Ask a silly question

By Bartleby

EVERY SO often, a tweet goes viral. The latest case is that of Olivia Bland, a 22-year-old Briton who suffered what seems like the “interview from hell” at a tech company called Web Applications which left her crying at the bus stop. Remarkably, she was offered a job after the ordeal. Very sensibly, she turned the offer down, saying that she “felt like being sat in the room with my abusive ex”. If a two-hour interview was bad, imagine what a working week would have been like.

The interviewer in question, Craig Dean, apologised on Twitter saying that

I have no desire to see anyone hurt; and can only apologise if anything I've done has had that effect; it was not my intent

That apology is hard to square with Ms Bland’s account of the interview which started, bizarrely, with Mr Dean scrolling through her Spotify play list and criticising her choices, had him calling her an “underachiever” and ended with him being sniffy about the way she sat. The company’s rating on Glassdoor, the site that allows workers to review employers, is just 2.7 out of 5 and only 41% of people approve of Mr Dean, the CEO.

There may be some people who think that an aggressive interviewing style is fair game and a good way of sorting out the best candidates. Bartleby once sat through a university interview in which he was placed in the centre of a room while one academic in front of him and one behind him fired questions; coming from an argumentative family, this was like being at home. But the broader problem is that many people have an idiosyncratic interviewing style, which may be unrelated to finding the best candidate.

Worse still, this may lead to persistent bias. Many orchestras moved to a system of asking candidates to play their instruments from behind a screen, so the panel did not know whether they were male or female; to aid the process, women removed their shoes so the sound of heels did not give the game away. Using the screen made it far more likely that a woman would get the position. Interviewers’ judgment can be affected by all kinds of irrelevant factors, some of which may be unconscious. One study found that men with a limp handshake were less likely to get the job.

A metastudy of personnel selection methods found that structured interviews were three times more useful than unstructured versions in predicting job suitability. Structured interviews involve technical questions, such as the candidate’s ability to use a type of software. But a key point is that all candidates are asked the same questions in the same order. Mr Dean would not have been able to conduct his Spotify review, since it would be entirely irrelevant. (But Bartleby’s suggestions for Mr Dean’s playlist are “You Can Take This Job and Shove It” by Johnny Paycheck and “She Works Hard For the Money, so you better treat her right” by Donna Summer.)

In a Harvard Business Review piece on “How to Take the Bias Out of Interviews”, Iris Bohnet wrote that

unstructured interviews consistently receive the highest ratings for perceived effectiveness from hiring managers, dozens of studies have found them to be among the worst predictors of actual on-the-job performance — far less reliable than general mental ability tests, aptitude tests, or personality tests.

So even if Mr Dean had oozed charm and praised Ms Bland’s eclectic Spotify combination of Barbra Streisand and Stormzy, he should not have been asking the question in the first place unless she was applying to be a DJ or music journalist. It is he, not Ms Bland, who needs to go away and do some training on his interview technique.