Confirmation Bias In UX Design
This is part 1 of a two-part series on topics related to "Bias in Design"
An overview of the most common forms of bias seen in interaction design
With the conceptualization and adoption of next-generation technology and disciplines such as AI, AR, and UX, eventually, the issues of racial and cognitive bias would seep in...
First and foremost as a developer and/or a designer you have to remember that regardless of where you were raised or how cultured you think you may be, you ARE still capable of exhibiting some form of bias. That's a fact.
There are many forms of cognitive bias that occur during the design and research process which could hinder the creation of the resulting product or service. This, in turn, puts the validity of the research and your work in general in a compromising position.
*** “I think it's outrageous if a historian has a 'leading thought' because it means they will select their material according to their thesis” *** *** ― Antony Beevor ***
One of the biggest forms is known as Confirmation Bias. This type of bias leads to designers, developers and of course ordinary people to intentionally take into consideration any data that only supports their argument or desired methods. This is done based on a number of factors including holding the belief that what they know and their experience is a fact without regard for the opinions of others.
*** “People put a lot less effort into picking apart evidence that confirms what they already believe.” *** *** ― Peter Watts, Echopraxia ***
You see this occur during A/B testing for example, when we want to see the results that we desire and stop at that point. Self Validation is the objective and it often leads to us making errors we try to ignore by making excuses. In UI/UX you must be careful to avoid this form of bias as it can cloud your judgment and cause damage to the usability testing and user research you conduct for your design patterns.
You must always keep in mind that in the design process you are not the end-user and that you are building patterns to be put to use in order to satisfy the needs of the targeted audience.
Here are some of the ways that you can prevent or minimize Confirmation Bias from taking hold during the design process:
1. Mapping the Journey: All this means is that you need to create a solid qualitative research plan so that you can interview potential users, analyze their motivations and their behavior so that you can deliver reliable yet personalized experience based on the data you obtained from them.
2. Diversify Your Feedback: This is where it gets tricky and what many designers often fail at. You need to make sure that the pool of respondents is not from the same background as yourself. While this can't always be done in every situation, the effort should be made nevertheless to make sure there is as many unique perspectives as possible for surveys and focus groups.
3. Look Out for the Naysayers: Yes you read it right and yes you have to include them. The ones who disagree with you, who don't like the design, or have issues with the intuitiveness of the product are important. They help you see if there are mistakes that you may have missed or exposed to new issues. The user feedback that comes from them is priceless.
4. Always refer to the Data: The data is your friend. You need it to succeed. The data from the research you conduct allows you to ask the questions that really matter to determine how a product is well received by users. Tracking usage data with quantitative metrics enables this to happen, the right way.