Counselling and Psychotherapy Research Journal

Perspectives

Using thematic analysis in counselling and psychotherapy research: A critical reflection

Victoria Clarke, University of the West of England, Bristol, UK

Virginia Braun, The University of Auckland, Aotearoa/New Zealand

It has been just over a decade since we published a paper outlining a new approach to thematic

analysis (TA) entitled Using thematic analysis in psychology (Braun & Clarke, 2006). Our approach to TA

has become widely used both in and beyond psychology, and particularly in applied research areas,

including counselling and psychotherapy. The popularity of our approach also seems to have

prompted growing interest in TA more broadly, with a section or chapter on TA a common feature of

many recently published research methods texts (e.g. Vossler & Moller, 2014), and the latest editions

of established texts (e.g. McLeod, 2015). TA is recognised as a useful method for psychotherapy

process research (Mörtl & Gelo, 2015) and as a method that is “flexible, straightforward and

accessible” (McLeod, 2011, p.146) for counselling and psychotherapy researchers. TA has most

commonly been used by counselling and psychotherapy researchers to explore the experiences and

views of specific groups of clients and therapists (e.g. Carew, 2009; Hunt, 2013), typically drawing on

data from interviews (e.g. Hunt, 2013), and, less often, focus groups (e.g. Carew, 2099). However, there

are other ways of using TA. For example, our counselling psychology students have used a post-

structuralist, queer and feminist theory-informed TA to interrogate therapists’ discursive constructions

of heterosex (Shah-Beckley, 2017) and an attachment theory-informed TA to analyse psychotherapy

transcripts (Willcox, 2017).

Although TA as a distinct analytic method has increased in popularity over the last decade, we

continue to see evidence of confusion about TA – what it is, what philosophy underpins it, and what

‘best practice’ looks like. In this short commentary, we address some of the main areas of confusion

and poor practice in counselling and psychotherapy research. Our aim is to support counselling and

1
psychotherapy researchers, research supervisors and research methods teachers to improve the

understanding and implementation of TA (and our approach particularly) in this field.

We intended our approach to TA to be a fully qualitative one. That is, one in which qualitative

techniques are underpinned by a distinctly qualitative research philosophy, that emphasises, for

example, researcher subjectivity as a resource (rather than a problem to be managed), the importance

of reflexivity, and the situated and contextual nature of meaning. Kidder and Fine (1987) dubbed this

orientation ‘Big Q’ qualitative – qualitative research conducted within a qualitative paradigm (they

contrasted this with ‘small q’ qualitative – the use of qualitative tools and techniques within a positivist

paradigm). We assumed that most readers of our paper would understand this intent. We were

wrong! In counselling and psychotherapy research, and in other research areas, we see many instances

of our approach cited alongside other, often very different (particularly with regard to underlying

philosophy), approaches to TA, and our approach ‘mashed-up’ with a positivist research sensibility

and analytic procedures. What is particularly troubling is that these ‘mash-ups’ seem to be unknowing,

reflecting some degree of confusion about what qualitative research is, rather than active, ‘knowing’

choices.

One major point of clarification therefore is that TA is not a term for one approach to qualitative

analysis, but many. Indeed, TA is best thought of as an umbrella term for a wide variety of approaches,

which share some assumptions in common (typically that TA is a method, not a methodology, and

flexible in terms of theoretical application), but also vary in terms of analytic procedures and guiding

philosophy. In more recent publications, we have distinguished between three ‘schools’ of TA: (1)

‘coding reliability’ approaches underpinned by a positivist philosophy and involving the use of a more

structured approach to coding, with an emphasis on ensuring the reliability and accuracy of coding

(e.g. Boyatzis, 1998), which we classify as a ‘small q TA’; (2) approaches like ours that are located within

a qualitative paradigm and emphasise an organic approach to coding and theme development, with

quality coding resulting from depth of engagement (‘Big Q TA’); and (3) ‘codebook’ approaches that

combine the structured coding procedures of small q TA with the underlying qualitative philosophy of

2
Big Q TA; which might be dubbed ‘medium Q TA’. This final school includes template (e.g. Brooks,

McCluskey, Turley & King, 2015) and framework (e.g. Smith & Firth, 2012) analysis, among others (see

Braun, Clarke, Hayfield & Terry, in press).

If drawing on different ‘doing TA’ resources, counselling and psychotherapy researchers need to

understand any differences in philosophy and procedure and explicitly discuss how different

procedures have been reconciled, and actively negotiate any tensions in underlying philosophies. It is

important to stress that we do not advocate, even though it is often claimed that we do (!), the use of

codebooks and coding frames, an approach to coding based on developing a singular ‘consensus’.

and coding reliability measures. Why? Because these do not cohere with the qualitative sensibility that

underpins and shapes our approach.

The various schools of TA outlined above differ in how the ‘theme’ is conceptualised. This brings us to

another important point of clarification – if using our approach, we do not conceptualise themes as

‘domain-summaries’ (see Connelly & Peltzer, 2016); summaries of (often divergent) responses on a

particular issue or topic. We have read many instances of counselling and psychotherapy papers

reporting themes with titles like ‘Experiences of…’ and ‘Barriers to…’, or one-word titles like ‘Causes’

and ‘Barriers’, in which the theme consists of a descriptive summary of what the participants said

about these causes or barriers, without any sense of whether any or what underlying patterning ties

the analytic observations together. In writing our 2006 paper, we again took for granted that most

readers would understand what a (fully realised) theme is, and how this differs from a summary of

participant responses to a particular data collection question, or in relation to a particular area or

‘domain’ of the data (we and others have identified such ‘themes’ as instances of an under-developed

analysis; Connelly & Peltzer, 2016). In small and medium q approaches to TA, this is often how themes

are understood – another distinction between schools.

In Braun and Clarke (2013) and Braun et al. (2014) we outlined our notion of a ‘central organising

concept’ to attempt to more clearly articulate our conceptualisation of a theme. In our approach to

TA, themes can perhaps be usefully thought of as key characters in the story we are telling about the

3
data (rather than collection pots into in which we place everything that was said about a particular

data domain). Each theme has an ‘essence’ or core concept that underpins and unites the

observations, much like characters have their own psychological makeup and motivations. We find

DeSantis and Ugarriza’s (2000) discussion of themes particularly useful. It highlights the ways in which

themes are active creations of the researcher (rather than just passively ‘emerging’ fully formed from

the data) that unite data that at first sight might appear disparate, and often capture implicit meaning

beneath the data surface. For example, Moller, Timms and Alilovic (2009)’s research on trainee

practitioners’ perceptions of personal therapy reported two themes: ‘personal therapy helps me to be

a better practitioner;’ and ‘personal therapy costs me’. Each theme is organised around a central

concept, which together ‘told a story’ about two contrasting ways the trainees’ made sense of the

topic. This approach to themes has the potential to highlight shared meaning, as well as contrasts or

disjunctures in meaning, more clearly than summaries which compile divergent views.

This brings us to another common area of confusion about the purpose and limits of TA. Despite

countless examples of researchers describing TA as merely a descriptive method, useful for

summarising only surface meaning, and of use just in research focused on participants’ experiences

and subjective meanings, TA is not simply a method for data description and reduction. TA can be

used to describe and summarise – and there is nothing inherently wrong with this, if appropriate to

the research aims. But more importantly, rich analysis typically moves from simple summation-based

description into interpretation; telling a story about the ‘so what’ of the data. And TA can be used in

‘critical’ qualitative approaches, informed by poststructuralist, social constructionist and discursive

theory, which are never (just) descriptive (Clarke & Braun, 2014). This means that although TA is widely

used, its potential as a method is often under-appreciated, perhaps due to misreading TA as

atheoretical, rather than theoretically flexible and able to be used within different theoretical

frameworks.

So, finally, we emphasise that theory is not optional in TA! Although TA is flexible in terms of the

theoretical framework(s) underpinning analysis (with the exception of theory-as-paradigmatic – small,

4
medium or Big Q), it is never conducted in a theoretical vacuum, and thus should not be thought of as

atheoretical. Yet we have read numerous papers in counselling and psychotherapy research that

contain little or no discussion of the philosophy underlying and shaping the use of TA. This is

problematic in TA, because theory is not ‘inbuilt’ as part of a complete package – as it is for many

other ways qualitative data are analysed. Imagine shopping for a teddy bear... Most teddy bears are

ready-made: the manufacturer has made the bear, stitched together the fabric, filled it with stuffing,

added eyes and a nose and perhaps tied a ribbon around the neck. The ready-made bear is akin to

such approaches – like interpretative phenomenological analysis (IPA) and grounded theory – which

offer qualitative methodologies, rather than (just) methods. In methodologies, theory is inbuilt, and

ideal research questions, methods of data collection, and sampling procedures are defined or

delimited. TA is not like that. This means using TA is like shopping at a ‘make your own bear’ shop.

The bear you take away has all the same elements as the ready-made bear (fur, stuffing, eyes, nose

and ribbon), but you select the specifics – the type of fur, eyes, nose and so on. The final bear is a

product of your choices. TA needs to be underpinned by theory, as much as IPA and grounded theory

do, but the researcher must choose the theories that inform their use of TA, and how exactly they

implement TA. McLeod (2015: 147) suggests that for this reason, TA “is a good choice for researchers

who feel confident that they know what they are trying to achieve”. TA, then, requires researchers to

think about aspects of the research process that can potentially be side-lined if using a methodology

like IPA or grounded theory, and we see this as a good thing! One that, underpinned by a sound

understanding of TA, can lead to a conscious and reflexive application of approaches and procedures.

We hope readers of CPR find this brief discussion around some key ‘traps’ around TA useful, and we

look forward to seeing how the use of TA in counselling and psychotherapy research evolves in the

future.

Further Reading

Our thinking around TA has evolved considerably in the last decade, so we encourage readers of CPR

interested in our approach to read some of our more recent writing, alongside our original 2006

5
paper. We particularly recommend a chapter that discusses TA in the context of counselling and

psychotherapy research (Braun, Clarke & Rance, 2014), our qualitative textbook (Braun & Clarke,

2013), which locates our approach within a broader qualitative research philosophy, and a recent

chapter, which situates our approach in relation to the history and wider terrain of TA (Terry, Hayfield,

Clarke & Braun, 2017). Our TA website provides a comprehensive list of our publications, as well as

FAQs and a checklist for reviewers and editors on evaluating published TA papers:

https://www.psych.auckland.ac.nz/en/about/our-research/research-groups/thematic-analysis.html.

References

Boyatzis, R. E. (1998). Transforming qualitative information: Thematic analysis and code development.

Thousand Oaks, CA: Sage.

Braun, V. & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in

Psychology, 3(2), 77-101.

Braun, V. & Clarke, V. (2013). Successful qualitative research: A practical guide for beginners. London,

United Kingdom: Sage.

Braun, V., Clarke, V. & Rance, N. (2014). How to use thematic analysis with interview data. In A. Vossler

& N. Moller (Eds.), The counselling & psychotherapy research handbook (pp. 183-197). London, United

Kingdom: Sage.

Braun, V., Clarke, V., Hayfield, N. & Terry, G. (in press). Thematic analysis. In Liamputtong, P. (Ed.),

Handbook of research methods in health and social sciences. New York, NY: Springer.

Brooks, J., McCluskey, S., Turley, E. & King, N. (2015). The utility of template analysis in qualitative

psychological research. Qualitative Research in Psychology, 12(2), 202-222.

Carew, L. (2009). Does theoretical background influence therapists' attitudes to therapist self-

disclosure? A qualitative study. Counselling and Psychotherapy Research, 9(4), 266-272.

6
Clarke, V. & Braun, V. (2014). Thematic analysis. In T. Teo (Ed.), Encyclopaedia of critical psychology (pp.

1947-1952). New York, NY: Springer.

Connelly, L. M. & Peltzer, J. N. (2016). Underdeveloped themes in qualitative research: Relationships

with interviews and analysis. Clinical Nurse Specialist, January/February, 51-57.

DeSantis, L. & Ugarriza, D.N. (2000). The concept of theme as used in qualitative nursing research.

Western Journal of Nursing Research, 22(3), 351-372.

Hunt, J. (2013). An initial study of transgender people’s experiences of seeking and receiving

counselling or psychotherapy in the UK. Counselling and Psychotherapy Research, 14(4), 288-296.

Kidder, L. H., & Fine, M. (1987). Qualitative and quantitative methods: When stories converge. In M. M.

Mark & L. Shotland (Eds.), New directions in program evaluation (pp. 57-75). San Francisco, CA: Jossey-

Bass.

rd
McLeod, J. (2015). Doing research in counselling and psychotherapy, 3 ed. London, United Kingdom:

Sage.

nd
McLeod, J. (2011). Qualitative research in counselling and psychotherapy, 2 ed. London, United

Kingdom: Sage.

Moller, N.P., Timms, J. & Alilovic, K. (2009) Risky business or safety net? Trainee perceptions of

personal therapy: A qualitative thematic analysis. European Journal of Psychotherapy & Counselling,

11(4), 369-384.

Mortl, K. & Gelo, O.C.G. (2015). Qualitative methods in psychotherapy process research. In O.C.G Gelo,

A. Pritz & B. Rieken (Eds.), Psychotherapy research: Foundations, process and outcomes (pp. 381-428).

Wien, Austria: Springer-Verlag.

Shah-Beckley, I.S. (2016). Therapists' and non-therapists' constructions of heterosex: A comparative

story completion study. Unpublished doctoral thesis, University of the West of England, Bristol, United

Kingdom.

7
Smith, J. & Firth, J. (2011). Qualitative data approaches: The framework approach. Nurse Researcher,

18(2), 52-62.

Terry, G., Hayfield, N., Clarke, V. & Braun, V. (2017). Thematic analysis. In Willig, C. & Stainton-Rogers

nd
(Eds.), The Sage handbook of qualitative research in psychology, 2 edition. London, United Kingdom:

Sage.

Vossler, A. & Moller, N. (Eds.), (2014). The counselling & psychotherapy research handbook. London,

United Kingdom: Sage.

Willcox, R.S. (2017). An attachment theory-informed thematic analysis of bereaved families' narratives.

Unpublished doctoral thesis, University of the West of England, Bristol, United Kingdom.