Comparing Emotions in Class and During Online Video Lectures in a Flipped Classroom

• In Event: Emotions in Technology-Rich Learning Environments: Theory, Measurement, and the Future

Abstract

Flipped classrooms are designed to have students watch lectures online before class and then participate in more active learning activities during class time (Baepler, Walker, & Driessen, 2014). Although this is a promising new instructional approach, little research has investigated the role of emotions in flipped classrooms, which is especially important given recent research highlighting the role of emotions in education (e.g. Pekrun & Linnenbrink-Garcia, 2014; Schutz & Pekrun, 2006). The purpose of the present study is to compare emotions reported in an undergraduate flipped physiology and anatomy course in two instructional contexts (i.e. in-class versus online video lecture) at two time points. As such, this study compares the structural validity of self-reported in-class and online video emotions, examines the longitudinal direction of causality between emotions while watching online video lectures and those reported in class, and investigates the differential predictive validity of emotions in these two contexts on self-reported behavioral engagement, attentional control, and academic achievement.

This study is framed within the control-value theory of achievement emotions (CVT), which posits that environmental factors predict students’ appraisals and emotions, and that emotions in turn predict learning and achievement related outcomes (see Pekrun, 2006). Specifically, the present study compares the structure, the longitudinal relation, and the predictive validity of emotions within two environmental contexts, namely emotions in-class with emotions while watching online video lectures, and investigates how emotions in these two contexts predict behavioral engagement, attentional control, and achievement. As such this study extends CVT by comparing emotions in a technology based learning environment with emotions in a face-to-face learning environment.

Participants (N = 278) reported their emotions, attentional control, and behavioral engagement associated with in-class and online lecture viewing at two time points in a large flipped undergraduate anatomy and physiology course (see Table 1). This flipped course consists of video lectures posted to an online video platform (i.e. Panopto©) that students watch online prior to the in-class active learning component. Exam grades were collected directly from the instructor.

Confirmatory factor analyses indicated that the data fit a two-factor model of emotions (i.e., in-class and online video lecture emotions separately) substantial better than a one-factor model (i.e., emotions across contexts were modeled on one factor) at both time points (see Table 2). Longitudinal structural equation models indicated that emotions are consistent across time points and contexts, and there is a spread of activation from in-class emotions across time points and contexts (see Figure 1). Correlational analyses generally suggest that context specific emotions predict context specific behavioral engagement and attentional control, with higher correlations found within contexts than between contexts (see Table 3). Furthermore, video lecture emotions are better predictors of academic achievement than lecture based emotions.

Results extend the CVT by highlighting differences between in-class and online video lecture emotions in terms of their structure, longitudinal relation, and predictive validity. The current conceptualization of environmental factors within the CVT needs to be extended to include technology-based learning environments.

Authors

• John Ranellucci, Michigan State University

• Kristy Amber Robinson, Michigan State University

• Joshua Michael Rosenberg, Michigan State University

• William Saltarelli, Central Michigan University

• Cary J. Roseth, Michigan State University

• Lisa Linnenbrink-Garcia, Michigan State University