UX research comes in many shapes and forms, and it often borrows techniques from other fields such as psychology, anthropology, and market research. While you don’t have to learn every single research method out there (there are way too many), you should get yourself familiar with the different ways in which research can be classified.
Qualitative research is based on observation and feedback collection techniques. It provides insights that can be used to investigate human behaviour, user needs, motivation and goals. Examples of qualitative research methods are user interviews and think aloud studies, in which you collect feedback directly from the user or by observing them interact with a product.
Quantitative research is based on the collection and analysis of quantifiable information. It provides data that can be used to define metrics, measure user experience, and make comparisons. Examples of quantitative research methods are surveys, in which you collect feedback in a measurable way, and A/B testing, in which you compare two variants of something to see which one performs better.
Generative (or exploratory) research is aimed at studying a little-known phenomenon in order to find problems worth solving and generate new ideas to spark innovation. Evaluative research is aimed at evaluating or testing something to uncover its strengths and weaknesses, usability issues, and opportunities for improvement. Think of generative research as research conducted to study a problem, while evaluative research is conducted to test a solution.
User Research vs Design Research
Both user research and design research are subsets of UX research, but there is a very subtle difference between the two: the former focuses on the user, the latter on a product. Remember that you don’t need to test a product in order to conduct user research, especially if you are conducting generative research – interviews, focus groups, and observation techniques are often used for this. Likewise, you don’t necessarily need users to conduct design research: the design of a product can be evaluated by means of heuristic evaluation (i.e. by assessing it against a set of guidelines and best practices), benchmarking (i.e. comparing it with other similar products). That said, design research also includes the typical usability testing methods which do require users (and you really cannot conduct UX research without them).
Cross-Sectional vs Longitudinal Studies
Cross-sectional studies are conducted with a set of participants at a given point in time. For the vast majority of usability tests, this means having 5-10 users test a product once, after which the study has ended.
In longitudinal studies, the same user is observed interacting with a product for a longer period of time (usually several weeks). This is helpful to understand changes in user behaviour and discover issues that may not emerge right away in a cross-sectional study.
Between-subjects vs Within-subjects research
These are two types of study design which are especially important to understand when working in academia and conducting scientific experiments. When you are testing something with two variables A and B, you have two options: in a between-subjects setup, you are going to show variable A to half of the participants and B to the other half (that is, you are comparing the two variables between two different groups of participants); in a within-subjects setup, you are going to expose each participant to both variable A and B (that is, you are comparing the two variables “within” each participant).
Between-subjects studies have the advantage of minimising bias – each participant will only see one version of whatever you are testing. The main disadvantage of this type of setup is that it requires a much higher number of participants, and it can be very time consuming. In a within-subject study, you can test with fewer people, but whatever variable is shown first to each participant will affect their perception and feedback of the second one. This is why it is good practice to change the order in which you expose participants to each variable (i.e. some participants will see A then B, while others will see B then A). When it comes to usability tests, this applies also to the order of the tasks you give your participant, as they will expose them to different parts of your user interface.
In-house vs On-site vs Remote Studies
Finally, where you conduct the study is also important. In-house studies are conducted in the “lab” (usually a room that is audio and video recorded). These have the advantage of taking place in a controlled environment but which is also artificial. Some users may not be entirely comfortable or feel they are being tested, or in general behave differently from how they would in a natural environment. On-site tests address this by taking place “in the field”, that is, where the product being tested would normally be used. In many cases this could be a user visit taking place at the user’s home. You can also conduct “guerrilla tests” recruiting random passersby from the streets – this has the advantage of simulating the real experience of using a smartphone in a crowded place, with background noise and other distractions that you wouldn’t find in a research lab.
Remote tests are becoming more and more popular, as they allow a researcher to conduct a study with participants in a separate location (making it especially useful for recruiting international participants). Today there are many remote UX research tools that allow you to conduct moderated and unmoderated studies. Moderated studies are, for example, interviews – for which a remote setup is a no-brainer and has very little disadvantages, as it simplifies logistics and the recruitment process significantly. As for unmoderated studies, the researcher prepares a study in advance, then sends it to a number of participants who will take the study from the comfort of their home, and send the result back to the researcher. The disadvantage here is that the researcher cannot intervene if the user does not understand a question or has any issues with their tasks.