Three evolving thoughts about flipped learning

8617442811_567440ed8c_mWhile specifications grading continues to unfold in class, I’m also still using and refining the flipped learning model. Recently I had time to reflect on how I’m implementing flipped learning in my classes, and I noticed that some of my thoughts on flipped learning have evolved over the last few years, including some breaks from things I’ve written here on the blog. Here are three of those thoughts that stood out for me.

What I used to think: Pre-class activity in a flipped learning model is about mastering content-oriented instructional objectives.

What I think now: Pre-class activity is for generating questions.

I attended a talk by Jeremy Strayer last year, and he said something that stuck with me: that the purpose of pre-class work in the flipped classroom is to “launch” the in-class activity. In flipped learning we certainly want students to pick up fluency with basic content and learning objectives prior to class. But I think Jeremy’s point is that content delivery shouldn’t be the primary purpose of pre-class work. Rather, it should be to prime the student intellectually to engage in whatever high-level tasks we have devised for the in-class meeting.

This point was echoed in this study from Stanford which suggests that while the flipped learning model in itself is an improvement over a standard lecture-oriented model, there are even stronger learning gains among student when their pre-class work consists of open-ended explorations of concepts that precede a more text-based study of those concepts. The Stanford study suggests that “flipping the flipped classroom” in this way is the best approach.

So based on all of this, I’ve started learning away from “content delivery” as the main purpose of pre-class work and toward the notion of question generation. For example, when we get around to studying Eulerian paths in the discrete structures course, the pre-class activity will go like this:

  • Read the section in the book about Eulerian paths and circuits and pay particular attention to the definitions. I often provide some optional additional videos for students to use if the book is too opaque.
  • Look at a graph and identify whether a given path/circuit is Eulerian or not. (Thereby ensuring that students know the basic terminology.)
  • Now, go play with the following script in Sage:
    g = graphs.RandomGNM(10,20)
    print g.degree_sequence()
    g.is_eulerian(path = TRUE)

Run this script as many times as you need, changing the 10 and the 20 to adjust the number of nodes and edges if necessary, until you get TRUE for the is_eulerian output. What do you notice about the numbers in the degree sequence list? What’s different about this degree sequence from the degree sequence for a graph that isn’t Eulerian?

In this way, students may not have mastered a smaller list of basic learning objectives than they used to, but they are coming to class more invested in the main idea of the section (the fact that a graph has an Eulerian path if and only if it has exactly two vertices of odd degree) and with their antennae up, so to speak.

Here’s another evolving thought:

What I used to think: Students in a flipped classroom need to have some graded measure of accountability when they arrive at class (an entrance quiz, etc.) to ensure that they do the pre-class work.

What I think now: Accountability doesn’t have to look like a quiz.

I used to take it for granted that in order to get students to do any sort of work, I needed to attach a grade to it. My mind changed on that last semester in the first-semester discrete structures class I was teaching.

I was giving entrance quizzes that covered the basic learning objectives from Guided Practice, pretty much one of these every class day, just as I had done with almost all previous flipped classes. But I began to notice that the quizzing was causing more problems than it solves. Students were telling me – both directly through their comments and indirectly through body language – that they were tired of being quizzed all the time.

So I decided to try something radical: Don’t give them a quiz at all. Just let them do the Guided Practice and let that be that. What surprised me was that not only did the completion rates for Guided Practice not get worse, the students’ comprehension of basic ideas improved, and their work in class improved as well. And we had more time in class for the high-level work inherent in a flipped class meeting.

This may not work for all student demographics, but I also have to say these students were mostly freshmen and sophomores, and not all intrinsically motivated by the material. And reducing the quizzing improved their work, at least to my eye.

So in my classes now, there is accountability but without graded quizzes. It really looks more like responsibility than accountability. I’ve been sticking to a simple message that students are adults – and preparing for a class is an adult responsibility that they are expected to perform; giving them detailed guidelines and lots of help in getting prepared; and spelling out what failure to prepare does to their learning. If students show up unprepared, we move on as if they had prepared – no exceptions. The “accountability” consists in holding hard boundaries on what we will and will not do in class.

And so far, I’ve had no problems with students showing up unprepared. Maybe some do show up unprepared, but they’ve been able to participate and learn anyway. (I have clicker question data to prove it.)

I think this, along with my shift to specs grading, signals a more general shift in my teaching toward the concept of andragogy as opposed to pedagogy – treating students as responsible adults rather than as children who need constant supervision and rule-setting. I have more to say about that later.

Finally, a third evolving thought:

What I used to think: The in-class instruction in a flipped class should focus primarily on active student work with little to no lecture.

What I think now: The in-class instruction should focus on two things: Answering questions, and engaging students in high-level tasks – and lecture can play an important role in both.

This post from Kris Shaffer left me in thought for days, especially his thoughts about the “geographical flip” versus the “timing flip” and his preference for the latter, which is also rooted in that Stanford study I linked earlier. Kris does something here which I wish more of us would do, which is to transcend the shopworn “lecture sucks” narrative and instead try to craft the best pedagogy that combines the most effective uses of several modalities.

I’ve found myself being much more amenable to lecture in class these days. I plan those lectures: They are short, surgically targeted at the most common misconceptions or aimed at specific student questions that were voiced during the pre-class assignment.

One of my colleagues once told me that he loved teaching because as a mathematician, he loves difficult problems, and teaching is a problem whose solutions generate even more and harder problems. I enjoy this aspect of teaching too, along with the continuous evolution of thought that it requires.

Want to continue the conversation? Follow me on Twitter (@RobertTalbert) or on Google+, and share this article on your networks using the social media buttons above.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s