsing several personality traits into one behavioral competency or overall “job-fit” score. Thus the I-O psychologist is able to gain insight into competency potential of an individual rather than focus on individual personality traits. The FFM has itself been mapped to empirical competency frameworks such as the Great Eight (Bartram, Reference Bartram2005). Such competency approaches seem to be reflective of research that has demonstrated interactive relationships between the five factor model traits. For example, Witt Reference Witt(2002) found that a combination of extraversion and conscientiousness showed stronger prediction of job performance in jobs that require a great deal of interaction, leading to incremental validity. The behavioral competency approach to assessment, alongside findings on incremental value of personality constructs, suggests that personality constructs are of greater value as predictors of job performance, when considered alongside other personality constructs as part of larger competency scores, rather than single scores. Again, this adds to the evidence for personality assessment inclusion in selection and recruitment.
Conclusion
Although we applaud the introduction into the more broad discussion of PD and the use of psychometric assessments for selection, we would like to negate the fear factor that the focal article may have created. With the correct application there should not be anything to worry about when applying trait-based personality measures for selection and hiring decision.
Although there is clearly some uncertainty around the shift in DSM-V toward the FFM model and dimensional diagnoses, this should not be causing alarm; rather I-O psychologists should be applauding the field of applied psychology for embracing this long overdue move. I-O psychologists and employers alike should take note that evidence of maladaptive personality traits alone cannot be used to form diagnoses for PD or even identify someone who is “at risk” of having a PD. They also must trust that exclusion of people with extreme trait scores (with or without PD) can be legally justified for specific jobs, if that personality trait is deemed essential for successful performance in the job role. However, we acknowledge that advancement in research and knowledge in this area may help HR professionals and I-O psychologists feel more confident in tackling this challenging area and bridge the divide between the two fields of clinical and I-O psychology.
You have
Access
The growing support for the merging of “normal” and “abnormal” personality conceptualizations within the field of applied psychology has culminated in the adoption of the five-factor model (FFM) of personality within the DSM-V as part of a new approach to diagnosing personality disorders (PD). The focal article (Melson-Silimon, Harris, Shoenfelt, Miller, & Carter, Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) discusses the possibility that the shift in PD diagnostic procedures may lead to a rise in legal challenges in the I-O psychology field as the line becomes blurred between normal personality and pathological, disordered personality. However, we argue that the line between normal and abnormal personality model structures should not have been drawn in the first place, and we do not feel this shift is cause for alarm for those using personality testing in employee recruitment and selection. It is possible to employ the same over-arching theoretical framework to meet different goals. The fact that comprehensive trait-based, work-related personality measures and similar trait-based measures in clinical diagnostics are related to the same model structure does not mean the resulting measures can be or are being used interchangeably.
Although we do not necessarily agree with the authors’ level of concern that the link of PD to the FFM will result in increased noncompliance with ADA, we welcome their introduction into the more broad discussion of mental health (including PD). This is in line with a considered drive to increase awareness of mental health globally in employment settings in recent years. However, by inciting “cause for concern for legal action” we feel the topic is approached from the wrong angle. After all, the reason for strict legislation like the ADA is to eliminate any discrimination in the workplace and to remove the divide, not to emphasize the divide. Work-related personality measures, common within the field of industrial and organizational (I-O) psychology, are not designed to diagnose PD. In our argument, we pose that a shared theoretical model does aid in bridging the gap between the two fields of I-O psychology and clinical psychology, and we clarify the complexity of PD diagnostics, of which a personality measure is only one aspect, albeit a very valuable one. Despite the shift in the DSM-V to use dimensional, trait-based personality assessments, we argue that trait-based personality measures within the workplace remain non-clinical and highly predictive of job performance, and as such are valuable tools for employers.
Is there a line at all?
Prior to DSM-V (2013), a categorical medical model of personality disorders had been used for diagnosing personality disorders—a model that was criticized as early as the 1980s (Eysenck, Wakefield, & Friedman Reference Eysenck, Wakefield and Friedman1983; Kato, Reference Kato, Mezzich and von Cranach1988). The fact that the DSM adopted a trait theory approach in 2013 was therefore a long time coming. Dimensional approaches have been found as less arbitrary, more informative, and more consistent with empirical data on the nature of personality disorders (Trull, Widiger, & Guthrie, Reference Trull, Widiger and Guthrie1990; Widiger & Frances, Reference Widiger, Frances, Costa and Widiger1994). The DSM-V move toward a dimensional approach emphasizes the continuity with normal variations in personality traits and facilitates an integration between “normal” and “abnormal” personality, fields that have traditionally been quite separate. The consensus that has been reached over the years regarding the relationship between normal and abnormal personality structure is that abnormal personality generally represents extremes on continua in common with normal personality, as summarized by Markon, Krueger, and Watson Reference Markon, Krueger and Watson(2005). The most important historical hurdle in combining the two fields however was the lack of a universally accepted model of personality. This hurdle has long been overcome. Work by authors such as Costa and McCrae (Costa & McCrae, Reference Costa and McCrae1992; McCrae & Costa, Reference McCrae, Costa, John, Robins and Pervin2008) over the past 20 to 30 years has resulted in a consensus on personality structure. In particular, the FFM has been found to account for most of the variance from a wide variety of personality theories.
The FFM can be viewed as an overarching personality “umbrella,” which describes the full breadth of personality. However, it is an overview of the functioning of a person across their lifespan and omits many specifics that a complete theory of personality would include (McCrae & Costa, Reference McCrae, Costa, John, Robins and Pervin2008). Although normal and abnormal personality are on a continuum, there remains an adaptive and maladaptive variant of each factor within the FFM (Widiger, Costa, Gore, & Crego, Reference Widiger, Costa, Gore, Crego, Widiger and Costa2002). These adaptive and maladaptive variants are well illustrated in the Five Factor Form (Widiger, Reference Widiger, Leary and Hoyle2009). To give an example of the difference between adaptive and maladaptive variants, take for example “Activity,” a subfacet of Extraversion. It is generally adaptive and beneficial to be energetic (high in activity) but not to the point where a person would become frantic (maladaptive in high activity). On the other end of the scale it can be adaptive to be slow paced (low in activity) but when this moves into lethargy and sedentary behavior, it slides into a maladaptive variant of being low in activity.
Thus it follows that specific personality measures for either maladaptive or normal personality, including more specific job-related personality traits, are different branches of the same tree. Even though both are measuring at dimensional trait level and most if not all of the variance will be accounted for by the FFM, we have to consider the direction of the relationship between psychopathological and normal/work-related personality. A measure of psychopathology is looking for a different outcome to a workplace personality measure and therefore assesses different characteristic adaptations and dynamic processes than what a work-related measure would assess. Although the theoretical lines are blurring as measures belong to a common continuum, we cannot draw the conclusion that typical workplace personality testing could thus be regarded as a suitable alternative for maladaptive personality diagnostics and vice versa.
This links directly to the ADA, which states very clearly that an employer cannot use a medical examination as a test to screen applicants before a conditional offer of a job has been made. Thus, it is unsuitable and potentially illegal to use a dimensional trait-based measure designed for assessing pathological maladaptive personality in a work-related setting. It is not surprising that some of the few case-law studies mentioned by the authors refer to the MMPI, which was not designed to directly measure job-related behavior but actually to assess people for mental health (Butcher et al., Reference Butcher, Graham, Ben-Porath, Tellegen, Dahlstrom and Kaemmer2001). This highlights the need to continue to draw a line between the I-O psychology and clinical fields in the sense of making sure the appropriate measure is used for the given situation. Moreover, international guidance on employment and hiring practices such as the EEOC’s Guidelines (1978), the SIOP Principles (2018), the APA Standards (AERA, 2014), and ITC Guidelines (2014) all emphasize that measures used for selection and hiring decisions must be job relevant and predictive of successful performance of the job in question. We will return to this point later on, but first let us consider the more specific premise the authors make regarding the relationship of normative personality testing and the diagnosis of personality disorders in particular.
Using personality measures and the diagnostic procedure in the DSM-V
We propose that a lack of understanding on how the novel dimensional DSM-V diagnostic procedure would be applied may have led to unwarranted concern among I-O psychologists. Within this new approach, there are two criteria that a clinician must use to diagnose a PD. One is a maladaptive personality traits section, where clinicians are able to rate people on the occurrence of maladaptive traits—largely based on the FFM. People must exhibit a specified number of such traits for any given PD diagnoses. However, evidence of maladaptive traits must be combined with the second diagnostic criterion, which is evidence of adaptive functioning impairment; persons with PD must indicate that they are at least moderately impaired in at least one area of personality functioning: identity, self-direction, empathy, or intimacy. Furthermore, the person must show that his or her impairment is inflexible; that is, it occurs persistently and across most contexts.
A psychometric assessment measure that collects data on personality traits could in principle be used as part of a clinical diagnosis. However, what differentiates using this information for selection and recruitment versus clinical diagnoses is that the presence of maladaptive personality traits needs to be combined with evidence of adaptive functioning impairment. Assessment of such personality functioning would definitely not be included in any legally defensible pre-hire assessment, nor be administered by an I-O psychologist. Given these additional criteria, it is safe to conclude that an I-O psychologist or HR professional would not be able to label someone with a PD based on his/her personality trait assessment results alone. It is also important to remember that, from an ADA and legal perspective, a plaintiffs can only win judgments if they can prove that they have been classified within a group that has protected status and that this classification has been used to discriminate against them. Within the current discussion, plaintiffs would need to prove that they have been labeled as having a PD under the mental health umbrella and that their assessment scores were used to determine this classification. This seems an unlikely scenario based on the criteria we have outlined above.
Occurrence of maladaptive traits and extreme scores
In addition to postulating that nonclinical personality measures could be used for clinical classification and lead to direct discrimination, the authors of the focal article argue that screening out individuals with extreme trait levels may produce disparate treatment against individuals with PD—a form of indirect discrimination. They pose that typically for selection decisions, candidates with extreme trait levels are sifted out either overtly (via PD classification) or indirectly; someone with a PD, who would show more extreme trait scores, is more likely to be sifted out and excluded from the hiring process. Hence, their recommendation is to avoid assessing extreme traits (such as dark side traits, which could be seen as subclinical measures of PDs), as this would avoid this type of discrimination. This is an interesting position when considering research has found that aspects of dark-side personality in certain contexts can be associated with negative workplace outcomes, such as deviant and counterproductive work behavior (Judge & LePine, Reference Judge, LePine, Langan-Fox, Cooper and Klimoski2007, Penney & Spector, Reference Penney and Spector2005).
The maladaptive personality traits stated in the DSM-V that form part of PD diagnoses are indeed at the extreme end of the FFM factors. We are not refuting the possibility that, using either clinical or nonclinical personality measures, a person with a PD would probably score at the extreme ends on some of the traits measured. We are refuting that this would lead to discrimination because an I-O psychologist following strict standards and guidelines for assessment design (as mentioned before, i.e., the APA standards/ITC guidelines) should not sift out candidates on any trait unless this is directly job relevant and predictive of future job performance. For certain jobs this could result in people with PDs being screened out as they exhibit extreme scores on domains that were deemed job relevant, but this is legally defensible. Following I-O best practice guidelines, it would be highly unlikely that someone could accidentally discriminate against a protected group (including candidates with PD) by using personality measures.
The future of personality measures used for selection and hiring decisions
Using personality measures for selection and hiring decisions remains an appropriate and valid tool, and we should not be fearful of litigation when using personality measures in the workplace. However, using the right personality tools and using them properly for selection is critical. The original authors have provided a platform for beginning the discussion of how to approach PD and workplace personality measures, and they cite some practical recommendations. Below we summarize our own recommendations, some in line with the original authors’ but provided against the perspective we have shared above.
Recommendation 1: Conduct a job analysis
This step should be taken not simply due to an attempt to avoid ADA challenges but because a job analysis is a fundamental start to any recruitment, selection, or development trajectory design. Inclusion of a personality assessment for every job should not be customary but based on the outcome of the job analysis. This is especially relevant when considering research findings such as those from Barrick, Mount, and Judge Reference Barrick, Mount and Judge(2001) that have shown that the FFM does not predict overall job performance across all occupations, and prediction of job performance using personality traits has been found to be context-dependent. However, substantial research also promotes the use of personality measures in predicting specific aspects of job performance (e.g., risk). Studies have found low agreeableness and conscientiousness to be strongly associated with work accidents for instance (Clarke & Robertson, Reference Clarke and Robertson2008). Thus, when appropriately administered, personality assessments have been shown to be significant predictors of job related performance and thus should not be routinely discounted.
Recommendation 2: Use a tool specifically designed for work-related personality measurement
Ensure you select measure(s) that are supported by evidence of reliability and validity for their intended purpose (employee selection/hiring), as required by the standards and guidelines (i.e., SIOP, APA, ITC). Although the authors recommend that practitioners should avoid measuring constructs that are closely related to PDs, we have argued that there is no divide between normal and abnormal personality structure that makes avoiding measuring all constructs associated with PD challenging. However, what can be avoided is using an inappropriate measure of personality in the wrong context. Thus the only division that should remain is the application of clinical and nonclinical personality measures.
Recommendation 3: Do not avoid the use of dark-sided personality traits due to fear of ADA challenges
The authors have taken a stance against the use of dark-side personality measures, as dark-side traits are often subclinical measures of PD and could be used for discrimination. However, we believe that the growing body of research and findings suggests that the dark side of personality in organizations is worth paying attention to. Moreover, it is legally defensible and appropriate for some jobs to specify extreme personality traits (that may form part of these dark-side personality constructs) that would result in hiring exclusion (e.g., risk taking and being a surgeon, hostility and being a customer service representative). Again, this highlights the need to document the appropriate personality traits required for the job and the correct, nonclinical assessment(s) to measure them.
Recommendation 4: Consider reporting personality test results in a behavioral-based competency framework
Prehire assessment of potential job performance through personality constructs tends to be done by collapsing several personality traits into one behavioral competency or overall “job-fit” score. Thus the I-O psychologist is able to gain insight into competency potential of an individual rather than focus on individual personality traits. The FFM has itself been mapped to empirical competency frameworks such as the Great Eight (Bartram, Reference Bartram2005). Such competency approaches seem to be reflective of research that has demonstrated interactive relationships between the five factor model traits. For example, Witt Reference Witt(2002) found that a combination of extraversion and conscientiousness showed stronger prediction of job performance in jobs that require a great deal of interaction, leading to incremental validity. The behavioral competency approach to assessment, alongside findings on incremental value of personality constructs, suggests that personality constructs are of greater value as predictors of job performance, when considered alongside other personality constructs as part of larger competency scores, rather than single scores. Again, this adds to the evidence for personality assessment inclusion in selection and recruitment.
Conclusion
Although we applaud the introduction into the more broad discussion of PD and the use of psychometric assessments for selection, we would like to negate the fear factor that the focal article may have created. With the correct application there should not be anything to worry about when applying trait-based personality measures for selection and hiring decision.
Although there is clearly some uncertainty around the shift in DSM-V toward the FFM model and dimensional diagnoses, this should not be causing alarm; rather I-O psychologists should be applauding the field of applied psychology for embracing this long overdue move. I-O psychologists and employers alike should take note that evidence of maladaptive personality traits alone cannot be used to form diagnoses for PD or even identify someone who is “at risk” of having a PD. They also must trust that exclusion of people with extreme trait scores (with or without PD) can be legally justified for specific jobs, if that personality trait is deemed essential for successful performance in the job role. However, we acknowledge that advancement in research and knowledge in this area may help HR professionals and I-O psychologists feel more confident in tackling this challenging area and bridge the divide between the two fields of clinical and I-O psychology.