Attentive or Not? Toward a Machine Learning Approach to Assessing Students’ Visible Engagement in Classroom Instruction

• DOI: 10.1007/s10648-019-09514-z
• Publication Date: 2019-12-18T09:03:40Z
• Authors: Patricia Goldberg; Ömer Sümer; Kathleen Stürmer; Wolfgang Wagner; Richard Göllner; Peter Gerjets; Enkelejda Kasneci; Ulrich Trautwein
• Keywords: Situational ethics; Educational Psychology; Student Engagement

Abstract Teachers must be able to monitor students’ behavior and identify valid cues in order draw conclusions about actual engagement learning activities. Teacher training can support (inexperienced) teachers developing these skills by using videotaped teaching highlight which indicators should considered. However, this supposes that (a) of are known (b) work with videos is designed as effectively possible reduce the effort involved manual coding procedures examining videos. One avenue for addressing issues utilize technological advances made recent years fields such machine improve analysis classroom Assessing attention-related processes through visible (dis)engagement might become more effective if automated analyses employed. Thus, present study, we validated a new rating approach provided proof concept vision-based evaluated on pilot recordings three lessons university students. The system was significantly correlated self-reported cognitive engagement, involvement, situational interest predicted performance subsequent knowledge test. approach, based gaze, head pose, facial expressions, good estimations ratings. Adding synchrony feature improved correlations ratings well prediction posttest variables. discussion focuses challenges important next steps bringing classroom.