Course Overview

Artificial intelligence is rapidly changing what is possible in qualitative research. However, using AI-assisted qualitative analysis well requires more than knowing how to write a prompt. Researchers need to understand where AI can strengthen qualitative inquiry, where it can introduce methodological and ethical risks, and where human interpretation and judgment remain essential.

This interactive two-module course provides practical, hands-on experience using generative AI across qualitative research and analysis. Participants will work with qualitative data to explore how AI can support activities such as preparing and organizing data, developing and refining analytic frameworks, coding, identifying patterns and themes, interrogating interpretations, comparing analyses, and developing evidence-based findings.

Throughout the course, participants will critically examine the strengths and limitations of AI-assisted analysis rather than treating AI output as inherently authoritative. Particular attention is given to confidentiality, bias, methodological alignment, transparency, researcher reflexivity, and verification of AI-generated interpretations.

By the end of the course, participants will have both practical experience using leading AI tools and a framework for deciding when, how, and whether AI should be incorporated into their own qualitative research practice.

What You Will Learn

Participants will learn how to integrate AI into qualitative research in ways that are purposeful, methodologically appropriate, and transparent. Through demonstrations, guided practice, and hands-on analysis, participants will learn to:

• Make informed decisions about whether and how AI fits with different qualitative methodologies, epistemological commitments, research contexts, and analytic purposes.
• Identify appropriate and inappropriate roles for AI across the qualitative research and analysis process.
• Prepare qualitative data for responsible use with AI, including considerations related to confidentiality, de-identification, data security, and platform choice.
• Develop effective prompts and instructions for qualitative research tasks and understand how seemingly small changes in instructions can affect analytic outputs.
• Use existing codebooks with AI and explore AI-assisted approaches to deductive and inductive coding.
• Use AI to identify patterns, relationships, themes, divergent perspectives, and potential alternative interpretations within qualitative data.
• Critically assess AI-generated coding and interpretations rather than accepting outputs at face value.
• Compare human and AI-supported analyses and use discrepancies as opportunities for deeper analytic thinking.
• Recognize potential hallucinations, biases, decontextualization, and other limitations in AI-assisted qualitative analysis.
• Develop strategies for verification, documentation, transparency, and maintaining an appropriate human role in interpretation and meaning-making.

Course Format

This interactive course is offered through two 3-hour online modules or one day in-person, and combines instructor-led demonstrations, discussion, guided experimentation, and hands-on exercises.

Participants work directly with qualitative data throughout the course rather than simply watching demonstrations. Exercises allow participants to experiment with AI, examine its outputs, compare analytic approaches, identify problems, and refine their methods.

Short pre-course materials will help participants configure relevant AI tools and prepare for the hands-on work. As a result, live course time can focus primarily on analysis, methodological decision-making, and discussion.

Participants will have opportunities to work with and compare leading generative AI platforms. The course emphasizes transferable approaches and analytic practices rather than dependence on any single AI product.

Module Breakdown

Module 1: Foundations and AI-Assisted Coding

The first module establishes a foundation for responsible and methodologically informed use of AI in qualitative research. Participants begin by considering what generative AI can and cannot do with qualitative data. They also examine different positions on the appropriate role of AI in qualitative inquiry.

Participants then move into hands-on work. They explore preparing qualitative data for AI-assisted analysis, protecting confidentiality, developing effective analytic instructions, and using AI with qualitative transcripts and codebooks. Exercises introduce AI-assisted coding and allow participants to examine how different prompts, tools, code definitions, and analytic choices influence results.

Rather than focusing only on whether AI can produce codes, participants consider a more important question: What would make an AI-assisted analysis trustworthy and useful? Participants learn to interrogate AI output, identify questionable classifications and interpretations, and distinguish efficiency from analytic quality.

Module 2: From Coding to Interpretation, Validation, and Insight

The second module moves beyond initial coding into the more interpretive stages of qualitative analysis. Participants use AI to explore patterns across coded data, examine relationships among concepts, identify divergent and negative cases, develop and challenge candidate themes, and generate alternative interpretations.

Participants also experiment with using AI as an analytic interlocutor rather than simply an automated coder. For example, they ask it to challenge an emerging interpretation, identify evidence that does not fit a proposed theme, compare competing explanations, or trace claims back to the underlying qualitative evidence.

A central emphasis is validation and human judgment. Participants compare outputs, investigate discrepancies, and examine the evidence supporting AI-generated claims. They also consider when AI assistance strengthens analysis and when it may undermine methodological integrity.

The module concludes by helping participants develop a defensible approach to incorporating AI into their own qualitative workflow. This includes decisions about documentation, transparency, verification, researcher reflexivity, and appropriate boundaries for AI assistance.

Who Should Attend

This course is designed for evaluators, researchers, applied social scientists, consultants, students, and other professionals who conduct or use qualitative research and want to make informed decisions about incorporating generative AI into their work.

The course is appropriate for participants who are relatively new to AI as well as those who already use tools such as ChatGPT, Claude, or Gemini. It will help them move beyond basic prompting toward more systematic and methodologically defensible qualitative applications.

Prerequisites

There are no technical or programming prerequisites.

Participants should have a basic understanding of qualitative research and analysis. Familiarity with concepts such as coding, codebooks, themes, and interpretation will be helpful. The course focuses on applying and critically evaluating AI within qualitative analysis rather than teaching qualitative methods from the beginning.

Participants will receive instructions before the course for setting up access to the AI tools used during the hands-on exercises. Course exercises will use provided qualitative data so participants can experiment in a shared environment without uploading sensitive or identifiable research data.