Qualitative Research

News Blog readers are referred to a BMJ article signed by 76 senior academics from 11 countries.[1] The paper is led by the redoubtable Trisha Greenhalgh. The signatories include three members of the CLAHRC WM Scientific Advisory Committee (Jeffrey Braithwaite, Mary Dixon-Woods, and Paul Shekelle), a further CLAHRC WM collaborator (Russell Mannion), and the Director of a peer CLAHRC (Ruth Boaden). The authors make a strong case for the BMJ to publish more qualitative research, but, in a separate article, the BMJ editors hold their ground.[2] The CLAHRC WM Director thinks qualitative information is extremely important, and numbers alone seldom provide everything needed to theorise and take decisions. He would be happy to append his name to the signatories of the Greenhalgh letter. The authors provide examples where qualitative research has been influential, and the CLAHRC WM Director has written a reader on important qualitative research.[3] Most of the Director’s personal work involves collecting and synthesising both qualitative and quantitative data. As a Bayesian he spends a lot of time quantifying belief in order to accomplish such synthesis.[4] [5] The BMJ should be prepared to consider qualitative research on its merits and adopt a more welcoming posture.

Qualitative research is great, so what’s the problem? The CLAHRC WM Director does not like the way qualitative research is often carried out. Here are some piquant thoughts on how it might be improved:

  1. Do not determine sample sizes by theoretical saturation without saying what is to count as “saturation”. This should be described quantitatively, as discussed in a recent post.[6]
  2. Drop the argument that qualitative research preserves data in all their original complexity. No. The whole point of science is to represent underlying mechanisms and this means abstracting a theory from the messy observed world.
  3. Drop the irritating statement that qualitative research provides ‘rich data’. What is so rich about qualitative data, but not a large, in-depth, epistemological study like BioBank?
  4. Do not say that sampling can be purely purposive and leave it at that. Purposive sampling results in subgroups, and you should say how you will sample within the subgroup (unless there is only one person in the sub-group; the prime-minister for example).
  5. Combine qualitative and quantitative data whenever possible. Mixed methods research appears more often in protocols than in research papers. Yet this is a case where the total really can be much more than the sum of its parts.
  6. Drop the idea that qualitative research is not subject to bias, since it describes peoples lived experience, which is inviolate. No, there is shedloads of evidence that responses are labile, turning on the context on which discussions occur.[7] Here we start to lap against the real Achilles’ heel of some qualitative research – it is tainted by (sometimes subsumed by) notions of constructivism.

The heart of the problem is the idea that qualitative and quantitative research are quite different things that should be governed by a distinct set of laws. This veil of distinction is bogus. For a start, numbers and qualitative data are not that different. They are frequently inter-convertible – witness an informative Bayesian prior. And measurement theory is not dependent only on numbers, but on homology. Quantitative researchers should not allow qualitative researchers to bamboozle them with talk of different ‘paradigms’. Qualitative researchers should ensure that they grasp abstract, but crucial, epidemiological ideas, such as regression to the mean and the distinction between differences in relative and absolute risk. Classicists learn Latin and Greek, and health researchers should learn quantitative and qualitative, since a narrow understanding of one or the other is just that – a narrow understanding. So, shed any constructivist paradigm cloak and consign it to trash. When John Paley and I submitted our article to the BMJ, arguing that qualitative research need not (indeed should not) be constructivist in its epistemology,[3] the editors wrote back to say that BMJ qualitative researchers were not so silly – or rather that they took a more pragmatic stance. But, we replied in stout defence of our submission quoting a BMJ paper that said, “Most qualitative researchers today share a different belief about knowledge, called ‘constructivism…’.”[8] Our argument was not that knowledge was not socially constructed, but that it was not necessary, indeed it was undesirable, for qualitative researchers to buy into constructivism as a paradigm. Accepting that knowledge is constructed does not, in our view, mean that it can’t be constructed well or badly, or that all constructions are equivalent, or that there is not a reality ‘out there’ that can be at least partly understood by means of knowledge constructed under guidance from the scientific canon. In short, qualitative research is in, the two world view idea of qualitative and quantitative research is out.

— Richard Lilford, CLAHRC WM Director

References:

  1. Greenhalgh T, Annandale E, Ashcroft R, et al. An open letter to The BMJ editors on qualitative research. BMJ. 2016; 352: i563.
  2. Loder E, Groves T, Schroter S, Merino JG, Weber W. Qualitative research and The BMJ. BMJ. 2016; 352: i641.
  3. Paley J, & Lilford R. Qualitative methods: an alternative view. BMJ. 2011; 342: d424.
  4. Yao GL, Novielli N, Manaseki-Holland S, et al. Evaluation of a predevelopment service delivery intervention: an application to improve clinical handovers. BMJ Qual Saf. 2012; 21(s1):i29-38.
  5. Hemming K, Chilton PJ, Lilford RJ, Avery A, Sheikh A. Bayesian cohort and cross-sectional analyses of the PINCER trial: a pharmacist-led intervention to reduce medication errors in primary care. PLoS One. 2012;7(6):e38306.
  6. Lilford RJ. Sample size for qualitative studies: two recent approaches. CLAHRC WM News Blog. 11 March 2016.
  7. Lincoln YS, Guba EG. Paradigmatic controversies, contradictions, and emerging influences. In: Denzin NK, Lincoln YS, eds. The landscape of qualitative research: theories and issues. 2nd ed. Sage, 2003:253-91.
  8. Kuper A, Reeves S, Levinson W. An introduction to reading and appraising qualitative research. BMJ. 2008; 337: a288.

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s