On a glorious run through the scrub brush in the wild lands of Wyoming, my brother-in-law remarked between breaths, "Interviews are worthless. Totally, utterly pointless. They don't actually give you any reliable information. We only think they do. And we're completely wrong." Now, my dear brother-in-law is something of an adorable, argumentative curmudgeon and sometimes stakes out a controversial position just to move the conversation in an interesting direction. I thought perhaps this was one of those times. But he was emphatic and cited numerous books and studies to support his assertion that traditional job interviews are, in fact, pointless.
I checked his facts, and he's right. The overwhelming body of evidence concerning job interviews supports his assertion. Interviews do not give the interviewer an accurate sense of whether or not a potential applicant can do a good job. A solid interview does not predict success. So why do we still do them?
For one, they serve as a kind of security blanket. We want to be sure we "like" the candidate. But Don Moore -- associate professor at the Haas School of Business at UC Berkeley -- notes, "Interviews favor candidates who are attractive, sociable, articulate, and tall. They also favor manipulative candidates, or ones who know how to make a positive impression even in a brief interview. But those aren't always the best job performers." We all have examples of "likeable" hires who interviewed well but turned out to be complete disasters.
When Moore's MBA students conducted interviews with two other classmates and then guessed who would perform better on the midterm exam, their success rate was not much better than simply flipping a coin: just over 50 percent. "We all want to believe we are good judges of character," he explains, but we don't bother to collect the real evidence we need to determine job performance. We instead rely on wholly unreliable "gut instinct."
Google did their own internal assessment of the efficacy of their interviews and concluded they were deeply flawed. Laszlo Bock, Google's senior vice president of people operations, explains, "We looked at tens of thousands of interviews, and everyone who had done the interview and what they scored the candidate, and how that person ultimately performed in their job. We found zero relationship." Bock describes it as a "complete random mess." Supervisors were not very successful at picking the right members for their teams.
Interview teams also bring inherent cultural, racial, and gender bias with them into an interview. A 2003 experiment conducted by Columbia School of Business professor Frank Flynn and New York University professor Cameron Anderson illustrates our prejudicial assumptions. Flynn and Anderson gave students identical applicant profiles to assess; the only difference was that one applicant was named Heidi, the other was named Howard. The students rated both applicants equally competent, but Howard was viewed as more favorable a candidate. Men and women alike categorized "Heidi" as aggressive and difficult to work with -- simply because of her successful track record.
I witnessed this unconscious gender bias firsthand when I was on an interview team. A female candidate answered a direct question about her accomplishments, citing many examples of successful programs she'd instituted. A male candidate answering the same question was personable but did not give a direct answer. During the debrief, many on the team forgot his evasive answer, but the female candidate was labeled "aggressive" for tooting her own horn.
From a job-seeker's standpoint, too, sometimes interviews are a waste of time. My worst interview ever? One in which members of the interview committee held unconcealed hostility toward each other. I felt I'd walked into a dysfunctional Thanksgiving supper -- without the wine and promised dessert. As each day passed after the interview, I recalled more awkward and unprofessional comments from that cringe-worthy ordeal. Although it always stings when you don't get a job, in this instance the pain of rejection was assuaged by the feeling that I had just dodged a bazooka.
Megan McArdle, writing for the Daily Beast, argues that although interviews are a good way to assess whether someone appears to be sociable or charming, they don't help determine if a candidate is "hardworking, honest, or supportive of their team members when things go wrong." Many Human Relations directors have argued that although this may be true, the thinking has always been that at least they won't hurt a good candidate's chance of being hired.
Not so, says Jason Dana at the Yale School of Management. His research indicates that "unstructured interviews are worse than invalid; they actually decrease accuracy." Dana says this is because we are prone to "dilution" and "sensemaking." When we are bombarded with too much non-diagnostic information, our brains "dilute" the valuable information. We don't ignore extraneous information, and we start to reduce our reliance on "good information." We seek out information that proves a preconceived impression of a candidate. Dana explains that if a candidate offers a response that is inconsistent with the interviewer's impression, the interviewer will reformulate their "sensemaking" until the candidate's answers seem to match their impression of the candidate. We want a narrative -- a story -- that makes sense, so when we don't hear one, we create one for ourselves.
So what's an employer to do? If you're not ready to scrap the interview entirely -- despite the convincing evidence -- create a more structured interview process in which each applicant is asked the same questions. Employ a scoring rubric that lists the specific skills, qualities and experience needed in the job. Each candidate can then be measured against this score sheet and not simply assessed by the wattage of their sparkling smile or their ability to turn on the charm.