The most-read route from raw qualitative data to a named theoretical contribution.
Thematic analysis is the most popular technique for analysing qualitative data, but researchers are often confused about how to present their final outcome as a conceptual model. STA answers that problem. It is termed “systematic” because it follows a structured, sequential approach: themes are developed from research gaps, objectives, questions and theoretical underpinnings — not merely from surface commonalities in the data. The process is adaptable across constructivist, positivist, grounded theory, ethnographic, narrative and interpretive phenomenological methodologies.
Transcribe, familiarise, and select quotations that carry participants’ voices
Extract powerful, meaningful words from the chosen quotations
Label data segments with short phrases capturing their core message
Group codes into patterned meanings linked to research questions
Interpret keywords, codes & themes into defined concepts
Synthesise concepts into a named model of arrows and boxes
Immerse yourself in the data. A selective transcript is more useful than an exhaustive one: foreground quotations that resonate with the research objectives, reflect diverse participants’ voices, and balance readability with authenticity. Power quotes — “the most compelling bits of data you have” — illustrate your points most effectively.
A novel step in thematic analysis. Identify recurring, meaningful words and phrases within the chosen quotations that encapsulate participants’ experiences and perceptions. Keywords pave the way for code development and amplify the depth and quality of interpretation.
Assign short phrases that capture each data segment’s core message and significance. Coding transforms complex textual data into a theoretical form and converts raw data into insightful, manageable units.
Organise codes into meaningful groups. Themes are more than recurring elements — they embody patterned meanings that link the research questions and the data, moving the analysis from detail to abstraction.
Interpret keywords, codes and themes together to define the concepts emerging from the data, assessed for clarity, accuracy, reliability, applicability and contribution to theory and practice.
Synthesise all concepts into a named model that answers the research questions and states the study’s contribution. As the paper notes: “Theorizing is also work” — many studies fail at publication because this conceptual work was never done.
The paper demonstrates every step using a real published study — Naeem and Ozuem’s (2022) research on COVID-19 panic buying — showing how statements such as “buying groceries means welcoming infection in our family” became keywords (“welcoming infection”), then codes (“provoke perception”), then themes (“perceived consequences”), and finally the “Panic buying as social practice” model grounded in Protection Motivation Theory and the Theory of Rumor Transmission.
Braun and Clarke’s reflexive thematic analysis identifies themes as patterns of shared meaning. STA differs in three ways: it adds keyword selection as an explicit step between quotations and codes, creating an auditable chain of evidence; its themes are anchored to research gaps, objectives, questions and theoretical underpinnings rather than commonalities alone; and it does not stop at themes — its defined end point is a conceptual model that makes a theoretical contribution.
Yes. The 2023 paper demonstrates that STA is adaptable to grounded theory, ethnographic and narrative approaches and, with some adaptation, to more descriptive positivist methodologies as well.
Quotations are the foundation of the analysis. They should symbolise robust patterns in the data, incorporate a diverse range of participants’ voices, and be presented with enough context for readers to comprehend their essence.
The article is open access in the International Journal of Qualitative Methods at https://doi.org/10.1177/16094069231205789 — one of the journal’s most-read methodology papers.