Analysing Qualitative User Data

A quick how to guide on qualitative data analysis

Most UXers I talk to love interviewing users but dread what comes next; analysing qualitative data! This can be especially daunting for designers and junior UX researchers. In this article, we'll explore the process of analysing qualitative user data, including different types of qualitative data analysis, and how to choose the right analysis method.

Qualitative data analysis is the process of examining non-numerical and unstructured data, such as interviews, focus groups, or observation notes, to identify patterns, themes, and insights. Unlike quantitative data, which can be easily quantified and analysed using statistical methods, qualitative data requires a more nuanced approach. Analysing qualitative data typically involves manual coding, in which the researcher reads through the data and identifies key themes or concepts. These themes are then used to identify patterns and insights.

Benefits of Qualitative Coding

Qualitative coding can provide several benefits, including:

1. Helps to identify new insights: By categorising the data and identifying patterns, UX professionals can uncover new insights that may not have been apparent before.

2. Increases the validity of the study: By systematically assigning codes to segments of data and organising them into categories, it is possible to identify patterns and themes in the data that may not have been apparent otherwise. This helps reduce the subjectivity of the analysis and ensures that the findings are based on the data itself rather than the our biases or preconceptions

3. In-depth insights: Qualitative data can provide rich insights into users' behaviours, needs, and motivations that cannot be captured by quantitative data alone.

4. Decreases bias: Coding qualitative data helps us be aware of potential biases in the way we analyse the data. Even though we cannot fully get remove subjectivity from the equation, by qualitative coding we decrease it.

5. Flexibility: Qualitative coding allows for flexibility in data analysis. Researchers can adapt their approach as they uncover new insights or unexpected patterns in the data.

6. Transparency: It enables other researchers to methodically and systematically review your analysis.

7. Contextual understanding: Qualitative data can provide a more nuanced understanding of the context in which users are operating, including social and cultural factors that may influence their behaviour.

   

   This is Midjourney’s interpretation of a UX Researcher analysing qual data…

How do you Code Qualitative Data?

There are several steps involved in coding qualitative data:

1. Transcribe your data: If you are working with interviews or focus groups, you will need to transcribe the audio or video recordings into text format. Thankfully, most researchers have access to transcription tools these days and this step is done by AI! A few years ago, it was common for researchers to manually transcribe data! Can you imagine how time consuming that can be? Please note that even though transcription tools have improved, it’s always worth checking the transcript once it’s generated to avoid embarrassment (I have many horror stories)...

2. Familiarise yourself with the data: Read through the transcripts or observation notes to get a sense of the data. This step might take some time depending on the amount of data we have.

3. Identify themes: Start identifying key themes or concepts in the data. This can be done manually (e.g. using sticky notes or software like Miro) or using software tools like NVivo, which employ AI to help identify common themes.

4. Code the data: The fun begins! At this step we categorise each piece of data into one or more themes. This is an area that could be supported by AI but before you start using ChatGPT check your organisation’s rules! Also, be warned that most AI tools are not very good at this… yet.

5. Review and refine your codes: Review your codes to ensure they are consistent and make sense in the context of the data. This can take a while and it’s normal to change some of your initial codes at this stage.

Types of Qualitative Data Analysis

This was a very quick overview of qualitative data coding! There is more than one way to conduct this analysis and each has its own strengths and weaknesses. Some of the most common types of qualitative data analysis include:

1. Content analysis: Content analysis involves analysing the frequency and distribution of themes in the data by turning non-numerical data, such as words or descriptions, into numerical data. This is done by breaking down large amounts of text into smaller, meaningful segments and assigning codes to each one. These codes are then grouped into categories, and the frequency of each category is calculated to identify themes and patterns in the data. However, content analysis can be limited in its ability to provide a deep understanding of the context in which the data was collected and can be time-consuming if done manually. This approach is commonly used when we analyse survey open-text questions (e.g., NPS) and customer reviews left on third party websites.

2. Thematic analysis: Thematic analysis involves identifying and describing themes in the data. Thematic analysis can be done inductively, starting with the data and identifying themes, or deductively, starting with a pre-defined set of themes. Thematic analysis can be useful for providing a detailed understanding of the context in which the data was collected. It is flexible and can be applied to a wide range of data types, including interviews, focus groups, and surveys. In fact, it’s the most commonly used method by UX researchers and it can be done in UXR tools like Dovetail and Condens.

3. Discourse analysis: Discourse analysis involves analysing the language and communication patterns used by participants. This can be useful for understanding the ways in which participants construct meaning and the social and cultural factors that influence their behaviour. However, discourse analysis can be time-consuming and requires a high level of expertise.

4. Grounded theory: Grounded theory is a method of analysing data that involves developing a theory or explanation for the data based on the patterns and themes that emerge from the data. This method is useful when the research question requires a more exploratory approach to data analysis. Grounded theory is time-consuming, and it requires a high level of expertise, but it can lead to rich and nuanced insights.

How to Choose the Right Analysis Method

Choosing the right analysis method depends on several factors, including the research question, the type of data collected, and the available resources. To choose the right analysis method, consider the following steps:

1. Define your research question: What do you want to learn from the data? This is the most important thing you need to establish before choosing a method. You need a clear research question in order to choose an appropriate analysis method. In fact, I would suggest thinking about your analysis when you start working on your research plan not after you collect your data!

2. Identify the type of data you have: What type of qualitative data have you collected? Is it interview data, focus group data, or observation data? This will help you choose an analysis method that is most appropriate for your data.

3. Consider your available resources: Do you have access to software tools for qualitative analysis? Do you have the expertise to conduct a specific type of analysis? How much time do you have to complete the project? Consider the resources you have available to help you choose an analysis method that is feasible for your project. There’s no point spending a long time analysing data with the best possible method if by the time you finish the analysis the data are no longer relevant!

Analysing qualitative user data is a crucial step in the UX research process that allows us to gain in-depth insights into users’ behaviours, needs, and motivations. Qualitative coding is an essential part of this process, as it helps us to organise the data into categories and identify patterns and themes. This article only scratches the surface but, hopefully, it gives you an overview of the methods available to you.

If you want to learn more check out the following:

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative research in psychology, 3(2), 77-101.

Saldaña, J. (2015). The coding manual for qualitative researchers. Sage Publications.

Miles, M. B., Huberman, A. M., & Saldaña, J. (2014). Qualitative data analysis: A methods sourcebook. Sage publications.