The Future of Diversity Training
Since 2016, we’ve taken square aim at Diversity and Unconcious Bias Training. At a time when it was normal to share these schemes as flagship change initiatives, our opening gambit was always loud and clear: “Unconscious Bias Training is not an effective way to reduce or mitigate long-term bias in your organisation”. In December 2020, the scientific evidence finally reached newspaper headlines as UBT was scrapped from the Civil Service:
“Unconscious bias training does not achieve its intended aims. It will therefore be phased out in the civil service. We encourage other public sector employers to do likewise.”
This means it’s time to update our position for 2021.
We stand by our bluntness over the past 5 years. The financial resources spent on diversity training were never justified by the evidence for its impact, and this is especially true considering the wide array of behavioural science tools that are often cheaper and more effective in promoting behaviour change.
However, as practitioner knowledge has grown, we feel able to outline a more nuanced position. “Diversity and bias training doesn’t work” is an unclear and oversimplified representation of the mixed evidence across a variety of outcomes. It remains possible that well-designed diversity, bias or anti-racism training programmes could have a positive effect, especially on learning outcomes that do not assume long-term behaviour change.
As we see a renewed vigour around discrimination and systematic racism, we want to make sure that if organisations do consider training in their suite of interventions, they design them in accordance with the evidence and robustly test their effect. To achieve this goal, we invite the D&I experts of these organisations to join us in amping up the research on effective diversity training.
The state of the research
Before we jump into what organisations can do to achieve the best outcome through Diversity Training, let’s look at the state of the scientific literature. Currently, systematic research is lacking and there are several questions that need to be answered even to conduct useful studies:
What are the most reliable behavioural measures?
How do we make sure that those who most need Diversity Training sign up?
What elements of the training need to be the same across studies?
The most recent high-quality meta-analysis criticised the lack of progress made on these fronts. However, the authors were still keen to emphasise that successful diversity training does occur, depending on the outcomes you are trying to achieve:
All in all, the state of the research signals that there is a lot of work to be done. This is especially relevant when we look at outcome measures and experiment design. But while new studies are underway, there are several learning methods that show promise:
What makes diversity training effective?
Firstly, we need to answer the burning question above: what’s the point? If long-term behaviour change is your response, you may want to look elsewhere (see our article on Inclusion Moments). But what about cognitive learning and attitude change? Effective diversity training programs on these grounds follow the path of everyone's favourite university classes: They run over a longer period of time, with mixed teaching methods and integrated into the wider organisational culture. They also involve the perspectives of all partaking employees and empower them to form allyships.
Let’s take a closer look at each of these elements.
Make it long-term
Research on learning and development is clear: longer-term programs have been shown to result in more effective information retention and skill development. The literature shows that this is no different when it comes to diversity training. Bezrukova et al.'s meta-analysis showed that longer diversity training programs are associated with positive reactions, better diversity knowledge, and positive attitudes towards the minority groups. These relationships become stronger with the hours of training received.
Inviting participants to multiple training occasions will create more opportunities for contact and exposure to diversity issues. Coupled with a well-structured syllabus including skill development, longer diversity programs also result in lasting behaviour change.
Use multiple learning methods
Effective learning occurs when a learner "touches all the bases" and combines different learning styles (e.g., feeling, thinking, acting, or reflecting), thus maximizing learning opportunities (Kolb & Kolb, 2005).
The meta-analysis found that trainees had the most positive reactions to diversity training when it employed several teaching types. As a similarly positive consequence, behavioural learning slightly increased in multiple method training. However, there is no evidence that this minimal change would extend beyond a few days after the session.
Make it an integrated effort
It is often said the training is 'only one part of the puzzle', meaning that diversity training needs to be considered as a part of the wider D&I strategy. Evidence for the beneficial effect of integrated approaches dates to the early 2000s when Bendick et al. (2001) found that integrated diversity programs were highly effective in a US sample of multiple organisations. Most recently, Bezrukova and peers confirmed that integrated diversity training can have a significant positive effect on attitudinal and behavioural change.
Still, there is little consensus on how this would be best achieved. We need to work harder to determine the interaction between learning programmes and other diversity initiatives, especially as this seems to be a promising avenue for exploration.
Watch the framing
The way diversity initiatives are framed can affect feelings of inclusion and fairness of traditionally overrepresented groups, which can have important implications for the outcomes of diversity training and intergroup relations. For instance, Plaut et al.'s (2011) research showed that diversity programs presenting multicultural ideas with the awareness of White employees' cultural background have a significantly positive effect on reactions to the training. Consequently, the positive evaluation of diversity training can result in a stronger attitude and behaviour change.
It is critical that organisations are cognizant of the framing of their training initiatives -particularly considering the risk of 'backlash' presented in the literature. The effectiveness of the same initiative may well differ depending on the matter in which it is promoted.
Let’s not repeat the same mistakes of the last decade - we need to experiment.
The above tools are promising avenues for more effective training but they require more robust evaluation. There is much to learn about how to optimise diversity training and integrate it with a broader suite of interventions to bring about lasting behaviour change. And most importantly, organisations shouldn’t just resume training with the principles introduced, but they need to pioneer and embrace an experimental approach.
“There is a lack of rigorous experimental work on diversity training, especially when applying a behavioural science lens.”
Chang et al.’s 2018 study on online diversity training provides a blueprint we can use to kickstart further research. The authors ran a randomised controlled trial to assess the effect of a brief module on behaviour and attitude change. They found that the training only had a behaviour change effect on those already supportive of minority groups.
These findings support the small effect sizes previously reported in the diversity training literature. This is no surprise as they did not assess any of the principles we have detailed. This gives the motive to replicate their study and test the effectiveness of existing tools that have been already suggested to be more effective. In addition, the marginal success arising from a single virtual module reinforces us to think that research is possible and warranted even while we are all connected via Zoom.
The bottom line and our next steps.
Overall, the evidence suggests that training methods can be helpful in eliciting positive effects. Nonetheless, there is much to learn about how to optimise behavioural, attitudinal and cognitive outcomes, all while integrating them with a broader suite of interventions. To that end, MoreThanNow has developed a partnership with an online learning provider that allows us to randomise learners into different training experiences. This simple functionality will allow us to conduct robust trials in this space on behalf of our organisational partners, and we call for support and collaboration to help us push forward this work.