ChatGPT Validation of an Antiracism in Medical Education Survey (Poster 6): Division H - Section 1: Applied Research in Schools, 11:51 AM

• In Event: e-Lightening Ed-Talk Session 17

Abstract

The study examined the use of artificial intelligence, specifically ChatGPT versions 3.5 and 4.0, in survey creation and validation. Traditionally, survey development is time-consuming and prone to biases. The researchers adapted an anti-racism in medical education survey from a validated scale and prompted ChatGPT with the survey items. ChatGPT identified weak areas and provided suggestions for improvement. It also helped assess if the questions targeted the intended themes. Compared to a student focus group, ChatGPT proved to be a powerful and efficient tool for initial survey validation before pilot testing. This study highlights ChatGPT as an affordable and potentially less biased approach to creating comprehensive and concise surveys in various research fields.

Authors

• Simran Shamith, Drexel University

• Carolyn Giordano, Drexel University

• Elizabeth K. Kachur, Medical Education Development

• Beverley Crawford, University of Pennsylvania