You’ve come across Thematic Analysis according to Braun and Clarke (2006) and are wondering what this qualitative research method is all about?
No problem.
In this article, you’ll learn:
- The 6 steps of Thematic Analysis according to Braun and Clarke.
- The different types of Thematic Analysis you can do.
- The difference between Thematic Analysis and qualitative content analysis.
- The types of research projects for which Thematic Analysis is particularly well-suited.
By the end of this article, you won’t just have added another qualitative method to your toolkit, but you’ll also know when to best employ each one.
Thematic Analysis according to Braun and Clarke (2006)
Thematic analysis is one of the most popular qualitative research methods out there. Since Braun and Clarke published their paper “Using Thematic Analysis in Psychology” in 2006, it has been cited over 150,000 times. Therefore, the method has gained recognition far beyond the realms of psychology and is used across various disciplines.
The reasons for the popularity of thematic analysis are manifold.
Unlike Grounded Theory, it represents a specific method rather than a methodological approach. This means that there are concrete steps for its execution that have been clearly and explicitly defined.
Moreover, thematic analysis still offers a certain flexibility, which is essential for qualitative approaches.
The method has evolved very little since 2006, meaning the guidelines by Braun and Clarke are still highly relevant.
However, since then, the duo has distinguished between three different kinds of thematic analysis. This differentiation arose because the method was sometimes interpreted in ways different from what they originally intended.
The three types are as follows:
Reflexive Thematic Analysis
This is the method as Braun and Clarke envisioned it. It’s based on a constructivist mindset, meaning subjective interpretations of the data are at the forefront.
Positivist Thematic Analysis
In this variant, researchers compute a reliability measure to check for agreement among them. This version follows more of a positivist mindset and isn’t quite what the two originally had in mind.
Thematic Analysis with a Codebook
This third variant is neither one extreme nor the other. It involves working with a codebook, which can contain predefined categories but can also be expanded spontaneously.
What is a “Theme”?
A Theme is either…
…a summary of the content
or
…a central concept that encapsulates the meaning of similar content.
Themes cannot be discovered or found within the content. They have to be generated by you. So never write: “I identified 5 themes…” but instead say “I developed 5 themes…”
6 steps of reflexive thematic analysis as proposed by Braun and Clarke (2006)
What follows are the 6 steps of reflexive thematic analysis as proposed by Braun and Clarke (2006).
#1 Familiarize yourself with the data
First, transcribe your data if you have it only in audio or video format.
Then, read the entire dataset twice from start to finish. This gives you a good overview of all your material. It’s better than starting to evaluate a transcript without knowing the rest.
Try to fully immerse yourself in the situation described in the transcripts. However, always maintain an analytical perspective.
Take notes as you read. You can also take notes right after conducting an interview or while visiting a company on-site if that’s your research context. All your notes are for your personal use; you don’t need to share them later. However, they will assist you in the evaluation later on.
In your notes, jot down your initial reactions. These can be analytical or purely intuitive.
#2 Generate initial codes
Now it’s time to start coding. The codes that emerge at this stage are categories, but not themes yet!
What are categories?
Try to code all your data according to the same schema. That is, find categories on a consistently similar level of abstraction.
An example of a category would be “democratic decision-making within the team.”
Another category on the same level could be “open discussion about the integration of new technologies.”
Two categories that aren’t on the same level might be “hierarchy” (too abstract) or “weekly meetings where personnel decisions are made” (not abstract enough).
With the categories, you can certainly venture a preliminary interpretation, like “democratic decision-making”. This exact phrase wasn’t in the data; it’s something you interpreted.
But what’s the purpose of the categories?
The categories reduce the volume of your data and group your analytical units.
What’s essential for Thematic Analysis as per Braun and Clarke is that you don’t code EVERYTHING. Instead, you should only form categories that are relevant to your research question.
In the data, you’ll find many sections that just aren’t interesting and won’t help answer your research question. You don’t need to code these sections.
The naming of your categories should be chosen such that they precisely describe what’s relevant to your question. A category doesn’t have to consist of just one word; it can be a bit longer (3-6 words).
How to Code?
You can either work digitally with software like NVIVO or use pen and sticky notes. I’m more of a software person. But everyone has their own preferences.
Even while coding, you can and should continue to take notes that you can use later on.
#3 Generate the First Themes
The reflexive Thematic Analysis by Braun and Clarke operates inductively. Your themes should arise exclusively based on your data and, at this point, based on your codes.
Now, group the codes. Which ones are thematically related? This will lead to clusters of categories. Each cluster will then become a theme.
Here, you can also work with mind maps and visually develop the clusters. It’s also possible that within a larger cluster, you have smaller clusters (or “subthemes”). However, try not to make it too complicated.
In the end, having 3 to 6 themes is a good amount to work with.
Avoid Thematic Buckets
The biggest mistake in coding and also in generating themes is the use of so-called “buckets”. A classic bucket includes categories like “Advantages”, “Disadvantages”, “Barriers”, and “Challenges”. It’s crucial to steer clear of these.
#4 Review Your Themes
Once you’ve finalized all the themes, create a final mind map featuring all the themes, potential subthemes, and categories.
Check if everything forms a coherent overall picture and accurately reflects the content of the data. Ask yourself the following questions for each theme:
- Is this more than just a category?
- Does this theme encapsulate multiple categories?
- How does the theme relate to the research question? Are there overlaps between themes?
- Is there sufficient data supporting the theme?
- Is the theme too broad or too specific?
If you encounter issues with these questions, such as overlapping themes, take a step back and rephrase the themes or rearrange the structure.
The Thematic Analysis by Braun and Clarke isn’t a linear process; you can always move forward and backward as needed.
#5 Define and Name Your Themes
Write a detailed description for each theme, comprising 5 to 6 sentences.
Also, finalize the specific designation for each theme.
If you encounter issues while describing or naming, it typically indicates that the theme isn’t distinct enough. In such cases, revert a step or two and reconsider.
#6 Write Down Your Findings
The final step for your Thematic Analysis according to Braun and Clarke involves drafting your report.
In most instances, this will be an academic paper.
Now, you will integrate your findings with existing literature and align the motivation, research question, results, and discussion.
In your methodology section, ensure to cite Braun and Clarke (2006) and explain how you approached your thematic analysis.
In the results section, introduce all the themes at a glance and then delve deeper into each specific theme. Provide quotes from your data that represent each theme.
Certainly, let the quotes speak for themselves. They can even be a bit lengthy. However, simply stringing together quotes isn’t sufficient. Between them, you must jot down your interpretation and establish the connection between the data and the theme.
What’s the difference between Braun and Clarke’s Thematic Analysis and Qualitative Content Analysis?
Answering this question isn’t that straightforward. The approach suggested by Braun and Clarke is simply their perspective on systematically evaluating qualitative data.
With qualitative content analysis, there are different variants too, for example by the German social scientists Mayring or Kuckartz.
A content analysis would probably be more suited if you have a big qualitative data set and would like to count your categories or if you want to develop a codebook for other researchers to use.
The procedures of thematic analysis and inductive content analysis are quite similar and differ by maybe one or two steps and their respective labels.
When selecting your method, consider the target audience for your research. For more interpretative research and an English-speaking audience, choose thematic analysis.
For a more structured approach and some quantification of your qualitative data, choose content analysis.