Journey to Interleaved Practice #2
In my last post, I gave some background to the study that I am undergoing with my calculus students this term. In this post, I want to share some of the tools and methods I used to make the path towards interleaving clearer to me.
I have been a fan of interleaved practice for some time since it is well-known in the scientific community to be a successful strategy for learning (here I am thinking about learning as a flexible and long-term change in long-term memory that can be measured through test performance). However, when thinking about how to successfully implement interleaved practice it feels like a very daunting task and there are a lot more questions compared to answers:
How many questions do I assign at each step?
How do I best mix-up all the questions?
Should some topics be more weighted compared to others?
In what order to I teach the topics? Should I also interleave the way I teach the topics?
So what I did was draw some inspiration from a Slack work-group where Yana‘s husband Fabian (congrats on the recent wedding!) put together an Excel worksheet that gave a potential teaching and quizzing structure using an interleaved approach. If you open this link, you can see space to enter the topics, as well as the number of classes you have, and finally the number of questions you want per quiz. Hitting the “Do Quiz” button will create two lists: one that suggests topics to teach during any particular class, and one that suggests the topics for each quiz (which I assume happens at the beginning or end of each class).
I took this basic structure and decided to create a list of potential topics for my integral calculus course. I divided this list into six “strands” each with a certain number of “lessons” (note: I am not done finalizing this list yet – it is a work in progress). Basically I sat down, went through each chapter of the textbook and made a map of how the topics were interconnected. For example, the Sequences and Series section of the textbook discussed geometric series. Well, I could easily do this in Strand One so that students have an introduction to sigma notation before working with sigma notation with approximations. Then I could circle back to sigma notation later in Strand Six when working with Taylor Series, effectively spacing out our work with sigma notation throughout the semester. Each placement of a lesson within a strand was a calculated choice to try to space out the important ideas as best I could.
Now that I had a list of topics, I could input this information into Fabian’s worksheet and get an idea of how to interleave topics. I decided that I had already interleaved teaching topics as best I could, so I ignored the top output. I chose 4 questions per quiz and focused my attention on the bottom output. Using the output at the bottom as a model, I created a new page that listed the four questions I wanted to include on each quiz. See Sheet 3 of this workbook for that page (again a work in progress).
My final decision was not to use quizzes, but homework assignments instead. That is, at the end of each lesson, I give a PDF handout like this one to each student that is due at the beginning of the next class. This particular PDF came after the lesson on the Fundamental Theorem of Calculus Part II (FTC II). Notice that there are questions about the FTC II, but there are also questions on the topics of geometric series and the definition of the definite integral as well (the limit of the Riemann sum question).
To ensure that students complete each 4-question homework assignment to the best of their ability, I check for completion only at the beginning of class. We then take the questions up as a class – focusing on the “hard” questions that students are having trouble with. So far things have been going very well. The first test is coming up next week, and I will definitely try to blog about any interesting information I gather from looking at their responses.