Thread by @drvicclarke

1/ A thread with some tips on writing qualitative research dissertations - esp those using thematic analysis - including common problems to avoid (prompted by marking student projects). First tip - as @ginnybraun & I always say check local requirements! Broadly speaking, there

2/ are two styles of qual research reporting: 1) "add qual and stir" - default quant conventions slightly tweaked for qual: finding & filling the "gap" introduction & rationale, methodological critique of existing studies, separate "results" & discussion... 2) qual centric. The

3/ latter is far less well understood & recognised - I've had reviewers/editors insist on me reworking qual centric reports into something more conventional, examiners do the same to my students. So check what is required in your context. If a marker/examiner doesn't "get it"

4/ they may not like it! So for qual centric reporting if you're not positivist/realist in your research what does that mean for your introduction? Don't be a positivist/realist in the intro! Think carefully about framing & footing of quanty research - don't present quanty

5/ "findings" as facts but as claims. Don't engage in methodological critique based on quanty assumptions (odd!). Understand your intro as contextualisation & rationale for your study rather than simply a lit review & finding a "gap". Don't describe single study after study...

6/ try to overview & synthesise if discussing a body of lit. The best intros in my view make an argument for the research & frame it within relevant wider contexts, they flow beautifully - I always know why I'm being told something & where things are going. There's no jumping

7/ around to seemingly unrelated topics... Work out if your intro is the classic inverted triangle - start broader get more specific - or what I call "stacking boxes" - you have several different topics to discuss that aren't easily classified as broader/more specific - they

8/ are all roughly the same - so how do you order/stack the boxes? It's a judgement call - see what works best as you write. Definitely have signposting/an overview at the start to help the reader understand where things are going. Try to have linking sentences between topics/

9/ sections to signal transitions. We've been there now we're going here... End the intro with your research questions/aims. These should make sense/not be a surprise given the context you have presented. The reader should almost expect your research questions! Don't present

8/ hypotheses or discuss as a student was asked to recently what you expect to find! Don't formulate your research question in terms of the impact of X on Y - this is essentially quanty hypotheses in a bad disguise! You need a qual specific formulated question. Check out my and

11/ @ginnybraun's qual textbook for advice on qual specific research questions - uk.sagepub.com/en-gb/eur/succ…

12/ This paper has guidance on research questions specific to thematic analysis - uwe-repository.worktribe.com/output/7164974…

13/ Check your dissertation title to make sure it isn't implicitly quanty framed too. I can't tell you how many times I've got to the research questions expecting a quant study! There's been nothing in the title or intro to lead me to expect a qual study! In the method/ology make

14/ make sure you discuss your philosophical assumptions even if only briefly & esp. if using TA as it's a theoretically flexible method(ish) not a theoretically bound methodology. If you're an insider researcher (a member of the grp you are researching) Take a dive into the

15/ insider researcher lit - this paper by @drnikkihayfield & Caroline Huxley is a great place to start: tandfonline.com/doi/abs/10.108…

16/ In general look for methodological lit related to your design choices. Don't say I had to do video call ints because of Covid... draw on the video call methodological lit to provide a more robust rationale & discussion. Great chapter by @phannadr - cambridge.org/core/books/col…

17/ I often encounter students being told to cite more than 1 source on thematic analysis - peril!! Most aren't aware of the diversity in TA - don't mix & match incompatible approaches. Braun & Clarke and Boyatzis = NOPE! B&C and Joffe/Barbour/Guest = also NOPE! Check out the

18/ Conceptual & Design Thinking for Thematic Analysis paper linked to earlier for guidance. Do you have to explain why you didn't use other method/ologies to explain your choice of analytic method? Some like this but I think it's odd... I've seen utterly implausible alternatives

19/ presented & very poorly explained! Check out this paper to help you develop a robust rationale for your selected method/ology (free to read at the moment) - onlinelibrary.wiley.com/doi/epdf/10.10…

20/ If you used B&C TA don't just provide a generic description of the 6 phases - tell the reader what you actually did, how you engaged with the process! Avoid quanty style headings in the method/ology (Materials? Nope) & generally rationale (why) before procedure (how). In the

21/ Analysis (not Findings if you can avoid this discovery oriented heading) start with an overview (simple list of themes, a table or thematic map) - I'm often befuddled as to what/where the themes are! Don't confuse topic summaries with themes if doing B&C TA - see Conceptual

22/ and Design Thinking... paper linked to earlier for a discussion. Avoid one word theme names as they aren't very informative (& suggest topic summaries) & keep in mind that themes are complex, rich & multifaceted stories - you can't report loads of them & do them justice.

23/ We generally advise against loads of theme levels & subthemes - the latter are useful to highlight a facet of the central concept. If you want lots of themes/theme levels try template or framework analysis as these are designed for this. In qual centric an integrated "results

24/ & discussion" is all good but this seems to be something that more mainstream/positivist folks really struggle with... In a dissertation with an integrated R&D you still need a general discussion where you reflect on the study & look forward to future research. Things to

25/ avoid here - noting you *may* have influenced the analysis because of your positioning. Take it from me - you did! Try to reflect on *how* you did. Don't bemoan the lack of generalisability of your small sample - gaaaahhhh! You're evaluating qual here using quant standards

26/ And you're implicitly invoking a quant conceptualisation of statistical generalisability. There are qually forms - discuss these instead! Check out this fantastic paper by @BrettSmithProf - tandfonline.com/doi/abs/10.108…

27/ When you make suggestions for future research don't turn on the "random ideas generator"!! The suggestions should arise from your research - intriguing "findings" that need to be explored further, addressing the limitations of your study. If suggesting research needs to be

28/ conducted with other groups of people - try to provide an evidenced based discussion explaining why things might be different (and similar) for these groups as we do in this open access paper on gay fathers: tandfonline.com/doi/full/10.10…