Stop and Consider: Embedding Questions in Videos

During a recent CTL Technology Spotlight showcasing the approach one instructor had taken to flip her classroom, I had the pleasure of sitting around a table of educators from Chemistry and the question came up concerning how to help to get more students to actually watch the videos that the instructor created.  One solution given at our table was to utilize quiz like questions embedded directly into the video, however the instructor also expressed difficulty in the past with trying to create such videos on their own.  I was quick to point out that the new video system, Kaltura, actually had a built in system to make it relatively simple to add questions anywhere in the video and transform your video into a quiz if you desired.  While adding the questions to the video has become substantially easier, I had to stop and consider if it was the best thing to do or not.  So often with technology we’re given a multitude of features and options, but how often do we stop and consider if using those options will actually be best for our students.

To answer this question of whether or not quiz questions embedded into your videos will benefit your students, first you have to explore why it is students may not be watching videos in the first place.  During the workshop, the speaker offered anecdotal information she received when talking to her students whom admitted to skipping her videos.  They offered that they didn’t watch some of the videos because they didn’t have the time and/or they were not easily accessible.  Her solution to these issues were to send the video links inside her class announcements, so students could easily access them on their phones, and then to tell students to speed the videos up by 1.5x or when watching them and simply slow down if they didn’t understand something.  While this was just her own experience, the research seems to support this notion that students tend to have a limit when it comes to spending time watching a video. 

According to the research by Guo et al. (2014) whom examined student’s video watching habits across several different lengths of video in an online MOOC, the optimal length of a given video is around six minutes or less.  Anything over six minutes results in an exponential drop off wherein the student watches less and less of the video. However, recent researchers have argued that the exact optimal length of the video is a bit of a variable depending on other factors that influence the cognitive load.  They often cite research by Mayer and Moreno (2003) who found that cognitive load could be reduced with techniques like signaling, information weeding, conversational style, and matching modality.  The idea is that the optimal length of video a student will watch may be extendable beyond the six minute limit with proper usage of these cognitive load reducing techniques.

Considering some of these techniques, we can start to draw parallels as to the kind of benefit inserting a question or set of questions into a video may have.  The obvious answer is that it may help create a form of segmenting that is equivalent to cutting your video into shorter chunks that are posted separately.  In theory chunking a video allows students to stop and reflect before they begin the next video in the series and reduce their cognitive load; however in practice nothing really forces a student to take a break or prevents them from simply stringing together three or four short videos and end up “binge watching” a series of videos in one sitting that may have been just as long as the original video.  By segmenting a video with embedded questions instead, you are effectively forcing the student to stop and change gears from passively watching to actively thinking about a question and selecting an answer before they are able to move forward.  In this way, you are still chunking a video but are also ensuring that there is at least a small mental pause between segments.

This switch from passive learning to active learning is often what you seek to do in a class lecture, so it makes sense to incorporate those same techniques in your video lectures as well.  Depending on the type of questions you are asking, these kind of active learning stops can also help move a passive video from tier 1 of Bloom’s Taxonomy into higher tiers such as tier 2, if you are asking simply questions that check for understanding, tier 3, if you are asking questions that would force them to apply what they have just learned to respond to the question, or even tier 4, if your questions lead to a form of information analysis.

Signaling information is another technique prescribed by Mayer and Moreno (2003) that one could easily argue can be accomplished through embedded questions. By asking a question about something that was just covered in a video, you are clearly highlighting its importance to the overall topic and effectively asking student to pay special attention to that information.  If done consistently, students would learn to look for these questions and anticipate that they will help guide them to the most important information in a video.  In these ways, embedded questions seem to be able to fill a role as two different types of techniques that could reduce the cognitive load and this could result in students consuming more video content overall.  While the evidence seems to point in this direction, more research into the results of embedding questions in videos needs to be conducted to better understand the possible benefits to students.

While it’s usage as a means to get students to consume longer videos is still being researched, another possible use for video embedded questions is to supply students with more opportunities for quizzing overall. You can read more about the benefits of frequent quizzing in another recent CTL blog post “Testing is a Key to Better Learning” by Kate Williams.

References:

Guo, P. J., Kim, J., & Robin, R. (2014). How video production affects student engagement: An empirical study of MOOC videos. ACM Conference on Learning at Scale (L@S 2014); found at http://groups.csail.mit.edu/uid/other-pubs/las2014-pguo-engagement.pdf.

Mayer, R. E., &  Moreno, R. (2003). Nine ways to reduce cognitive load in multimedia learning. Educational Psychologist 38, 43-52.