: I like to define thematic analysis as the active recognition of patterns (in behaviours, experiences, expression, etc.) across a data set to generate themes that hold meaning in relation to a predefined theory and or set of questions.
: All of the emphasis in that statement is mine, because those words in italics are very important. Thematic analysis involves looking at patterns and, by extension, any exceptions there may be to those patterns. So, it is at least as much about seeing what is not there in the data as about seeing what is there. That involves active interpretation across your entire data set.
: It’s really like the concise framing of the insights that you are getting across the data set and then constructing them into a coherent whole. And you do need multiple sources in that data set to strengthen your evidence base. This is why, as a PhD student, you may find your supervisor needling you to broaden your references or your range of sources – because the broader your data set is, the more reliable the themes that you are generating are.
: So, across your data set, you generate themes that hold meaning in relation to a predefined theory and or set of questions. Now, this is not an attempt at a textbook definition. It’s something that you can maybe go back and reflect on to see how you can apply it in your own work. And then, when you have had a chance to do some thematic analysis yourself, you retrospectively realise that all these words and phrases are actually important when you’re talking about this method of analysis.
: Bear in mind that, from a critical thinking perspective, you also need to be explicit about your assumptions when you’re doing thematic analysis. Again, all of these building blocks of critical writing are interlinked.
: If you are conducting qualitative research, you are likely to gather the data you need through interviews, focus group discussions, document reviews – all of which you’ll probably find exciting and stimulating. However, your data often comes out the other end as plain, boring text.
: Even when you have original audio or video clips as your sources, you typically need to reduce all of that exciting verbal and visual information to text. And, sometimes, it can look like just like ink on paper, you know, the words seem little more than abstract print on paper. So, the data is sitting there and it just feels like a sea of words running into many, many, many pages sometimes.
: That is, until – and unless – you as a researcher and analyst begin to impose a structure on it. It is that process of formulating the patterns and looking across them that brings out the themes you need. They are not going to emerge on their own. The themes that you do draw out are contingent on your data, but also on your theoretical framework, the particular angle that you’re bringing to bear on the subject.
: This means that it is quite possible for two people to work with a similar data set, the same data set. But, if they’re looking at it from different perspectives, bringing in different theoretical frameworks, different research questions, they can end up constructing different meanings out of that same data set.
: This is why it is important to be explicit throughout your work about what frameworks you’re using and what questions you are asking and attempting to respond to – so that, in the context of your work, the themes that you’re drawing out do make sense and the conclusions follow logically from the evidence you have presented.
: The more you practice thematic analysis, the more intuitive it becomes for you and the more adept you become at it. However, frameworks exist in the literature that newcomers to the method may find especially useful. The framework I want to highlight in this module, just for its practicality, is one I adapted from the 2006 paper titled “Using thematic analysis in psychology,” by Braun and Clark.
: The framework consists of six consecutive, albeit overlapping, steps. One, know your data. Two, generate codes. Three, construct themes. Four, refine those themes. Five, substantiate the themes. And six, write up your thesis or paper. Now, I had been using thematic analysis in my work long before I happened upon this framework.
: But the way that my process fitted retroactively into these steps validated my approach and heightened my appreciation of the framework. It was rather gratifying to realise that there was a method to the madness I had been practising all along. So, I am not just highlighting this framework here because it’s nice – to add yet another formula or model to the tons you likely already have, but because I know from experience that it works.
: The framework gives us six simple steps that are to be held flexibly, partly because there is quite a lot of overlap in them, especially between steps four and six, but also because you ideally go back and forth in your analysis rather than following a series of linear steps. Again, the more you do this kind of analysis, the more intuitively it starts to come to you.
: To make these steps more relatable, I am going to be illustrating them with a sample action research project I conducted between 2018 and 2019 as I go along. The overarching aim of this project was to interrogate the extent to which policymakers in Nigeria’s energy sector deploy scientific evidence in their decision making, and to identify opportunities for enabling science-policy exchanges in the sector.
: This overall aim was then broken down into several smaller questions that lend themselves to qualitative inquiry. Question one: what factors do policymakers consider when choosing what household energy options to prioritise, and what are the relative weights of these factors in decision making?
: There are two other questions – questions two and three – that we will not be focusing on for the purposes of this unit, but just to highlight how this was a broader project that dealt with those questions. Question four: what, if any, needs exist for new forms of evidence, and how can these be incorporated into institutional decision making processes?
: Now, as I said earlier, questions one and four, in bold type, are the ones that most relate to the codes and the themes that I’m going to be highlighting later on in this unit and in subsequent units.