Date of Award
Master of Science (MS)
The use of online learning environments in higher education is becoming ever more prevalent with the inception of MOOCs (Massive Open Online Courses) and the increase in online and flipped courses at universities. Although the online systems used to deliver course content make education more accessible, students often express frustration with the lack of assistance during online lecture videos. Instructors express concern that students are not engaging with the course material in online environments, and rely on affordances within these systems to figure out what students are doing. With many online learning environments storing log data about students usage of these systems, research into learning analytics, the measurement, collection, analysis, and reporting data about learning and their contexts, can help inform instructors about student learning in the online context.
This thesis aims to lay the groundwork for learning analytics that provide instructors high-level student engagement data in online learning environments. Recent research has shown that instructors using these systems are concerned about their lack of awareness about student engagement, and educational psychology has shown that engagement is necessary for student success. Specifically, this thesis explores the feasibility of applying machine learning to categorize student posts by their level of engagement. These engagement categories are derived from the ICAP framework, which categorizes overt student behaviors into four tiers of engagement: Interactive, Constructive, Active, and Passive. Contributions include showing what natural language features are most indicative of engagement, exploring whether this machine learning method can be generalized to many courses, and using previous research to develop mockups of what analytics using data from this machine learning method might look like.
Stepanek, Nicholas R., "Towards Student Engagement Analytics: Applying Machine Learning to Student Posts in Online Lecture Videos" (2017). Student Work. 2917.