Tag Archives: learning

“Invisible learning” by David Franklin

2 Replies

Several things happened today.

  1. I had a lovely time reading in the hammock
  2. I tried to kill two birds with one stone (figuratively of course): writing a blog post about the book I read (which I really loved) and try a new-to-me format of Instagram posts: A caroussel, where one post slides into the next as you swipe (so imagine each of the images below as three square pictures that you slide through as you look at the post)

Turns out that even though I really like seeing posts in this format on other people’s Instagram, it’s way too much of a hassle for me to do it regularly :-D

Also a nightmare in terms of accessibility without proper alt-text, and for google-ability of the blog post. So I won’t be doing this again any time soon! But I’m still glad I tried!

And also: check out the book!

Invisible Learning: The magic behind Dan Levy’s legendary Harvard statistics course. David Franklin (2020)

How to learn most efficiently when participating in a MOOC

Leave a reply

How to learn most efficiently when participating in a MOOC? Yes, I’ll admit, that title promises quite a lot. But there is a new article by Yong and Lim (2016) called “Observing the Testing Effect using Coursera Video-Recorded Lectures” that tells us a lot about how (not) to learn. We have talked about the testing effect before: repeated testing leads to better results on examinations that repeated studying does. And it is confirmed again in this study.

Why am I so excited about this? Because both video-based studying and testing are becoming more and more common these days, and both are sometimes made out to be really bad ideas.

We find video-based learning in most aspects of our lives now (at least if we are talking about lives similar to mine ;-)) — I always follow one or two Coursera courses at the time, and I love watching TED talks. Most softwares I use have video tutorials, and in fact I talked about how I liked the video tutorials of the Monash simple climate model interface only on Tuesday. And whenever I get stuck with a task, I watch video tutorials on youtube to get me going again. And of course many of the lectures at my university are being recorded and many students rely on re-watching them when studying for exams. And, of course, there is the One Planet — One Ocean MOOC that I am involved in preparing. So obviously I see value in video lectures. Even though many people believe that re-watching a lecture does not provide the same experience as seeing it “live”, I don’t think that matters much for lectures in which there is not a lot interaction between lecturer and audience. If you can make yourself use them wisely, I think video lectures are a great substitute for lectures you — for whatever reasons — can’t watch live.

But this is also the biggest issue I have with video lectures: they can easily seduce us into believing that we are learning, when we in fact are not. For example, when I say that I am “following” those Coursera MOOCs, what that means is that I have videos playing while I do something else (like writing emails or cleaning my apartment), i.e. I am not listening carefully, and I never ever do the tests and quizzes they provide. Yet, I still feel like I am learning something. I might or might not* be, but in any case I am not using those resources as effectively as I could be, and in fact most people aren’t.

And testing, I get it: Educators typically don’t like designing tests, because it is really hard. And most students don’t like taking tests, again because it is really hard, so tests have a really bad reputation all around. Especially repeated testing and e-assessment (like we are developing for mathematics and mechanics) people really love to hate!

But this is where the Yong & Lim (2016) study comes in. They showed a short (<3min) Coursera lecture to their participants. Depending on the group, during study time, they showed the clip either once and then tested three times, showed it three times and tested once, or showed it four times. Initial recall right after the study period is best for the group that watched the same clip four times, but it turns out that both groups that test during studying perform significantly better on a test a week after the study period: testing as part of studying (and in contrast to just repeatedly watching a clip) helped anchor the new knowledge significantly better.

From this is it clear that we should definitely be taking advantage of the tests provided with video lectures! Or if there are no tests available, like with TED talks**, instead of watching a lecture over and over again, test ourselves on it: Can I remember the main points? What were the reasons for x or the steps in y? Why did she say z?

And, more importantly, as educators we should take these results to heart, too.  If testing is this important, we need to provide good tests to students, and we need to encourage them to use them to practice.

One scary fact to end this post with: Of the 30 idea units presented in the videos of the study, the best group retained on average only about half until a week after watching those videos. And the worst group only one-third. I didn’t see those videos so I can’t speak about how well they were made and whether the tests addressed all of those 30 idea units, but I wouldn’t bet on students remembering more of the videos I want them to learn from. Which really gives me something to think about.

*watching those videos and feeling good about doing something productive might actually just give me the illusion of competence

**or if we feel that the tests are really bad, which does happen

—

Yong, P., & Lim, S. (2016). Observing the Testing Effect using Coursera Video-Recorded Lectures: A Preliminary Study Frontiers in Psychology, 6 DOI: 10.3389/fpsyg.2015.02064

Will giving your students more structure make them need more structure?

Leave a reply

One of the arguments against offering students practice opportunities online and providing automated feedback right then and there is that that way, they will never learn to work independently. Since I am working on e-assessment a lot and with many different courses at the moment, this is a fear that I definitely need to take seriously. I don’t believe that the danger is as big as it is sometimes made out to be, but I do believe that there is a vicious circle to be aware of.

It all starts with the instructor having the impression that students are not able to organize their learning on their own. Since the instructor wants the students to succeed, she offers them a clear structure, possibly with bonus points or other kinds of rewards, so they have a safe space with instantaneous feedback to practice skills that are required later. So far, so good.
Now the students are given this structure, and get used to working on problems that are presented in small portions and with instantaneous feedback. They start believing that it is the instructor’s job to organize their learning in such a way, and start relying on the instructor to provide both motivation and bite-sized exercises.
Which the instructor, in turn, notices and interprets as the students becoming less and less able to structure their learning.
At this point it is very easy to fall in the trap of trying to provide an even better, more detailed, structure, so that the students have a better chance of succeeding. Which would likely lead to the students relying even more heavily on the instructor for structure and motivation.
Teufelskreis

It is easy to fall into a vicious circle where the instructor feels like they need to provide more and more structure and motivation, and the students feel less and less responsible for their own learning.

So what can we do? On the one hand we want to help students learn our content, on the other hand they also need to learn to learn by themselves. Can both happen at the same time?
I would say yes, they can.
The first step is recognizing the danger of entering into this downward spiral. There is absolutely no point in hoping that the students will take the initiative and not fall into the trap of relying on us, even if we point out that the trap is there. Of course they might not fall in, but whether they do or not is beyond our influence. We can only directly influence our own actions, not the students’, so we need to make sure to break the spiral ourselves.
The second step is to make sure that we resist the urge to give more and more detailed exercises and feedback.
The third step is to create an exit plan. Are we planning weekly quizzes as homework that students get a certain number of bonus points for? Then we should make sure that over time, either the number of bonus points will decrease, the time interval will become longer, the tasks become more difficult, or a combination of all three. The idea is to reward the behaviour we want just long enough that students establish it, but not any longer than that.
And of course, last but not least, instead of giving students more structure, we can help them learn the tools they need to organize their learning. Be it training skills to organize yourself, or helping them find intrinsic motivation, or teaching them to ask the right questions so they can walk themselves through complex problems until they find an answer.
It’s a pretty thin line to walk, and especially the fourth step might really be out of an instructor’s control when there is a lot of content to go through in very little time and the instructor isn’t the one deciding how much time is going to be spent on which topic. Most TAs and even many teaching staff won’t have the freedom to include teaching units on learning learning or similar. Nevertheless, it is very important to be aware of the vicious circle, or of the potential of accidentally entering it, to be sure that our best intentions don’t end up making students depending on us and the structures we provide, but instead make them independent learners.