Written and maintained by CASRAI Editorial Board
Last updated
Who this is for: a researcher with interview transcripts, open-ended survey answers or field notes who needs a defined sequence for analysing them and a free tool to do it in. After reading, you should be able to run the six phases of reflexive thematic analysis on your own dataset using Taguette and export what you need to write up.
Short answer: analyse qualitative data by reading the whole dataset, coding the parts relevant to your research question, grouping codes into candidate themes, reviewing and refining those themes against the data, and writing an analytic narrative with extracts. Braun and Clarke describe this as six phases of reflexive thematic analysis, and stress that the phases are not a rigid procedure.
What it is used for in research
Qualitative analysis turns non-numerical material (transcripts, documents, notes) into patterns of meaning that answer a research question. Thematic analysis is one widely used approach; others, such as content analysis, framework analysis, the Colaizzi method and grounded theory (see axial coding), follow different procedures. Before analysing, confirm how the material was gathered: see qualitative data collection techniques and the overview of qualitative research.
This page follows the reflexive thematic analysis process published by Virginia Braun and Victoria Clarke at thematicanalysis.net (read 2026-11-08). They call their approach “reflexive TA” and say it differs from most other approaches in both underlying philosophy and procedures for theme development.
The six phases, and what each one produces
| Phase | What you do (per Braun and Clarke) | What you should have afterwards |
|---|---|---|
| 1. Familiarising yourself with the dataset | Immerse yourself in the data and note early analytic ideas. | Notes on first impressions. |
| 2. Coding | Apply short labels to features of the data relevant to the research question, across the whole dataset, and collate codes with their extracts. | A set of codes, each with its data extracts. |
| 3. Generating initial themes | Group codes into candidate patterns of meaning and gather the relevant data for each. | Candidate themes with supporting extracts. |
| 4. Developing and reviewing themes | Check candidate themes against the coded data and the full dataset; split, merge or drop them. | A revised set of themes. |
| 5. Refining, defining and naming themes | Sharpen the scope of each theme and give it a name. | Named, defined themes. |
| 6. Writing up | Write an analytic narrative with extracts, placed in the literature. | The results section or report. |
Two cautions from the same source. First, “These phases do not prescribe a rigid process you must follow”; analysis is typically recursive, so you will move back and forth between phases. Second, the names of the phases have changed since they were first articulated in 2006, so check which version a method section or supervisor is citing.
Install or access (official source only)
Taguette is a free, open-source tool for qualitative coding, released under a BSD license. Download it from the official install page at taguette.org/install.html (read 2026-11-08).
| Platform | Official route |
|---|---|
| Windows | An EXE installer that includes Python, Taguette and Calibre. |
| macOS | A DMG; the first time you open it, right-click and choose Open. |
| Linux | Install with pip in a virtual environment (pip install taguette). Python 3 and Calibre are required; on Debian-based systems the page gives sudo apt install python3 calibre. |
To run it, double-click the Taguette icon, leave the terminal window open, and use the application in your browser at localhost:7465 (taguette.org/getting-started.html, read 2026-11-08).
Step-by-step: the six phases in Taguette
Taguette supports importing documents, creating tags (its word for codes), highlighting passages and exporting. The steps below attach those actions to the phases above. Menu names are as given in Taguette’s getting-started guide.
- Import your documents (supports phase 1). Open the Documents tab and choose Add a document. Accepted formats are .pdf, .docx, .txt, .odt, .md and .html. A name is required; a description is optional. Choose Import. Then read each document in full and keep your own notes on early analytic ideas.
- Create tags (phase 2). Open the Highlights tab and choose Create a tag. Give it a name and, optionally, a description. Taguette uses punctuation to build hierarchies: a tag named
tech.flosssits undertech. Tags are listed alphabetically. - Highlight and tag passages (phase 2). Open a document, click and drag across the text, then click the “new highlight” pop-up. Tick the relevant tags (or choose Create a tag to make one), then Save & Close. To remove a highlight, click it and choose Delete highlight. Code the whole dataset, not only the passages that stand out.
- Review one tag at a time (phases 3 and 4). Open a tag in the Highlights tab to see every passage coded with it. Use this view to decide which codes belong together as a candidate theme and which should be split, merged or dropped. Because tags support hierarchy through punctuation, you can rename tags to nest codes under a candidate theme as you go.
- Define and name themes (phase 5). Use each tag’s Description field to record the scope of the theme and the name you settle on, so the codebook you export documents your decisions.
- Export for writing up (phase 6). Export what you need for the analytic narrative (see the next section).
Steps 4 and 5 are an editorial reading of how to apply Braun and Clarke’s phases with Taguette’s features; the Taguette guide itself does not map its features to thematic analysis phases.
Output and how to read it
| What you want | Where to export it | Formats |
|---|---|---|
| The codebook (all tags) | Dropdown in the Project tab | QDC, CSV, XLSX, DOCX, HTML, PDF |
| All highlights for one tag | Open the tag in the Highlights tab, then Export this view | HTML, CSV, XLSX, DOCX, PDF |
| All highlights in the project | See all highlights, then Export this view | HTML, DOCX, PDF |
| The whole project | Project Info tab, then Export project | SQLite3 |
Read an exported tag view as the evidence file for one code: every extract that supports it, in one place. A theme is not just a heavily used code; per Braun and Clarke, themes are patterns of meaning developed through familiarisation, coding, and theme development and revision, and are checked against the full dataset in phase 4. For the write-up, choose extracts that illustrate the narrative rather than listing every highlight.
Troubleshooting
- The page at localhost:7465 will not load. Taguette runs from the terminal window that opens when you start it. Keep that window open while you work, then open
localhost:7465in your browser. - A document will not import. Check the file type against the supported list: .pdf, .docx, .txt, .odt, .md, .html. Convert other formats first.
- macOS refuses to open the app. The install page says to right-click the app and choose Open the first time.
- The Linux install fails. The install page lists Python 3 and Calibre as requirements and recommends a virtual environment for pip.
- You keep returning to earlier phases. That is expected: Braun and Clarke describe analysis as typically recursive.
Free alternatives and other software
Taguette is open source under a BSD license, which is why this walkthrough uses it. If you need to compare commercial packages, see NVivo vs ATLAS.ti vs MAXQDA. The thematic analysis process above does not depend on any particular program, and AI-assisted coding raises its own questions (see AI in qualitative coding).
Frequently asked questions
What are the steps to analyse qualitative data?
For reflexive thematic analysis: familiarise yourself with the dataset, code, generate initial themes, develop and review themes, refine, define and name themes, then write up. The authors say the phases are not a rigid procedure.
Do I need software to analyse qualitative data?
The phases describe a way of working, not a program. Software such as Taguette helps by holding documents, tags and highlights together and exporting them, but the interpretive work of coding and theme development is yours.
Is thematic analysis the only method?
No. Content analysis, framework analysis, the Colaizzi method and grounded theory follow different procedures; see the linked guides above. Choose the method that fits your research question and methodology.
Can I export my codes from Taguette?
Yes. The codebook exports as QDC, CSV, XLSX, DOCX, HTML or PDF, and the whole project exports as SQLite3.
Sources (official documentation)
- Braun, V. and Clarke, V., “The Reflexive TA Process”, thematicanalysis.net/doing-reflexive-ta/ (read 2026-11-08).
- Braun, V. and Clarke, V., “Understanding TA”, thematicanalysis.net/understanding-ta/ (read 2026-11-08).
- Taguette, “Getting started”, taguette.org/getting-started.html (read 2026-11-08).
- Taguette, “Install”, taguette.org/install.html (read 2026-11-08).
