Plenary Talk II Measuring Student Engagement in Early Engineering Coursework

Abstract

This talk describes recent efforts for quantifying students’ engagement in early engineering coursework, through designing, implementing, and testing a system to measure the students’ emotional, behavioral, and cognitive engagement states. Engineering programs suffer from a high rate of attrition in the freshman year, primarily due to poor engagement of students with their classes. The project plans to develop a sensor-driven, computational approach to measure emotional and behavioral components of student engagement. This information will be used to identify teaching strategies that increase engagement, with the goal of enhancing student success and retention in STEM education pathways. The project features a multi-disciplinary collaboration between faculty and undergraduate researchers in engineering, the physical sciences, psychological sciences, and education. The project involves students in first- and second-year engineering STEM subjects and the experienced faculty who teach these courses. Findings from the project could be a valuable step toward an early warning system to detect student disengagement and anxiety in STEM and non-STEM courses. Project goals include: (i) establishment of a robust network of non-obtrusive and non-invasive sensors in mid-size classes to enable real-time extraction of facial and vital signs, which will be integrated and displayed on instructors’ dashboards; (ii) identification of robust descriptors for modeling the emotional and behavioral components of engagement using data collected by the sensor networks; (iii) pilot testing of the system’s effectiveness in gathering meaningful data for subsequent work on emotional, behavioral, and cognitive metrics of engagement. The fundamental research question to be addressed relates to improving student learning by the automated capture of non-verbal cues of engagement: How can we use students’ expressions of engagement, based on non-verbal signs such as facial expressions, body and eye movements, physiological reactions, posture, to enhance learning? Findings from the project will constitute a foundation for multi-disciplinary research to incorporate novel machine learning and artificial intelligence-based models for measuring engagement in STEM classes. This project has been funded by the National Science Foundation (NSF). The talk will describe our latest discoveries in this long-term and multidisciplinary project.