Fact Check: NYTimes on Personality Tests
What the Times gets right and wrong, and what to look for in a valid workplace personality test
Employers are desperate to avoid the risk of making the wrong hire—which wastes money, time and often negatively impacts company culture. As personality testing at work becomes a more popular solution to this problem, it’s rightfully receiving more discussion and critiques.
A recent New York Times Style section article aims to blow the lid off the personality testing “industry” and features unvalidated “pop-science” personality tests to negate the entire concept. We agree with their assessment of pop-science tests which are unvalidated and put people into discreet categories. Still, the article sloppily lumps together both unvalidated, unscientific tests with research-backed, validated tests that meet the criteria laid out by the experts they cite in the article as “psychometric tests”.
The field of Psychometrics, which is part of personality testing, is the science of measuring mental capacities and processes. As in any scientific research, there are defined protocols and standards to ensure that the measurements are accurate and the results are validated.
Well-researched, proven, and validated personality tests help to break down the conscious and unconscious bias of traditional, unvalidated hiring practices that are speculative and proven to be less effective than validated personality tests in conjunction with other researched hiring methods. So what did NYT get right and wrong? And how can you tell if the research behind a personality test is legitimate? We break it down.
WHAT THE NEW YORK TIMES REPORTING GETS RIGHT
Pop science personality tests that put people into brightline categories are not validated and go against the principles of valid psychometric testing. The New York Times correctly states that psychometric testing should:
Be validated and proven: A properly validated and reliable psychometric test of personality will show that results correlate to some measure of performance or outcome variable. An abundance of research links traits from the Big Five or Five-Factor model to correlate to workplace performance, job satisfaction, teamwork, attention to detail, and even overall life factors such as career success and total monetary lifetime gain (the list goes on). There are also a slew of other statistical properties that a proper assessment should meet—like factor structure of the test, reliability, assessment item validity—which the DISC and Myers Briggs do not.
Measure traits on a spectrum and against a large data sample: A properly designed test will show where people fall on a spectrum. For example people cannot be neatly sorted into Extraverted or Introverted (and measuring job applicants against a few other candidates will essentially have this effect). There’s a huge gray area in between these two extremes and most people fall somewhere in between, as do the levels of this trait needed to be successful at a role. Most professional jobs require some degree of working alone and interacting with people that individuals within a wide range on the introversion to extraversion scale would be suited. The brightlines of the Disc or Myers-Briggs test do not properly account for the nuances inherent in individuals and the traits required for a role. We get it, the brain loves and is wired to categorize —this is why astrology, pop-science personality tests and Buzzfeed quizzes like “which Seinfeld character are you” are popular and can be fun (To no one’s surprise on the Hire an Esquire team, our very own I/O Psychology guru, Eric Fox, is a Jerry).
WHAT THE NEW YORK TIMES REPORTING GETS WRONG
There are valid psychometric tests which meet the criteria laid out by the experts cited in the New York Times article. The New York Times mistakenly:
Lumps together pop-science and scientifically validated tests as psychometric testing: There are pop-science tests which the Times positions as representative of Psychometric testing and there are scientifically valid methods of measuring personality, proven to correlate with workplace performance. There are many tests within the sphere of psychometric testing based on the Five-Factor Model test which is backed and validated by 85+ years of Industrial / Organizational Psychology research and meet the criteria laid out by the Time’s own experts.
Ignores psychometric tests that are currently in use and their scientific validity
Not unlike Gemini’s horoscope last month, the NYT article paints a grim and vague picture of the psychometric landscape. They have experts chime in about what a scientifically validated psychometric assessment would look like, and neglect to mention that these tests do in fact exist.
Tests backed by the Five Factor Model that are validated and meet the criteria laid out by the Time’s own experts include Hogan, Caliper, and Hire an Esquire’s psychometric testing. In addition to being based on valid research, they also measure candidates on a spectrum and compare the scores of a candidate to that of a relevant population (E.G. White collar professionals, Lawyers, Salespeople). Hire an Esquire’s psychometric testing also triangulates data from outside sources in addition to a personality inventory.
And the Five Factor model is not just used for workplace tests but also trusted in clinical settings as the basis for mental health tests like the MMPI. In addition to withstanding the test of time in research and medical settings, the five factor model has been in the news more recently for its effectiveness as a political weapon. Political junkies may remember the Cambridge Analytica scandal where misinformation was effectively spread targeting Facebook users. Facebook users susceptibility to believing and spreading misinformation was gauged via personality “quizzes” which were actually Five Factor model personality inventories.
At Hire an Esquire, we’re using personality inventories for good. We’ve spent >3 years building, testing, and proving our tests over thousands of placements. Since implementing, we’ve seen higher job satisfaction rates for our contractors and dramatic increases in client success metrics like repeatedly working with the same contractor and permanently hiring contractors —we also saw a significant increase in diverse placements. While we firmly believe validated personality tests make for better hiring decisions, we also use them and think that they should be used as just one additional data point in an overall hiring decision.
And New York Times, give us a call next time, we’d be happy to go on record.