Braun & Clarke state thematic analysis is primarily inductive, but in a recent primer, spoke about using a deductive approach. With that said, thinking about the first thematic analysis (not their latest reflexive), one can be hybrid if one explains their methods. Susan Sibley at MIT applied deductive and inductive to grounded theory, cycling back and forth; the thought, expounded by others as well is that all research includes the researcher to varying degrees.

Sometimes researchers think Braun and Clarke are the only game in town. Read this seminal article:

Fereday, J., & Muir-Cochrane, E. (2006). Demonstrating rigor using thematic analysis: A hybrid approach of inductive and deductive coding and theme development. International Journal of Qualitative Methods, 5(1), 80-92.

No, reflexive thematic analysis does not allow you to have prior hypotheses. But there are other "hybrid" versions of thematic analysis that allow you to begin with a deductively given set of codes and then include other inductive codes during the analysis process.

As David L Morgan has previously said, Braun and Clark's newer book from 2022 is now the authoritative source on Reflexive Thematic Analysis. It should clarify most of your questions!

Keep in mind that Braun and Clarke have explicitly allowed deductive approaches. My suggestion: Move beyond being either/or and recognize all data is data. (Page 4 of their article SPECIFICALLY talks about “the continuum.”)

Braun, V., & Clarke, V. (2021). One size fits all? What counts as quality practice in (reflexive) thematic analysis?. Qualitative research in psychology, 18(3), 328-352.

David L Morgan Does this also include the previous version of Braun and Clarke (2006)? Because I reviewed many articles and dissertations, and they employed this version inductively and deductively.

Many thanks

Helen Ali see my reference. Braun and Clarke, even in there 2022 book, absolutely endorse deduction as a possibility.

@david, I can see you have a hands on experience with thematic analysis. Kodus !

Hi Helen,

As David Coker said, there is no single thematic analysis approach. Braun and Clarke's is one of the key approaches out there.

From small q to Big Q, we have a coding reliability thematic analysis approach at one extreme pole (e.g., Boyatzis, R. E. (1998). Transforming qualitative information: Thematic analysis and code development; Fereday, J., & Muir-Cochrane, E. (2006). Demonstrating rigor using thematic analysis: A hybrid approach of inductive and deductive coding and theme), and Reflexive thematic analysis by Braun and Clarke at the other. The former is underlined by the small q approach driven by realism, seeing themes as a topic of summary and emphasising controlling bias and intercoder agreement. The latter is underlined by the Big Q approach driven by relativism, seeing themes as an output of the analytic process, a shared pattern of meaning.

Between these two poles, there is an in-between approach such as Template Analysis by King et al. (2018), or Gale et al. (2013). Framework Method analysis. These approaches tend to integrate the principles of both small q and Big Q, driven more by pragmatic principles. In this approach, there could be priori themes but you are allowed to generate themes as data emerges.

For your need, I think you can choose the in-between approach.

Hope this helps.

I have a different interpretation of RTA by Braun and Clarke from David Morgan's. I don't think in their Reflexive Thematic Analysis, prior hypotheses are not allowed.

They explicitly write this in their 2022 book:

“Coding doesn’t have to be either inductive or deductive, although it can be; your analysis can have elements of both orientations. What’s most important is that your coding orientation fits your purpose.”

And

“In reflexive TA, a deductive orientation refers to a more researcher- or theory-driven approach, where the dataset provides the foundation for coding and theme development, but the research questions asked – and thus the codes developed – reflect theoretical or conceptual ideas the researcher seeks to understand through the dataset” (emphasis is mine).

And this:

“No matter whether your analysis is more inductive or deductive, or indeed a blend of both, theory is still important – it’s what gives reflexive TA, and indeed all forms of qualitative analysis, both its foundation, and its analytic power ”

As such, I think there is a degree of flexivity in their RTA in regard to being deductive/inductive.

Best,

Kien Nguyen-Trung, like Braun and Clarke 2021, these authors specifically speak about the use of deductive and inductive. But what if they didn't! Researchers must be pragmatically flexible to apply methodologies and methods to their studies as long as the methods are systematic, rigorous, and coherent. One last point: We are all, to varying degrees, deductive. Our inductive ideas must come from our lived experiences.

Morse, J. M., & Mitcham, C. (2002). Exploring qualitatively-derived concepts: Inductive—deductive pitfalls. International Journal of Qualitative Methods, 1(4), 28-35.

I certainly agree with the claim that one can never be free from prior assumptions. I have made this point in an article about Grounded Theory and it would also apply to Reflexive Thematic Analysis.

In terms of deductive analysis in Reflexive Thematic Analysis, I can't see Braun and Clarke allowing a prior hypothesis to guide any part of their 6-step framework.