Implicit bias and assumptions in user recruitment: a call for collaboration
When I got married I decided not to change my surname. To me this is an outdated tradition that harks back to a time when a woman became property of a man when she married. This constantly causes confusion as people always assume that my husband and I will have the same name. I appreciate that I am in a minority in the UK, and as it was my choice to do this, the assumptions don’t really bother me.
Why is this relevant to user research?
Assumptions can cause offence, and can be damaging to user research as they may result in people not putting themselves forward to participate in research. Often assumptions are due to our implicit bias.
Implicit bias is something I came across as a result of a podcast which mentioned Project Implicit, a non-profit organisation and international collaboration between researchers who are interested in implicit social cognition. Project Implicit are investigating thoughts and feelings that exist outside of conscious awareness or conscious control, an area that feels very relevant to user recruitment.
During the last 5 years I have been working in user recruitment the UX sector has changed dramatically. The impact on user recruitment is that we regularly recruit people based on very subtle behaviours. Or we are asked to source people who have low digital confidence but often people who meet that criteria don’t necessarily identify with that label. Very rarely are we recruiting just based on criteria such as ‘ideally people who shop online’.
As a result, the available pool of suitable people for many projects is very small, and we must consider how we can best approach all groups of people and make our recruitment inclusive.
What are we doing to address assumptions and implicit bias in user recruitment?
I am based in Bristol, and this means I am surrounded by people who can help tackle this challenge.
This task model was created by Nic Price and Ben Cubbon as part of a project researching the participants journey when taking part in research. This task model shows that, for the participant, the research experience starts way before the session, and there are a number of opportunities to introduce bias in this journey. For example what message do people see when they initially search for research opportunities. You can find out more about this research on the Participant needs Medium page, the task model has been published under the creative commons licence.
There is an organisation in Bristol called Women’s Tech Hub who work with organisations to increase diversity in the tech workforce. I have been in touch with them to talk about the external content we use to promote research and the language we use when asking people to come forward and participate in research. This is the start of the journey to understand gender bias and how we might be able to appeal to women in different ways.
These are two smalls steps we are taking to change how we recruit users in the hope that this will increase response rates to adverts we put out. As highlighted in this story by Tomer Sharon, there are lots of ways people recruit participants for research, I am hoping to start a conversation about how the process can be more open and hopefully easier.
This is an open call, if you work in user research, UX or design, then get in touch with me/us via email. I am looking for feedback from anyone who has advice on how we can tackle implicit bias in user recruitment (or research), or if you are trying to do something similar and want to compare notes.
Please follow me on here, get in touch on Twitter, or visit the People for Research website for further contact details.
This story is based on a lightning talk for UX in the City 2018.
