Several people have approached me to ask if it is possible to use coding in a WPR analysis. Coding in this context refers to the labelling of discrete items in research materials (contrast “learning to code” in Research Hub, 30 Oct 2023). The espoused goal is to find a way to organize a large mass of material, which seems unmanageable if one simply applies the WPR questions. The topic addressed here is relevant to earlier Research Hub entries on data, where I ask specifically about the place of data in a WPR analysis (30 May 2022, 29 June 2022, 29 Aug 2022). 

The most common way to approach qualitative studies that involve large amounts of material is thematic analysis. I have on occasion positioned WPR as opposed to or in contrast with thematic analysis. I wish to explain the basis of this claim and then to proceed to consider what RTA (reflexive thematic analysis) may offer WPR researchers. 

The title of this entry refers to my recent article introducing WPR to the study of music education where I list several “themes” and organize the paper around them (Bacchi 2023). The term “themes” is used incorrectly in this paper – hence the “oops”! As Braun and Clarke explain, I should have referred to “topics” rather than to “themes”. To qualify as a theme, the selected items would need to capture “some level of patterned response or meaning” (Braun and Clarke 2021: 341). I pursue this distinction between “topics” and “themes” later. 

On what basis have I distanced WPR from thematic analysis? 

To preview the discussion to follow, I contrast the types of analysis performed by RTA and WPR. The former seeks to identify themes; the latter identifies and interrogates problem representations. In RTA, themes are “picked out” by researchers; in WPR problem representations form part of an analytic strategy. 

Note, I am not saying that themes simply emerge from the data. Braun and Clarke (2021: 343) make it very clear that they have never said that themes simply emerge from the data:

An account of themes “emerging” or being “discovered” is a passive account of the process of analysis, and it denies the active role the researcher always plays in identifying patterns/themes, selecting which are of interest, and reporting them to the readers (Taylor and Ussher, 2001) 

They justifiably correct St Pierre for her characterisation of thematic analysis as a form of analysis “in which themes somehow miraculously emerge from the data” (St. Pierre 2019: 4). Rather, Braun and Clarke (2006: 80) have consistently identified themes as part of an interpretive process involving the researcher. 

My concern is that inadequate attention is paid to the central role played by research questions in this interpretive process. That is, RTA and WPR ask distinctly different forms of research question, informed by specific theoretical premises (see below). This contrast in theoretical perspective makes it difficult to bring the two approaches together. 

Braun and Clarke (2006: 95) attest to the powerful role played by research questions in their comments on a possible “mismatch between theory and analytic claims”: 

“if you are working within an experiential framework, you would typically not make claims about the social construction of the research topic, and if you were doing constructionist thematic analysis, you would not treat people’s talk of experience as a transparent window on their world.”

A recent worked example of RTA by Byrne (2022) illustrates the close ties between “themes” and one’s research questions. Likewise, Braun and Clarke (2006: 88) emphasize the goal of identifying what is “interesting” in the material, driven clearly by the researcher’s research questions. For this reason, researchers are enjoined to “spell out” their theoretical assumptions. 

I take this insight one step further. If research questions “manage” or “control” the analysis to the extent just described – if one’s research questions shape the analysis – it follows that the common distinction drawn between “inductive” (“data-driven”) and “deductive” (“analyst-driven”) accounts seems spurious (Braun and Clarke 2006: 83-84). Since all researchers are necessarily engaged in asking research questions and, since all research questions are theoretically informed, all research is “analyst-driven”. And, if this is the case, the selection of codes and the naming of themes are necessarily shaped by the research questions.

Following this reasoning, the kinds of research questions posed in any study play a prominent role in the kind of analysis produced. It is at this level that I detect a tension between thematic analysis and WPR. 

The politics of research questions

Alvesson and Sandberg (2013) provide a useful guide to the contrasting theoretical premises in varieties of research questioning.

They show, for example, how research questions problematise different sorts of assumptions underlying existing literatures. Their typology of assumptions includes: in-house assumptions, root metaphor assumptions, paradigmatic assumptions, ideological assumptions and “field assumptions” (2013: 54). To generate research questions “the focal point in problematization as a methodology” is “to illuminate and challenge those assumptions underlying existing theories (including one’s own favorite theories) about a specific subject matter” (Alvesson and Sandberg 2013: 53; emphasis in original). 

As an example of an in-house assumption, Alvesson and Sandberg (2013: 54) explain how researchers who question the status of certain characteristics as leadership traits, remain, theoretically, within “trait theory” as a theoretical proposition. “Trait theory”, as a framing logic, is considered to be uncontroversial. It is simply not mentioned. 

This example illustrates that the analysis (of leadership “traits”) necessarily reflects the researcher’s theoretical positioning and hence their research questions. In this sense, research questions shape what will be considered relevant. Trying to identify “themes” through coded terms will tell us little if the researcher’s own assumptions are not considered. 

Braum and Clarke (2006: 78) stress the need for researchers to “make their (epistemological and other) assumptions explicit”. However, the central role of research questions in shaping the parameters of what is analysed is underdiscussed. The framework for analysis is generated through research questions that are based upon assumed theoretical premises. It follows that the very different kinds of research questions produced by WPR and RTA means that blending the two forms of analysis faces significant obstacles.

Comparing research questions

WPR asks questions about problematizations (or problem representations), and these are tied to the specifics of practical texts (Bacchi 2009: 34). That is, to initiate a WPR analysis, the researcher identifies “proposals” or “proposed solutions” in policy and other forms of “text” and works backwards to draw attention to the implicit problem representations they produce. “Proposals” are taken from “practical” or “prescriptive texts” that provide guides to conduct. By telling us what to do, the proposals within “texts” indicate what needs to change and hence what is enacted as “the problem”. In this way WPR offers an analytic strategy built upon problematisations. The questions a WPR researcher asks will invariably be questions about problematisations, their presuppositions, origins and effects. 

Thematic analysis (and Reflexive Thematic Analysis) starts from different kinds of questions. While Braun and Clarke (2006: 77 Abstract) argue that RTA is a useful strategy in disciplines “beyond psychology”, the kinds of questions asked tend to presume the existence of states of being or states of mind. Byrne (2022) for example focuses on deciphering the “opinions” and “attitudes” of research subjects. 

Key premises in this approach sit uncomfortably with a WPR way of thinking. The presumption of sovereign subjects with “attitudes” is put in question in a WPR approach to subjectivity (Bacchi and Goodwin 2016: 49-53). Indeed, WPR treats “psychology” as a governing knowledge and a contingent historical creation that needs “to be interrogated rather than enshrined as ‘truth’” (Bacchi and Goodwin 2016: 5).

The tensions between WPR and RTA also appear in relation to the kinds of material used for an analysis. In RTA and other forms of thematic analysis, the scripts from interviews and focus groups commonly form the basis of the analysis (Byrne 2022). In WPR interviews are not a straightforward source of “opinions”. Rather, interview texts are subjected to a style of questioning that looks to identify the presuppositions underlying “statements” (Bacchi and Bonham 2016). 

In other words, RTA and WPR rely on contrasting theoretical premises which shape the questions they ask. These contrasting premises make it difficult to imagine the use of RTA to identify “themes” in a study wishing to use WPR. 

“Coding”, “topics” and “themes”

In my paper introducing WPR to music education I discovered the usefulness of “topics” as opposed to “themes” (Bacchi 2023). In preparation for the presentation, I engaged with and read the critical literature in the field. This immersion alerted me to the areas of controversy under discussion by those involved in music education. Using WPR I wanted to illustrate how it was possible to recognize “proposals” and work backwards to identify the problem representations implicit within them. I chose “proposals” within the National Plan for Music Education 2022 that aligned with the areas of controversy I had identified: progression/development, inclusion and diversity, talent and creativity, teacher training and professionalism, and evidence-based policy. I mistakenly called these areas of controversy “themes” whereas they were (simply) topics.

Is anything gained from identifying topics, as I did in the music education paper? I would see this approach as useful in a “first run” through the material, a way of indicating that the selected text (National Plan for Music Education 2022) included many proposals for change in music education programs. I noted that, in the Plan, the word “should” was used some 250 times, illustrating the large number of proposals in this highly prescriptive text (Bacchi 2023: 6). 

My decision to highlight five “topic areas” allowed me to introduce WPR thinking and to encourage a more critical questioning of the Plan. I believe listing topic areas proved a useful strategy for these purposes. Hence, it is completely possible that, in your own work, identifying “topics” may contribute to your analysis. I would stress, however, that these topics are not themes, and that they did not require coding. 

“Themes” within problem representations? 

Forms of analysis depend, in the first and last instance, on research questions. Moreover, as argued above, the sorts of research questions associated with WPR are different in kind and scope from the sorts of research questions guiding thematic analysis. A WPR analysis targets governmental problematisations. It would be inappropriate, therefore, to try to identify “themes” prior to seeking out and interrogating those problematisations. 

However, might it be possible to use coding and theme identification to organizeidentified problematizations/problem representations? If you start your analysis with the WPR questions to identify problem representations, could some form of coding be useful in their analysis? Is the counting of word usage – e.g., the 250 mentions of the word “should” – in effect, a form of coding or thematic analysis? Given that there are often plural problem representations in any practical text and allied texts, might it be useful to assign different problem representations a code? 

I invite readers to send me examples where such an analysis has been performed. The example I wish to offer applied a conventional WPR analysis despite the large amount of material the study involved. Identified problem representations were then organized into categories, as outlined below, around “themes”. Does this case-study illustrate a way in which thematic analysis may be compatible with WPR? 

Suicide prevention initiatives targeting Sámi in Nordic countries 

A 2021 article by Jon Petter Stoor et al. set out to map and examine suicide prevention initiatives among Sámi. Seventeen initiatives targeting Sámi were identified during 2005–2019, including nine in Sweden, five in Norway, one in Finland and two international initiatives. Applying the WPR questions, the authors identified “40 problematizations regarding how to prevent suicide among Sámi” (Stoor et al. 2021: Abstract). 

The authors started their analysis from “proposals” in the selected texts and worked backwards to identify problem representations. Given the large number of identified problem representations (40), they introduced five categories to organize the material in a meaningful way: “pertaining to shortcomings on individual (5), relational (15), community/cultural (3), societal (14) and health systems levels (3)”. These categories can be described as themes (rather than topics) since they capture “some level of patterned response or meaning” (Braun and Clarke 2021: 341). Moreover, it is possible to imagine how some form of coding might have facilitated the organizing of the material into these categories/themes.

Importantly, Stoor et al. (2021) did not stop the analysis there. They produced a Supplementary Table listing the 40 problem representations that had been “read off” from identified proposals. 

I offer a sample of the identified problem representations to indicate the richness of the material.

Supplementary table 3. Problematizations, category and level of intervention suggested, yielded through applying the “What is the problem represented to be?”-approach on suicide prevention initiatives targeting Sámi in Norway, Sweden and Finland. 

Problematizations:  

Young Sámi men do not have enough tools for emotional regulation 

Sámi (and non-Sámi) youth at risk of suicidality do not have an active enough lifestyle 

Sámi (and non-Sámi) young kids do not have enough coping skills to deal with life’s challenges, conflicts and mental health issues 

Young reindeer herders do not have good enough skills to take care of themselves/increase mental well-being 

Young male reindeer herders do not have good enough conflict management skills 

Sámi youth do not have enough access to peer-support 

Young male reindeer herders do not have enough access to peer support 

Young reindeer herders do not have enough access to peer support 

Young Sámi men do not have enough access to peer support 

The Supplementary Material Table conveys something of the complexity involved in thinking through how suicide prevention among the Sami was problematized. Making this amount of detail available renders it possible to raise questions about the five categories produced by Stoor et al. Does the sample of problem representaions (above) suggest another schema for organisation, perhaps by age categories? The point is that, by listing the 40 problem representations, other ways to envision the “problem” could be extracted from the material. In this way, the authors protect against the dangers of simplification that can accompany coding and thematic analysis. 

Conclusion

To summarize, research questions shape an analysis. WPR and RTA consist of theoretically distinct research questions. Hence, it is inappropriate to seek out RTA-style themes prior to applying the WPR questions. it may on occasion suit your purposes to single out different topics within a body of material, as I did in the music education paper. This analytic intervention is not, however, a thematic analysis. 

Once (plural) problem representations have been identified, it may be possible to use themes to organize them. The Stoor et al. (2021) example illustrates how problem representations can be organized by themes. Importantly, the authors developed their framework of five categories (themes) only after they had applied the WPR questions. They applied WPR thinking as a first stage in the analysis to identify problem representations and then considered how to make this material more meaningful by categorizing the different approaches to suicide prevention. While such categorization may be useful, it is important to find ways to qualify such modes of simplification. Stoor et al. use Supplementary Material to this end. 

I see the possibility of this form of thematic analysis and even the use of coding to help organize large amounts of research material in the manner applied here. I stress, however, the need to acknowledge that any imposed categorisation could be otherwise and to provide, to the best of one’s ability, the full richness of the identified problem representations. The goal in a WPR analysis is to find ways to highlight the complexity of the heterogeneous factors involved in making lives and worlds and to resist the temptation to simplify for the sake of academic conventions. 

HAVE A HAPPY HOLIDAY SEASON!

REFERENCES

Alvesson, M., & Sandberg, J. 2013. Constructing research questions: Doing interesting research. London: Sage.

Bacchi, C. 2009. Analysing Policy: What’s the Problem Represented to be? Frenchs Forest: Pearson Education.

Bacchi, C. 2023. Bringing a “What’s the problem represented to be?” approach to music education: A National Plan for Music Education 2022. Music Education Research, DOI: 10.1080/14613808.2023.2223220 

Bacchi, C. and Bonhan, J. 2016. Poststructural Interview Analysis: Politicizing “personhood”. In C. Bacch and S. Goodwin, Poststructural Policy Analysis: A Guide to Practice. NY: Palgrave Macmillan. pp. 113-122. 

Bacchi, C. and Goodwin, S. 2016. Poststructural Policy Analysis: A Guide to Practice. NY: Palgrave Macmillan. 

Braun, V. & Clarke, V. 2006. Using thematic analysis in psychology, Qualitative Research in Psychology, 3:2, 77-101, DOI: 10.1191/1478088706qp063oa 

Braun, V. & Clarke, V. 2021. One size fits all? What counts as quality practice in (reflexive) thematic analysis?, Qualitative Research in Psychology, 18:3, 328-352, DOI: 10.1080/14780887.2020.1769238 

Byrne, D. 2022. A worked example of Braun and Clarke’s approach to reflexive thematic analysis. Quality & Quantity, 56: 1391-1412. 

St. Pierre, E. A. 2019. Post qualitative inquiry, the refusal of method, and the risk of the new. Qualitative Inquiry. online first doi: 10.1177/1077800419863005.

Stoor, J. P. A, Eriksen, H. A. and Silviden, A. C. 2021. Making suicide prevention initiatives targeting Sámi in Nordic countries. BMC Public Health, 21:2035. https://doi.org/10.1186/s12889-021-12111-x  

Taylor, G.W. and Ussher, J.M. 2001. Making sense of S&M: a discourse analytic account. Sexualities 4, 293/314