Personality tests and job interviews... yawn

Some companies test for "personality", some don't. Some say it works, some have no idea. Some do it just because they can. Fact is, there's no evidence it works at all. Fact is, behavioural interviews work better, but not by a lot. Fact is, nothing really works to predict job performance. We cull out the candidates we don't "think" fit the picture, then we interview to weed out the people we don't like the look of, or to choose (somehow) between close candidates. We interview because we are people and we are hiring people. We judge things quickly, face to face, and back it up with logic later. However we may test for "personality" in order to discriminate in yet another way, and just because we can. Not because it's a better way.

So how come we read this at BNET: "Sweeney claims that his testing methodology is based upon some pretty heavy-duty science. While I’m naturally skeptical of psychological research, I have no particular reason to disbelieve him. So I’m afraid that I think your irritation may be misplaced. In all probability, the company in question has discovered that testing is a better way to recruit successful sales employees than the standard interview process." Sweeney is a marketing guy, BTW.

Yep, interviews tell us little about how someone will perform in a role. But what correlation is there between personality and job role performance? None. So what is this "heavy duty science"? Well who knows, it's not disclosed here and it's not in any literature I've found. I'll keep looking...

But in the meantime... as far as personality testing goes, consider that you can guess what a company wants to see - honesty, integrity, an outgoing nature. You can also guess that overdoing it and appearing 'perfect' would be spotted, so you lie strategically. Lying, or telling falsehoods if you prefer, is smart. It's human. Sure, some people will be diligent in their answers, but even so it comes down to interpretation. Honest people will mark themselves down, some will simply misinterpret the question... in either case the validity will be doubtful. And then you want to somehow correlate personality with job performance? Over what time scale? A year? 3 years? What exactly will you accept as a positive correlation? Because I can assure you if it was that easy we'd all be doing it! In the real world a motivated person with the "wrong" personality may very well run rings around a less motivated person with the "right" personality. For a while, anyway. They will have ups and downs. Work performance is the sum of many parts, and good teams are made up of many different types of people. Selecting for "sameness" is just another way to shoot yourself in the foot. Avoid it.