Relationships between UX design, ethnography, and the scientific method
As an academic social scientist turned UX researcher, I have often thought about how these fields both gather knowledge and form actionable insights. Are scientists and designers doing dramatically different things, or are we just speaking different languages?
The Scientific Method is seen as the gold standard for rigorous inquiry within the Western world. It emerged from Enlightenment ideas of empirically understanding how the observable world works using reason.
For anyone who can’t quite recall their secondary school science class, here is a basic rundown of the scientific method:
Form a research question: What do you actually want to know?
Hypothesize an outcome: There are two parts to a good hypothesis—a variable and outcome. How the variable affect the outcome is essentially your hypothesis.
Collect data: This can be in a lab setting, ethnographically, or with any other method. Remember, “data” doesn’t just equal numbers. It can be qualitative as well, or a mix of qualitative and quantitative.
Analyze the data: Systematically organize and piece apart the data into themes or categories in order to answer the research question.
Accept or reject the hypothesis: Did the variable create the expected outcome? This is your discovery, and your insight.
The idea is that with a controlled environment where only one variable is at play. The goal is to create new knowledge that answers an open research question about the world.
Designers use a surprisingly method, the UX Design process. (Note: there are a few slightly different versions, but I’m referencing this one from the UX Design Institute.)
Conduct user research: Though not always explicit, many designers tend to follow the scientific method when they create research questions, systematically investigate users’ behaviors, needs, and opinions. Once data is collected, it is analyzed (pulled apart into components) and/or synthesized (put together with other data streams and alongside business needs).
Design solutions: In other words, innovative creation of new features, systems, processes, or products that answer the users’ needs. This is the step where you take the insights from research and translate them into the real-world with sketches, information architecture, and other components. The design is the “blueprint” of the solution.
Create a prototype: This is where you bring the design to life, with a low- or high-fidelity prototype.
Validate the design: Here is when the scientific method is used implicitly again! As a designer, you conduct another research sprint to test your prototype with actual users, and iterate the solution based on their feedback. Again, hypotheses are often used (e.g.: “If we add a button to the top, people will see it sooner and click into the next page.”).
Now that we have our definitions and context, let’s dig into these two processes. I argue here that the scientific method’s focus on hypothesis testing can be used within the UX design process. It’s especially useful to do so explicitly when describing design decisions to business stakeholders, because the scientific method is more commonly understood of the two.
However, the emphasis on testing a single hypothesis in a controlled setting doesn’t always work with real users going about their lives. People perform differently in lab settings than in “the wild.” When this distinction is especially important, or you are beginning the process of generating foundational research, ethnography is the best process to create actionable knowledge. It involves observing and participating in people’s daily lives and communities of practice, in order to uncover patterns, unspoken social rules and conventions, and how people navigate processes and decision-making. Ethnographers can form hypotheses, but it is not possible to isolate a single variable within people’s lives, because the point is that we live interconnected, messy lives. However, the insight is the description of that messiness and how it can be strategically incorporated into designs, and a more nuanced, systems-level understanding of users’ needs.
So which is the best process, and are they really that different? I would say that they all have their strengths and weaknesses, and each has its purpose.
My point is that designers need both the scientific method in order to systematize knowledge creation, as well as ethnography in order to actually understand users’ real lives, and it’s important to be able to articulate the relationships between all three. Designers and scientists are not totally different species, they just speak different dialects.