Thematic analysis examines patterns of meaning in qualitative data. This guide uses Braun and Clarke’s reflexive approach: engage with transcripts, code relevant meanings, develop themes and interpret them in relation to the question. The researcher actively develops the analysis. A frequent word or an interview-question heading is not automatically a theme.
Which approach are you using?
Thematic analysis is a family of approaches. Name your chosen version and explain its theoretical assumptions. Braun and Clarke’s guidance describes reflexive work as recursive rather than a rigid sequence. Do not combine procedures from different traditions without justification.
Agreement between independent coders is not a universal requirement for reflexive thematic analysis. The authors discuss inappropriate universal quality criteria in One size fits all?. Collaboration may enrich interpretation, but explain its purpose rather than treating consensus as the only sign of rigour.
How should you prepare the transcript?
Check it against the recording, use consistent identifiers and protect personal information. Explain transcription decisions when they affect interpretation. Read individual interviews and the dataset as a whole. Record initial observations, uncertainties and assumptions about the topic. An extract can mean something different when removed from its conversational context.
What is the difference between a code and a theme?
A code labels an analytically relevant meaning in an extract. An extract may receive several codes. A theme develops a broader pattern around a central organising idea. The authors’ FAQ distinguishes this from simply summarising answers about the same topic.
These three statements are synthetic teaching material. They are not real interviews or participant data. The example question concerns working students’ experiences of accessing consultations.
| Synthetic extract | Possible code |
|---|---|
| The only appointment is during my work shift. | Appointment conflicts with work |
| When I miss it, I feel the university assumes we do not work. | Support assumes control over time |
| I can attend an online appointment during my break. | Online access reduces the timing conflict |
Candidate theme for this exercise: Access to support depends on control over personal time
The candidate theme illustrates a line of interpretation. Three invented extracts cannot establish a theme for a real research project. Review your own themes against the complete dataset and question, including differences that challenge the initial account.
How do the six phases guide practice?
Familiarise yourself with the data; code across the dataset; develop initial themes; review and develop those themes; refine their scope and names; and write an analytical account using extracts and literature. These phases allow movement backwards and forwards. Explain actual decisions, including themes you split, combined or abandoned, rather than listing phase names alone.
Keep reflexive notes about how your position shapes interpretation. Software can help organise extracts, but storing material in a programme does not produce an analytical argument.
How do you write the findings?
Explain each theme’s central meaning and relationship to the question. Use extracts to support an analytical narrative rather than presenting a quotation list. In the example, online access may reduce one scheduling conflict; this does not establish that it solves access for every student.
Avoid presenting the most common words as findings or copying interview questions as theme names. Justify dataset scope and analytical depth rather than borrowing a universal interview or theme count. The downloadable sheet contains the same synthetic exercise. For real transcripts, follow participant agreements and institutional requirements for confidentiality and approved tools.
FAQ
Can one extract have several codes?
Yes, where they capture relevant meanings. Codes may change during analysis.
Is an interview question a theme?
Not automatically. Reflexive themes develop patterned meaning around a central idea.
Must I calculate inter-coder agreement?
Not as a universal reflexive-analysis rule. Quality criteria must fit the chosen approach.
Are the example statements real data?
No. They are synthetic material for explaining analysis, not research findings.
Downloadable worksheet
Related guides
- Case Study Research for Your Dissertation: Design, Structure and Worked Example
- Dissertation Aims and Objectives: How to Write Them with Examples
- Tables and Figures in a Dissertation: Titles, Sources and Interpretation
- Qualitative or Quantitative Research: Choosing a Design for Your Dissertation
- The Discussion Section in a Thesis: How to Interpret Results