Thematic analysis is a method for identifying, analyzing and reporting patterns of meaning, called themes, in qualitative data such as interview transcripts. The most widely used approach, developed by Virginia Braun and Victoria Clarke, has six phases: familiarize yourself with the data, generate codes, construct initial themes, review the themes, define and name them, and write the report. This guide explains each phase with an interview example.
When to use thematic analysis
Thematic analysis suits research questions about people's experiences, views, perceptions and practices, for example:
How do first-year students who work part-time describe balancing work and study?
It is flexible: it works with interviews, focus groups, open survey answers, diaries and documents, and with many theoretical approaches. That makes it a popular choice for bachelor's and master's theses. See qualitative vs quantitative research.
Key terms
- Code: a short label for a feature of the data relevant to your question, such as "working late shifts" or "guilt about missed lectures".
- Theme: a broader pattern of shared meaning across the data, built from several codes, such as "Study always comes second".
- Inductive analysis: themes come from the data.
- Deductive analysis: codes and themes are guided by existing theory or concepts.
The six phases
1. Familiarize yourself with the data
Transcribe the interviews (or check automatic transcripts carefully), read them several times, and note first impressions.
Several students mention feeling "always tired" and describe their manager's attitude to exam periods.
2. Generate codes
Work through the data systematically and label every segment relevant to your research question. Code inclusively; one segment can have several codes.
| Extract | Codes |
|---|---|
| "I work Friday and Saturday nights, so Monday lectures are a write-off." | Weekend night shifts; missing lectures; tiredness |
| "My manager lets me swap shifts in exam weeks if I ask early." | Flexible manager; planning ahead |
| "I feel guilty skipping seminars, but I need the money for rent." | Guilt; financial necessity |
3. Construct initial themes
Group related codes into candidate themes and collect the extracts for each.
- "Money comes first": financial necessity, rent, guilt.
- "The tired student": night shifts, tiredness, missing lectures.
- "Negotiating flexibility": flexible manager, planning ahead, swapping shifts.
4. Review the themes
Check each theme against its extracts and against the whole data set:
- Is there enough data to support it?
- Is it distinct from the other themes?
- Does it tell a coherent story about your research question?
Merge, split or discard themes as needed. A thematic map (a diagram of themes and their links) can help.
5. Define and name the themes
Write a short definition of each theme, explaining its core idea and how it relates to the others and to your question. Choose names that capture the meaning, often using participants' own words.
6. Write the report
Present each theme with a clear explanation and well-chosen quotations, and interpret what it means. Do not just list codes or summarize what participants said: explain the pattern and connect it to your research question and the literature.
Participants described their paid work not as a choice but as a condition of studying at all. As one student put it, "I need the money for rent." This sense of necessity shaped how they justified missed classes...
Reflexive thematic analysis
Braun and Clarke now call their approach reflexive thematic analysis, to emphasize that themes are actively constructed by the researcher, not simply "found" in the data. In your methodology chapter, reflect on how your own background and assumptions may have shaped your interpretation.
Software
Thematic analysis can be done with highlighters and a spreadsheet, or with qualitative data analysis software such as NVivo, ATLAS.ti, MAXQDA or the free QualCoder. Software organizes your codes; it does not do the analysis for you.
Ensuring quality
- Keep an audit trail of your coding decisions.
- Discuss codes and themes with your supervisor or a peer.
- Use quotations from a range of participants, not just the most eloquent.
- Describe negative cases that do not fit a theme.
Writing it up in your thesis
- Methodology: explain why you chose thematic analysis, which approach you followed (for example Braun and Clarke's six phases), whether your coding was inductive or deductive, and how you ensured quality. See how to write the methodology chapter.
- Findings: one section per theme, with quotations.
- Discussion: relate your themes to the literature. See how to write the discussion chapter.
Start with good questions
Strong research questions make analysis easier. The free research question generator suggests qualitative questions for your topic.
Frequently asked questions
How many themes should I have?
Usually two to six for a bachelor's or master's thesis. Fewer, well-developed themes are better than many thin ones.
What is the difference between a code and a theme?
A code labels a specific feature of the data. A theme is a broader pattern of meaning that brings several codes together.
Can I use AI to code my interviews?
Some tools can suggest codes, but interpretation is the researcher's job, and uploading interview data to an AI tool may break your ethics approval or data protection rules. Check with your supervisor and ethics committee first.
Is thematic analysis the same as content analysis?
No. Content analysis often counts how frequently categories occur; thematic analysis focuses on patterns of meaning.