Thematic analysis is one of the most commonly used methods for analysing qualitative dissertation data — and one of the most commonly misunderstood. Students often produce a neat set of topic summaries and call them “themes,” which reads as descriptive rather than analytical to an examiner. Here’s how Braun and Clarke’s six-phase framework actually works, and where it most often goes wrong.
What Thematic Analysis Actually Is
Thematic analysis identifies, analyses, and reports patterns of meaning — themes — across a qualitative dataset such as interview transcripts, open-ended survey responses, or focus group discussions. Developed by Braun and Clarke (2006), it remains one of the most widely cited approaches to qualitative analysis across research disciplines, largely because it’s structured enough to teach clearly while staying flexible enough to fit very different kinds of research questions. (See the peer-reviewed overview published in ScienceDirect for the full academic framework.)
The Six Phases
Always confirm your specific department expects Braun and Clarke’s framework specifically, since other qualitative approaches exist — but where it’s used, the process runs through six distinct phases:
- Familiarisation with the data — transcribing (if needed), then reading and re-reading your data thoroughly before any coding begins. This stage is easy to rush, but genuine immersion in the data is what makes later interpretation possible.
- Generating initial codes — systematically labelling interesting or relevant features across the entire dataset, not just the sections that seem obviously important on a first read
- Searching for themes — grouping related codes together into candidate themes that capture a genuine pattern of meaning
- Reviewing themes — checking whether each candidate theme actually holds up against the coded extracts it’s built from, and against the dataset as a whole
- Defining and naming themes — refining exactly what each theme captures and giving it a clear, specific name
- Producing the report — writing up your themes with supporting evidence, woven into a coherent analytic narrative rather than presented as a disconnected list
The Mistake That Costs the Most Marks: Themes vs Topic Summaries
This is the single most common weakness examiners flag in thematic analysis chapters. A genuine theme captures something meaningful in relation to your research question — a patterned response or shared meaning across your data. A topic summary just groups data by subject matter without that deeper interpretive layer. “Participants discussed workload” is a topic, not a theme; “Workload as a source of identity threat among early-career professionals” reflects genuine analytical interpretation. If your “themes” could be generated just by skimming your interview questions, they’re probably topic summaries in disguise.
Deductive vs Inductive Coding
Decide early, and state explicitly in your methodology, which approach you’re taking:
- Deductive coding — codes derived from existing theory or your research questions, applied to the data
- Inductive coding — codes that emerge directly from the data itself, without being driven by a pre-existing framework
- A combination of both is common and entirely legitimate — many studies use existing theory to guide initial coding while remaining open to genuinely new patterns emerging inductively
Using NVivo (or Doing It by Hand)
NVivo is widely used for managing thematic coding, especially for larger datasets — it helps organise codes, track how often they appear, and manage the sheer volume of text involved in even a modest set of interview transcripts. That said, thematic analysis doesn’t require software; many dissertations code manually using highlighting and a structured spreadsheet, particularly with smaller datasets. The software manages your coding — it doesn’t do the actual interpretive thinking for you.
It’s Iterative, Not Linear
Despite the numbered phases, thematic analysis is not a one-pass checklist. Braun and Clarke themselves describe it as a genuinely iterative, interpretive process — expect to move back and forth between phases, revising codes as your understanding of the data deepens and refining themes as new ones interact with ones you’d already defined. A first attempt that gets reworked substantially isn’t a failure; it’s how the method is meant to work.
Writing Up Your Analysis
Your write-up, typically in your results or findings chapter, should include:
- A brief overview of your analytic process and why Braun and Clarke’s framework suited your specific study
- A clear description of each theme, illustrated with well-chosen, representative quotations from your data
- Interpretation showing what each theme means in relation to your research question — not just what participants said, but why it matters
How ASH Helps
A thematic analysis chapter can be analytically sound and still lose marks if the write-up reads as descriptive rather than interpretive, or if theme names and structure aren’t clearly and consistently presented. ASH’s editing service checks exactly this kind of clarity, alongside the language corrections covered in our guide to why international students lose marks for English in the UK, and connects directly to your methodology chapter and results chapter.
If you want a second opinion on your thematic analysis chapter before you submit, message ASH on WhatsApp:
👉 Chat with ASH on WhatsApp: +92 313 1624960
(This WhatsApp number is ASH’s official business line.)
Related Guides
- How to Write a Dissertation Methodology Chapter
- How to Present Your Dissertation Results
- How to Write a Literature Review for Your UK Dissertation
- ASH Research Support Services
FAQs
What’s the difference between a theme and a topic summary?
A theme captures a genuine pattern of meaning in relation to your research question; a topic summary just groups data by subject without that interpretive layer. This distinction is where examiners most often flag weak thematic analysis.
Do I need NVivo to do thematic analysis?
No — NVivo helps manage coding for larger datasets, but thematic analysis can be done manually, especially for smaller sets of interviews, using highlighting and a structured spreadsheet.
Is thematic analysis a strictly linear, six-step process?
No — it’s genuinely iterative. Expect to move back and forth between phases as your understanding of the data develops, rather than completing each phase once and moving on permanently.
Should I use deductive or inductive coding?
Either can be appropriate, and many studies combine both — state your choice explicitly in your methodology and justify it against your specific research question.
Want a second opinion on your thematic analysis chapter before you submit? Message ASH on WhatsApp: +92 313 1624960.