The mother (and father) of all research types
As researchers, we have all come across various types of research. We choose one over the other based on the requirement. I have used different research methods and types, but somehow, because research is not often a theoretical practice, I have never known the technical names of the types of research I use. Don’t get me wrong, I know a lot of people who have conducted research in an academic setting and then made career transitions to the UX space. I am pretty sure these are the guys who really KNOW research in theory. I guess I have been so caught up in the research practice that I haven’t really “researched” the theory enough. Recently, I learned to categorize research I conduct into two major types: 1. Exploratory research; and 2. Descriptive research.
I am not going to stick to conventions and state the definitions by the book, of each of these types. Rather, I will explain them in my own words, as I have conducted them.
Exploratory research I usually conduct exploratory research when the exact problem areas are not well-defined. It’s also useful to conduct this research in the initial stages of product and customer discovery. Exploratory research is mostly qualitative in nature.
The process I follow is quite simple. I first like to define the problem, or rather, what we think is the problem, and state the research questions we are trying to address. I also note down the assumptions and hypotheses at this stage, and keep referring to them as I progress in my research. When I conduct exploratory research for a new product or service, there’s usually no problem areas that need addressing. In such cases, I only define the research questions, and state the assumptions and hypotheses.
I then go on to explore the research questions and other problem areas, if any, through various research methods — it could be just one or a combination. This is the part where I validate the problems.
I then move on to the discovery stage, gathering insights from the previous stage. This involves discovering the actual problems, if they are different from the “assumed problems”. Discovery also consists of solutions to the actual problems, or suggestions and recommendations. In many cases, researchers are not asked to propose solutions, but I personally believe that researchers can also be strategists.
An example of an exploratory research project I worked on is identifying the problems in accommodation services at a disability services and support organization in Australia. This involved exploring the day-to-day activities of the staff and clients, the challenges they face, and what challenges have a trickle-down effect.
Descriptive research This type of research is more rigid in nature and highly structured. Although some of the methodologies used may be qualitative, the analysis is usually quantitative in nature. I use this type of research when I have a well-defined research task or problem in hand.
Descriptive research involves a lot more pre-planning, especially in defining what we want to measure and how we will measure it. Unlike exploratory research, in the descriptive type, it’s best to stick to the research design because we want to make sure we collect all the data we need for analysis.
The next step is very straight-forward — execute the research plan — whether it’s an interview, survey or a diary study, I always follow the existing plan!
The final stage in this process is analysis of data. If I am conducting interviews, I usually do a sentiment analysis by looking for keywords, and developing a heat map. Any data I have, even if it’s qualitative in nature, I numerize it. Because the research is descriptive in nature, the end result is a conclusive study, describing functions and characteristics.
A descriptive research I conducted is for an online car sales platform in Australia, the purpose of which was testing the beta website with the target audience. This research involved task analysis and longitudinal study. I looked for covariance of two variables: whether the type of cars people buy varies by income range and gender. I also looked for correlation between variables: whether mothers are increasingly looking for cars with safety features in them.
Causal research This is another type of research I came across in my literature review, however, I strongly feel that it’s a subsect of descriptive research, rather than being one of it’s own type. By definition, this type of research is suitable to identify if there’s a causal relationship between two or more variables. While I have conducted causal research in an academic setting, I haven’t done so in my professional research career.
What’s the point of this article if I don’t state my favorite type of research? It’s exploratory research, mostly because of it’s flexible nature and the surprising elements that emerge in the process. I have often discovered insights that were never a part of the plan, and that geeks me out. :)