Many years ago my boss and me interviewed several people for a job which I've no doubt any of them could have done well. All scored near enough the same. A colleague from an social science background (LA) insisted that we/I should re-interview them with some new criteria and so on to ensure we weren't influenced by their dress, looks, accent, mannerisms and so on as this was the 'professional' way of dealing with it. I told him we'd make the decision by a totally unbiased method and so my boss simply tossed a coin and eliminated one by one until there was only one person left on the list. They got the job and they were perfectly fine.

Hi Teresa

I got some interesting responses on this thread:

www2.cipd.co.uk/.../68227

I still haven't found a way to address social mobility specifically but lots to think about on the subject of bias generally.

I can't make the link work. It keeps taking me to the CIPD main site but if I hover over it to get the address, the address doesn't take me to the site either.

In reply to Elizabeth Divver:

Hmmm, I'll repost it here: behavioralscientist.org/.../

Alternatively, go to www.behavioralscientist.org and it's the current lead story, or search for 'debias'. Note, behavioral is missing the 'u', and debias is one word... two countries separated by a common language :)

In reply to Teresa:

That works, thank you.

It's a very interesting article, particularly their finding that it is better to remove the opportunity for bias than to attempt to challenge the bias. I'm still torn on that point. It might be an effective tool to use in a recruitment campaign, but without the brave souls that challenged bias in the people around them, our society would be a very different place today. For one thing, neither of us would be in professional or managerial jobs.

It's a useful article - once we get as far as the interview. What it doesn't address is the candidates who are missing - having selected themselves out on the basis of the initial advert/vacancy notification. That might include the almost 2 million people who will not apply if the job does not specifically offer flexible working; or the women who might consider the recruiter is looking for men because of the wording used.
(What I'm trying to say is that hiring bias starts long before we get to an interview.)

In reply to Anna:

Oh I agree! It's very difficult not to unintentionally cause a bias, for instance the commonly used 's/he' - meaning she or he - is commonly read as 'she' with the 'he' bit being overlooked.

Thanks for sharing, Teresa. This just goes to (further) show the importance of taking a proactive, thoughtful, evidence-based approach to recruitment.

In reply to Elizabeth Divver:

Interesting conversation.
Ummmm, I lean towards Elizabeth's view for this reason. Who can 'challenge the bias' If the folk who are being discarded/overlooked/deselected are not present within the decision making group?

In reply to Anna:

Agreed. And/or even selected themselves out on discovering who will be doing the selecting and/or methods being used.

Very interesting question and topic.

It's fascinating to see organisations actively working towards de-biasing the hiring process. It's a very complex thing to do because, as mentioned by Anna in this thread, many biased 'things' are subconscious like wording, application length, name of the hiring manager, etc.

In my field, artificial intelligence and conversational software, I've had the chance to work with forward-looking recruitment teams on this very topic.

The idea is that an artificially intelligent recruitment software (aka chatbot) would be a good alternative to avoid as many biases as possible. We've seen it work in two ways:

1. All-encompassing & welcoming conversations.

Instead of posting on job boards for very specific jobs (which includes lots of potentially biased wording), corporations would welcome all applicants to simply talk to the chatbot.

The chatbot would, of course, be geared to ask relevant questions to the users and look for specific skills. When potentially relevant candidates arose, it would then drive them down the recruitment funnel and eventually hand over to a human.

In this process, we avoid most of the formatting biases a job ad might display.

2. Skill-focused.

This one works as sort of the flip side of my previous point.

Since the applicants all talk to a chatbot instead of a human, the chatbot focuses on finding the best candidate for the job.

We are all guilty (consciously or unconsciously) of giving into biases. With a machine doing this job, the idea is it eliminates much of the biases recruiters might feel when reading applications or discussing directly when them.

I keep finding the progress we're making on this quite fascinating. I doubt we'll ever completely get rid of these biases, but there is work that can be done at this stage to, at the very least, reduce them to a tiny minimum.