Generalizability in qualitative research: misunderstandings, opportunities and recommendations for the sport and exercise sciences

Abstract

Generalisation in relation to qualitative research has rarely been discussed in-depth in sport and exercise psychology, the sociology of sport, sport coaching, or sport management journals. Often there is no mention of generalizability in qualitative studies. When generalizability is mentioned in sport and exercise science journals it is often talked about briefly or highlighted as a limitation/weakness of qualitative research. The purpose of this paper is to provide a detailed discussion of generalisation in order to dispel any misunderstandings or myths about generalizability in qualitative research and offer guidance about how researchers might consider generalisation. It is emphasised that it is a misunderstanding to claim that qualitative research lacks generalizability. It is highlighted that statistical types of generalizability that inform quantitative research are not applicable to judge the value of qualitative research or claim that it lacks generalizability. Reasons as to why researchers might consider generalizability in qualitative research are then offered. It is emphasised that generalisations can be made from qualitative research, but just not in the same way as quantitative results are. To help guide how generalisation might be considered, four different types of generalizability are presented: naturalistic generalisation, transferability, analytical generalizability and intersectional generalizability. Practical strategies are also offered for considering generalizability when seeking to publish qualitative research or reflect on already published work. The paper concludes with a set of recommendations to support high quality and rigorous qualitative research for scholars – including journal editors and reviewers – in relation to generalizability.

Keywords:

• Qualitative research
• rigour
• naturalistic generalisation
• transferability
• analytical and intersectional generalizability

Acknowledgments

Thank you to the reviewers for their insightful comments that strengthened the paper.

Notes

1. In laying out these goals, I certainly appreciate that for some researchers the various points raised about generalizability are ‘old hat’. For these researchers I hope the paper is an opportunity to productively expand on what is said in it, thereby further expanding visions of generalizability. At the same time, whilst points here may be well known to some, a close examination of the sport and exercise sciences literature, numerous conversations with colleagues, many debates witnessed at academic conferences, countless meetings with government policy-makers and sport organisation leaders, a check at what is said on social media (e.g. twitter) by new and established researchers who share editorial journal decisions or reviewer comments, the debates that occurred in the British Medical Journal (BMJ) on qualitative research (Greenhalgh et al. 2016) and the various responses to it on the BMJ website and social media, and in my experience of handling and reviewing over 1500 papers (see Endnote 2), would together strongly suggest that the points here very much need to be said and amplified.

2. Given my motivation to stimulate conversations about generalizability and/or an expanded understanding of it, I purposefully avoid singling out articles for critique. I should also note that, as the former editor of Qualitative Research in Sport, Exercise and Health, and formerly an editorial board member of the Sociology of Sport Journal, plus currently an associate editor of Psychology of Sport and Exercise and Psychology of Sport, Exercise and Performance Psychology, an editorial board member of 7 other journals (e.g. International Journal of Sport and Exercise Psychology), and ad hoc reviewer for numerous other journals, I have handled and reviewed over 1500 papers in the last 10 years. In my experience at least, it is common to find researchers stating in their submitted manuscript for possible peer-review that a limitation/weakness of their qualitative research is a lack of generalizability.