Papers: "Teaching problem solving through cooperative grouping. Part 1: Group versus individual problem solving", Heller, Keith, & Anderson 1992; "Teaching problem solving through cooperative grouping. Part 2: Designing problems and structuring groups", Heller & Hollabaugh, 1992
Presenter: Shanna Shaked
What are we trying to teach with group problem solving?
The availability of group resources allows teachers to assign more difficult problems that probe deeper levels of knowledge. These are not the sort of problems typically found in textbook problems, which can typically be solved through simple equation manipulation.
How can we encourage the use of prescribed problem-solving techniques?
Put difficult, context-rich problems on the exams for which the techniques taught in class will be useful.
Does Mastering Physics work?
It works well for teaching the equations and getting students used to manipulating variables. It also contains videos that provide background context for the use of equations/concepts. Would Sapling Learning be better? It's not attached to a textbook, and the problems are designed by physics education researchers.
Strategies for teaching problem solving:
1) ask for a "guess" at the answer before applying equations (to stress the conceptual nature of the problem).
2) FIND, KNOW, PLAN, CALC, CHECK (Shanna's method)
3) Design a flow-chart to explain problem-solving strategies to students
4) Design worksheets for students that encourage each step of the process
Students often do not respond well to this - many feel that they can just plug in the numbers they see in the problem and go from there. Should students start in groups so that they have to get used to justifying their methods, then try individual problem-solving?
Why do we teach problem-solving in groups?
Groups were able to solve higher-order problems correctly more often than individuals, even the highest-performing students in a class. Groups containing high, middle, and low performing students works the best. High ability students provide leadership while middle and low performing students ask questions that usually lead to the correct solution.
Students were trained to problem-solve in groups did better as individuals than students who were trained to problem-solve individually.
The magic number for groups is 3 students - one high, one middle, and one low performing student. It should be homogeneous in gender or 2 females and 1 male.
OK, so let's get this going at UCLA. How do we get the students' buy-in?
Try having part of the exams be done in groups. Students will then be motivated to practice working in groups in order to do well on their exams. To cut down on inter-group conflict, have the students report on how well their group functioned, and identify ways they can increase their group performance.
Make group-work part of the classroom culture! Does it really matter if students are enjoying themselves, or are we only concerned with the way they are learning? If you are trying something you think students won't like, make discussion sections mandatory to encourage them to stick with it, even when it gets difficult.
What if your group member doesn't show up, and you've carefully arranged groups? Have a practice group session the week before the exam, and if you don't show up to that, you get a 0 on the group portion of the following exam. For the students who prefer individual work, the students can "opt in" to discussion section for various sections of the quarter. That would mean that the students would be committing to coming to discussion sections during that time, and would be penalized for absences. Could you offer an alternative discussion section for students that don't want to do this sort of problem-solving, and just want a traditional discussion section (with students doing individual work).
What about introverts or people with social anxiety? Are there people that just don't learn well in groups? How valid are these findings? How can we as physics teachers evaluate the validity/usefulness of physics education research? Maybe all introverts should just learn to deal with the relatively small amount of group learning that has to happen in a college classroom.
Group roles matter! Each member of the group should have a well-defined role (preferably with examples of what to say to the group). Roles include: manager, recorder/checker, skeptic, summarizer. Has anyone tried this? Does it feel silly in a college classroom? It might seem silly at first to the students, but it has been shown to work well, even for higher education.
Classroom structure matters! Discussion sections work best in classrooms where desks are not nailed to the floor so that the groups can move around. There are only so many classrooms available. How can we get around this?
How can you be certain that any gains in understand are not due to the introduction of new problem-solving techniques, or to the use of group-work in discussion sections? No studies have been done to show that group work is better than individual work for college students (that we know of). There has been some research that shows that students understand material better after having discussed it with peers (peer-learning is better).
Teaching Physics to Life Science Majors
Are they prepared for physics classes? It varies, but most don't care about the physics, or don't see the value in it. Students are most motivated by the grade in the class. How can we get them to be motivated by the usefulness of physics in their future careers? Maybe give them context-rich problems that relate to life sciences. Maybe show them an MCAT problem that relies on physics knowledge. Could we use the Physics section of the MCAT as a pre-test to show them why they actually need this class.
In the life sciences at UCLA, 1 in 4 classes are not experimenting with "flipped" classrooms, though the same innovation is not necessarily taking place in the discussion sections.