Tuesday, October 19, 2021
2:00 – 4:00 PM (PDT) in BWW 1212 and Zoom:
In March of 2020, the COVID-19 pandemic led to an emergency switch to remote instruction at post-secondary institutions in the United States. Several surveys have emerged, shining light on the impacts of this switch on student identity, course efficacy, and well-being as reported by students and faculty. In this work, we present the first learning analytics-based study of the effects of remote instruction under COVID-19 on learning at an institution of higher education. In part one of the study, we explore the nature of instructor and student engagement during the pandemic with a large-scale analysis of learning management system data from 100,000 students taking 28,000 courses, comparing analytics from the emergency remote instruction semester of Spring 2020 to those from the previous three spring semesters. In the second part of the study, we use student grade data from the semester following the emergency switch to remote instruction, in addition to course prerequisite information, to gauge remote instruction's impact on preparation for future learning as compared to preparation that occurred in semesters prior to emergency remote instruction. Challenges and successes of this experience with remote instruction will be discussed as well as the limitations of the data and analysis.
Zach Pardos is an Associate Professor at UC Berkeley studying adaptive learning and AI in the Graduate School of Education. His research focuses on knowledge representation and recommender systems approaches to increasing credit mobility and degree attainment in higher education using behavioral and textual data.