ChatGPT Early Adoption in Higher Education: Variation in Student Usage, Instructional Support, and Educational Equity
Richard Arum, Maria D. Calderon Leon, XunFei Li, Jomar Lopes
Abstract
This study examines the early adoption of ChatGPT in late 2022 and early 2023 by identifying variation in student awareness, academic use, and perceived instructional support for the technology at a diverse U.S. public research university. Specifically, we investigate (a) how individual awareness and academic use of ChatGPT vary by student characteristics and field of study, (b) how instructor encouragement or discouragement varies across courses and student demographics, and (c) how academic use patterns differ between science, technology, engineering, and mathematics (STEM) and non-STEM fields. Data from 938 undergraduates, merged with administrative records, revealed disparities in ChatGPT awareness and use, with underrepresented minority, first-generation, and international students less likely to know about or use the tool academically. Course-level analysis highlighted that instructor encouragement is more prevalent in STEM fields and upper-division courses but decreases in classes with higher underrepresented minority representation. Open-ended responses showed distinct patterns of ChatGPT use, with STEM students favoring conceptual assistance and coding and non-STEM students engaging in writing and instructor-assigned activities. These findings underscore early inequities in access and use of emerging educational technologies and call for institutional strategies to promote equitable awareness and skill development aligned with academic and professional goals.