Developing AI Agents for Qualitative Survey Analysis in Youth Enterprise Education
Developing AI Agents for Qualitative Survey Analysis in Youth Enterprise Education
-
Co-developing and evaluating an AI agent to enable scalable, trustworthy qualitative analysis of survey responses
Overview
Young Enterprise supports over 500,000 young people each year through programmes that develop enterprise, employability and financial literacy skills. Every programme generates rich qualitative feedback from participants, offering invaluable insights into young people's learning experiences, skill development and aspirations. However, analysing thousands of open-ended survey responses manually is prohibitively time-consuming, meaning much of this evidence remains underused.
This project will address that challenge by developing an AI-powered qualitative analysis agent that enables YE to analyse learner feedback accurately, efficiently and at scale. Rather than replacing human judgement, the project will explore how AI can responsibly augment qualitative research by combining state-of-the-art large language models with rigorous human validation.
Researchers from The Open University and Young Enterprise’s Research & Evaluation team will work collaboratively throughout the project to co-design, test and refine the AI agent using real-world evaluation data. A central component of the project will be a systematic comparison of AI-generated thematic coding with expert human analysis, identifying where AI performs reliably, where human oversight remains essential, and how trustworthy AI-assisted qualitative analysis can be implemented in practice. In doing so, the project will generate new methodological evidence on the strengths, limitations and responsible use of AI for social science research, and AI-assisted programme evaluation.
The role of IET
The Open University's Institute of Educational Technology (IET) brings a strong track record in researching how technology shapes learning, evaluation, and educational practice, and this expertise underpins the academic rigour of the project. IET's involvement ensures the AI agent being developed is grounded not just in technical capability, but in sound qualitative research methodology — from the initial scoping of YE's survey data through to the design and interpretation of the human-versus-AI coding experiment that sits at the project's methodological core.
As IET is also leading the Research Capability Hub, which is focused on transforming research training for social scientists, IET offers a natural route for embedding the resulting framework of the project into wider methodological training and dissemination.
