A realist approach to thematic analysis: making sense of qualitative data through experiential, inferential and dispositional themes

ABSTRACT

Thematic analysis (TA) is the most widely used method for analysing qualitative data. Recent debates, highlighting the binary distinctions between reflexive TA grounded within the qualitative paradigm and codebook TA with neo-positivist orientations, have emphasized the existence of numerous tensions that researchers must navigate to produce coherent and rigorous research. This article attempts to resolve some of these tensions through developing an approach to TA underpinned by realist philosophy of science. Focusing on interview data, we propose the use of three types of themes (experiential, inferential and dispositional) and the use of corresponding validity indicators (empirical adequacy, ontological plausibility and explanatory power). Using an illustrative example, we outline the conceptual foundations of a realist approach to TA and present recommendations for conducting it in practice. This approach, we claim, reconciles several existing binaries between distinctive types of TA by incorporating the contributions of both for the development of different types of themes.

KEYWORDS:

• Realist analysis
• critical realism
• paradigms
• small q / big Q qualitative
• data analysis
• validity
• study design
• research quality
• interdisciplinary research
• methodology

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 We use the term ‘realism’ here for simplicity and because it is inclusive of different traditions of realism, but we acknowledge that much of our understanding is informed by authors associated with the label of critical realism (e.g. Bhaskar 1975, 1978; Archer 2007; Danermark et al. 2002; Sayer 1984). Our account is based on features across the cited scholarship that we believe are compatible and can guide qualitative researchers in navigating thematic analysis.

2 Other descriptions of qualitative data analysis refer to this process as ‘inductive’ coding (see Braun and Clarke 2006). However, we later use the term ‘inductive’ to refer to a type of logical reasoning (see Danermark et al. 2002) so we use ‘data-driven’ to avoid confusion.

3 Other description of qualitative data analysis refers to deductive coding as ‘theory-driven’ or ‘top-down’ thinking (Braun and Clarke 2006). We use the term ‘deductively’ as it is used in logic to refer to the process of testing the truth of an emerging premise based on the truth of an observable conclusion (e.g. Sports coaches generally experience X (premise 1). This particular participant is a sports coach (premise 2). Therefore, this participant experiences X (conclusion).)

4 In some ways, assessing whether a particular experience for one individual can be said to be true for another individual at all could be a contentious step for researchers depending on their approach. Taking the example of where our two participants expressed a feeling of enjoyment in seeing their athletes achieve their goals, on the one hand it is important not to assume they are having precisely the same experience of enjoyment and for precisely the same reasons. On the other hand, however, it is important not to assume their experiences are entirely unique. Our approach was to be sensitive to the possibility of uniqueness in each participant’s experiences but also pragmatic in the sense that the value of much qualitative research relies on the assumption that at least some individual experiences can be meaningfully said to be shared by others. Indeed, grounded in our realist ontological assumptions we adopt the view that the similarities between people’s experiences are made possible by the shared underlying reality within which a particular individual’s experiences resides.

5 We use the term inductive to mean the reasoning involved in generating a more general claim from what is known about a particular case.

Additional information

Notes on contributors

Gareth Wiltshire

Gareth Wiltshire is a Lecturer at Loughborough University, UK. He is a social scientist with broad interests related to the social determinants of health. His work is largely informed by concepts and theories derived from sociology but is proactive in engaging with researchers, practitioners and ideas across different disciplines.

Noora Ronkainen

Noora Ronkainen is a Senior Researcher at the University of Jyväskylä, Finland. She is an interdisciplinary scholar with main research focus on learning, development, identity and meaningful experiences in sport and physical cultures. She has studied the role of sport and movement in peoples’ lives across the life span in various national and movement culture contexts.