Diversity in Study Participants and a Critique of the “Representative Sample” in Human-Computer Interaction Research

We reflect on our experiences in improving the diversity of participants in our research, focusing on geographic diversity and countering WEIRDness. Our reflections are grounded in four studies conducted over two years, with more than 100 total participant engagements across more than 100 hours of user studies. Our samples included participants from the UK and the USA, but also from the Republic of Ireland, the Netherlands, Kenya, Nigeria, Ghana, Armenia, Israel, and Japan. We reflect on some of the challenges we encountered and what we have learnt about the benefits of geographic diversity. Finally, we discuss the scientific ideal of “representativeness” and consider whether it is possible, or appropriate, in small sample studies ($\mathbf{n} \sim 20$). We propose that representativeness is antagonised by the ideal of diversity. We seek alternative ways of understanding and articulating the epistemic value of diversity in Human-Computer Interaction (HCI) research.