‘Unconscious bias training’ doesn’t work

I’ve spent most of my working life in London – one of the most culturally diverse cities in the world. And yet many of my employers have struggled to reflect that diversity in the people they hire. As for recruiting more women, the gender population balance is 50:50 pretty much everywhere; and yet, in many roles (especially the more senior ones) women are somehow hard to ‘find’.

Can we blame this on bad old-fashioned prejudice? For the most part, no. Certainly not among middle-class metropolitans, where to express a racist or sexist opinion risks social and professional ostracism.

But what if bias in the recruitment process is unconscious? How can recruiters deal with prejudices and blindspots they’re not even aware they might have?

One answer is the growing practice of unconscious bias training – plus related strategies. Let the professionals help you identify and dismantle those hidden inhibitions and you too can have a diverse workforce!

But, says Odette Chalaby for Apolitical, there’s a problem with that:

“Despite its vast popularity, there’s no evidence that unconscious bias training actually changes behaviour or improves workplace equality. Diversity training, another common tactic, fares as poorly: research in the US has found that it either does not change the number of women in management positions, or actually reduces it.

“Having women alongside men on selection panels sometimes improves women’s chances of being recruited, but sometimes it does the opposite. And there’s no high-quality evidence that leadership development training programs help women progress.”

There is one approach that does seem to work though:

“But it’s not all bad news. Evidence shows that skills-based assessment tasks (where candidates are given tests that replicate the work they’ll actually do on the job) and structured interviews (where all candidates are given the same questions in the same order) have a positive impact on diverse recruitment. Unstructured interviews are more likely to allow unfair bias to creep in.”

Job interviews are horrible. It’s bad enough being the interviewee, but most people I know who have also been the interviewer say that the latter is more stressful.

That’s because there’s more at stake. If for a particular job the interviewee messes up, they can shake it off and move on. The interviewer, however, will have to live with the consequences of their decisions – and so will their colleagues.

If the two sides both mess up by accepting / making an unwise job offer – it is usually much easier for an employee to sever the relationship than the employer.

This asymmetry is compounded by asymmetries in information. Generally, it’s in the interest of the interviewer to accurately communicate the requirements of the job. The interviewee, however, has an incentive to present his or her ability to do it in, ahem, the best possible light.

Interviewers therefore find themselves looking to establish a deeper connection with the interviewee – one that enables them to decide whether or not to accept surface appearances.

As in most situations where strangers find themselves thrown together – whether it be a job interview, chatting to a fellow passenger on a plane, meeting a friend-of-a-friend for the first time, or going on a blind date – what people try to do is establish what they have in common. In the absence of verifiable objective information, we figure people out subjectively and by analogy to what one knows about one’s own existence.

The problem with that though is that the process is facilitated by shared life experiences – with obvious implications for diversity (and not just in terms of sex and ethnicity, but also class).

I think that may be why unconscious bias training appears to have so little impact – the issue is not a dislike of difference, but a reliance on familiarity.

It’s also why an interview process based on objective tests is much to be preferred over subjective first impressions.

Chalaby provides an excellent example:

“…in 1970 just 5% of musicians in the US’s top five orchestras were female, today it’s more than 35%. That’s down to a simple behavioural fix to weed out recruiters’ bias: the use of a curtain in auditions to hide whether a man or woman was playing.”

Of course, it may be that the position you’re seeking to fill doesn’t lend itself to an objective test of competence. But, then, if that’s the case, you might want to ask whether it’s worth filling at all.

---

Peter Franklin is Associate Editor of UnHerd. He was previously a policy advisor and speechwriter on environmental and social issues.

peterfranklin_