1/ To all those advocating saturation as *the* criterion for determining qual "sample" size (instead of Gender & Society's positivist qual 35 int minimum) please note that saturation has been critiqued for bloody decades as realist/positivist & not working for all qual. Here's
3/ Here's me & @ginnybraun critiquing the use of saturation as information redundancy in thematic analysis research - arguing that saturation only makes sense in positivist/realist forms of TA. For our reflexive approach it simply doesn't work: tandfonline.com/doi/abs/10.108…
4/ Before anyone asks what do I do instead - check out Malterud et al.'s critique of the saturation concept and their alternative - information power (this is not without criticism but it is way less problematic than saturation): journals.sagepub.com/doi/abs/10.117…
5/ Purlease acknowledge in advocating for saturation you are taking a position, one relying on particular theoretical assumptions. Qual researchers must get better at recognising & owning their positionality!! I'm fed up of positioned takes on qual being presented as definitive!
6/ Some other examples of critiques of saturation - there are LOADS beyond this - a fab paper which "challenge[s] the unquestioned acceptance of the concept of saturation & consider[s] its plausibility and transferability across all qualitative approaches" journals.sagepub.com/doi/10.1177/14…
7/ This is a fab paper by Low which argues saturation is "a logical fallacy, as there are always new theoretic insights as long as data continue to be collected" - tandfonline.com/doi/full/10.10…
8/ A really interesting critical discussion of differemt approaches to determining "sample" size in qual - tandfonline.com/doi/full/10.10…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
1/ I've been having a lot of discussions about getting started with familiarisation & coding in reflexive thematic analysis recently. So here's a 🧵 with the highlights. Familiarisation is about getting to know the contents of your data - if you interacted with participants to...
2/ generate them & you transcribed them this gives you a head start. It's also about starting to engage with your data analytically - what's going on here? & reflecting on your emotional responses to the data/participants - how do you feel when you read them/certain participants?
3/ I encourage students to write familiarisation notes - on individual data items & overall, & after each round of familiarisation - for themselves, not to share with anyone. And they're not limited to writing notes, students have audio recorded notes, written poetry, doodled...
1/ The language & concepts of nonpositivist/Big Q qualitative research - what do we take from quantitative/positivist research, what do we rework, what do we leave behind?
A 🧵 of musings starting with generalisability - a term often associated solely with quant.
2/ I have often equated generalisability with statistical generalisability & argued that it's a concept that doesn't hold relevance for qual. But this paper by @BrettSmithProf convinced me that generalisability is a broader concept & can be reworked: tandfonline.com/doi/abs/10.108…
3/ I've started to avoid data *collection* as it implies things exist as data before researchers arrive on the scene & we are relatively (& ideally in small q qually) passive/unobtrusive in the process - keeping our "influence" to a minimum. For me data *generation* acknowledges
1/ @ginnybraun & I made it to the Build-a-Bear Workshop in Bristol! Here's why we think it works as a great 'metaphor' for thematic analysis & for challenging common misconceptions of TA. We mentioned this in our webinar yesterday eve which you can watch on YouTube (link below).
2/ People often assume that TA can only be used for basic descriptive/summative analyses, that it's atheoretical like qualitative content analysis often claims to be or only realist/essentialist, lacking the sophistication of grounded theory or IPA. Students often contact us...
3/ after being told that TA isn't sophisticated enough for a Masters or Doctoral project... These are all problematic assumptions. TA is different from approaches like grounded theory, discourse analysis & IPA because it is closer to a method than a methodology - a theoretically
1/ I didn't link to *that* paper because of prudishness - I have led/taught on a module on sexuality since the early 2000s & give lectures on explicit topics including masturbation... I didn't link to it for several reasons including not want to encourage nudge nudge "humour"...
2/ which communicates discomfort with the topic... there is a role for humour in talking about sex but not here, not now. The topic is distressing in various ways and doesn't need air time on Twitter. The primary concerns are ethics, why this research was allowed to go ahead...
3/ why the supervisor apparently sanctioned this, why UoM appear to be funding this research, why this research has UoM ethical approval or doesn't, how this paper got through peer review, why the QR editors apparently didn't ask any of these questions & published the paper...
1/ A thread on piloting data generation "tools" in qually research & different ways of thinking about piloting in small q & Big Q qually. First up, what's small q and Big Q qually? Small q is where qually is defined by collecting & analysing qual data but the underpinning values
2/ default to disciplinary norms - typically positivism. It seems to be most often practiced unknowingly - the assumption is that this is what constitutes good practice & there isn't an awareness of other possibilities for qual. Big Q qual involves both qual techniques & the
3/ distinct values & traditions that have developed around qual methods (e.g. interpretavism, phenomenology, constructivism, narrative to name a few). Big Q is typically a rejection of positivism. And this shapes research practice in various ways including around piloting.
1/ A thread on ensuring/assessing quality in qualitative research & whether checklists/guidelines that aspire to be universally applicable have a role to play.
The first problem with universal guidelines is that there isn't a widely agreed on definition of what qually research is
2/ A - over simplified - definition of qually research (well any research) is that it involves tools & techniques for collecting & analysing data & research values (paradigms, 'ologies) that tell you what the data represent, what you can access through them: contextually situated
3/ sense-making, a universal truth of experience, discourses, narratives, social constructions etc. It's very hard to develop a definition that works for all forms of qualitative research. So lots of guidelines/checklists are based on partial definitions but these aren't...