Harnessing active engagement in educational videos: Enhanced visuals and embedded questions

The prevalence of online instruction highlights the importance of videos in education. Pedagogies that include elements that actively engage students are accepted as an improvement over more passive modes of instruction. How can we transfer the advantages of active engagement to instruction via video? Previous research on instructional videos has shown that there are a number of principles, the adherence to which benefit student learning by maximizing productive cognitive processing. To understand the impact of combining such principles we designed and produced four different versions of the same physics demonstration video, varying levels of “visual enhancement” designed around these principles and the amount of active engagement across the different versions. Using pre-post video testing, we compared how much viewers learned across the four different versions. We found that actively engaging students by embedding questions throughout the video increases student learning. We also found that physics videos are most effective when they include enhanced visuals and embedded questions. Notably, it is the combination that matters most; the learning effect from embedding questions is increased when the video also includes enhanced visuals. This study represents an important step towards understanding how instructors can design and refine their videos to maximize student learning.

Figure 1

Summary of four versions of videos used in the study.

Figure 2

Shoot the monkey setup schematic.

Figure 3

Screenshots from the shoot the monkey demo video versions which had enhanced visuals (versions 1 and 3).

Figure 4

Screenshot from the shoot the monkey demo video versions which (a) had enhanced visuals (versions 1 and 3) and (b) did not have enhanced visuals (versions 2 and 4).

Figure 5

Images of two tracks in the high road, low road demonstration.

Figure 6

Screenshot from the high road, low road demo video versions which (a) had enhanced visuals (versions 1 and 3) and (b) did not have enhanced visuals (versions 2 and 4).

Figure 7

Experimental design.

Figure 8

Level of physics background of Mechanical Turk participants in the study ($N=300$).

Figure 9

Average postvideo test score for participants who watched each of the four versions of the videos: version 1: with enhanced visuals and embedded questions; version 2: without enhanced visuals, with embedded questions; version 3: with enhanced visuals, without embedded questions; version 4: without enhanced visuals, without embedded questions. Error bars represent the standard error of the mean.

Figure 10

Model 2 interaction between enhanced visuals and embedded questions in predicting postvideo test scores.