Personality Test Faking Actually Improves Performance Prediction

Apparently, smart people lie so much on personality questionnaires it becomes detectable through factor analysis, and because it’s smart people that do it, this “Ideal-Employee Factor” enhances predictive validity.

Perhaps the most common objection to using personality questionnaires in employee selection is the issue of social desirability and faking. Understandably so, after all, who wouldn’t lie to land their dream job? What’s more, who would intentionally sell themselves short when completing a personality questionnaire?

Let’s be honest, if you were give a question like this, how would you answer?

Or this?

What’s worse, every attempt to solve this problem has failed miserably.

Adjusting scores using social desirability scales fails to improve the validity of personality assessments (Ones & Viswesvaran, 1998), and the adoption of ipsative or forced rankings reduces their utility in employee selection (Johnson, Wood, & Blinkhorn, 1988).

But let’s step back for a second and ask ourselves a few empirical questions on the subject:

• Do candidates really fake their responses?
• If so, how much does it affect their scores?
• And if it affects them a lot, what is its effect on selection process validity.

Imagine, for a moment, the ideal employee. Hard working, resilient, likeable, considerate, self-reliant, organised etc. Who wouldn’t want to hire someone like that?

So naturally, when job savvy candidates are completing personality questionnaires, and they see a question that appears to measure one of the above traits, what do they do?

Well the research seems suggests that at least some candidates are playing the game and present themselves as the ideal employee, giving rise to the so called “Ideal Employee Factor” that appears in factor analytic studies. In particular, trait conscientiousness and emotional stability i.e. hard-working and resilient, are over represented together in certain people, with other desirable personality traits also displaying unusually high scores. These personality traits which are normally unrelated to one another begin to cluster in certain people, making them identifiable through factor analytic techniques.

So overall, not great news for using personality questionnaires in selection… or is it?

The evidence suggests that faking doesn’t reduce criterion-related validity, and that adjusting scores based on social desirability scales doesn’t increase their validity either.

This research also suggests that ATIC is more strongly associated with general cognitive ability than any other variable measured in the study (Klehe, Kleinmann, Hartstein, Melchers, König, Heslin, & Lievens, 2012).

Although the correlation itself was only moderate, it does mean that ATIC is a partial predictor of cognitive ability. As a result, ATIC expresses some of cognitive ability’s validity, and the research very clearly shows that cognitive ability is a far stronger predictor of performance than any personality trait (Schmidt & Schaffer, 2016).

Also, who says that faking is such as bad thing?

Today’s personality questionnaires aren’t that easy to see through and keeping your lies consistent isn’t a cognitively simple activity. Why wouldn’t you want to hire candidates that are smart enough to play the system?

Impression management is an important competency for many roles, how does that differ from so called “fakability”?

If you are better able to identify performance criteria in selection, surely this translates to identifying (and thus expressing) performance criteria in the workplace.

So overall, perhaps the issue of candidate faking shouldn’t be a conversation stopper when discussing personality questionnaires.

If a candidate “does well” on a personality questionnaire, it means one of two things:

• They are legitimately high scoring on traits which are relevant to performance.
• They have “impression management” skills, which are also useful in the workplace.

No need for social desirability scores, or ipsative questionnaires, or the development of “objective” measures of personality, as even when the candidate lies it doesn’t reduce the validity of the score.

So overall, good news in a roundabout sort of way.

It stands to reason however, that the ability to fake may not always be predictive of performance in all roles.

For example, many roles require individuals to be honest and truthful, and hiring fakers may increase counterproductive work behaviours.

By that same token however, in roles which require aggressive impression management, perhaps fakability is an even stronger predictor of future performance.

Lastly, perhaps there is an association with job performance, but what does this mean for job-fit, retention, and employee engagement? Are high fakability candidates likely to stay, and would you even want them to?

As always, I would love to hear your thoughts!

Johnson, C. E., Wood, R., & Blinkhorn, S. F. (1988). Spuriouser and spuriouser: The use of ipsative personality tests. Journal of Occupational Psychology, 61(2), 153-162.

Klehe, U. C., Kleinmann, M., Hartstein, T., Melchers, K. G., König, C. J., Heslin, P. A., & Lievens, F. (2012). Responding to personality tests in a selection context: The role of the ability to identify criteria and the ideal-employee factor. Human Performance, 25(4), 273-302.

Komar, S., Brown, D. J., Komar, J. A., & Robie, C. (2008). Faking and the validity of conscientiousness: A Monte Carlo investigation. Journal of Applied Psychology, 93(1), 140.

Ones, D. S., & Viswesvaran, C. (1998). The effects of social desirability and faking on personality and integrity assessment for personnel selection. Human performance, 11(2-3), 245-269.

Schmidt, F. L., Oh, I. S., & Shaffer, J. A. (2016). The Validity and Utility of Selection Methods in Personnel Psychology: Practical and Theoretical Implications of 100 Years… Fox School of Business Research Paper.