A Comprehensive Guide to Thematic Analysis in Qualitative Research
What is Qualitative Data?
What do all the methods above have in common? They result in loads of qualitative data.
If you're not new here, you've heard us mention qualitative data many times already. Qualitative data is non-numeric data that is collected in the form of words, images, or sound bites. Qual data is often used to understand people's experiences, perspectives, and motivations, and is often collected and sorted by UX Researchers to better understand the company's users.
Qualitative data is subjective and often in response to open-ended questions, and is typically analyzed through methods such as thematic analysis, content analysis, and discourse analysis. In this resource we'll be focusing specifically on how to conduct an effective thematic analysis from scratch!
Qualitative data is the sister of quantitative data, which is data that is collected in the form of numbers and can be analyzed using statistical methods. Qualitative and quantitative data are often used together in mixed methods research, which combines both types of data to gain a more comprehensive understanding of a research question.
UX Research Methods
There are many different types of UX research methods that can be used to gather insights about user behavior and attitudes. Some common UX research methods include:
- Interviews: One-on-one conversations with users to gather detailed information about their experiences, needs, and preferences.
- Surveys: Online or paper-based questionnaires that can be used to gather large amounts of data from a broad group of users.
- Focus groups: Group discussions with a moderated discussion to explore user attitudes and behaviors.
- User testing: Observing users as they interact with a product or service to identify problems and gather feedback.
- Ethnographic research: Observing and interacting with users in their natural environments to gain a deep understanding of their behaviors and motivations.
- Card sorting: A technique used to understand how users categorize and organize information.
- Tree testing: A method used to evaluate the effectiveness of a website's navigation structure.
- Heuristic evaluation: A method used to identify usability issues by having experts review a product and identify potential problems.
- Expert review: Gathering feedback from industry experts on a product or service to identify potential issues and areas for improvement.
Introduction to Thematic Analysis of Qualitative Data
Thematic analysis is a popular way of analyzing qualitative data, like transcripts or interview responses, by identifying and analyzing recurring themes (hence the name!).
This method often follows a six-step process, which includes getting familiar with the data, sorting and coding the data, generating your various themes, reviewing and editing these themes, defining and naming the themes, and writing up the results to present. This process can help researchers avoid confirmation bias in their analysis. Thematic analysis was developed for psychology research, but it can be used in many different types of research and is especially prevalent in the UX research profession.
When to Use Thematic Analysis
Thematic analysis is a useful method for analyzing qualitative data when you are interested in understanding the underlying themes and patterns in the data. Some situations in which thematic analysis might be appropriate include:
- When you have a large amount of qualitative data, such as transcripts from interviews or focus groups.
- When you want to understand people's experiences, perspectives, or motivations in depth.
- When you want to identify patterns or themes that emerge from the data.
- When you want to explore complex and open-ended research questions.
- When you are interested in understanding how people make sense of their experiences and the world around them.
Some UX research specific questions that could be a good fit for thematic analysis are:
- How do users think about their experiences with a particular product, service or company?
- What are the common challenges that a user might encounter when using a product or service, and how do they overcome them?
- How do users make sense of the navigation of a website or app?
- What are the key drivers of user satisfaction or dissatisfaction with a product or service?
- How do users' experiences with a product or service compare with their expectations?
It is important to keep in mind that thematic analysis is just one of many methods for analyzing qualitative data, and it may not be the most appropriate method for every research question or situation.
A key part of a UX researcher's role is being aware of the most appropriate research method to use based on the problem the company is trying to solve and the constraints of the company's research practice.
Types of Thematic Analysis
There are two primary types of thematic analysis, called inductive and deductive approaches.
An inductive approach involves going into the study blind, and allowing the results of the data-capture to guide and shape the analysis and theming. Think of it like induction heating-- the data heats your results! (OK, we get it, that was a bad joke. But you won't forget now!) An example of an inductive approach would be parachuting onto a client without knowing much about their website, and discovering the checkout was difficult to use by the amount of people who brought it up. An easy theme!
On the flip-side, a deductive approach involves attacking the data with some preconceived notions you expect to find in the qualitative data, based on a theory. For example, if you think your company's website navigation is hard to use because the text is too small, you may find yourself looking for themes like "small text" or "difficult navigation." We don't have a joke for this one, but we tried.
To get even more nitty-gritty, there are two additional types of thematic analysis called semantic and latent thematic analysis. These are more advanced, but we'll throw them here for good measure.
Semantic thematic analysis involves identifying themes in the data by analyzing the exact wording of the comments made used by participants. Latent thematic analysis involves identifying themes in the data by analyzing the underlying meanings and actions that were taken, but perhaps not necessarily stated by study participants. Both of these methods can be used in user research, though latent analysis is more popular because users often say different things than what they actually do.
Steps in Conducting a Thematic Analysis
Let's jump in! As mentioned before, there are 6 steps to completing a thematic analysis.
Step One: get familiar with your data!
This might seem obvious, but sometimes it's hard to know when to start. This might take the form of listening to the audio interviews or unmoderated studies, or reading the notes taken during a moderated interview. It's important to know the overall ideas of what you're dealing with to effectively theme your study. While you're doing this, pay attention to some big picture themes you can use in step two when you code your data.
Break out key ideas from each participant. This might take the form of summarized answers for each question response, or a written review of actions taken for each task given. Just make sure to standardize it across participants.
Step Two: sort & code the data.
Now that you have your standardized notes across your participants, it's time to sort and code the collected qualitative data! Think of the themes from before when you were taking your notes. Think of these codes like metaphorical buckets, and start sorting! Every comment that fits a theme in a box, put it there.
Back to our navigation example: some codes could be "small text" or "hard to use." We could put a participant action of "squinting" into the bucket for "small text," or a comment from another mentioning they had trouble finding "tents" in "hard to use."
Step Three: break the codes into themes!
Try to think of each theme as a makeup of three or more codes. For the navigation example, we could put both "small text," and "hard to use" into a theme of "Difficult Navigation."
Step Four: review and name your themes.
Now is the time to clean up the data. Are all your themes relevant to the problem you're trying to solve? Are all the themes coherent and straightforward? Are you comfortable defending your theme choices to teammates? These are all great questions to ask yourself in this stage.
Step Five: Present!!
To have a cohesive presentation of your thematic analysis, you'll need to include an introduction that explains the user problem you were trying to identify and the method you took to study it. Use the terminology from beginning of this resource to identify your research method. Usually for something like this, it will be a user survey or interview.
You also need to include how you analyzed your participant data (inductive, deductive, latent or semantic) to identify your codes and themes. In the meaty section of your presentation, describe each theme and give quotations and user actions from the data to support your points.
Step Six: Insights and Recommendations
Your conclusion should not stop at your presentation of your findings. The best user researchers are valuable for both their insights and recommendations.
Since UX researchers spend so much time with participants, they have indispensable knowledge about the best way to do things that make life easy for the company's users. Don't keep this information to yourself!
On the final 1-3 slides of your presentation, state the "Next Steps & Recommendations" that you'd like your team and leadership to follow up on. These recommendations could include things like additional qualitative or quantitative studies, UX changes to make or test, or a copy change to make the experience clearer for readers. Your ultimate job is to create the best user experience, and you made it this far-- you got this!
And there you have it! That's everything you need to complete a thematic analysis of qualitative data to identify potential solutions or key concepts for a particular user problem.
But don't stop there! We recommend using these principles in the wild to conduct research of your own. Identify a question or potential problem you'd like to analyze on one of your favorite sites. Use a service like Sprig to come up with non-bias questions to ask friends and family to try and gather your own qualitative data. Next, complete and document yourself completing the 6-step analysis process. What do you discover? Be prepared to share on interviews-- hiring managers love to see initiative! Good luck.
Caulfield, J. (2022, November 25). How to Do Thematic Analysis | Step-by-Step Guide & Examples. Scribbr. https://www.scribbr.com/methodology/thematic-analysis/