Qualitative Data Analysis: Using AI and Human Judgement

Qualitative data can reveal insights that structured data cannot, but analysing large volumes of open-ended text has traditionally been slow and difficult to do consistently. AI can now code and categorise qualitative data quickly and at scale, but human judgement remains essential to ensure the results are meaningful, accurate and relevant.

This one-day, hands-on course uses a real dataset to build a practical workflow from raw text through to coded, validated and summarised results. Participants use AI tools to accelerate coding and categorisation, while applying their own judgement to define useful categories, validate the results and account for ambiguity, sarcasm and context. The course is designed for professionals who work with survey comments, reviews, interview transcripts or other open-ended feedback. Participants leave with a practical approach to analysing qualitative data and insights they can trust and act on.

• Compare AI-based and traditional qualitative coding to determine when AI is the better approach.
• Develop category and theme structures by defining them yourself or using AI to derive them from the data.
• Write effective prompts to produce consistent and well-defined coding schemes.
• Code open-ended text at scale using AI tools to categorise large volumes of qualitative data efficiently.
• Validate AI-generated coding by checking reliability and cross-referencing results across different AI tools.
• Identify content AI is likely to misinterpret including sarcasm, ambiguity, and cultural or domain-specific language.
• Combine coded qualitative data with quantitative data using pivot tables to reference, summarise and explore findings.
• Apply human judgement appropriately by recognising where AI can be trusted and where human oversight remains essential.