I would say that thematic analysis (TA) is indeed quite useful and suimple to apply. However, its advantage is also its disadvantage. TA does not go into the complications of reflective research, it does not deal with the complexities of interpretation and the hermeneutic aspects of unraveling meaning and the ongoing process of sense-making. This is wonderful for those who want simple, clear-cut, categorical answers to research questions. It is nevertheless lacking for those who feel the complexities of sense-making are crucial for the validity of the research as a whole. For instance if we understand that categories are not part of the data but are rather a creation of the researcher then we may ponder how we found the themes we "found", or why we labled them the way we did.  Moreover, we may want to understand the difference between "theme" and "category". We might also want to demonstrate how diferent pieces of data may fit two ore more themes as they depend on our manner of construing them.

I might also add that the seemingly simple, and not reflective aspect of the method. may be of great importance when one wants to get published in positivist oriented journals (most journals I would say). 

To sum up, I would recomend the method for beginner researchers, but would urge them and more experienced researchers to take it with a grain of salt, or to the least with a grain of reflexivity.

My tuppence is that I agree that this particular article has its limitations. I've used Liamputtong for guidance in thematic analysis and found it simple, informative and detailed enough to go forward with. Two references are useful:

Liamputtong, P., & Serry, T. (2010). Making Sense of Qualitative Data. In P. Liamputtong (Ed.), Research Methods in Health: Foundations for evidence-based practice (pp. 369-432). South Melbourne: Oxford University Press.

Liamputtong, P. (2013). Qualitative Research Methods (4th ed.). South Melbourne: Oxford University Press.

Cheers. 

I would say that thematic analysis (TA) is indeed quite useful and suimple to apply. However, its advantage is also its disadvantage. TA does not go into the complications of reflective research, it does not deal with the complexities of interpretation and the hermeneutic aspects of unraveling meaning and the ongoing process of sense-making. This is wonderful for those who want simple, clear-cut, categorical answers to research questions. It is nevertheless lacking for those who feel the complexities of sense-making are crucial for the validity of the research as a whole. For instance if we understand that categories are not part of the data but are rather a creation of the researcher then we may ponder how we found the themes we "found", or why we labled them the way we did.  Moreover, we may want to understand the difference between "theme" and "category". We might also want to demonstrate how diferent pieces of data may fit two ore more themes as they depend on our manner of construing them.

I might also add that the seemingly simple, and not reflective aspect of the method. may be of great importance when one wants to get published in positivist oriented journals (most journals I would say). 

To sum up, I would recomend the method for beginner researchers, but would urge them and more experienced researchers to take it with a grain of salt, or to the least with a grain of reflexivity.

I think the Braun and Clarck article is very good in that it covers a lot of the different theories/schools that utilize Thematic Analysis (TA). I agree that it does not go into detail on how to actually do the analysis. One can find more information on the actual steps that need to be taken when looking at some of the "schools of thought" mentioned in the article (i.e. grounded theory). The most important concept for identifying a theme is the idea of "saturation" which is not mentioned in this article. They also do not mention "inter-rater reliability" which cuts down on the the potential subjectivity in the analysis. I am attaching an article by Guest, Bunce, Johnson 2006 on data saturation that students might find helpful. 

I do have to take a moment to defend TA because it is so central to qualitative research in general. [Stepping onto soapbox... feel free to keep walking].  In reality, there can be a lot of subjective interpretation in statistical research as well as TA; both on the front end, by pre-determining which issues are important to consider and during analysis i.e. identifying the strength of correlation between variables from a "fan-shaped" pattern on a scatter plot. I can understand why students might feel insecure about identifying the categories and themes on their own. That is the reason there "inter-rater reliability" is recommended; it helps cut down on the subjectivity and verify with someone else. Ideally, multiple people would be able to come to similar categories. TA has the potential to elucidate what most quantitative cannot: 1) what do the participants themselves see as the issue worth discussing, 2) why is there an issue/ what causes the issue, 3) what do the participants think can be done about the issue/problem. Most often in quantitative research, the topic/ issues have been predetermined by the researchers as they create the survey instrument. The results often leave the second and third questions (about causes and explanations) very limited by preconceived categories. TA allows researchers to try to learn from the participant's own words.  [Steps off soapbox]

For what it is worth, and I am no expert in the field, I think Braun & Clarke's article on Thematic Analysis is a great tool. I think what I got out of the paper is the importance of stating at the outset factors that may influence the creation of themes, for example epistemology, personal experience etc, which in a lot of published papers using this method are sadly missing (most probably due to lack of jornal space). From my own experience of using his method being up front about these things and reflecting on them gave me confidence to make some of those leaps and feel confident in the themes I had constructed 

The article by Braun and Clarke provides a flow charts about stages of analyses, very similar to grounded theory! TA is iterative so you have to go back and reflect on themes that you created and how the jigsaw fits together. It is not as prescriptive as grounded theory and much more flexible but still challenging for novice researchers, based on my experience!

I see that John Ireland's "cynical and jaundiced [view of] hunting for themes" has received 4 down-votes, which probably would not surprise him. Still, I have some similar concerns about Thematic Analysis, since I don't think it addresses the age-old question of how much the results of qualitative analysis originate in the data or in the researcher.

In particular, it seems to me that TA relies on a version of "naive induction" where the themes are there, in the data, waiting to emerge. But I think we all acknowledge that analysts play a major role in influencing what we ultimately "find." Unless we explicitly engage with this issue, I believe that we do leave ourselves open the kind of criticism that John Ireland describes.

The comparison to Grounded Theory is instructive in this regard. With the development of Constructivist Grounded Theory, especially in the work of Kathy Charmaz, there has been much more attention to the role that the researcher plays in driving the research process. So maybe we need a more "constructivist" version of Thematic Analysis?

As much as I can remember, Braun and Clarke clearly stated that in TA, inductive and deductive approaches can be applied. They stated that it depends on whether the research is a theory-driven one or otherwise. They also stated that themes do not just emerge; thus, the analysts should reflect on the analytical procedures so that readers will know how the themes were arrived at. In my on-going doctoral research, I found the article excellent and have recommended it to others. 

We just discussed Braun & Clarke in my graduate seminar on qualitative Data Analysis. As part of that, I had the students construct two "concept maps" -- one each for the strengths and weakness of Thematic Analysis.

The central concept on the strengths side was the flexibility of the method. In particular, it can be used for both inductive and deductive approaches, and it doesn't require strong assumptions about epistemology.

For weaknesses, there were several statements about the overall lack of clarity in conducting the method. In particular, despite the six steps involved, the students felt they were still uncertain about what they would actually do if they were undertaking a Thematic Analysis.

After that discussion, I personally am even more convinced that TA is basically a "generic" form of qualitative induction. In an extreme version, I would say that the six steps amount to:

1. Look at your data thoroughly (immerse yourself)

2. Pay even more careful attention to the data (code it)

3. Try to make sense of the data (search for themes)

4. Check what you've found (review tentative themes)

5. Come to some conclusions (finalize themes)

6. Write it up (write it up)

I don't mean to be as cynical as this undoubtedly sounds, since I think most qualitative analysis has followed a similar pattern over the years. So, if this is what we commonly do, then it is useful indeed to what is now a well-branded label for it. The price we pay, however, is that if a paper simply says they did Thematic Analysis, and then cites Braun & Clarke, then there will still be a great deal of uncertainty about what actually happened during the analysis.

Hello David

A great idea to discuss Braun & Clarke in a seminar with students to see what they think- I might do the same if I get the chance.

I agree that having a 'brand label' can be useful but that simply saying you did Thematic Analysis, and then citing Braun & Clarke is not very informative. I don't think using the label in that way was the authors intent as paper does impress upon the need to explicitly consider and discuss issues such as how people went about analysing their data, or what assumptions informed their analysis so that the finished product contains an account of what was done, and why. If when we are asked to review papers that claim to have used Braun & Clarke's brand of thematic analysis, it may be worth ensuring the authors have made explicit some of these issues.   

I guess as in most things in life there is a trade off. In this case between having a flexible method accesable to researchers with results that may be followed by lay readers and the weakness of not giving specific direction as to higher level analysis that some lay reader may not follow.

David, I totally agree with your observation. TA is a generic approach to content analyses. Moreover, one never truly follows the stesps as perscribed. There is alway a non-methodical moment(s) in inteerpretation. There are always epipanies and insights that come from the reflective stages (in the shower, in bed, etc.). The power of memmo making is no less important than methodical analysis. Moreover, as Rennie (2012) demostrates, TA is in actuality a translation of hermeneutic process into a practice. As he calls it "methodical hermeneutics". 

TA simplifies comparative qualitative research - that is both its strength and weakness.

You all are right about thematic analysis but it takes too much time to conduct and complete it as comparison with other analysis.

On the whole I think TA is what you make it. Students may start off with a simpler content analysis that they can develop to see the connections etc. As for themes, that surely depends on what you are researching? Themes might be theory-driven or inductive or even abductive. How about calling it realist TA? Taking what is there as reality rather than trying to interpret what might/not be there. This is an issue with most qualitative methods and I have read articles and wondered how on earth the researcher has come to that conclusion based on the evidence! 

An imperfect but a necessary research method that seeks to discover ideal types. Yes, these are deliberately exaggerated caricatures, but in social scientific research we need to use them in order to escape what Max Weber famously described as the ´realm of the vaguely felt´. Paul

I agree with the opinion David L Morgan. Considering high citation, Thematic Analysis is a good, universal, applied research method. Best wishes

I invite you to the discussion.

I concur with all the positive/favourable comments on Thematic Analysis (TA). Nevertheless, an understanding of TA’s advantages/disadvantages/pitfalls should be able to help in an effective/efficient application of TA (see: Braun and Clarke, 2006, pp. 94-95; Javadi and Zarea, 2016, pp. 38-39; Nowell et al, 2017, p. 2; Riger and Sigurvinsdottir, 2016, p. 36).

• Braun, V. and Clarke, V. (2006) Using thematic analysis in psychology, Qualitative Research in Psychology, 3, 2, pp. 77-101.
• Javadi, M. and Zarea, K. (2016) Understanding Thematic Analysis and its Pitfall, Journal of Client Care, 1, 1, pp. 33-39.
• Nowell, L. S., Norris, J. M., White, D. E. and Moules, N. J. (2017) Thematic Analysis: Striving to Meet the Trustworthiness Criteria, International Journal of Qualitative Methods, 16, 1, pp. 1-13.
• Riger, S. and Sigurvinsdottir, R. (2016) Thematic Analysis, in Jason, L.A. and Glenwick, D.S. (eds.) Handbook of Methodological Approaches to Community-Based Research: Qualitative, Quantitative, and Mixed Methods. New York, NY: Oxford University Press, Chapter 4, pp. 33-41.