One of the most popular fads in education from kindergarten through grad school these days is “group work”. The reasons given for group work are often non-pedagogical—the most common being that students will have to work in groups in the real world, so they need to get used to it. Of course, a lot of work in the real world is done by individuals not working in groups, so this is a rather bogus reason. But not completely—big projects are done in groups, and students do need to learn how to work together on a common goal.
If done badly (as it often is), group work serves no function educationally. Top students in college have learned to avoid group projects at all costs, since they end up doing all the work, often at the last minute after other members of the group have failed to deliver on their promises. Even when the other students are diligent, the top students have to redo or correct the work of the less competent, with the result being that the top students end up doing more work for a less-good result than if they had done it by themselves. This naturally leads top students to be suspicious of working in groups, which does not prepare them for group-work in the “real” world, thus defeating one of the key goals of the pro-group-work advocates.
Even for students in the middle of the class, the extra work of coordinating effort and keeping everyone on task increases the effort so much that doing the work separately would still be faster for most projects.
Devlin makes an argument for “progressive” group work as opposed to “traditional” teacher-directed instruction. He provides evidence from a couple of studies by Boaler, but ends up rather dismissing one problem: “Of course, teaching math in the progressive way requires teachers with more mathematical knowledge than does the traditional approach (where a teacher with a weaker background can simply follow the textbook – which incidentally is why American math textbooks are so thick). It is also much more demanding to teach that way.” I suspect that the differences Boaler was measuring could as easily have been how much math the teachers knew, rather than the difference in teaching style. Devlin claims that “real work” consists almost entirely of collaborative group problem solving—he has no use for people thinking by themselves or for themselves.
But this post is not a polemic against group work. I often teach classes (generally at the grad school and senior college level, where most of my teaching has been) that require group work. The key, for me, is that the projects must be large enough that a single student can’t complete them alone—not just that the bottom-of-the-class students can’t complete them alone, but that the top-of-the-class students can’t.
Of course, few math or science projects in K–12 are that big—even 6-month-long science-fair projects are usually more effectively done by one student than by a team. Theater, dancing, music, and sports all provide real group efforts that cannot be duplicated by an individual (though each also has opportunities for solo work as well). Even writing classes offer opportunities for newspapers, literary magazines, and other group efforts in which multiple people can contribute effectively. In math and science, the opportunities for real group effort are much harder to find, particularly at the lower grade levels. About all I’ve seen that really uses team efforts are the FIRST robotics competitions.
Even at the college level, it is rare to have real group projects that aren’t more easily done by individuals until the senior level. By then, students have often been so burned by badly designed “group” exercises that they can’t work together effectively. So the challenge for teachers is how to teach students to work effectively in groups, without making a fake project that works against the pedagogic goals.
In one senior-design engineering course I’ve co-taught, the project is split into two parts. The first semester is dedicated to students forming teams and writing proposals for their projects, and the second is dedicated to actually doing the design and prototyping work. We required the book Teamwork and Project Management for the first semester and had students do group presentations on the content of the text (each chapter being presented by a different group). This undoubtedly got them to read the chapter they needed to present, and probably helped them with their presentation skills, but I was not very impressed with the book. In fact, I found it so badly written that I never got more than halfway through. If anyone knows a readable book on putting together teams to do engineering projects, I’d be interested in hearing about it.
I’ve also coached a middle-school math team, where the students individually were top of their classes (though still with a pretty wide range of ability), but had never worked together on math. Giving them problems that were challenging enough to get all participating was hard. Most of the time, what happened was that one of the students would be so far ahead of the rest that there was no group effort, just one student thinking out loud. I considered a partial success when I got two students thinking out loud at the same time. Full success was getting the students to check each other’s work and correct mistakes. This is a much more modest goal than real group effort, but the math team’s goal was to do well in a timed math contest, not to do real collaborative work on a project. Their best strategy was to have the top kids thinking out loud while the rest checked their work and made sure that every question was answered fully. (As it turned out, this strategy did work—they won the county math team contest. Two of the five also won the county individual math contests for their grades.)
Riley Lark has some good advice on training students to do group work for K-12 teachers, giving specific roles for each member of the team. He does not address how to design group work that is actually easier to do in a group than individually. I would be very interested to hear his suggestions on that more vexing problem.