QualCoder Pro (CAITA)

QualCoder Pro is a local Streamlit web app for qualitative researchers who need to code interview transcripts without commercial CAQDAS licences. Upload transcripts, run them through a three-stage coding pipeline, and export the results as Excel workbooks. The app runs entirely on your own machine as a local Streamlit process, so transcripts are never sent to an external server or third-party API.

Features

• Three-stage coding pipeline (qualcoder_core.py): Stage 1 extracts participant responses from a transcript and assigns an initial code to each sentence using a keyword codebook; Stage 2 groups related initial codes into broader categories; Stage 3 maps coded segments to research questions to build a thematic framework.
• Multi-format transcript upload: DOCX, PDF, and TXT files, with batch processing of multiple files in one run.
• Codebook configuration: use the built-in default codebook, which covers educational-technology themes such as LMS use, virtual teaching platforms, and assessment practices, or upload a custom JSON codebook.
• TF-IDF keyword suggestions: scikit-learn's TfidfVectorizer scans uploaded transcripts and suggests candidate domain keywords, which can be combined with manually entered keywords.
• Results dashboard: per-file segment previews, per-file and aggregate statistics (segment counts, unique codes), and downloadable Excel files for each stage plus a ZIP of the full project output.
• Citation export: download a citation for the tool in RIS, BibTeX, EndNote XML, CSL JSON, or MODS format from the app sidebar.

Installation

Requires Python 3.7 or later.

git clone https://github.com/MuhammadTayyabIlyas/qualcoder_app.git
cd qualcoder_app/qualcoder_app
pip install -r requirements.txt

Running

streamlit run app.py

The app opens in your browser at http://localhost:8501. Upload a transcript (or try the included examples/sample_transcript.txt), enter at least one research question, and start the analysis from the Analysis tab.

Docker

A Dockerfile and docker-compose.yml are included for containerised deployment:

docker build -t qualcoder-pro .
docker run -p 8501:8501 qualcoder-pro

Tests

python -m pytest tests/ -v

Background

The default codebook and thematic categories (LMS use, virtual teaching platforms, technology integration barriers, professional development) reflect the author's own PhD research into digital technology adoption in higher education. QualCoder Pro grew out of the need for a coding workflow that fit that project without the cost of commercial qualitative analysis software.

License

MIT. See qualcoder_app/LICENSE.

Author

Built by Tayyab Ilyas, AI Agent Engineer and PhD researcher in Barcelona.