Several interviewees mention training, but they do not all mean the same thing. One wants help learning the software, another needs time away from customers to practise, and a third worries that asking for training will make them look incapable. Filing every extract under “training needs” preserves the topic while losing much of what the interviews revealed.
Thematic analysis develops and interprets patterns of meaning across qualitative material. It can work with interviews, diaries, observations and other records, helping researchers move from individual accounts towards an explanation that addresses the research question.
Choose an approach before applying its procedures
Thematic analysis is a family of approaches, rather than one universally agreed recipe. Some use structured codebooks and seek consistency between coders. Reflexive thematic analysis, associated with Virginia Braun and Victoria Clarke, treats the researcher’s interpretive work as central to developing themes.
Their explanation of thematic analysis distinguishes these approaches. The choice matters because quality checks should fit the method: inter-rater agreement is not a universal requirement for reflexive analysis, and asking several researchers to discuss their interpretations is not the same as testing whether they independently assign identical codes.
Describe the approach you are using and why it suits the study. A pragmatic product study can be systematic without claiming to follow a methodology whose assumptions and procedures it does not use.
Move from extracts to a pattern with meaning
Begin by becoming familiar with the full material, including context that may be lost in a transcript. Code passages relevant to the question, then explore relationships among them. Candidate themes need to be examined against both the extracts and the wider dataset, refined and given names that communicate the analysis. Braun and Clarke’s six-phase guide describes this as a recursive process, with movement back and forth as understanding develops.
In the fictional training example, “learning depends on permission to be inexperienced” might become a candidate theme if several accounts support it. That interpretation is more specific than “training”, but it also demands evidence. The researcher would need to examine whether it fits the material, whether it obscures other experiences and what it helps explain about the original question.
A more interpretive theme is not automatically better than one closer to participants’ explicit accounts. The appropriate depth depends on the research purpose, the material and the analytical commitments of the study.
Keep the explanation accountable to the evidence
Preserve links from claims to relevant extracts and record important analytical decisions. Examine cases that complicate the emerging account rather than choosing only quotations that make it sound convincing. A theme can be important without being the most frequently mentioned topic, especially when it explains a consequential experience.
Counts can sometimes provide descriptive context, but they need a clear unit and should not be mistaken for population prevalence. Ten comments may come from one participant, and silence on a topic may simply mean nobody was asked about it.
AI-generated summaries can assist navigation through material, but check them against the source and retain responsibility for the interpretation. Affinity mapping may help explore relationships visually; grouping notes is a working step, not evidence that a finished thematic analysis has taken place.
The final account should explain what the patterns mean for the research question, with enough evidence and context for a reader to assess the argument. A collection of headings with quotations beneath them is only useful when the analysis connects the two.