A case for (premeditated) diversity in UXR

A thought experiment

I know what you are thinking… Another post about the need for diversity in UX? Is it really necessary? By now, you’ve probably learned about the many good reasons for advocating for diversity in UX. You have probably noticed the growing number of Diversity, Equity and Inclusion (DEI) units and initiatives within all kinds of organizations. If you are a designer, you might have encountered articles with specific recommendations and guidelines on how to design with diversity in mind. This topic is becoming a staple of pretty much every UX conference (and rightfully so). But the point I want to make here is not about how important is for minorities in UX to be seen and heard (although I am all for that). Rather, I want to make a point about why is not only beneficial for teams to be diverse (you should read this great piece on the topic, btw) but crucial enough to be purposefully pursued, particularly when it comes to research. Ok, hear me out:

…one of the reasons proper research tends to be trivialized in the UX design process is because it suggests that empathy is something that can be tokenized and generalizable.

The problem with UXR

As I have written elsewhere, the purpose of research is not just to find the right answers but to do so while avoiding bias. Understanding this is crucial to developing a research mindset for oneself and within an organization but that is easier said than done. Moreover, User Experience Research has the added challenge of having to be fast and cost-effective. Take for example the Lean Research Framework (LRF) by the MIT D-Lab, a really useful attempt at summarizing all the conditions of “good research” in a relatively simple framework. According to the LRF, good research should be rigorous, respectful, relevant and right-sized. The fourth point, “right-sized” addresses the speed and cost-effectiveness mentioned above through the assessment and elimination of unnecessary protocols. Unfortunately, quite often in “the real world”, even a relatively simple framework as LRF can take “too much time” for people to follow, and research ends up being bootstrapped.

Whose empathy?

Like anything else, Design Thinking (the methodological backbone of UX) has many flaws. Some of those flaws seem to work against the very idea of diversity. In this piece from 2020, Darin Buzon makes a compelling case for the role that Design Thinking has played in maintaining an industry historically white and male exactly where it is. Vivianne Castillo, the founder and CEO of HmntyCntrd addressed this very issue at UXRCONF this year.

Besides these fundamental issues (which deserve to be discussed at length), I am of the opinion that one of the reasons proper research tends to be trivialized in the UX design process is because it suggests that empathy is something that can be tokenized and generalizable. Let me explain. The idea that one can design for a community in which they are not really embedded is laughable. Empathy, you know, putting yourself in someone else’s shoes, is simply not enough to create a solution for a so-called “wicked problem”. The mere idea of using empathy as a synonym for research is highly questionable. Discursively, it means lowering the bar to the level of suggesting that anyone can design for everyone and that imagining yourself in someone else’s situation is good enough. And, according to a recent study by Zippia in the US, those anyones are most commonly men (59.6%), white (76.1%) or straight (88%).

The trap of heuristics

As a field, UX relies heavily on heuristics, cognitive shortcuts, rules of thumb. You are probably familiar with the idea that 5 people are enough for testing, this is a good example of a heuristic. Privileging heuristics makes a lot of sense in a rapid-paced field that has embraced the idea of releasing quickly and iterating often but it also aligns perfectly with the idea of producing insights quickly, which justifies the use of words like empathizing and researching interchangeably. The problem with heuristics as cognitive resources and decisions making tools is they are the shortest paths toward bias, and the problem with empathy as a mindset for research (again, answer-finding while bias-checking) is that it can only take us that far. For example, empathy cannot replace lived experience because one simply can’t acknowledge what they don’t know.

If we have to do better research but still keeping it quick and cheap, while maintaining the whole idea of empathy as it is and keep privileging heuristics, we need to diversify the lived experiences of UX and UXR teams by design.

Not just UXR

Now, just to be clear, between finding answers and avoiding bias, the latter is way harder not just for UXR but for everyone. Academia, where one certainly has more time to be mindful of avoiding bias, has the exact same problem. An analysis of social and behavioural studies conducted by researchers at the University of British Columbia in 2010 found that 80% of the studies analyzed were conducted on participants from Western, educated, industrialized, rich and democratic societies (the authors used WEIRD as an acronym). The problem is that the results of these studies are often presented as universal when the representativity of WEIRD in the world population is only 12%. But why did this happen? Well, because this is the population that attends the universities where these studies are conducted and if you decide to snowball sampling within the university to conduct your study you’ll end up with the same kind of people. A terrible combination of empathy and heuristics, if you ask me.

So, let’s do a quick summary:

1. Research is inherently hard, it requires finding answers while avoiding bias
2. UX demands research done fast and cheap
3. UX Research champions the notion of empathy, which is inherently limited and problematic
4. UX as a field relies heavily on heuristics, which are prone to bias
5. Empathy and heuristics are a bad combination when the field is dominated by traditionally privileged demographic groups

…we would have to recognize that our lived experiences are in fact research assets and that the aim is not to focus on hiring individuals but to build teams.

Funds of Knowledge

It would be naive to assume that the field of UX will change fundamentally in the next few years, but I do think that there is at least a way to tackle some of the problems that we currently have while maintaining the methodological underpinnings of UX somehow intact. If we have to do better research but still keep it quick and cheap, while maintaining the whole idea of empathy as it is and keep privileging heuristics, we need to diversify the lived experiences of UX and UXR teams by design. I am, of course, aware of the potential issues around proposing something like this. There are good reasons against hiring with particular demographics in mind but the opposite is also true. For example, research conducted by Christopher Chabris, Anita Wolley and Thomas Malone has demonstrated that simply increasing the number of women in a team increases its overall intelligence. However, I am not proposing to hire based on individuals’ identities but rather, on their funds of knowledge.

The concept of funds of knowledge was developed by educational researchers Luis Moll, Cathy Amanti, Deborah Neff, and Norma Gonzalez to refer to “an individual’s skills and knowledge that have been developed historically and culturally within their communities”. For this to work, we would have to recognize that our lived experiences are in fact research assets and that the aim is not to focus on hiring individuals but to build teams. This would have an immediate effect on the potential of a research team to empathize because empathy would come from very different points of view, the way we think about heuristic approaches to solutions would also change because heuristics are culturally dependent as well. We would have to stop thinking about diversity as something that is just right but also operationally and (ultimately if you want to take that route) financially beneficial for organizations.

So, that’s my pitch. I am aware that this can be a controversial proposition and that there is a lot more thinking to do to avoid issues of tokenization, for example. But that’s why this is just a thought experiment, one that requires more funds of knowledge. Your funds of knowledge.