The Data is Overwhelmingly Clear: Unconscious Bias Trainings Do Not Reduce Bias

Bias Reduction Protocol
Written By Michael Snelgrove

At Protocol65, we study the gap between business practices and behavioral science. The goal of our work is to uncover instances where business and behavioral science aren't aligned and then help address the discrepancy. Sometimes, this research can be tricky and we invest hundreds of hours researching a particular topic. When these deep research efforts pay off, they produce awesome moments of personal achievement. But other times, hours of research aren't needed; instead, it's a quick 15 minute Google search and a discrepancy appears. When we uncover a gap this fast, it produces an emotional blend of curiosity and disappointment:

If we can find out in less than 15 minutes that a common organizational practice doesn't work, why in the world is the practice still so common?

This is the situation we currently find ourselves when it comes to unconscious bias training, and it's worse than you might think.

Unconscious Bias Training Does Not Reduce Bias

It's worth repeating this header again - unconscious bias training does not reduce bias. The research is so unequivocally clear and well-articulated that it feels redundant to write another article on the topic, but the current prevalence of the training makes it worthwhile.

While unconscious bias and diversity training can take many forms, the training to which we're referring here is the most common (the type that most of us have attended as part of a mandatory requirement). These trainings normally consist of four parts:

  • A description of how the brain works and the distinction between conscious and subconscious thought processes.

  • A discussion of the most common types of biases and an explanation of how these biases, combined with stereotypes, lead to harmful outcomes.

  • A series of examples of how these biases impact our thoughts, behavior, and interactions with others.

  • A dialogue on the importance of being aware of our biases, with advice on how to combat bias in our decision-making moving forward.

Study after study has concluded that this type of training fails to reduce bias in employee behavior and organizational processes (e.g., hiring, promotions, etc.). Sociologist Musa al-Gharbi summarized the conclusions well in his December 2020 article: "Generally speaking, they [bias training programs] do not increase diversity in the workplace, they do not reduce harassment or discrimination, [and] they do not lead to greater intergroup cooperation and cohesion."

This conclusion appears over and over again, and it's not hard to understand why this type of training doesn't work. Simply bringing awareness to a topic doesn't guarantee any actual change in behavior.

Imagine hosting a 2-hour workshop that provides a detailed overview of the cardiovascular system, and then expecting participants to have lower blood pressure the next day. It simply doesn’t work.

To note, I've seen people accept these conclusions, but retain the belief that the specific training they offer is effective due to slight variations from the example above. In these cases, it's critical to ask whoever leads the training: "how do you know this works?" Due to the strength of the evidence, I highly recommend that you default to the training not being effective unless someone can prove otherwise.

The issues don't end with the lack of positive outcomes. Unfortunately, unconscious bias training can also trigger people to become more biased. You read that correctly, unconscious bias training often leads to more biased behavior. There are at least three reasons why this happens:

  • The training reinforces norms of bias and prejudice – Discussions in many training programs normalize bias and prejudice as part of the effort to bring awareness to unconscious bias. The thinking goes: “if we tell people it’s common, they will be more likely to accept that they have unconscious biases.” Unfortunately, a side effect of these discussions is that they reinforce bias and prejudice as the norm. Norms have a large impact on our behavior and strengthening the norms of bias/prejudice can lead to an increase in biased beliefs and behavior.

  • The training enables moral credentialing – In a truly weird psychological phenomenon, attending unconscious bias training can enable the most prejudiced attendees to become more biased. Simply put, by attending the training, they can add an item to their “I’m a good person” balance sheet. With another item in the “good person” category, this unlocks people to behave in a more biased or prejudiced manner and still maintain a positive self-perception. This concept is known as moral credentialing or moral licensing.

  • The training strengthens stereotypes - Many unconscious bias trainings mention or shine light on particular negative stereotypes, such as those based on race, gender, or socioeconomic status. This makes sense, right? You have to name the stereotype to address it. Unfortunately, there are real negative side effects to this approach, which can reinforce prominent stereotypes due to the illusory truth effect or create new stereotypes where they didn't previously exist.

There's Real Harm in Continuing Unconscious Bias Training

In addition to unconscious bias training potentially leading to more biased behavior (with no evidence that it reduces bias), they also cause harm to organizations in other ways:

  • First, unconscious bias trainings consume both attention and resources from other diversity, equity, and inclusion (DEI) solutions. Especially when unconscious bias is mandated across the organization, the scale of the program is likely to consume organizational resources (in both money and employee time) while also convincing leaders that they're doing enough.

  • Second, when employees discover the lack of evidence supporting unconscious bias workshops, they can become quite cynical. We've seen this play out recently in employee engagement forums. What kind of message is your organization sending if you're investing in a DEI solution that is well known as being counterproductive? Spoiler alert: it's not a good one.

  • Third, the continuation of ineffective bias trainings can lead people to question whether unconscious bias is real. KPMG's UK Chair was brazen enough to make this point during a February 2021 virtual meeting, openly saying: "There is no such thing as unconscious bias, I don’t buy it. Because after every single unconscious bias training that has ever been done, nothing’s ever improved.” This is an unfortunate conclusion for a leader to reach, one that is undoubtedly detrimental to legitimate efforts to reduce bias.

The strong evidence against unconscious bias training doesn't appear to have impacted its popularity - it's still common in large corporations and is frequently used as a response to high-profile incidents of prejudice or harassment (one reason these types of trainings are so widespread is that they have long been used for compliance and liability protection, despite their ineffectiveness at changing behavior).

However, awareness of the ineffectiveness and likely harm caused by unconscious bias training is growing. It's now widely written about in popular venues like Harvard Business Review and appears in many recent and well-known books, such as Jennifer Eberhardt's Biased. As a result, more leaders and DEI practitioners are becoming aware of the research and its clear conclusion. As this happens, organizations will be faced with an interesting question:

How much are we paying to increase bias in our organization?

We hope that the answer to this question is eventually $0 and organizations make the decision to massively revamp their bias reduction efforts. But we're not yet there. This is one gap between business and behavioral science that unfortunately is holding strong.

The Path Ahead - Solutions that Reduce Bias

At the start of the article, I wasn't exaggerating when I wrote that a 15 minute Google search will provide all the information necessary to reach the conclusions I've mentioned in this article (just copy and paste "does unconscious bias training work" in the search bar). While there is broad alignment that unconscious bias training does not work, finding out what does work to reduce bias is a bit more complicated due to the wide proliferation of unsupported solutions.

At Protocol65, we're committed to advancing the discussion of evidence-based solutions for bias reduction. We've tirelessly researched the topic and have created a summarized report to answer the key question: "What does work to reduce bias?"

>> Check out the Bias Reduction Protocol <<

-----

Sources: This post is based on a number of articles and papers that provide an in-depth review of the research on unconscious bias workshops. The top sources are:

Michael Snelgrove
Previous

Smart Beyond IQ, Four More Factors That Influence Your Thinking
Next

My Journey to Applied Behavioral Science