https://orcid.org/0000-0002-3435-091X
![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
ABSTRACT
Despite the persistent dominance of a ‘scientific psychology’ paradigm in health psychology, the use of qualitative research continues to grow. Qualitative approaches are often based on fundamentally different values from (post)positivistempiricism, raising important considerations for quality, and whether qualitative work adheres to, and is judged by, appropriate publication standards. Thematic analysis (TA) has become a particularly popular method in qualitative health psychology, but poor practice is widespread. To support high quality, methodologically coherent TA practice and reporting, we critically reviewed 100 systematically selected papers reporting TA, published in five prominent health psychology journals. Our review assessed reported practice, and considered this in relation to methodological and quality recommendations. We identified 10 common areas of problematic practice in the reviewed papers, the majority citing reflexive TA. Considering the role of three ‘arbiters of quality’ in a peer review publication system – authors, reviewers, and editors – we developed 20 recommendations for authors, to support them in conducting and reporting high quality TA research, with associated questions for reviewers and editors to consider when evaluating TA manuscripts for publication. We end with considerations for facilitating better qualitative research, and enriching the understandings and knowledge base from which health psychology is practiced.
Acknowledgements
Thanks to the UWE Faculty of Health and Applied Sciences for internally funded research time to support the writing of this paper, and Kieran McCartan for his helpful comments on and support with the application for this resource. Thanks also to Annabelle McDonald and Shanuki Jayamaha in the School of Psychology at Waipapa Taumata The University of Auckland for support with the review and compiling references. Finally, thanks to the excellent anonymous reviewers whose engaged feedback provoked us to think and reflect further.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Sparkes and Smith (Citation2009, p. 491) concluded that ‘criteria should be viewed as lists of characterizing traits that are open to reinterpretation as times, conditions, and purposes change’ – so we note the specific time in which we have reviewed these papers and written these recommendations.
2 Whether there are quality considerations specific to publishing qualitative and specifically TA research in open access journals (obviously, a very diverse grouping) is an important question, but beyond the scope of this paper.
3 We included papers reporting a set of themes, as well as papers using the specific term ‘thematic analysis’, because of the blurred boundaries between qualitative content analysis and thematic analysis – evident in terms like ‘thematic content analysis’ – and a history of using grounded theory techniques to develop themes from qualitative data in psychology and other disciplines (see Braun & Clarke, Citation2006, Citation2021a).
4 Hutchison et al. (Citation2011) made a similar observation about the diversity within grounded theory being poorly understood.
5 These terms reflect the psychoanalytic leanings of early proponents of content analysis (e.g., Krippendorff, Citation2018), the forebear to TA.
6 The notion of ‘searching for themes’ (Braun & Clarke, Citation2006) was our unintended contribution to ‘themes-as-diamonds’. We have settled on the phrase ‘generating initial themes’ (Braun & Clarke, Citation2022b), to better capture our conceptualisation of themes and (phase three of) the reflexive TA analytic process.
7 Being mindful of power dynamics, we follow Smith’s (Citation2011) use of anonymity, describing in general terms the hallmarks of bad practices, instead of using actual examples from the reviewed papers. As there is value in real-world examples, we quote examples of good practice from the reviewed papers in the supplementary materials (see File B).
8 We didn’t measure quality by journal in any specific way. However, our impression was that the TA published in JHP was generally of a higher quality than that published in the other journals. With the obvious word count constraints restricting authors in PHM, it’s probably unsurprising that we judged the TA published in that journal as the most problematic.
9 We also encourage more honesty about why we choose a particular method – rather than inadvertently evoking the (positivist) idea of a perfect method (that will reveal the truth). In so doing, we could help reveal some of the ‘behind-the-scenes’ mess of research-as-practice rather than the refined version of research-as-published (here, we evoke the mismatch between the ‘official’ account of science and the everyday practiced reality of science, as discussed by Gilbert & Mulkay [Citation1984] in their classic work).
10 Reporting is not just useful from a quality perspective, but also contributes to better practice among a research community, revealing the ‘challenges’ posed by group research that are ‘often undiscussed in research manuscripts’ (Linabary et al., Citation2021, p. 720), and the ‘messiness of [qualitative] knowledge production’ (p. 733). This can also be a useful mode for navigating difference and disagreement, which do not need to be unified into a singular consensus-based story, and for reflexive TA can highlight that the analysis is a story not the only story from the data.
11 Sometimes themes names consisted of one- or two-words that identified the domain/topic that was the focus (e.g., pain; social support), or suggested a topic summary (e.g., relationships with healthcare professionals; experience of the intervention) when they actually focused on shared meaning. Such names are not best practice for reflexive TA as they don’t identify what it is about pain or the nature of relationships with healthcare professionals that is the focus of the theme.
12 These include concepts and practices like accuracy, bias, researcher influence (something ‘external’ that may impact), saturation, participant validation of transcripts and findings, triangulation, and multiple coders.
13 Part of a knowing practice includes understanding and acknowledging the limits and partiality of one’s knowledge and understanding, and seeking clarification or advice where needed.
14 We thank an anonymous reviewer for noting the importance of editorial boards, and acknowledge we are not the first to suggest this – it was discussed, for instance, at the International Society of Critical Health Psychology (https://ischp.net/) annual conference in 2019.
15 It is also ethically problematic for majority group/outsider researchers not to disclose their positionality, particularly when researching and representing participants from minoritised communities and/or who are subject to stigma in a way that the researchers are not (e.g., a straight-sized researcher interviewing fat participants). Careful, considered reflexive ethical consideration is needed here.
16 Sandelowski and Leeman (Citation2012) argued that for qualitative ‘findings’ to be accessible to, and actionable by, practitioners – something important for health research – they must be organised into (shared meaning) thematic statements (see also Campbell et al., Citation2021). In this they are arguing against the use of topic summaries, and instead for shared-meaning based themes.
17 We thank an anonymous reviewer for encouraging the broader scope of thinking, beyond publication, to more fundamental questions around what the discipline is, and how health psychology is taught. Although full consideration of these is beyond the scope of this paper, these provocations align nicely with the insights and recommendations generated through our critical review.