6.2. Qualitative Analysis and Thematic Coding

Review all transcripts, observation notes, and open-ended survey responses.

The first step in qualitative analysis involves gathering all the relevant data from various sources such as transcripts from user interviews, notes taken during observations, and responses from open-ended survey questions. By compiling this data, you create a comprehensive set of information that reflects the user experience.

Imagine you are a detective trying to solve a mystery. You’ll collect different pieces of evidence like witness statements, forensic reports, and security footage. Similarly, in qualitative analysis, these data points help paint a complete picture of the user experience.

Conduct open coding by assigning initial labels to discrete ideas (e.g., 'unclear icon,' 'slow loading').

Open coding is the initial process of identifying key concepts or ideas from your collected data. You look through the transcripts and notes and label discrete ideas or observations to categorize the data. For instance, if many users mention confusion with an icon, you would label this feedback as 'unclear icon.' This helps break down the information into manageable pieces for further analysis.

Think of this process like a gardener sorting through plants. Each unique idea is like a different type of flower, and the gardener labels each type to better understand the variety they have. This helps when deciding what to plant next or how to care for each type.

Progress to axial coding by grouping labels into broader categories ('navigation issues,' 'system feedback').

Axial coding involves taking the initial labels created during open coding and organizing them into larger themes or categories. For example, all the labels concerning user navigation confusion may be grouped under the broader category 'navigation issues.' This helps to identify patterns and relationships between different pieces of feedback, making it easier to understand user experiences as a whole.

Imagine you are sorting laundry. After you’ve collected all the clothes, you group them by color—whites, darks, and colors. Each group helps you better manage the laundry process, just like axial coding helps manage the qualitative data.

Finally, use selective coding to identify core themes that explain the majority of user experiences.

Once you have categorized the data through open and axial coding, selective coding focuses on identifying the main themes that emerge from the data. This step synthesizes all the analysis and finds the central ideas that represent user experiences most significantly. These core themes should encapsulate the essence of users’ feedback and help you understand their overall needs and behaviors.

Think of this as putting together the pieces of a puzzle. After you’ve grouped pieces by color and shape, selective coding helps you find the pieces that are essential to completing the picture. In user studies, these themes complete your understanding of the user journey.

Create an affinity diagram on a whiteboard or digital canvas: cluster similar codes spatially and draw connections to illustrate relationships between themes (e.g., navigation difficulties linked to information architecture inconsistencies).

An affinity diagram is a visual tool that helps organize and represent your qualitative analysis. By clustering similar codes together, you can see relationships and patterns between different themes. This method allows you to illustrate how various issues, like navigation difficulties, might stem from broader problems in the design, such as information architecture. This visual arrangement aids in the analysis and presentation of findings.

Imagine setting up a family calendar. You put all the birthdays, anniversaries, and holidays in one place and use colors or symbols to show relationships. Each event can be linked, helping you see when you might have busy months or when family gatherings are likely to happen. Similarly, an affinity diagram helps you visually organize and relate different user feedback themes.

--