Unconscious Bias Training Doesn't Eliminate Bias. Here's What Works Better.
Two women partners at a major Brazilian services firm reached out to me for help: "In the past few years, we launched numerous diversity initiatives, events and roundtable conversations. We even organized unconscious bias and sexual harassment training for our male partners but we still face pushback and lack of interest in promoting women-focused initiatives at work," they shared with me.
I was very sympathetic but not at all surprised. Many leaders in the United States and Brazil tell me proudly about their companies' unconscious bias training. It has become a staple in the list of key initiatives a progressive company should have in its diversity and inclusion portfolio. In fact, nearly half of midsize companies in the United States and nearly all the Fortune 500 companies use some form of diversity training. Yet, evidence exists that the effects of such training are limited and that sometimes it can backfire by reinforcing bias or generating backlash.
To investigate this issue further, a group of Wharton professors conducted a clever large field experiment with over 3,000 participants in one global organization (61.5% male; 38.5% located in the United States; 63 countries represented). They wanted to test "whether a brief science-based online diversity training can change attitudes and behaviors toward women in the workplace." The training itself incorporated all the best practices available in the market, including individual feedback on own biases and actionable strategies to counteract them.
The training was voluntary.
Five months after conducting the training they measured the change in participants' attitudes and real workplace decisions. Some of the findings included:
The training was more likely to lead to a change in attitudes rather than a change in behaviors (i.e. awareness did not translate into action)
Participants in the United States, on average, didn't exhibit much change in attitudes towards supporting women
International participants saw a significant improvement in attitudes towards supporting women (but they started from a lower baseline)
The biggest impact was surprisingly (or maybe not so) on women, especially in the United States. There was an increase in their attitude towards supporting women at work and it prompted them to seek mentorship from senior colleagues to overcome barriers at work
The above results are very much consistent with what I had observed while leading discussions in companies and through conversations with individuals who went through such training around the world. There is value in the training in terms of validating the experience of those who are the target of biased behavior and give them some coping tools. However, they rarely lead to actual behavioral change on the part of those who are in majority power positions.
It's clear that a short training intervention is not sufficient in terms of changing people's long-held beliefs and behavioral patterns. The researchers offer several actions that companies can take to replace or supplement them:
Recruit more women and under-represented minorities, especially into leadership roles
Change processes to mitigate the effects of stereotyping and bias (here's an example of one company that changed its hiring process)
Develop longer term programs and multiple interventions (here's an example of a strategic framework you can apply)
This article was originally published in my column for Forbes Careers.