,
17 tweets,
4 min read
Read on Twitter
1/ A third & final thread on writing a discussion section. @ipanalysis did a great thread on discussions on 29 March - so check that out too. We probably say something diff things - which highlights there's no (secret) rule book for, or 1 right way of, writing qual research
2/ so you need to work out what works for your research. You have the option of a separate results & discussion or an integrated R&D + a short 'overall discussion' (where you evaluate the study, make suggestions for future research - the discussion 'classics' etc). Some think
3/ integrated is best - but I think it depends on the nature of yr research, approaches used etc. A rule of thumb that can help you decide is 'am I making lots of micro connections with existing research & theory?' - does this small analytic point resonate with existing lit?
4/ If yes then integrated is prob a good idea, to avoid basically repeating the analysis in the discussion. If the connections are broader (with connections across rather than within themes) then separate R & D might be more effective. I'm not a fan of the theme-by-theme...
5/ contextualisation in relation to the lit in the discussion. It can be a bit BORING! You've just read my analysis, well, here it is again with some lit thrown in for good measure... Try to contextualise across themes and draw out broader insights, implications & contributions
6/ When it comes to the study evaluation - obvs you need to reflect on the limitations of your research, but if you say the 'findings are generalisable because of the small sample' I will cry :( Quant researchers don't bemoan the lack of subjective sense-making in their research
7/ So evaluate yr qual research on its own terms & reflect on its limitations as qual research. Qual research can still be generalisable just not in a statistical sense. Check out this fab paper by @BrettSmithProf for more on this: tandfonline.com/doi/abs/10.108…
8/ Also if you're arguing that because of the limitations of the sample, things might be different for different groups - can you substantiate this speculation in any way? Is there any evidence that might point to or suggest there are differences? It can be helpful to think about
9/ the study evaluation as a reflection on 'what I have learnt about the topic, my methods etc & what other researchers need to know'. e.g. don't recite general, acknowledged limitations of methods & leave it at that - we know this already! What have you learnt about this method?
10/ One of my students has used an online qually survey in her research - a fairly novel technique, in this paper she has a great, brief reflection on using Q surveys. This is the type of thing that makes study evaluations sing & actually make them useful: onlinelibrary.wiley.com/doi/abs/10.111…
11/ When making suggestions for future research don't switch on the RANDOM IDEAS GENERATOR! Step away... don't press that lever... & don't say 'we need research with bigger samples so we can generalise...' because I will be so, so sad. Make suggestions that arise from yr research
12/ So one of my students is researching a particular health condition right now & this wasn't something we asked directly about but a few mentioned their hopes for how treatment might develop/how it could be different. This would make for a great suggestions for future research
13/ If we'd asked about this directly the participants might have said a lot more about this. So reflect on if you were going to do the study again what you'd differently or what you'd want to explore next given what you have learnt. Avoid assuming things are obvious - I've...
14/ spoken to women, so men are next... Why? (they might well be, but why?) One thing I got asked this week is 'should I have some personal reflections in my discussion?' Maybe - if relevant. e.g. if you're a practitioner-researcher how has yr practice changed as a result of...
15/ doing this research? Finally, end with a POW! Yr conclusion is yr final chance to explain to the reader why yr research is awesome & remind them of the key message. If U had to describe your main finding, contribution etc in 1 sentence - what would that B? That is yr conclu!
16/ Build your conclusion around that & don't introduce new material - it's summing up, leaving the reader with a clear 'take home' message. Good luck!
Aren't generalisable!!
