Background
Men in prison interact with a series of services daily. For example, they order their food, get in touch with loved ones and sign up for a course.
In the UK Ministry of Justice, we observed a phenomenon in digital teams: they were each creating their own personas even though they were all designing for the user group. Not only that but these personas were reinforcing stereotypes and creeping in biases.
Task
Staff do not often meet prisoners to break some of the assumptions we all have about people in custody – so we needed to do that for them. We could not have bad personas and be ok with that. It was time for this problem to be addressed in the research and design community.
I joined a small team of researchers to identify patterns of behaviour, attitude, and values on men in prison to encourage a less biases view of our users. We did this as part of our dedicated time for learning and development.
My role and approach
I started soon after the researchers had agreed and drafted a research plan. Initially, my contribution would be on designing any artefacts needed for the project. However, I soon became more involved in the analysis. Throughout the project:
I transcribed interviews and supported researchers on coding transcripts using Dovetail
I worked with the team to draft and iterate a codebook
I affinity sorted insights and visually organised them so the team could draw conclusions
I led the creative direction of the project, including the test of different visual languages
I designed the persona cards and any other artefacts such as quote cards, a website about the project and an insights library
Codebook drafted by the team with examples of what is and is not included in each tag
Transcribing and tagging transcripts based on the codes from the codebook
Insights library: a Google Sheet digital teams can use to source verbatim quotes
Project website: a Google Site to explain the rationale and the outputs of this work
Result
Five personas have emerged from quantitative analysis. We knew there was a risk that not a single persona would be uncovered from the data we collected. And that was ok. We led the data guide us.
Preview of the five personas uncovered through qualitative data analysis
These personas do not have any:
Demographic data, like ethnicity or age
Photographs
Salaries or interests
Names, which can say a lot about you
Gender
This is quite unusual but we believe that by getting rid of these elements we would encourage teams to think about archetypes instead of stereotypes – and that is what this project have achieved.
We provoked digital colleagues to challenge their practice and tools and to think of prisoners more empathically – as Greek Gods who made mistakes or became a product or their environment.
Myself and my colleague Amber Westerholm-Smyth (who conceptualised this project) delivered a few talks about this work in different conferences, including Service Design in Government and the UX research conference UXR.
We also wrote a blog about this work.