What matters when lectures move online? Measuring the impact of fluency, modality, and feedback on students' metacognitive judgments and learning

ABSTRACT

Due to precautions associated with the COVID-19 pandemic, many courses were forced to shift online, thus quickly increasing the prevalence of online courses and video-recorded lectures. Recent data on student learning indicate that student engagement suffers when courses are delivered virtually;anecdotally, this finding is an agreement with instructors' experiences. Perhaps the decrease in student engagement could be due to their involvement in other activities while they "attend" lectures online (e.g., doing dishes, folding laundry). In three experiments, I evaluated the impact of various factors that could be influencing learning online-specifically, I manipulated lecturer fluency, information modality, and feedback and measured their effect on students' metacognitive judgments and actual learning performance. All experiments were conducted online. In Experiment 1, lecturer fluency (fluent or disfluent) and information modality (audio or video) were manipulated in a repeated measures design (Time 1 and Time 2). Students in the fluent lecturer conditions not only reported higher judgments of learning (JOLs) but also performed better on a measure of actual learning compared to those in the disfluent lecturer conditions. Additionally, students performed better when they listened to the lectures rather than viewed them (i.e., an audio superiority effect was observed). Further, student-participants' reported lower JOLs and exhibited higher learning performance at Time 2 (versus Time 1). Therefore, Experiment 2 aimed to further explore the impact of instructional modality (replicating elements of Experiment 1) in addition to feedback. Participants were provided false positive feedback, false negative feedback or were not provided any feedback after being presented a video or audio lecture. As in Experiment 1, JOLs decreased while actual learning performance increased at Time 2. Additionally, there was a significant effect of false feedback on participants' JOLs. Participants that received false negative feedback demonstrated lower JOLs at Time 2 than those that did not receive any feedback. However, unlike Experiment 1, there were no significant effects of modality on JOLs or learning performance. I next sought to investigate how feedback would impact study choice. Thus, Experiment 3 was a replication of Experiment 2, but participants were asked whether they wanted to restudy the lecture material prior to taking a test at both Time 1 and Time 2 (with feedback only preceding their Time 2 choice). While there were no significant main effects of the manipulated variables (i.e., modality or feedback), participants that chose to restudy the materials showed increased learning performance compared to those that chose not to restudy. The findings from these three experiments have important implications for online learning. For instance, feedback might be ineffective in improving students' learning unless it is individually tailored and concrete. The results also indicate that courses should be designed such that the same test format is administered consistently. (PsycInfo Database Record (c) 2023 APA, all rights reserved)