Asynchronous online teaching involves the use of pre-recorded lectures, which students can watch in their own time. Lecturers might hope that students watch those recorded lectures in the same way as they would live lectures, taking careful notes and avoiding distractions. However, in the real world, students may well be watching the recorded lectures on double speed (or faster), or listening to them while simultaneously playing video games or completing homework tasks.
Putting aside the distractions for the moment, a reasonable question is whether the speed at which students watch video recordings matters for their learning. That is the question that is addressed in this recent article by Dillon Murphy (UCLA) and co-authors, published in the journal Applied Cognitive Psychology (sorry, I don't see an ungated version online).
Murphy et al. conducted several experiments with undergraduate UCLA students. In the first experiment, students were randomly allocated to watch videos (on conducting real estate appraisals, and on the Roman Empire) at normal speed or at faster speeds (up to 2.5x), and then tested on their content knowledge (using multiple choice and true/false questions), both immediately after the videos and a week or so later. They then compare students who watched videos at different speeds, in how well they perform on the tests. Their results are nicely summarised in Figure 2 from the paper:
As you might expect, performance decreased with video speed. However, the only statistically significant difference was between the slowest group (1x) and the fastest group (2.5x), where
...the 1x group performed better than the 2.5x group... but there were no other pairwise differences...
The effect of video speed is pretty small (the effect size was d=0.24) though, so on the surface that suggests that we probably shouldn't be too concerned about whether students are watching videos at double speed. Moreover, watching videos at double speed means that students could watch each video twice in the same amount of study time, which could improve learning.
Recognising this, in the second experiment, Murphy et al. compare students randomly assigned to watch videos at normal speed once, and students who watched the same videos at double speed twice in immediate succession. Comparing those two groups, they find that:
...an independent samples t test on test performance did not reveal differences in comprehension between participants watching the videos a single time at 1x speed... and participants watching the videos twice at 2x speed...
However, in the real world few students watch videos multiple times immediately. Most would watch a second time much later, probably just before a major assessment. So, in the next experiment, Murphy et al. tested how well students performed a week after watching the videos for the first time. Students who watched at normal speed watched only once (a week before the test), while students who watched at double speed watched once a week before the test, and again just before the test. Comparing those two groups, Murphy et al. find that:
...participants watching the videos twice at 2x speed... performed better on the comprehension tests than participants watching the videos a single time at 1x speed...
The problem with this experiment is that it conflates the effect of delayed testing with the effect of video speed, because the double-speed group watched the video the second time just before the test. It might have been better for the normal-speed group to watch the video just before being tested (or Murphy et al. could have compared results with the results from their first experiment).
In the next experiment, Murphy et al. look at how well students perform when watching videos multiple times at different speeds. Specifically, they compare a group of students who watched at normal speed and then double speed, with a group of students who watched at double speed, and then normal speed. When students were tested immediately after watching both videos in succession, Murphy et al. find that:
...an independent samples t test on test performance did not reveal differences in comprehension between participants watching the videos at 1x speed then again at 2x speed... and participants watching the videos at 2x speed then again at 1x speed...
And, when testing students a week or so after they watched the videos the first time, and just after their second viewing, Murphy et al. find a similar lack of any statistically significant difference in test performance.
So, what can we take away from this study? On the surface, it suggests that video speed makes little to no difference to student learning, at least up to double speed. That means that students could study effectively in half the time, by watching the videos at double speed. That frees up time for the students to do further study, including re-watching the same recordings. Watching videos twice does appear to improve on performance, but more research on that question is needed before we can conclude that students should double-down on double-speed video watching.
However, there is a key problem with this study. It assumes that the mode of teaching in video lectures is simply provision of key facts that can be memorised and regurgitated in a later test. That might be suitable for some lower-level skills, but most of the time we want students to do more than learn and repeat key facts. Of course, there are some key facts that it is important for students to know, but ideally we want them to do stuff with those facts, and that is harder for students to learn from a video since it requires practice. For that, more active styles of lecturing are required, and as noted in this post from last week, interactive teaching approaches also may be best in terms of mitigating any negative impacts of online learning. Interactive teaching isn't really consistent with an asynchronous learning environment or pre-recorded lectures. So, this research really answers the question of whether we should be worried about how students are watching pre-recorded lectures, and not whether lectures should be provided pre-recorded or not. And, the research doesn't address the key issue of other distractions that students may have while watching pre-recorded videos (at normal or double speed). That might be an interesting follow-up study.
Overall, if lectures don't pass the 'laundry test', then we should not be surprised that students watch lecture recordings at double speed. Fortunately, it appears that the speed of watching isn't doing harm to their learning, over and above the lack of interaction in that style of online teaching.
[HT: Marginal Revolution]
Read more:
No comments:
Post a Comment