Is Unconscious Bias Training doing more harm than good?

Unconscious bias training (UBT) is popular among a number of workplaces. Often mandatory (which research suggests is harmful) it is packaged as a one-off, hour-long session which fits easily at the end of a busy day.

But have you come across the mounting evidence to suggest that UBT isn’t effective and may even be harmful?

If not, here’s a quick summary of some of the most compelling research:

• Comprehensive reviews of UBT found that this type of training isn’t associated with lasting changes in behaviour. A systematic review of UBT examining 492 studies, involving more than 87,000 participants is cited in many of the papers we’ve linked below.

• While UBT can have an impact on implicit measures (eg. attitudes, stereotypes), it hasn’t been shown to have a long-term effect on explicit measures (eg. conscious thoughts).

• UBT often takes the form of a one-off workshop or learning session and there’s little evidence to suggest that these types of interventions change peoples’ behaviour.

• UBT has been around long enough for us to conclude that the evidence for its effectiveness just isn’t there. 

• Measurement of how effective UBT training is, often relies on implicit association tests (IAT), which measure the strength of association between concepts and evaluations and stereotypes. However the effectiveness of IAT as a measure of implicit or unconscious bias is highly contested, not least because of a lack of retest consistency. 

Here’s why UBT could actually be harmful:

• These training sessions can contain messages suggesting that stereotypes and bias are unchangeable - asking people to suppress stereotypes tends to bring them clearly to mind, making them more cognitively accessible. What happens if I tell you not to think about giraffes?! 

• They can make people feel overconfident: research suggests that people whose workplaces have anti-bias training in place can be complacent about their own biases.

• People can more readily dismiss claims of bias in a workplace that explicitly shows a commitment to EDI training. UBT as a demonstration of this commitment, without further structures and policies in place, can be harmful.

• Framing bias as ubiquitous can make it seem acceptable, the narrative ‘everyone has bias so it’s not that bad’ can emerge as a harmful result of UBT. One study looked at this by framing racial bias as ‘implicit and unconscious’ and ‘explicit and conscious’. The participants who had received the ‘implicit’ framing considered a racial incident as having less intent, and therefore judged the incident as less harmful and were more lenient in their view of the perpetrator. 

• Making UBT mandatory can cause backlash from the majority group creating hostility or animosity towards other groups. We’re not for a minute suggesting prioritising the comfort of the majority group, but it is important to consider this as one way UBT could cause harm. We also know that people can be very resistant to any message that is delivered in mandatory training. 

• There isn’t enough evidence of the effectiveness of UBT from those who are subject to bias. The research to date has focused on the potential perpetrators of bias and not those who could be on the receiving end of this bias. This could mean that we are missing important accounts of the potential back-firing effects of UBT.

Very briefly, you might be wondering ‘Well, what should I replace UBT with?’ That’s something we’re keen to talk more about but to keep it brief here are some things which have been found to be effective:

• Approaches which involve mental effort, goal setting and integration of other ideas.

• Inter-group contact: spending time with people who are different from you is one effective way to reduce your bias towards that group. Structured mentoring is just one way to do this.

• Training that moves beyond awareness and towards an approach which targets changing behaviour and centres action, what to do, rather than the underlying reasons for bias.

Want to read more? Here are some links to the studies and research cited:

• A meta-analysis of procedures to change implicit measures

• Written Ministerial Statement on Unconscious Bias Training - GOV.UK

• Uncommon sense Why Doesn’t Diversity Training Work?

• Developing & delivering effective anti-bias training: Challenges & recommendations 

• Unconscious bias training

• Invalid Claims About the Validity of Implicit Association Tests by Prisoners of the Implicit Social-Cognition Paradigm - PMC 

Join the conversation on our LinkedIn!

Want to make sure you stay in the loop and never miss insightful posts like this? Sign up for our newsletter!