The Executive Selection Problem
Bryan B of HR Tests brings up a very interesting discussion about the role Industrial/Organizational Psychology plays in executive selection. The short of it, is that we don’t do such a great job. This is an area that is very difficult for us to tackle as rigorously as other areas. Sadly, it is also an area that we desperately need to address.
It is certainly true that we treat executive selection differently that we do other forms of selection. The reasons for this are manifold; to my knowledge, the principle factor in this difference in approach is that the most predictive of our selection tools fail us at this level in the organizational hierarchy. Cognitive ability tests are the “gold-standard” where selection methods are concerned. However, most candidates vying for executive level position are among the best and brightest individuals the organization has to offer. The cognitive ability among this group of people just doesn’t vary enough for researchers to identify a statistically meaningful difference between them.
To fill this gap, practitioners of Individual Assessment fill make use of a variety of additional objective and subjective measures. Personality tests, Interest Inventories, Assessment Centers play a role and occasionally projective tests (like the TAT) do as well. How well do these work? You’re going to have to ask someone else. Neither executive selection, nor individual Assessment is my areas of expertise. What I can say, is that using a character approach is not likely to be much more help than cognitive ability tests. Why? I’m glad you asked!
The underlying premise of assessing a candidate’s “character” is to identify traits that enable us to predict future behavior. This assumes a few things:
1) Traits are relatively stable
2) Traits predict behavior
3) Available trait (personality, character, values, etc.) assessments can actually measure executive’s traits
What does research tell us about these three assumptions? The Magic 8 Ball says, “Reply hazy, try again.” First off, we don’t have a lot of research about the stability of traits. What is out there suggests some measure of temporal stability, but it depends on what time spans we are talking about. I also suspect recent events and cognitive salience play a larger role in trait-based test responding than we normally acknowledge… call it a hunch.
The assumption about behavioral prediction is more problematic. Trait tests tend to be fairly general and the executive behaviors we are trying to predict are pretty specific. Work from the literature on attitudes suggests this is the worst possible combination. We also have evidence suggesting that external information (versus internal) influence behavior in many situations [Note: check out this neat summary of the impact of contextual information in behavior prediction].
The final assumption also gives us cause for worry about the role of trait tests on executive selection. In the best of times, personality tests are prone to faking. While there are some things we can do to reduce faking (I won’t give this away, if you don’t already know it. We need some tricks in our tool bag, after all), faking is a legitimate concern. Many people can make fairly good guesses about how to “fake good” on personality tests. I have reason to suspect that the best and brightest candidates will have little problem with this.
So where does this leave us. I suppose if I knew the answer, I would be paying somebody else to type this out for me while I wrecked yet another jet ski. The best I can say is that the status quo is not a great place for the scientific selection of captains of industry. We are long in questions, and short on answers. The good news is that now is a fantastic time for a brilliant person to find (and market) the solution. In the mean time, try not to buy any expensive snake oil; if you must, I can assure you that at least my brand is delicious (tastes suspiciously like a Venti Americano).