Towards Personalization of Student Learning and Engagement in a First-Year Undergraduate Course

Abstract

Advancements in classroom technology and data collection have allowed for new studies into how students interact with course material. This paper presents the development of a new tool designed to process timestamp information from a learning management system in a remote, synchronous course to analyze patterns of behaviour and predict student outcomes in the course. The timestamps are arranged to create a personalized timeline of activity for individual students, focusing on the length of time between successive interactions. Preliminary analysis of the timestamp intervals across a class of students over an entire term is also presented. The lengths of time between successive course interactions follows a long-tail distribution with peaks occurring at approximately 24-hour periods, implying that students were most likely to access course material at daily or multi-day intervals.