PERC 2024 Abstract Detail Page
Previous Page | New Search | Browse All
| Abstract Title: | Quantifying qualitative analysis: AI-assisted thematic exploration in physics education |
|---|---|
| Abstract Type: | Contributed Poster Presentation |
| Abstract: | Artificial intelligence (AI) utilizes advanced language models to analyze language patterns, extract meaningful insights, and make informed decisions across a wide range of applications. The question now arises: can AI be utilized to perform a basic qualitative data analysis, extending beyond the traditional realm of human-only analysis? This study seeks to determine whether leading large language models can effectively perform a qualitative thematic analysis, one of the most basic types of qualitative analysis. The data consists of interviews where introductory life science majors are asked to explain what happens when they drop a ball. The control group is two experienced human qualitative researchers which will be compared against the different AI models. The metrics of success will include the AI tools' ability to identify themes, the time required for analysis, ease of use, and the accuracy in trend identification, juxtaposed with human qualitative research methods. |
| Session Time: | Poster Session 2 |
| Poster Number: | B96 |
Author/Organizer Information | |
| Primary Contact: |
Praveer Sharan Purdue University |
| Co-Author(s) and Co-Presenter(s) |
Liam McDermott, Rutgers University and University of Connecticut Dan Young, University of Delaware Rebecca Lindell, American Association of Physics Teachers and Tiliadal STEM Education: Solutions for Higher Education |
