The COVID-19 pandemic forced many university lecturers to experiment with video-based teaching. Although the use of pre-recorded, asynchronous video had already been adopted by proponents of the flipped classroom, only a small proportion of teachers had any previous experience of creating or using pre-recorded lectures. The results of those first hasty attempts at using video were understandably somewhat hit-and-miss. Fast-forward to today and the majority of lecturers have returned to their original ways of working, but having seen the possibilities (and of course the problems) of putting lectures online, some of us may be considering using pre-recorded materials in our courses. Now, with time to plan properly, many lecturers will have questions about how to construct better video input from an educational perspective. Two newly-published articles in Educational Research Review and the Journal of Computing in Higher Education put forward evidence for two separate techniques that lecturers can use to increase student learning from asynchronous videos: instructor presence and inclusion of cognitive learning prompts.
The first study, published in Educational Research Review, was a meta-analysis collating 27 years of research into the inclusion of an instructor in video lectures. A total of 32 studies met the research criteria. One of the main goals of the researchers was to attempt to settle an argument in the literature over whether the inclusion of an instructor in videos does or does not affect actual learning outcomes. Here, there are two theoretically opposed positions. The split-attention effect suggests that including an instructor will raise cognitive load and reduce the time spent engaging with the material to be learned. In contrast, the Cognitive-Affective-Social Theory of Learning in digital Environments (CASTLE), suggests that since social cues enhance social, emotional, and motivational processes, the inclusion of an instructor can potentially promote learning outcomes.
The second study examined the inclusion of learning prompts into video lectures. Here, 253, US undergraduate physics students across three separate courses were each shown a version of the same 30-minute video. Participants were randomly assigned to watch one of three versions of the video: the first contained cognitive prompts (n = 86) such as, Write a brief summary of the content of the video you just watched. The second version contained metacognitive prompts (n = 81) such as, After watching this part of the video, what questions do you still need answers to? The final video was a no-prompt, control condition (n = 86).
The inclusion of instructor video was shown to increase motivation, aid retention and reduce the students’ perceived cognitive load. However, the so-called dwell time—that is the time students spent attending to the learning materials—did decrease due to the split attention effect.
The inclusion of cognitive prompts led to 10 procent better performance on a test taken directly after watching the video compared to the control condition. However, there was no significant effect found for the inclusion of metacognitive prompts.
From reading the two papers, it seems that the following recommendations can be made: The inclusion of the instructor in online videos has a range of positive effects for; how the video is judged, student motivation and the perceived cognitive load (how difficult students feel learning the material is). There is also evidence that the retention of content can be positively affected in some cases. There remains debate about exactly how the instructor should be included to gain the best results, with debate focussing on the size and position of the instructor, whether the instructor should always be visible and where the instructor should be placed on the screen. The effects of the inclusion of intermittent, full-screen instructor are not dealt with in this meta-analysis.
The advice about the inclusion of cognitive learning prompts in online video is much more robust in terms of learning outcomes, with students in the cognitive prompts research group averaging 10% better scores than the control group. The inclusion of such prompts should therefore be considered by anyone producing online video content.
Comment: In our newsletters we endeavour to highlight new research, and as such, the two papers described above are both recently published examples. However, there is, of course, a lot of earlier research that has been carried out that does not get mentioned, simply because it isn’t brand new.
When it comes to the use of video in teaching, we have a course in advanced video production for lecturers that will run during the spring term: Video produktion för universitetslärare (in Swedish).
There is also a general resource page on our website where you can find out about using sound and video in your teaching: Berika din undervisning med ljud och video (in Swedish)
Text: John Airey, Department of Teaching and Learning
Keywords: Asynchronous video lectures, instructor presence effect, embedded learning prompts, flipped classroom
