Explainer · Plain-language

What Is Thematic Analysis? Steps, Types & Examples | CASRAI

Thematic analysis is a qualitative research method for identifying, analysing, and interpreting patterns of meaning (themes) within a dataset. It is one of the most widely used methods in qualitative research, particularly in health, psychology, and social sciences.

CASRAI plain-language explainers — clear answers to recurring research-administration questions

The step most authors miss

Doing CRediT right? Don’t stop at the statement.

A CRediT statement credits you inside one paper. The recognition CRediT was built for happens when those roles are tied to you, persistently. Sign in with your ORCID — free — and claim your CRediT contributions on casrai.org, the home of the standard. They become a verified, portable part of your identity, not a line that disappears into one PDF.

Free: claim your contributions, then export a journal-ready CRediT statement, schema.org structured data, JATS XML, CSV or BibTeX — and preview your public profile. A membership publishes that profile publicly and verifies the journals you serve.

Braun and Clarke’s six-phase framework

Virginia Braun and Victoria Clarke’s 2006 paper “Using thematic analysis in psychology” in Qualitative Research in Psychology is among the most cited qualitative methods papers in the social sciences. Their six-phase framework provides a structured but flexible approach: (1) familiarising yourself with the data (reading, noting initial ideas); (2) generating initial codes (systematically labelling features of the data relevant to the research question); (3) searching for themes (collating codes into potential themes); (4) reviewing themes (checking themes against the dataset — do they work?); (5) defining and naming themes (producing clear definitions and memorable names); (6) writing up (producing the analysis that tells the story of the data). This framework transformed thematic analysis from an implicit, taken-for-granted practice into a named, teachable method with explicit procedures.

Inductive vs deductive, and latent vs semantic themes

Thematic analysis can be conducted inductively (themes emerge bottom-up from the data, without a prior theoretical framework — also called data-driven TA) or deductively (themes are identified top-down from an existing theory or framework — also called theory-driven or directed TA). This maps onto the inductive–deductive dimension of qualitative research more broadly. Themes can also operate at two levels: semantic (explicit, surface-level content — what participants say) or latent (underlying meanings, assumptions, and ideologies that shape what participants say). Latent thematic analysis is more interpretive and associated with constructivist or critical epistemologies. Braun and Clarke distinguish their approach from content analysis by its interpretive rather than counting orientation, and by the theoretically flexible way themes are constructed.

Reflexive thematic analysis — Braun and Clarke’s later development

In the 2010s, Braun and Clarke refined their approach into “reflexive thematic analysis” (RTA) to distinguish it from the codebook or template approaches that had borrowed their six-phase structure but applied it in a more positivist, reliability-focused way. RTA treats themes as constructions rather than discoveries — as interpretive accounts of the data, not pre-existing patterns waiting to be found. Researcher reflexivity (explicitly examining how the researcher’s perspective shapes the analysis) is central. The analyst’s active role in generating themes is a feature, not a bias to control. RTA rejects inter-rater reliability as a quality criterion, arguing that agreement between coders is irrelevant when the goal is interpretive depth rather than frequency counts. This debate clarified a genuine methodological fault line in how thematic analysis is used.

Quality, rigour and common errors

Quality in thematic analysis is assessed against criteria suited to qualitative research rather than reliability statistics. Braun and Clarke propose reflexivity, transparency, and fit between the analytical approach and the theoretical framework as key criteria. Common errors include: mistaking a topic for a theme (a theme makes a claim about the data, e.g. “participants experienced stigma as isolating” — not just “stigma”); weak themes that do not capture a coherent pattern; using quotations as themes rather than as evidence for analytical claims; conducting TA without specifying the theoretical position; and applying a deductive framework while claiming to be inductive. Yardley’s criteria (sensitivity to context, commitment and rigour, transparency, and reflexivity) are also widely cited.

Key facts

At a glance

Definition: Identifying, analysing and reporting patterns of meaning (themes) in qualitative data
Key paper: Braun & Clarke (2006) — most-cited qualitative methods paper
Six phases: Familiarise, code, search for themes, review, define, write up
Two axes: Inductive vs deductive; semantic vs latent themes
Later work: Reflexive thematic analysis (Braun & Clarke) — themes as constructions
Quality: Reflexivity, transparency, fit with theory — not inter-rater reliability
Common error: Mistaking a topic for a theme — a theme makes an analytic claim

Common misconceptions

What people often get wrong

Often heard: Thematic analysis and content analysis are the same method.

Actually: No — content analysis counts and categorises textual features (quantitative or qualitative); thematic analysis constructs interpretive accounts of patterns of meaning. TA is interpretive throughout; manifest content analysis is primarily descriptive and can be quantitative.

Often heard: Themes in thematic analysis are patterns that exist in the data independently.

Actually: No — in reflexive thematic analysis (Braun & Clarke), themes are constructed by the analyst rather than discovered. The analyst’s theoretical position, research question, and interpretive lens actively shape what counts as a theme.

Often heard: Thematic analysis requires inter-rater reliability to demonstrate rigour.

Actually: No — Braun and Clarke explicitly argue that inter-rater reliability is inconsistent with the interpretive, constructivist foundations of reflexive TA. Rigour is demonstrated through reflexivity, transparency, and the quality of the interpretive account, not through coder agreement.

Going deeper