statistics can’t predict the individual

That position you’re trying to fill? All those candidates you’re interviewing and assessing, scrutinizing and evaluating to find the very best person for the job? I’ve got some bad news for you.

I’ve probably never met you. Certainly don’t know the position you’re trying to fill or the candidates you’re looking for but I do know one thing. It is impossible to predict whether an individual will excel at the job or not. Can’t be done.

We want to. We want to know that we’re hiring the right person. We want to believe we can look them over and just know. Hiring managers think they can tell something by the way a person shakes hands or looks them in the eye or where they went to school or their GPA in Junior High or how nicely dressed they are or where they have worked in the past or the recommendation of a friend of a friend’s friend. Vendors really want us to believe that if we purchase their assessment, their interview guide, their hiring secrets book that we’ll suddenly know the perfect match for the job. But, no matter how good we are overall, we can’t predict the outcome of any one individual.

If you go to a doctor and get diagnosed with a life endangering disease, the doctor cannot predict your chance of survival. This is important: they can only tell you the survival rate of people with a similar set of symptoms. They can tell you that, as a group, X% survive, but they cannot tell you your exact chance of survival. There are just too many individually specific factors at play such as genetics, skill of the doctor/medical facility, resources, your state of mind, willingness to fight, etc. Statistics can’t predict the individual.

Credit scores are used to predict how likely someone is to pay their debt based on past history, current debt load, etc. The strongest we can say is that people with X credit score tend to be a safe credit risk. But it can’t say how likely an individual is to pay their debt. Again, there are just too many uncontrollable variables: a person with a great credit score might lose their job, have a financially catastrophic medical emergency, go through an ugly divorce, develop a drug habit – who knows? Likewise, although people with low credit scores tend to be more of a credit risk, it’s impossible to predict what a specific person with a low credit score will do. After all, there are plenty of people with low credit scores who are determined to turn it around. Statistics can’t predict the individual.

I can tell you that the average height of a professional basketball player is right at about 6’7” (thank you Wikipedia). I don’t know much about basketball, but I do know that height is an advantage. Yet, there have been 24 NBA players shorter than 5’9” including Hall of Famer Calvin Murphy who was right at 5’9” and 5’3” (!) Tyrone “Muggsy” Bogues. Statistically speaking, there’s no chance of a 5’3” person being successful, but statistics can’t predict individual results. Again, too many variables, including talent, drive, determination, creativity, etc. The strongest we can say is that the most successful people in the NBA tend to be tall, averaging 6’7” but we cannot say that any particular individual will be successful l due to their height. Statistics can’t predict the individual.

“Improves the odds.” That’s really all a good hiring system does. We try to accurately identify demands of the job and skills, knowledge, and experience required to be successful at the job. They we try it identify people who might have a chance at being successful and we measure a lot of different things in different ways and try to remove any evaluator  bias from the process (or at least cancel it out). All this to try to determine which of the candidates is most likely to be successful.

“Most likely to be successful.” That’s it. A great selection system will do a good job of identifying who is most likely to be successful BUT it cannot predict that any particular person will be successful. There are just too many other factors. We try to minimize those other factors with a well thought out selection system, but there are still too many uncontrollable variables. Someone who was a superstar might have family troubles, not get along with their boss, or not fit well with the company culture. And, there’re those who get weeded out by the select system but would have been fantastic.

Does this mean we shouldn’t create rigorous hiring processes? Just the opposite. I am a very strong believer in minimizing the variables and improving the odds when hiring. The more data and the more measures and the bigger the sample size, the more accurately we can predict. But, despite all the best efforts, there may be some who just don’t work out and there may be some phenomenal people that get missed.

Statistics can’t predict the individual.

Your thoughts? Please leave a reply.

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s