top of page

Unconscious bias training won't increase diversity

Environment versus human

In a great article in this month's Cambridge University Cam magazine, Professor Therese Marteau, Director of the Behaviour and Health Research Unit at Cambridge University, shows that we place too much faith in the ability of humans to change their behaviour. It's the environment that influences our actions much more than we would like to accept. In her line of work, it is clear that larger wine glasses make us drink more.

The same applies to making diverse recruitment and promotion decisions. Making our senior leaders attend unconscious bias training courses does little to fix the diversity problem in itself. Just as the knowledge about the health implications of too little exercise won’t significantly shift our health behaviour, so won’t being aware of unconscious biases significantly increase our leaders' likelihood to hire and promote more diverse talent.

Professor Iris Bohnet's fabulous work at Harvard University makes the same point. She talks about the importance of hard wiring diversity into our HR processes. She encourages us to think about ow can we design processes that make it less likely that we make biased decisions.

Human versus Machine

An interesting recent study by Professor Danielle Li, shows that in situations where managers decided to hire candidates despite test results indicating that they may not be suitable hires, these new hires tended to stay less long than those candidates who were clearly indicated as good hires through the test results. Professor Li and her team found that, on average, workers hired by managers who overrode test result recommendations less frequently (those in the bottom quarter of exception rates) stayed on average 120 days; whereas those hired by managers who overrode the test result recommendations more frequently (those in the top quartile of exceptions) stayed only 100 days.

Are we really so bad at making objective decisions? As Clodagh O'Reilly shows in this fabulous blog, we are all biased. For example, we tend to recall information that wasn't there (in Clodagh’s experiment, the word sun) as we expect it to be there after hearing lots of other related information (hearing words such as bright, rays, green, trees, flowers, summer). We use these shortcuts to help us make sense of a very complex world.

The Solution?

Wherever possible, we need to find HR solutions that stop us from falling prey to our biases and keep us from taking short cuts in decision-making. We can do this by using more objective data, such as translating our definition of high potential into psychometric tests and using these tests to inform our decision making, rather than relying on managers to make this decision unaided.

Furthermore, we need to ensure that these objective, multi-method recruitment or promotion methods are applied for every appointment, regardless of seniority. I often see that organisations’ recruitment processes become less rigorous and therefore more prone to bias the more senior the role is.

I like Prof Bohnet’s example of batch recruitment. She uses a fabulous analogy. It goes like this: If I asked you what you want to eat each day at the start of the month you are likely to give me a list a different lunchtime choices as you are likely to be find it difficult to imagine wanting to eat the same food every day. If, on the other hand, I asked you what you want to eat at lunchtime at the start of each day over the course of the next month you are much more likely to choose your favourite dish over and over again. The same applies to hiring. If we hire important senior roles in batches then it becomes much more obvious that we tend to hire the same type of candidate for each role.

Have you decided yet what you will have for dinner?

This article was first posted on LinkedIn on 23rd March 2016

bottom of page