thematic-analysis

To turn interview transcripts into candidate themes, the reflexive thematic analysis process applies open meaning-unit coding to raw text, aggregates codes into a pool, and clusters them into 5–8 structured candidate themes with boundary-uncertain flags.

Reflexive thematic analysis differs from grounded coding by enforcing descriptive, in-vivo preference coding at the meaning-unit level to structure candidate themes, rather than developing theory iteratively from the data ground up.

Yes, you can cluster existing initial codes into themes without raw interview text by using input mode B, which skips transcript coding and directly aggregates your compiled code list into structured candidate theme clusters.

To review candidate themes for internal consistency and research-question relevance, the workflow evaluates thematic structures across external distinction checks, ensuring each cluster maintains clear boundaries and aligns with the core research query.

Braun and Clarke thematic analysis provides multiple researcher-focused naming suggestions for each candidate theme, allowing researchers to select the most appropriate label while explicitly flagging boundary-uncertain coding labels for review.