If You Want Fewer Bad Hires, Don’t Interview Candidates — Here is Why
April 11, 2017
The unstructured interview—where the interviewer asks personal questions to "get to know" a candidate—is a staple of the American hiring process. However, there's evidence that unstructured interviews increase rather than decrease the likelihood of a bad hire.
A recent article on this subject in the New York Times cited a case where 50 college applicants, who had been rejected due to an interview, were subsequently added as students. In the end, their academic performance turned out to be was identical to that of the applicants who'd interviewed well. The interviews, in other words, did not predict future performance.
In another study, interviewers were given a list of "yes/no" and "this/that" questions to ask, while half of the candidates were told to answer at random. The interviewers rated the interviews who gave random answers higher than the ones that answered truthfully. Put another way, unstructured interviews increase the likelihood of a bad hire.
Why are unstructured interviews ineffective?
Well, first, there's confirmation bias. Most of us assume we're excellent at assessing the character of people we meet. However, humans are merely adept at confirming the first impression emerged from their own personal history.
For example, I once had an instructor who took an instant dislike to me because (as he admitted at the time) we shared the same first name. Fortunately, he was eventually able to overcome this prejudice and, twenty years later, is now one of my closest friends.
Another reason unstructured interviews don't work is that they favor candidates who have a similar appearance and background to the interviewer. This innate tribalism was a useful evolutionary adaptation for primitive societies, where anyone from outside the tribe was automatically your enemy.
In our more heterogeneous world, however, this "friend or foe" thinking subconsciously drives us to select people based upon their accidents of birth rather than whether well suited for the job at hand.
Unstructured interview questions also tend to elicit rote answers. All but the most clueless candidates have learned and rehearsed solid answers to the typical unstructured interview questions. For example, a question like "what's your greatest weakness?" is virtually guaranteed to produce a canned humblebrag. Similarly, "where do you expect to be in 5 years?" is simply an invitation to regurgitate the expected career path for the position the candidate is seeking.
Finally, unstructured interviews put individuals who are naturally introverted at a disadvantage because introverts are often ill-at-ease when meeting and interacting with strangers. By contrast, sociopaths are often superb candidates because they have learned how to manipulate others by "hitting the right notes," simulating the emotions and affect most likely to impress the interviewer.
Rather than unstructured interviews, recruiters and hiring managers should use more of the following:
Assessment tools. While psychological testing may not be the panacea that testing companies often claim it to be, there is significant evidence that assessment tools can help assess a candidate's likelihood of success at a certain job.
Job-related scenarios. Rather than asking unstructured questions, set the candidate to work upon projects or explain how to handle situations that are typical of the job that the candidate will be doing if hired.
Standardized questions. According to the New York Times article, job interviews are more effective is candidates are asked the same questions in the same order, especially when the questions are job-related rather than of the "get to know you" variety.
For even more strategies, read Dr. John Sullivan’s post “Job Interviews Have Become Predictable and Ineffective – Here Are 10 Ways to Change That”
To receive blog posts like this one straight in your inbox, subscribe to the blog newsletter.