In the past couple of years I have done a great deal of thinking and reading about the topic of homework and wrestling with the design of homework in my classes, particularly in light of what I have learned about how people learn from attending several Learning and the Brain conferences. I have also become a huge fan and user of a “flipped classroom” model which necessitates rethinking the role and purpose of homework entirely. As a result of my new knowledge and new pedagogical approach, most of the assignments I have used in the past have undergone significant change and hardly resemble what they looked like even two years ago. I wanted to share one example with the hope that my experience may encourage others to rethink how they are using homework.

I currently team teach a Molecular Biology course at the Loomis Chaffee School that is a two term course for seniors and juniors who have already taken biology and chemistry and are looking to continue their study of biology at an advanced level. Since the students are coming from very different chemistry backgrounds, in the first couple of weeks of the course, we review basic acid/base chemistry, weak acid equilibria and then apply these concepts to the study of amino acids and proteins. In the past we have given the students a problem set to work on during this first unit that includes problems on all of the basic concepts and several that require some extension and transfer to new situations. The students were encouraged to work collaboratively on the problems and frequently small groups of them did get together and helped each other to understand and solve the problems. Ultimately, each student was required to hand in his/her own set of solutions to the questions. Here is where the problem came in…we really had no way knowing for sure who had truly mastered the concepts and who had not. A high grade on the problem set was not a good indicator of mastery since we had little way of knowing anything about the path each student took to his/her answers. Did they solve them on their own with no outside assistance? Did they work with a peer who did most of the “heavy lifting?” If they got a problem wrong, can we figure out where they are stuck conceptually and help them see their mistake? Not likely.

The next year, we tried a different approach with the same problem set to try and gain more insight into where the students were getting stuck and to better monitor who was doing the work and who was “along for the ride” when it came to actually solving the problems. We had the students work in small groups on the problem set in class where we could go around and listen in and assist when needed. This was certainly better because it allowed us a chance to clear up misconceptions before they became “permanent” and guaranteed that the students would spend at least the amount of class time we dedicated to working on the problems. While this format was “better”, it was far from perfect. We still had no way to make sure that all of the students were mastering the concepts along the way since we did not collect and correct their work until the end.

This year we tried a third approach that has gotten us closer to our goal of knowing that each student has mastered the concepts and at the same time allowed us to monitor the path to mastery so that we could clear up misconceptions early and provide feedback often. Here is what we did: We looked at all of the problems on the problem set and first scaffolded them so that the problems were ordered by difficulty level and built better on one another. We had to rewrite several and scrap other ones we have used in the past. We then divided the students up into groups of 3 and instead of giving each group the entire problem set to work on, we handed the problems out one at a time on slips of paper. Each group only received one copy of the problem. They were instructed to work together on it and let us know when they thought that they had solved it. In order to get the next problem, one person from the group (selected randomly by the teacher!) had to explain the group’s answer. If the answer was incorrect or the selected person could not adequately explain how they had gotten to their answer, they had to go back and make sure of their answer and make sure that everybody understood it since the second attempt might have to be explained by a different person. Once the group had the right answer and could adequately explain it, they were given the next problem to work on. As the groups were working, we circulated around and were able to answer questions or clear up any misconceptions. As a result of this approach, the groups were frequently working on different problems and at different paces. Because we were asking different students to answer for the group each time, there was a built in mechanism to encourage collaboration and mastery from every member of the group. We also were able to hear from each and every student at some point and have a better sense of which students are struggling to master the material.

I have to point out that this approach takes significant class time and would not be nearly as easy for us to do if we were not also using a flipped model for instruction. We rarely use class time for content delivery, instead the students watch short video lessons we have prepared to introduce them to the content for “homework.” As a result, we can structure our class time as I have described above and learn much more about how the students approach the concepts and how they wrestle with coming to mastery.

While it is not perfect, it is definitely different. We have been pleased with many aspects of the change and are looking at all of our assignments and asking if there are ways in which we can improve them and how we can better use them to reach our goals for the students. Please feel free to post your thoughts or comments. I would love to hear what others out there are doing with respect to homework in their classes.