CLASSROOM ATTENDANCE BEHAVIOR OF PHYSICS UNDERGRADUATES IN THE ERA OF ARTIFICIAL INTELLIGENCE: A SURVEY-BASED EMPIRICAL ANALYSIS

With the spread of artificial intelligence and online digital learning resources, as well as the growing prevalence of research training for undergraduate physics majors, the learning styles of physics students have changed. Attendance patterns in face-to-face courses have taken on new characteristics, and traditional classroom management now faces new challenges. Based on questionnaire data from 509 undergraduate students and semi-structured interview data from 19 students in the School of Physics, this study employed Spearman's correlation analysis and a three-level text coding method to examine student attendance patterns, the determinants of absenteeism, and the relationship between absenteeism and academic performance. Results indicate that nearly two-thirds of respondents show varying degrees of absenteeism, and there were significant differences in absenteeism patterns among students in different grades. Statistically, there was no significant negative correlation between absenteeism and academic performance, and most students compensate for missed classes through digital learning resources. Absenteeism is driven by multiple factors, including teaching quality, online resources, and individual time and energy. Based on these empirical findings, this paper proposes five targeted strategies: improving the teaching quality, optimizing course assessment, establishing academic support mechanisms, implementing flexible regulations, and developing students' awareness of self-directed learning. This research offers empirical references for foundational-physics curriculum reform and the development of fundamental-science talent in the AI era.