Thematic analysis is the most widely used method for making sense of qualitative data — and the most often done loosely. Rigour comes from following a transparent process, so a reader can see how you moved from raw transcripts to the themes you report.
The six phases (Braun & Clarke)
- 1Familiarisation — read and re-read the data; note first impressions.
- 2Coding — label meaningful segments systematically across the whole dataset.
- 3Generating themes — cluster related codes into candidate themes.
- 4Reviewing themes — check themes against the coded data and the full dataset.
- 5Defining themes — name each theme and pin down what it captures.
- 6Writing up — build the narrative with vivid, representative quotes.
Codes are not themes
A common mistake is treating codes as themes. Codes are granular labels ('fear of failure', 'lack of feedback'); a theme is a broader pattern of meaning that several codes build toward ('perceived lack of support'). Themes are constructed, not found lying in the data.
Keep an audit trail — your codebook, sample coded extracts, and notes on how themes evolved. It's what lets you defend your analysis as systematic rather than impressionistic when an examiner asks how you got here.
Tools and support
Software like NVivo helps organise codes and keep the audit trail, though the interpretation is always yours. Build clean data first with strong interview and focus group guides, and see qualitative research mentoring and the data analysis hub for end-to-end help.
Frequently asked
How many themes should I have?+
Typically three to six main themes for a study — enough to capture the data's richness without fragmenting it. What matters is that each theme is well-supported by coded evidence, not the count itself.
Do I need software for thematic analysis?+
Not strictly — small datasets can be coded manually or in a spreadsheet. Tools like NVivo become valuable for larger datasets, team coding and maintaining a clear audit trail, but they organise your thinking rather than replace it.
phdguide's mentors are senior academics, former supervisors, statisticians and publication specialists with 25+ years of combined experience guiding MBA, MPhil and PhD scholars from topic to viva.
Ethical, compliant guidance: We provide academic support, mentoring, analysis, editing and structuring — not authorship. Your work stays compliant with university policies.