Although Melson-Silimon, Harris, Shoenfelt, Miller, & Carter (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) raised a number of important points regarding the use of dark trait measurement tools in selection systems, we argue that administering these tests as part of a selection process is unlikely to result in violations of the ADA. Specifically, we argue against the suggestion that we should “avoid measuring personality constructs that are closely related to PDs … particularly those that assess known PDs and ‘dark-side’ traits that are often subclinical measures of PDs” (Melson-Silimon et. al., Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019, p. 129). To buttress our argument, we will critically evaluate various EEOC guidelines. Further, we will use relevant case law throughout to strengthen our perspective. Specifically, we will be referring to assessments such as the Self-Report Psychopathy Scale (SRP; Neumann, Schmitt, Carter, Embley, & Hare, Reference Neumann, Schmitt, Carter, Embley and Hare2012; Paulhus, Neumann, & Hare, Reference Paulhus, Neumann and Hare2016; Paulhus & Williams, Reference Paulhus and Williams2002; Williams, Paulhus, & Hare, Reference Williams, Paulhus and Hare2007) and the Narcissistic Personality Inventory (NPI; Raskin & Hall, Reference Raskin and Hall1979). Machiavellianism is often considered in concert with psychopathy and narcissism as part of the dark triad (LeBreton, Shiverdecker, & Grimaldi, Reference LeBreton, Shiverdecker and Grimaldi2018) and thus shares similar traits to psychopathy and narcissism. However, there is no clinical diagnosis of “Machiavellian personality disorder,” and we will therefore focus our arguments on the dark-triad traits that do have companion clinical diagnoses.
As Melson-Silimon et al. (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) noted, prior to extending a conditional job offer, the ADA prohibits an organization from using a medical examination to screen job applicants. And, Melson-Silimon et al. suggested that tests such as the NPI and SRP may be interpreted by the courts as medical examinations. This conclusion struck us as inconsistent with the EEOC guidelines and case law that Melson-Silimon et al. reviewed in their article. Indeed, we actually expected these authors to conclude that using these types of tests in a selection system is unlikely to be interpreted as a violation of the ADA.
Contradictory to Melson-Silimon et al. (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019), after reading the focal article, we were left to conclude that these tests would be permitted in selection. We arrived at our conclusion by simply comparing two exemplar measures, the SRP and NPI, to the EEOC criteria for determining whether a test is a medical examination. As Melson-Silimon et al. noted, if a test meets any of these criteria, then the test may represent a medical examination. The criteria, in relation to the NPI and the SRP, are as follows:
(a) Are these tests typically administered by a healthcare/medical professional? The answer is “no” for both the NPI and SRP. In a selection setting, both tests are likely to be administered by an HR professional.
(b) Are these tests typically interpreted by a healthcare/medical professional? The answer is “no” for both the NPI and SRP. In a selection setting, both tests are likely to be interpreted by an HR professional.
(c) Were these tests originally designed to reveal an impairment or an applicant’s current mental or physical health? The answer is “no” for both of these tests. Both tests were developed and validated using normal (i.e., healthy) populations.
(d) Are these tests invasive? Neither test is invasive. Both rely on a standard paper-and-pencil self-report assessment format. In addition, they are interpreted as stand-alone tools (i.e., they are not interpreted in conjunction with mental health or forensic records).
(e) Do these tests measure physiological responses to a job-related task? The answer is “no” for both tests. These tests are simply surveys asking individuals to self-report on their thoughts, feelings, and actions.
(f) Are these tests typically used in medical settings? The answer is “no” for both tests. Occasionally, these tests might be used as initial screening inventories in a psychological clinic. However, elevated scores are insufficient for rendering a diagnosis of a personality disorder. Instead, such diagnoses would only be rendered after extensive interviewing with a trained clinical psychologist. Such clinical interviewing is not part of the standard application of these surveys in a selection context.
(g) Do these tests involve the use of medical equipment? The answer is clearly “no” for both tests.
Therefore, both the NPI and SRP fail to meet any of the criteria articulated by the EEOC for defining a pre-employment medical test. We assert that a personality measure that does not meet any of the EEOC criteria for medical exam need not be excluded from a selection process, assuming that the measured trait is job relevant. Consistent with this conclusion, the focal article noted in the case of Thompson v. Borg-Warner (1996), the courts found that use of the PASS-III (a test that inquires about mental impairment) did not violate the ADA because it did not meet any criteria for a medical examination.
Additionally, Melson-Silimon et al. (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) noted that the EEOC guidelines stipulated that medical examinations include “psychological tests that are designed to identify a mental disorder or impairment” (EEOC, 2000). First, we want to emphasize that the EEOC guidelines refer to tests that were specifically designed to diagnose personality disorders as medical examinations. However, as noted above, tests such as the NPI and SRP were developed and validated using healthy, normal populations with the goal of measuring nonclinical levels of these traits (Paulhus, Neumann, & Hare, Reference Paulhus, Neumann and Hare2016; Raskin & Hall, Reference Raskin and Hall1979). In contrast, other measures such as the Minnesota Multiphasic Personality Inventory (MMPI) were developed specifically to differentiate between those with and without clinical levels of psychopathology, including personality disorders. Thus, when developing this test, a population of clinically diagnosed individuals was used as a comparison group against a “normal” population (Hathaway & McKinley, Reference Hathaway and McKinley1940). Therefore, it makes sense that in cases such as Karraker v. Rent-A-Center (2005), courts ruled against the use of the MMPI in selection processes. In contrast, tests such as the NPI and SRP that are not designed to diagnose or measure clinical impairment and were not developed and validated using clinical samples, would not fall under the same classification as the MMPI.
As Melson-Silimon et al. (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) noted, the EEOC also provides examples of psychological tests that are not considered medical examinations. These tests include assessments of “honesty, preferences, and habits” (p. 6), which represent standard content on measures such as the SRP and NPI. Specifically, both the NPI and SRP include questions on topics such as honesty, lying, and scamming others, as well as habits such as risk-taking behaviors and tendencies to be manipulative and exploitive of others. Therefore, this provides additional evidence that such measures are likely to fall under the EEOC’s description of psychological tests that are not medical exams.
The final argument we want to make providing support for the measurement of nonclinical dark traits (e.g., the NPI and SRP) in selection relates to the case of Barnes v. Cochran (1997). In this case, even though the defendant organization used the MMPI, the court ruled in favor of the defendant because the plaintiff was unable to show that the select-out decision was actually a pretext for discrimination. In other words, if measures such as the NPI or SRP were used in a selection context, a cut score would be determined before the tests are administered. Every applicant falling above that cut score would be selected out of the hiring process. We agree with Melson-Silimon et al. (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) that if a narcissistic or psychopathic individual were to take these assessments, it is possible that his or her score would fall above the cut score, and thus he or she would be selected out. However, we do not agree that this necessitates the abolishment of using these tests in selection.
Do not throw the baby out with the bath water. Yes, these tests cast “wide nets,” and they may occasionally catch someone with a clinical personality disorder in that net. However, as seen in Barnes v. Cochran (1997), a selection process measuring dark traits might remove a large proportion of applicants (e.g., the top 15%,), but a high score does not equal a personality disorder. Additionally, using a high score as a cut score is not a pretext of illegal discrimination (e.g., such a selection process would not solely select out applicants with clinical personality disorders). By analogy, we would never recommend that an organization cease using tests of cognitive ability because individuals scoring below a cut score may suffer from intellectual disabilities. The reason that cognitive ability tests are being used is because they have been shown to be relevant to performance across a wide array of jobs. They are not being used to discriminate against a disabled group of individuals. Likewise, we would only recommend the use of surveys measuring the dark triad if the traits were judged to be job relevant.
In sum, although we agree that a job analysis must show job relevance of a test, and that selection tests must comply with the ADA and EEOC guidelines, we do not agree that measurement of nonclinical “dark-traits” need always be a concern. Specifically, we (a) argue that measures such as the NPI and SRP are in compliance with EEOC guidelines, and (b) have used previous case law to show how measures akin to the NPI and SRP can legally be used in selection. Our argument is similar for other dark traits, such as Machiavellianism. Further, our argument is consistent with the common use of cognitive ability testing in selection systems.
You have
Access
Although Melson-Silimon, Harris, Shoenfelt, Miller, & Carter (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) raised a number of important points regarding the use of dark trait measurement tools in selection systems, we argue that administering these tests as part of a selection process is unlikely to result in violations of the ADA. Specifically, we argue against the suggestion that we should “avoid measuring personality constructs that are closely related to PDs … particularly those that assess known PDs and ‘dark-side’ traits that are often subclinical measures of PDs” (Melson-Silimon et. al., Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019, p. 129). To buttress our argument, we will critically evaluate various EEOC guidelines. Further, we will use relevant case law throughout to strengthen our perspective. Specifically, we will be referring to assessments such as the Self-Report Psychopathy Scale (SRP; Neumann, Schmitt, Carter, Embley, & Hare, Reference Neumann, Schmitt, Carter, Embley and Hare2012; Paulhus, Neumann, & Hare, Reference Paulhus, Neumann and Hare2016; Paulhus & Williams, Reference Paulhus and Williams2002; Williams, Paulhus, & Hare, Reference Williams, Paulhus and Hare2007) and the Narcissistic Personality Inventory (NPI; Raskin & Hall, Reference Raskin and Hall1979). Machiavellianism is often considered in concert with psychopathy and narcissism as part of the dark triad (LeBreton, Shiverdecker, & Grimaldi, Reference LeBreton, Shiverdecker and Grimaldi2018) and thus shares similar traits to psychopathy and narcissism. However, there is no clinical diagnosis of “Machiavellian personality disorder,” and we will therefore focus our arguments on the dark-triad traits that do have companion clinical diagnoses.
As Melson-Silimon et al. (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) noted, prior to extending a conditional job offer, the ADA prohibits an organization from using a medical examination to screen job applicants. And, Melson-Silimon et al. suggested that tests such as the NPI and SRP may be interpreted by the courts as medical examinations. This conclusion struck us as inconsistent with the EEOC guidelines and case law that Melson-Silimon et al. reviewed in their article. Indeed, we actually expected these authors to conclude that using these types of tests in a selection system is unlikely to be interpreted as a violation of the ADA.
Contradictory to Melson-Silimon et al. (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019), after reading the focal article, we were left to conclude that these tests would be permitted in selection. We arrived at our conclusion by simply comparing two exemplar measures, the SRP and NPI, to the EEOC criteria for determining whether a test is a medical examination. As Melson-Silimon et al. noted, if a test meets any of these criteria, then the test may represent a medical examination. The criteria, in relation to the NPI and the SRP, are as follows:
(a) Are these tests typically administered by a healthcare/medical professional? The answer is “no” for both the NPI and SRP. In a selection setting, both tests are likely to be administered by an HR professional.
(b) Are these tests typically interpreted by a healthcare/medical professional? The answer is “no” for both the NPI and SRP. In a selection setting, both tests are likely to be interpreted by an HR professional.
(c) Were these tests originally designed to reveal an impairment or an applicant’s current mental or physical health? The answer is “no” for both of these tests. Both tests were developed and validated using normal (i.e., healthy) populations.
(d) Are these tests invasive? Neither test is invasive. Both rely on a standard paper-and-pencil self-report assessment format. In addition, they are interpreted as stand-alone tools (i.e., they are not interpreted in conjunction with mental health or forensic records).
(e) Do these tests measure physiological responses to a job-related task? The answer is “no” for both tests. These tests are simply surveys asking individuals to self-report on their thoughts, feelings, and actions.
(f) Are these tests typically used in medical settings? The answer is “no” for both tests. Occasionally, these tests might be used as initial screening inventories in a psychological clinic. However, elevated scores are insufficient for rendering a diagnosis of a personality disorder. Instead, such diagnoses would only be rendered after extensive interviewing with a trained clinical psychologist. Such clinical interviewing is not part of the standard application of these surveys in a selection context.
(g) Do these tests involve the use of medical equipment? The answer is clearly “no” for both tests.
Therefore, both the NPI and SRP fail to meet any of the criteria articulated by the EEOC for defining a pre-employment medical test. We assert that a personality measure that does not meet any of the EEOC criteria for medical exam need not be excluded from a selection process, assuming that the measured trait is job relevant. Consistent with this conclusion, the focal article noted in the case of Thompson v. Borg-Warner (1996), the courts found that use of the PASS-III (a test that inquires about mental impairment) did not violate the ADA because it did not meet any criteria for a medical examination.
Additionally, Melson-Silimon et al. (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) noted that the EEOC guidelines stipulated that medical examinations include “psychological tests that are designed to identify a mental disorder or impairment” (EEOC, 2000). First, we want to emphasize that the EEOC guidelines refer to tests that were specifically designed to diagnose personality disorders as medical examinations. However, as noted above, tests such as the NPI and SRP were developed and validated using healthy, normal populations with the goal of measuring nonclinical levels of these traits (Paulhus, Neumann, & Hare, Reference Paulhus, Neumann and Hare2016; Raskin & Hall, Reference Raskin and Hall1979). In contrast, other measures such as the Minnesota Multiphasic Personality Inventory (MMPI) were developed specifically to differentiate between those with and without clinical levels of psychopathology, including personality disorders. Thus, when developing this test, a population of clinically diagnosed individuals was used as a comparison group against a “normal” population (Hathaway & McKinley, Reference Hathaway and McKinley1940). Therefore, it makes sense that in cases such as Karraker v. Rent-A-Center (2005), courts ruled against the use of the MMPI in selection processes. In contrast, tests such as the NPI and SRP that are not designed to diagnose or measure clinical impairment and were not developed and validated using clinical samples, would not fall under the same classification as the MMPI.
As Melson-Silimon et al. (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) noted, the EEOC also provides examples of psychological tests that are not considered medical examinations. These tests include assessments of “honesty, preferences, and habits” (p. 6), which represent standard content on measures such as the SRP and NPI. Specifically, both the NPI and SRP include questions on topics such as honesty, lying, and scamming others, as well as habits such as risk-taking behaviors and tendencies to be manipulative and exploitive of others. Therefore, this provides additional evidence that such measures are likely to fall under the EEOC’s description of psychological tests that are not medical exams.
The final argument we want to make providing support for the measurement of nonclinical dark traits (e.g., the NPI and SRP) in selection relates to the case of Barnes v. Cochran (1997). In this case, even though the defendant organization used the MMPI, the court ruled in favor of the defendant because the plaintiff was unable to show that the select-out decision was actually a pretext for discrimination. In other words, if measures such as the NPI or SRP were used in a selection context, a cut score would be determined before the tests are administered. Every applicant falling above that cut score would be selected out of the hiring process. We agree with Melson-Silimon et al. (Reference Melson-Silimon, Harris, Shoenfelt, Miller and Carter2019) that if a narcissistic or psychopathic individual were to take these assessments, it is possible that his or her score would fall above the cut score, and thus he or she would be selected out. However, we do not agree that this necessitates the abolishment of using these tests in selection.
Do not throw the baby out with the bath water. Yes, these tests cast “wide nets,” and they may occasionally catch someone with a clinical personality disorder in that net. However, as seen in Barnes v. Cochran (1997), a selection process measuring dark traits might remove a large proportion of applicants (e.g., the top 15%,), but a high score does not equal a personality disorder. Additionally, using a high score as a cut score is not a pretext of illegal discrimination (e.g., such a selection process would not solely select out applicants with clinical personality disorders). By analogy, we would never recommend that an organization cease using tests of cognitive ability because individuals scoring below a cut score may suffer from intellectual disabilities. The reason that cognitive ability tests are being used is because they have been shown to be relevant to performance across a wide array of jobs. They are not being used to discriminate against a disabled group of individuals. Likewise, we would only recommend the use of surveys measuring the dark triad if the traits were judged to be job relevant.
In sum, although we agree that a job analysis must show job relevance of a test, and that selection tests must comply with the ADA and EEOC guidelines, we do not agree that measurement of nonclinical “dark-traits” need always be a concern. Specifically, we (a) argue that measures such as the NPI and SRP are in compliance with EEOC guidelines, and (b) have used previous case law to show how measures akin to the NPI and SRP can legally be used in selection. Our argument is similar for other dark traits, such as Machiavellianism. Further, our argument is consistent with the common use of cognitive ability testing in selection systems.
Author ORCIDs
Sydney L. Reichin, 0000-0003-1143-6056