Building Themes: Learning to See the Bigger Picture

This week, I engaged in thematic analysis by transforming codes into themes and producing a thematic map that demonstrated the connection between them.  It helped me see how individual comments could come together to tell a larger story. Instead of just asking, “What is this person saying?” I began asking, “What does this tell me about the experience I am studying?”

The hardest part was deciding which ideas belonged together. Some codes seemed related at first, but a closer look showed that they described different experiences. Other codes used different words but shared a similar meaning. I had to slow down, review the examples, and think about what connected them. It was easy to use broad labels like “support” or “challenges,” but those labels did not always explain what participants were experiencing.

I also wanted the themes to feel organized right away. My first attempt looked more like a list of topics than an explanation of what the data meant. Creating the thematic map made the differences clear to me. As I joined themes, and subthemes, I noted where there was overlap and where a theme needed to be more focused. The map assisted me in verifying my reasoning rather than merely showing the end results.

I was surprised by how much I had learned when I analyzed the data. An idea that seemed important became less important at times. When I considered a small detail with other responses, it could become more meaningful. I understood I didn’t have to stay with my first ideas after using this visual artist.

Changing a theme did not mean my earlier work was wasted. It meant I was gaining a clearer understanding of the data.

This week’s work connects to my research interests in online learning and how AI tools might support student wellbeing. I want to understand more than whether students find a tool useful. I also want to learn how they describe support, trust, and stress, along with what makes them willing or unwilling to use the tool. Thematic analysis can help me explore those experiences without assuming all students have the same views about the technology.

Moving forward I will be giving myself more time to improve and revise my themes. I also will be using thematic maps during the process instead of waiting to the end. I know that this will help me think through the process and keep my findings better connected.

From Comparing Codes to Understanding Meanings

Image credit from: “Thematic Analysis: A Step-by-Step Guide,” by S. McLeod, 2026, Simply Psychology (https://www.simplypsychology.org/thematic-analysis.html).

During this week, I practiced moving from CCA to thematic analysis. The CCA phase involved comparison of items of data and a decision on how this will fit together. I analyzed similarities and differences, checked for existing codes, and revised categories when not a good fit. Using thematic analysis, I started looking closely at the larger meaning behind those patterns. I retained the practice of comparing responses. However, the focus shifted; instead of sorting information, I explained what participants’ experiences had in common. The hardest part was understanding the difference between a topic and a theme. Words like “stress,” “technology,” and “support” describe what participants discussed, but they do not explain their experiences. Dr. Johnson made this clearer to me in this evenings session, so now I have a better understanding. A theme such as “Looking for reassurance when online learning feels lonely” says more about what those experiences mean. This was challenging because grouping similar comments felt easier than explaining the connection between them.

I came to realize that my perspective shapes the themes I develop, so I need to be conscious of the assumptions I bring to the data.

I was glad to discover this week’s work which connects nicely to my research interests around AI-supported cognitive behavioral strategies and mindfulness in online learning. Students may describe using AI to manage stress, plan assignments, or feel supported. I could examine the relationship between these three things rather than seeing them as isolated issues. This effectively centers the broader need for emotional support around the management of one’s school responsibilities.

What feels different is that I am stepping back from individual comments and looking at the bigger picture. I am asking, “What are these students telling me about their experiences?” At the same time, I want to make sure my themes reflect their words. In the future, I plan to make notes explaining my code and how my ideas change. My focus is to point to the individual experience and then to meaningful patterns without losing dimension.

What I Learned About Constant Comparative Analysis: When to Use It, Its Challenges, and Its Connection to My Research

This week, I learned more about constant comparative analysis, or CCA. Before this week, I understood coding as reading through data and giving labels to important ideas. What I did not fully understand was how much comparing is involved. With CCA, you are not just coding one interview or response at a time. You are going back and forth between responses and asking whether participants are saying similar things, describing different experiences, or bringing up something that does not fit the categories you already have.

I think I would recommend CCA to another doctoral student if they are trying to understand an experience or process from the participants’ point of view. It seems useful when you do not want to make too many assumptions before looking at the data. For example, if I were studying online students and stress, I might expect students to talk about workload or time management. But after comparing responses, I may find that students are more concerned about feeling isolated, not knowing what the instructor expects, or having trouble separating school from home life. CCA gives the researcher room to notice those things instead of only looking for what they expected to find.

At the same time, I would not use CCA for every qualitative study. If the goal is to summarize what people say about a topic, then a simpler thematic analysis might be enough. I also would not choose it if I already had a strict set of categories that I needed to use. CCA seems to work better when the researcher is willing to change the codes and categories as they learn more from the data.

The hard part for me was realizing that the comparison does not stop after the first round of coding. You may create a code early on and then later realize it needs to be split into two codes, combined with another one, or changed completely. That could be frustrating because it feels like going backward. Still, I can see why it matters. If the researcher does not keep comparing the data, it would be easy to miss important differences between participants.

This relates to my own research in online education and student well-being. I can see how CCA could help me see how those experiences relate to one another instead of considering them as a separate topic. For example, I may find that students who feel disconnected from classmates also describe more stress and less motivation. This kind of pattern may help us to understand student needs in online learning environments.