Statistics may sound like a dry subject, but many judgments and decisions in everyday life would be improved by an application of statistical principles. Take the following scenario: a football scout hears of a player who has powered his team to a good win-loss record. His coaches think he’s one of the most talented players they’ve seen. But the scout is unimpressed by the one practice game he sees him in; he tells his manager it’s not worth trying to recruit the player.
Most sports fans would think that was a pretty foolish call. Athletic performance is too variable to base an important judgment on such a small sample. It’s not necessary to take a statistics course to get the correct answer to this problem.
But consider this problem: an employer gets an application from a junior executive with an excellent college record and strong references from his current employer. The employer interviews the applicant and is unimpressed. The employer tells his colleagues that it’s not worthwhile recruiting him.
Most people regard this as a reasonable sort of decision. But it isn’t. Countless studies show that the unstructured 30-minute interview is virtually worthless as a predictor of long-term performance by any criteria that have been examined. You have only slightly more chance of choosing the better of two employees after a half-hour interview as you would by flipping a coin.
Extroverts do better in interviews, but for many if not most jobs, extroversion is not what we’re looking for.
In both of these cases, predictions based on references – school reports, prior performance, letters of recommendation – give a 65-75% chance of choosing the better of the two.
Why do we get the athletic problem right and the employment problem wrong? Because in the case of the job, unlike for athletic performance, we haven’t seen hundreds of candidates in interviews of a particular type and seen how well performance in the interview corresponds to ultimate performance in the setting we’re concerned about. We haven’t seen that the guy who looks like a dunce in the interview turns out to be a whiz on the job and the guy who aced the interview turns out to be a dud. The only way to see that the interview isn’t going to be worth much is to be able to apply the “law of large numbers”, which prompts the recognition that an interview represents a very small sample of behaviour, whereas the references summarise a lot of behaviour.
The bottom line: there’s safety in numbers. The more recommendations a person has, the more positive the outcome is likely to be for the employer. Consider the job interview: it’s not only a tiny sample, it’s not even a sample of job behaviour but of something else entirely. Psychological theory and data show that we are incapable of treating the interview data as little more than unreliable gossip. It’s just too compelling that we’ve learned a lot from those 30 minutes.
One recommendation is not to interview at all unless you’re going to develop an interview protocol, with the help of a professional, which is based on careful analysis of what you are looking for in a job candidate. And then ask exactly the same questions of every candidate. It’s harder to develop such a protocol than you might guess. But it can really pay off.