Suppose you interviewed me for a job, and I gave essentially meaningless answers to your questions. If you’re like a lot of hiring managers, you’d probably interpret what I said as useful information, and you’d rely on my nonsense statements to decide whether to hire me.
We’ve actually known for a long time that job interviews are a persistent illusion. That is, we keep asking open-ended questions in these unstructured discussions because we think we’ll “get to know a person” better. The fact is, such exchanges don’t yield useful information. All we do is convince ourselves that what’s meaningless is worth basing decisions on.
In a study done by Jason Dana and his colleagues at Yale University (“Belief In The Unstructured Interview: The Persistence Of An Illusion”), job candidates gave random answers to an interviewer’s questions—but interviewers were confident that their resulting impressions of the candidates were accurate.
It’s not just contrived interviews in research settings that fool employers. Dana’s work suggests that unstructured interviews in real settings are poor predictors of success on the job.
We make mistakes when we interview this way due to a common cognitive phenomenon: We’re wired to try to make sense of information, no matter how little value it has. Dana says we have a “propensity for ‘sensemaking’”—we try “to make sense of virtually anything the interviewee says.”
How can we stop our brains from making such costly hiring mistakes? First, we should avoid hiring people because we like them. (See “CMO: Chief Manager Of Outcomes.”) We should stop listening to our feelings about candidates and learn to rely more on objective metrics.
Chatting with a job candidate might be fun, but if you believe you’re a good judge of people, it will likely lead you to make a hiring mistake.
Second, we should learn to rely on better information. More concrete, objective measures of a candidate will likely improve your hiring.
In an interview with CMO.com, Dr. Arnold Glass, a researcher in human cognition at Rutgers University, said, “It has been known since Alfred Binet … constructed the original IQ test in 1905 that the best predictor of job (or academic) performance is a test composed of the tasks that will be performed on the job.”
In other words, use a job interview to learn what a job candidate can do. Gather hard evidence. Even a candidate’s GPA, when you’re interviewing recent college grads, is a more objective, useful predictor of future success than your gut feel. But open-ended questions don’t cut it.
What you think about job interviews is probably wrong. Dana cautions that information about a candidate gathered during an unstructured interview is likely illusory. Worse, it “can interfere with the use of valid information” that you take the trouble to collect and that can actually help you make good hiring choices.
Curb your intuition and avoid hiring who you like. The more objective evidence you can get, the more likely your hires will be good ones.