5 Ways to Spark Reflections About Bias
Use these challenges to make your colleagues recognize the importance of bias.
Posted June 22, 2022 | Reviewed by Davia Sills
- Errors are common in the decision-making process.
- Sometimes they reflect systemic biases.
- Simple logic challenges can raise awareness of bias and spur positive changes.
Organizations are groups of individuals that work towards a common goal by making decisions. Unfortunately, the decisions made by those people are rarely perfect. To err is human. While some errors are just noise, others are systematic deviations. Whether it’s gender bias in hiring, excessive optimism in forecasting sales, or resistance to new ideas, many organizations strive to eradicate undesirable biases.
Regardless of the approach your organization may adopt to help employees make better decisions, individuals need to recognize their own biases first. The best way to do this is to create an “aha moment” that challenges your colleagues’ perceptions of themselves and others. Ideally, this is an insight that raises not just awareness but also a perceived need to change. Here are five tried-and-tested challenges you can use to achieve this.
You may already be familiar with this particular math challenge, as it’s become a very popular way to illustrate System 1 thinking—mental processes that are fast, automatic, and intuitive:
A bat and a ball together cost $1.10. The bat costs $1 more than the ball.
How much does the ball cost?
Intuitively, “10 cents” readily springs to mind. The correct answer is, of course, 5 cents. (5c ball + $1.05 bat = $1.10 in total.) This problem is also used in the three-question Cognitive Reflection Test that measures people’s propensity to think deeply and deliberately.
Have a look at the following short story:
A son and father are driving down a twisty road when they accidentally swerve off the side and hit a tree. Upon arrival, the medics notice that while the father has died, the son is still breathing. They bring him to a local hospital in hopes of saving his life. When they arrive, the surgeon looks down at the boy and says, “I cannot operate on this boy because he is my son.” How is this possible?
Many participants who read this do not think of the most plausible answer: The surgeon is the boy’s mother. Instead, they first come up with another explanation until their mistake dawns on some of them.
This is a particularly powerful exercise because it shows just how “naturally” and automatically we often make systematic errors—in this case, illustrating gender bias.
You can use this challenge to encourage friends or colleagues to think about either gender bias or unconscious assumptions. You may even want to design your own challenge with similar cognitive principles. For example, you could demonstrate how consumers automatically associate a category with a particular brand, such as an online search with Google.
Sometimes the most impactful insights come about interactively. The next challenge is based on the Wason Selection Task developed a few decades ago. In it, participants need to guess a rule about a number sequence. They are told something like the following;
I’m going to give you a sequence of three numbers, and I have a particular rule in mind that this sequence of numbers obeys. I want you to figure out what my rule is by proposing your own set of three numbers to which I will either say ‘yes’ it follows the rule or ‘no’ it doesn’t follow the rule. You can let me know what you think the rule is anytime.
Here are the three numbers: 2 – 4 – 8
Most people notice that the numbers seem to double in value and propose a sequence like 8-16-32. The answer to this would be: yes, it follows the rule. Participants rarely try sequences that might falsify their theory, such as 12-6-3. This illustrates confirmation bias, the fact that we often seek out information that supports rather than questions what we already know. In organizations, this may be evident in everything from a biased interpretation of employee performance to a general lack of innovation.
The Eiffel Tower
Whether it’s in our everyday life or in our profession more specifically, the judgments we make are constrained by degrees of uncertainty. This makes us vulnerable to bias.
Consider the following task:
Is the total number of steps to the top of the Eiffel Tower more or less than 7,000 steps?
Choose the right answer:
* More than 7,000
* Less than 7,000
What’s your best estimate? ___________
The correct answer in this particular case would be 1,665 steps. The more/less than question is added to create anchoring bias. People will, on average, provide significantly higher estimates if they first see this anchor (7,000) than if they are asked to guess the number without seeing the more/less question first.
The Eiffel Tower challenge works best with groups of people who complete the task either with or without the anchor. Feedback about the mean estimate of the anchored group will provide your employees with a powerful illustration of collective error and bias. You can make it even more relevant (and maybe even fun) if you use a number associated with your organization.
There is one remaining bias we have not yet addressed: overconfidence. This is often referred to as the mother of all biases because it’s ubiquitous, stubborn, and may prevent you from recognizing your biases in the first place.
A great way to show your colleagues that they, too, may be overconfident is to give them a quiz with a few trivia questions, such as:
- Absinthe is (a) a precious stone, (b) a liqueur.
- Robert Plant was the singer of what rock group? (a) Led Zeppelin, (b) The Who
- Singapore was occupied between 1942 and 1945 by the (a) Chinese, (b) Japanese.
At the end of the quiz, ask your respondents (privately) how many questions they think they got right. Overconfidence is simply the difference between correct answers and their estimate. Chances are that, on average, your colleagues will be overconfident.
There are other techniques you can use to reveal overconfidence. The best-known example is the above-average effect or illusory superiority, often illustrated by the finding that more than half of all drivers tend to rate themselves as above-average drivers. You can come up with your own question from the domain of work. Your question would ideally refer to a skill that is socially desirable or taken for granted by a worker. Having above-average productivity would be a perfect candidate, for example. How do you think your colleagues would fare?
Wherever there are human judgments and decisions involved, there is also error. If this error is systematic, it is bias. Obviously, this presents a range of problems to organizations. If you want your colleagues to think about their own biases and how they can be reduced, you need to trigger a conversation first.
You may already have data about your organization that shows an unwanted bias that you would like to target. While it’s important to share this with your colleagues, you can create an even more personal and relatable spark. To do this, present them with a simple challenge that creates a moment of insight. I have presented a few options in this post.