Diverse sampling in user research: Why it matters
Imagine this: you go to wash your hands before dinner at a local restaurant, and place your hand beneath the new automatic soap dispenser. You keep waving your hand, but still, nothing happens. Surely, the dispenser must be broken…
From behind, you hear the dispenser’s engine start to work- someone else easily got a palmful of soap. You walk over to try this dispenser instead. To your surprise, it does not respond to your hand yet again.
You are appalled as you come to realize — there’s nothing you could have done to make it work, because this dispenser is racist by design. I forgot to mention that this other person has light skin, and you do not.
This is an example of how bias is designed into technologies that are only inclusive to those that are “top of mind” for the companies and employees that create them. It is an example of our need to rethink the way products are designed and evaluated.
What if this soap dispenser had been evaluated beforehand? Would this problem have been caught before the product reached the user? If a diverse, representative sample had been recruited for testing, what might have happened instead?
The design process today is broken, and we desperately need to redesign our approach. We need to ask ourselves: who does this product fail and why?
IBM Design — User Research
At IBM, we have dedicated user researchers who conduct studies on different aspects of the user experience that ensure our products not only function seamlessly, but are also inclusive to every user. As the voice of the user, it is the researcher’s responsibility to ensure any bias or inequality that is unintentionally designed never makes it to production.
IBM user researchers are seeking to recruit underrepresented study participants. If IBM Design does not consider the demographic qualities that make up our users — such as race, nationality, gender, or diverse ability, we fail to investigate opportunities to create an equitable product experience.
Focusing on conducting user research studies with more inclusive samples makes business sense. In highlighting a wider range of voices in our studies, we open new avenues for innovation. It’s designing for the world and markets we want to serve, rather than defaulting to the status quo.
Research Participant Panel
This is why we would like to invite you to sign up for the Research Participant Panel. Based on the information you provide, we will contact you for design research studies that you are eligible to participate in. We distribute incentives such as digital gift cards as a thank you for your involvement in each study!
Sign up for the Research Participant Panel here.
Ora Peled Nakash, Theresa Faught, Gabriella Pascual, and Kimberly Bazemore are Design Researchers at IBM based in San Jose, CA, Austin, TX and Haifa, Israel. The above article is personal and does not necessarily represent IBM’s positions, strategies or opinions.
