Use of Triangulation in UX Research to enhance the credibility of Insights
Contributed by Kiranraj Govind — User Insights Manager Khatabook
As researchers, we have all come across skepticism around sample sizes and design when we present findings. Strategic decisions are made only when stakeholders believe in the insights. Hence, we are constantly looking at methods to weave a compelling story that flows organically, makes sense logically, and is easy for everyone to absorb. One such tool in a researcher kit is triangulation, which I personally feel has helped me as a researcher.
The basic idea behind triangulation is that one can be more confident and increase the credibility and validity of the findings when different methods yield the same results. Previously triangulation was viewed as a qualitative research strategy to test validity through the convergence of information from various sources.
While this is perfectly true, I would go a step further and say that when we triangulate data, we are trying to answer more questions with clarity. We answer a particular behavior and the why’s behind them.
Coming back to the theoretical aspects, Denzin (1978) and Patton (1999) identify four types of triangulation
- Method triangulation: Uses different qualitative and quantitative methods to check for consistency of the generated findings.
- Investigator triangulation: When there are several investigators in the analysis process, the findings of each are then compared to develop a better understanding of how every investigator views the issue.
- Theory triangulation: Use different theoretical perspectives to make inferences about a distinct data set.
- Triangulation of Sources: Examines the consistency of different data sources within the same method. For instance, at various points in time, in public vs. private settings, comparing people with varying points of view.
One common phenomenon noticed recently is that many researchers use Triangulation and mixed-method research interchangeably. In my opinion in principle, both of them are used to create credibility and improve the validity of the results. Mixed methods are focused on combining qualitative and quantitative methods. In Triangulation, the combination is not necessarily limited to research methods; it can be researchers, data, or methods of analysis.
Interestingly, At Khatabook, we have been practicing Triangulation for a long time, not because it is an industry norm but because it just helped to add more credibility to the story we wanted to tell. Like Khatabook, In my opinion, Triangulation as a phenomenon may have existed for a long time but has emerged as a “buzzword” more recently. Many researchers may say that Triangulation helps validate the data, which is true, but at Khatabook, it helps us fill in the gaps and helps create a holistic picture.
Data Triangulation need not be restricted to a combination of just interviews and surveys, as it can take different forms. You can get as creative as possible while keeping the objectives and benefits of different methods in mind.
One useful map which explains different research methods and can serve as a framework is the What and Why framework created by Nielsen-Norman- Group.
As a researcher, the first thing you find out is that, in most cases, what users say and do can be very different. It is because, in most of these cases, users themselves don’t know why they do something they do! The framework helps as a guide to choosing the methods depending on what aspects we are most interested in.
While triangulation works well when done well, it also calls for some amount of resources and time. This is a luxury that many UX researchers are unlikely to have. Hence, try and use this for high-priority studies where you want some added leverage to get your findings across more effectively.
