Read on Twitter
This is an interesting thread, if only because I strongly disagree with most of it. There is value in open quali research even (especially?) if it has different epistemological foundations than many quantitative approaches. Longish thread (1/12) https://twitter.com/jacasiegel/status/1358831397918695428
Note, I'll be focusing on open science initiatives, not replication and not rigid pre-registration. Two main aspects to open in quali is 1.) open data (ex. releasing de-identified textual data) (2/12)
And 2.) transparent links between data-analytic claims (ex. active citation, Moravcsik, 2014 ). Needless to say these have GREAT practical challenges (and no real incentive...), but that's a different thread (3/12)
Opening these two aspects greatly helps in the ultimate goal of reflexivity. Otherwise all people have to go by is your word that you were reflexive, and that you did so and so to reduce your bias. Openness can let your readers and peers agree or disagree with your claims (4/12
Anxious that your readers will see the open data and "read" the "wrong" thing off it? then add in notes from the field and explain the context in the supplementary materials! (5/12)
Similarly, we don't shut down openness in experimental psych just because we can't cover all the minute details of each setting, being open with the important things is still beneficial (6/12)
If you go for the participatory approach, this also has great promise of being able to let your participants see what others in their community has to say (and what they think about it), and verify not only your results but also your process of reasoning. (7/12)
Also the promise to greatly reduce data waste, imagine if we had access to as much (safe) raw data we can get from all the ethnographies done so far - no need to re-invent the wheel everytime, or cause vulnerable population fatigue from something we should have already knew (8/)
Further, as Dr. Siegel will agree there's no theory free quali research (Atkinson, 2017) there's always a theory - whether formal ones, or just general cultural expectations but (9/12)
that's exactly why auditing should happen - it puts multiple voices on the table (participants'', readers', peers') rather than just the authors' alone. (10/12)
Also, qualitative approaches (both in data gathering, and analysis) cannot be so fickle as it changes on who ate what for breakfast - if it does there can be no accumulation of knowledge, no coherence in the processes or the phenomena that we analyze. (11/12)
This is a longstanding war-chant for qualitative research: no, not anything goes. Not even if (or especially because?) it has different epistemological foundations. (12/12)
Sidenote: Not anything goes, but the variability (especially in analysis) is Real. How variable is it? What is the consequence of this variability? High time for a large scale collaboration: many analyst, one quali data set. May have one simmering, send dm if interested (13/12)
