If you are working on a qualitative research project — whether it is a PhD thesis, a master's dissertation, or a journal paper — chances are you will need to analyse textual data such as interview transcripts, survey responses, or field notes. Thematic analysis is one of the most widely used and accessible methods for doing exactly that. It helps you identify, organise, and report patterns (themes) within your data, turning raw qualitative information into meaningful findings.
This guide walks you through the entire thematic analysis process using the influential Braun and Clarke (2006) six-phase framework, which remains the gold standard cited in thousands of published studies. Whether you are new to qualitative research or looking to strengthen your methodology chapter, this article will give you the practical knowledge you need.
What Is Thematic Analysis?
Thematic analysis is a qualitative data analysis method that involves reading through a dataset — such as interview transcripts, focus group recordings, open-ended survey responses, or documents — and identifying recurring patterns of meaning. These patterns are called themes.
Unlike methods such as grounded theory or discourse analysis, thematic analysis is not tied to a particular theoretical framework. This flexibility is one of its greatest strengths. You can use it within a positivist, interpretivist, or critical realist paradigm. You can apply it inductively (themes emerge from the data) or deductively (themes are guided by existing theory). This makes it suitable for almost any discipline: education, psychology, nursing, management, sociology, and beyond.
Braun and Clarke (2006) originally described thematic analysis as a "method for identifying, analysing, and reporting patterns (themes) within data." They later refined their approach into what they call reflexive thematic analysis, emphasising that the researcher actively constructs themes rather than passively discovering them. This distinction matters because it places responsibility on you, the researcher, to be transparent about the choices you make during analysis.
When Should You Use Thematic Analysis?
Thematic analysis is a good fit when:
- You have qualitative data (text-based) and want to find common patterns across participants or sources.
- You are working within a flexible theoretical framework and do not need to commit to a single epistemological position.
- You are relatively new to qualitative research and need a method that is systematic but not overly prescriptive.
- Your research questions ask "what" or "how" questions about experiences, perceptions, or meanings.
- You want results that are accessible to a broad audience, including policymakers and practitioners.
It is less suitable when you need to analyse the structure of language itself (use discourse analysis), build new theory from data (use grounded theory), or focus on individual lived experiences in depth (use interpretive phenomenological analysis).
The Braun and Clarke Six-Phase Framework
The most widely cited approach to thematic analysis comes from Virginia Braun and Victoria Clarke. Their six-phase framework provides a clear, structured process without being rigidly linear. You will move back and forth between phases as your understanding deepens. Here is each phase explained in practical detail.
Phase 1: Familiarising Yourself with the Data
Before you code a single word, you need to know your data intimately. This phase is about immersion. If you conducted the interviews yourself, you already have a head start, but you still need to engage with the data systematically.
What to do:
- Transcribe your audio or video recordings verbatim. Include pauses, laughter, and emphasis where relevant. If you are working with survey responses, compile them into a single document.
- Read through the entire dataset at least twice. On the first pass, read without taking notes — just absorb. On the second pass, start jotting down initial observations and ideas.
- Keep a reflexive journal. Note your reactions, assumptions, and early impressions. This is especially important for international students who may be analysing data from a culture different to their own.
Common mistake: Rushing through this phase. Many researchers jump straight to coding and miss the big picture. Spending adequate time here will save you hours of confusion later.
Phase 2: Generating Initial Codes
Coding is the process of labelling segments of your data that are relevant to your research questions. A code is a short, descriptive tag that captures the essence of a data segment.
What to do:
- Go through each transcript or document line by line. Highlight anything interesting or relevant and assign a code to it.
- Be specific. Instead of coding a paragraph as "motivation," use codes like "intrinsic motivation from personal interest" or "extrinsic motivation from career pressure."
- Code inclusively — if in doubt, code it. You can always discard codes later, but you cannot recover data you skipped.
- Use software tools such as NVivo, ATLAS.ti, or even a simple spreadsheet to organise your codes. For smaller datasets, colour-coded highlighting in a Word document works fine.
Example: Suppose a participant says, "I chose this university because the fees were affordable and my cousin studied here." You might apply two codes: "financial accessibility" and "family influence on decision."
By the end of this phase, you should have a long list of codes across your entire dataset. It is normal to have dozens or even hundreds of codes at this stage.
Phase 3: Searching for Themes
Now you step back from individual codes and look for broader patterns. A theme is not just a code that appears frequently — it is a meaningful pattern that tells a story about your data in relation to your research questions.
What to do:
- Print out or display all your codes. Group related codes together. You can use sticky notes on a wall, a whiteboard, or a digital mind-mapping tool.
- Ask yourself: What do these codes have in common? What story do they tell together?
- Create candidate themes. At this stage, they are provisional. You might have main themes and sub-themes.
- Some codes will not fit neatly into any theme. Set them aside in a "miscellaneous" group for now — you may return to them later.
Tip: A good theme has a clear central concept, is distinct from other themes, and is supported by sufficient data. If a theme only has one or two data extracts, it may be better as a sub-theme or code.
Phase 4: Reviewing Themes
This is a quality-check phase. You need to verify that your candidate themes actually work in relation to both the coded extracts and the full dataset.
What to do:
- Level 1: Re-read all the coded extracts for each theme. Do they form a coherent pattern? If not, consider splitting the theme, merging it with another, or reworking it entirely.
- Level 2: Re-read the entire dataset with your themes in mind. Do the themes accurately reflect the meanings in the data as a whole? Are there important patterns you missed?
- Create a thematic map — a visual diagram showing how your themes and sub-themes relate to each other and to your overarching research question.
This phase often involves several rounds of revision. Do not be discouraged if your initial themes change significantly. That is a sign of rigorous analysis, not failure.
Phase 5: Defining and Naming Themes
Once you are satisfied that your themes are robust, you need to clearly define what each theme is about and give it a concise, informative name.
What to do:
- Write a detailed description of each theme (two to three paragraphs). Explain what the theme captures, how it relates to your research questions, and what story it tells.
- Identify the boundaries of each theme. What does it include and what does it not include?
- Choose theme names that are concise but descriptive. Avoid single-word names like "motivation." Instead, use phrases like "Financial constraints as the primary driver of university choice" that convey the theme's specific meaning.
- Determine the relationship between themes. Are they all at the same level, or do some contain sub-themes?
Phase 6: Writing the Report
The final phase is where you weave your themes into a coherent narrative. This is not merely a list of themes with supporting quotes — it is an analytical argument that addresses your research questions.
What to do:
- Present each theme with a clear topic sentence, followed by data extracts (direct quotes from participants) that illustrate the theme.
- Provide analytical commentary after each extract. Do not let the data speak for itself — explain what the extract shows and why it matters.
- Connect your themes back to your research questions, existing literature, and theoretical framework.
- Be transparent about your analytical process. Discuss how you coded, how themes were developed, and what decisions you made along the way.
Structure for your findings chapter: Start with a brief overview of all themes (a summary table or thematic map works well), then dedicate a section to each theme with sub-sections for sub-themes. End with a synthesis that shows how the themes work together to answer your research questions.
Inductive vs. Deductive Thematic Analysis
One of the first methodological decisions you need to make is whether your analysis will be inductive or deductive.
- Inductive (bottom-up): Themes are derived directly from the data without trying to fit them into a pre-existing framework. This is ideal for exploratory research where you do not have strong prior expectations about what you will find.
- Deductive (top-down): You approach the data with a set of pre-determined codes or categories based on existing theory or literature. This works well when you are testing or extending a known framework.
Many studies use a hybrid approach, starting with some deductive codes from the literature and allowing additional inductive codes to emerge during analysis. Whichever approach you choose, state it clearly in your methodology chapter and justify your choice.
Ensuring Rigour and Trustworthiness
Qualitative research is sometimes criticised for being subjective. You can strengthen the credibility of your thematic analysis by following these practices:
- Thick description: Provide rich, detailed accounts of your data and analysis process so readers can judge the quality of your interpretations.
- Reflexivity: Acknowledge your own positionality, biases, and how they may have shaped the analysis. Keep a reflexive journal throughout.
- Peer debriefing: Discuss your codes and themes with a colleague or supervisor. A fresh perspective can reveal blind spots.
- Audit trail: Document every decision you make during analysis — why you merged codes, split themes, or discarded data. This trail allows others to follow your reasoning.
- Member checking: Where appropriate, share your findings with participants and ask if the themes resonate with their experience.
Note that Braun and Clarke have cautioned against using inter-rater reliability (where two coders independently code the same data and compare results) as a quality measure for reflexive thematic analysis. They argue that coding is an interpretive act, not a technical one, and that agreement between coders does not necessarily indicate quality.
Common Mistakes in Thematic Analysis
Based on years of reviewing student dissertations, here are the most frequent errors we see:
- Themes that are actually just codes: A theme like "salary" is too narrow. Themes should capture broader patterns, such as "Economic considerations outweigh personal passion in career decisions."
- Using data collection questions as themes: If your interview guide has five questions, your themes should not simply mirror those five questions. Themes should cut across questions and reveal deeper patterns.
- No analytical depth: Simply paraphrasing what participants said is not analysis. You need to interpret the data, explain what it means, and connect it to your research questions and the wider literature.
- Ignoring data that does not fit: If some participants contradict a theme, do not hide this. Discuss it openly. Contradictions often lead to richer, more nuanced findings.
- Skipping phases: Each phase of the Braun and Clarke framework serves a purpose. Jumping from coding straight to writing without reviewing and refining your themes will result in a superficial analysis.
Tools and Software for Thematic Analysis
You do not need expensive software to conduct thematic analysis, but the right tools can make the process more efficient, especially with large datasets.
- NVivo: The most popular qualitative data analysis software. Excellent for managing large datasets, coding, and visualising relationships between themes. Available through many university licences.
- ATLAS.ti: A powerful alternative to NVivo with strong network visualisation features.
- MAXQDA: User-friendly with good mixed-methods capabilities.
- Microsoft Excel or Google Sheets: Perfectly adequate for smaller datasets. Create columns for participant ID, data extract, code, and theme.
- SPSS: While primarily a quantitative tool, researchers conducting mixed-methods studies often use SPSS for the quantitative component alongside thematic analysis for qualitative data. If your study involves both survey data and interview transcripts, combining statistical analysis with thematic analysis gives you a more complete picture.
How to Present Thematic Analysis in Your Thesis
Your methodology chapter should include:
- A clear statement that you used thematic analysis, citing Braun and Clarke (2006) or their updated reflexive thematic analysis framework (Braun & Clarke, 2019, 2021).
- Whether your approach was inductive, deductive, or hybrid, and why.
- A description of how you moved through the six phases.
- The software or tools you used for coding.
- How you ensured rigour (reflexivity, peer debriefing, audit trail, etc.).
Your findings chapter should present each theme with:
- A clear heading (the theme name).
- An opening paragraph explaining what the theme captures.
- Two to four illustrative quotes from different participants, each followed by your analytical commentary.
- A closing paragraph linking the theme to your research questions.
Key References for Your Literature Review
If you are using thematic analysis, you should cite these foundational works:
- Braun, V. & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101.
- Braun, V. & Clarke, V. (2019). Reflecting on reflexive thematic analysis. Qualitative Research in Sport, Exercise and Health, 11(4), 589–597.
- Braun, V. & Clarke, V. (2021). Thematic analysis: A practical guide. SAGE.
- Clarke, V. & Braun, V. (2017). Thematic analysis. The Journal of Positive Psychology, 12(3), 297–298.
Final Thoughts
Thematic analysis is powerful precisely because it is flexible. It can be applied across disciplines, theoretical frameworks, and data types. But flexibility does not mean anything goes. A rigorous thematic analysis requires systematic engagement with your data, transparent decision-making, and analytical depth that goes beyond surface-level description.
If you follow the Braun and Clarke six-phase framework carefully, document your process, and stay reflexive about your own influence on the analysis, you will produce findings that are credible, insightful, and publishable.
Need help with your qualitative data analysis, statistical testing, or thesis writing? Our team of experienced researchers can guide you through every stage. Explore our Data Analysis & SPSS services or reach out directly to discuss your project.