Digital exhaust: an analysis of students’ interactions with online learning tools (618.42)

Abstract

Learning management systems (LMS) such as Blackboard, Desire2Learn, and WebCT have become ubiquitous on college campuses. At a minimum, these platforms are an efficient way to provide access to course materials (e.g. course notes, syllabus, study guides) to all students. But the real power of LMS is in their potential to provide online tools to engage students outside of class in ways that support and reinforce the learning that occurs during class. When students interact with these online tools, they leave behind a digital “exhaust” that is collected in access logs. This clickstream data can be analyzed for patterns of student behavior from which we can make inferences about student learning. The overarching goal of this project was to analyze clickstream data associated with a large‐enrollment, introductory biochemistry course for science majors. Specifically, we analyzed students’ access logs related to their use of weekly Progress Checks. Progress checks are a series of online assessment tools released weekly to students that gauge their acquisition of key biochemical principles and skills as well as introducing them to upcoming material to be covered in class. We hypothesized that successful students would exhibit different clickstream patterns than non‐successful students. We gathered clickstream data, student performance data on in‐class activities, quizzes, and exams, incoming GPA, and pre/post assessment data. These data were analyzed for patterns of interaction that would help flag students who are not performing adequately within a course (i.e., an early warning system) to proactively target and support student learning.Grant Funding Source: Supported by the NSF