Purpose: This paper aims to explore university students' multitasking behavior in online classrooms and their influence on academic performance. Also, the study examined students' opinions. Design/methodology/approach: A total of 302 university students fulfilled an online survey. Ten questions were focused on demographic information, five items evaluated online class behavior of students, 9 items evaluated self-efficacy and four items measured academic performance. Findings: Multitasking behavior was found to negatively influence self-efficacy of −0.332, whereas self-efficacy showed a positive influence of 0.325 on academic performance. Cronbach's alpha and average variance extracted values were 0.780 and 0.527 (multitasking behavior), 0.875 and 0.503 (self-efficacy), 0.781 and 0.601 (academic performance). Outcomes of the bootstrapping test showed that the path coefficients are significant. Originality/value: The research findings may help university managers understand undergraduates’ online and face-to-face behavior and strategies to improve the behavior to ensure the best academic outcomes. The novelty is based on using the partial least square structural equation modeling technique.