Assessing faculty self-perceived knowledge in using generative AI to teach 21st-century skills

Abstract

This study examines faculty self-perceived knowledge for generative AI (GenAI) integration in higher education using the TPACK-21 framework. A survey was administered to 127 faculty members at a large Southern U.S. university. Self-perceived knowledge was assessed across seven TPACK domains, and relationships were analyzed by gender, age, and appointment type. Descriptive statistics, one-way analyses of variance, and Pearson correlation coefficients were used to examine perceived knowledge across the seven domains and the relationships among them. Faculty reported high self-perceived knowledge in traditional domains, including pedagogical knowledge, content knowledge, and pedagogical content knowledge. In contrast, lower self-perceived knowledge was observed in technology-integrated domains, including technological pedagogical knowledge, technological content knowledge, and overall TPACK. Statistically significant gender differences were found in the technological knowledge and technological pedagogical knowledge domains, with male faculty reporting modestly higher self-perceived knowledge; however, the effect sizes were small. No other significant differences were identified across demographic variables. Correlation analyses showed that technological knowledge of GenAI was strongly associated with self-perceived knowledge in technology-integrated domains. The findings suggest a gap between faculty members' established pedagogical strengths and their confidence in integrating GenAI into teaching and learning. Targeted, discipline-specific professional development programs may help bridge this gap and support faculty in designing GenAI-enhanced instruction that promotes 21st-century skills.