Bianca Chandel
Research Mentor: Amanda Brown
Mentor Department: Educational Studies, Education
Author(s): Bianca Chandel, Amanda Brown, Qiuyu Chen
Session: Session 1 (9:00 AM – 9:50 AM)
Presentation Type: Poster 95
Abstract
AI tools offer significant potential for efficiently extracting relevant information from large volumes of unstructured qualitative data. This project examines how different artificial intelligence tools can support qualitative analysis of educational meeting materials within a research lab setting. The study focuses on three widely used AI platforms—ChatGPT (GPT 5.2), Google Gemini 3 Flask, and NotebookLM—and compares how each responds to carefully designed prompts when analyzing the same set of meeting-related content (in this study, meeting transcripts and memos). Using a comparative analysis approach, prompts are designed through an iterative process of testing, comparison, and revision across platforms in order to identify which approaches produce the clearest and most useful responses. In total, 18 meeting transcripts are analyzed, and AI-generated outputs are compared with human interpretation to examine consistency and accuracy. The project explores how AI tools assist with common qualitative research tasks, such as identifying meeting participants, recognizing references to instructional or student-related content, and analyzing the density of participant talk or identifying materials used during professional development meetings. The analysis also considers how different types of input materials—such as full transcripts, written summaries, or visual records (e.g., screen captures of shared documents or presentation slides)—shape the quality of AI-generated responses. In addition, the project allows for open-ended analysis to capture relevant information that may not be anticipated in advance. Rather than treating AI tools as neutral or automatic solutions, this project highlights how prompt wording and input materials influence analytical outcomes. The goal of this work is to better understand the strengths and limitations of AI-assisted qualitative analysis and to offer practical insights for education researchers who are interested in using AI tools to support the analysis of instructional discussions and collaborative work.
