Eric Mazur, a trailblazer in teaching
Harvard professor Eric Mazur, who developed the peer instruction method, was at EPFL for Education Day. The physics professor – not one to settle for the status quo – shares his thoughts on teaching.
Eric Mazur started teaching at Harvard 35 years ago. He inherited a course no one wanted: teaching physics to non-physics majors. Students included future engineers and doctors, who viewed the class as a bothersome requirement. But the fledgling assistant professor threw himself at the challenge. Surprisingly, the students did well on their exams and rated Mazur’s teaching very highly. Mazur thought he must really be a good physics teacher. “But that was a complete illusion,” he asserts now. Mazur went on to become a pioneer in education, a field he continues to shake up. Interview.
Can you tell us more about peer instruction?
Peer instruction is interactive teaching by questioning, because education is not simply transferring information. You don’t learn to play the piano by listening to pianists. The most important part of education is for students to make sense of the given information, assimilate it, and develop skills, mental models and ways of thinking. So I decided to throw the information transfer part outside the class and focus on the assimilation process – the “aha!” moment in the classroom.
What led you to try out this way of teaching?
In 1990, I read an article in the American Journal of Physics that claimed that physics students who don’t stay in the field learn very little. I read that and thought, “That’s impossible.” But I’m a scientist, and I’ve learned not to make assumptions. So I decided to give my students the test mentioned in the article. When I looked at the results it was clear that I had a big problem. That was a turning point in my career: until that moment, I thought I was a very good physics instructor, but then I discovered that I was basically helping students learn tricks, and that they weren’t learning what I wanted them to learn.
What did you do?
After I gave that test, the students were shocked and began to worry about their exams. They asked me if I could explain the answer to every single question. I remember going over a problem that, to my mind, was very easy. I spent 20 minutes detailing my reasoning, but they still didn’t get it. I didn’t know what to do, but I did know that half of the class had come up with the right answer. In a moment of despair, I said: “Why don’t you just discuss this with each other?” Something happened that I had never seen in my career. The 250 students in the auditorium spoke among themselves, and after two minutes they figured the problem out. This is what we call “the curse of knowledge” or the “expert blind spot.” The more expert you become, the less aware you are of conceptual difficulties, and the harder it is to teach beginners. This experience led me to peer instruction.
How does a peer instruction class work?
Students read outside of class, and in the classroom I teach by asking a series of questions. Students have to vote. After that I ask them to find someone who gave a different answer and try to convince each other. After two or three minutes they have to vote again. The number of students with the right answer often goes way up. When students talk to each other, they articulate their thinking and understand better. I conclude by giving a short explanation and continue with the next question.
What changes did you observe with this approach?
The first thing is engagement. Everybody is talking to each other, no one’s sitting there half asleep. The other thing is the circulation of information, which is no longer only from the teacher to the students, since they give each other feedback. The learning gains doubled after the first year and tripled after two to three years.
What are learning gains?
First of all there’s content. If you figured things out by talking and had that “aha!” moment, you know that content for life – even if you forget the answer you have the reasoning and it sticks, because our brains don’t store facts but mental models. The other thing is that it inspires confidence and increases self-efficacy, which refers to the belief in your ability to succeed. That’s very important. It means that if students are confronted with physics problems in the future, they won’t have a negative reaction. Finally, articulating their thinking represents an extra step for the students. That is another huge benefit. This approach helps students with lifelong learning. The real key to education is not listening to somebody talking. It’s doing it, experimenting and thinking rather than taking notes.
What are the main challenges facing education in the coming years?
Higher education is almost like a cartel. To be an electrician you must have a license, to be a lawyer you have to pass a bar exam, to teach at the university you need a PhD. A PhD involves doing research and writing a thesis, but not teaching. And somehow at the end of your PhD you’re supposed to know how to teach. It doesn’t make any sense to me. Furthermore, by writing the exams that test their students’ knowledge – and the quality of their teaching – professors are in effect evaluating themselves. This doesn’t happen anywhere else in society, and it can’t go on. The reason higher education has changed so little over hundreds of years is because we have very little accountability and we don’t treat this very important profession like we do others.
What needs to change?
We have to drastically change our approach to assessing students. In most courses, exams isolate students from each other and cut them off from any outside source of information. Is this a situation they will encounter later in life? Never. So we are actually measuring something that has no value, because the students will never work that way. With that type of assessment, students focus on memorizing the material, and three days after the exam they’ve forgotten most of it. It would be much better to really engage their thinking skills, tapping into their creativity and spirit of innovation.
I also think that having education focus more on collaborative skills is crucially important, because that’s how society works. Education focuses on the individual, and as a result people are unable to work together.
Is that how you adapted your course?
Yes, I completely changed my course to be project-based. Students work in teams and have to carry out six one-month projects. I form new teams each time around, making sure that people don’t already know each other. I also make sure that the projects are so difficult that no single individual can complete them. By doing that I create a sense of social responsibility for learning. Moreover, I try to ensure the projects have a “social good” component – for society, an organization or a group of people.
How do you make sure everybody on the team does their part?
At the end of each project, I ask the students for feedback on themselves and each other. I distribute the credits on the basis of these peer evaluations. If everybody does their part, everybody gets the same number of credits.
Can you give us an example?
For the part of my course that deals with waves, insulation and sound, I suggest the students do a project on El Sistema, a program begun by the economist and musician José Antonio Abreu in Venezuela in 1970s that offers musical training to young people living in poor neighborhoods. They begin their training when they are children, practicing first on cardboard instruments. The instruments don’t make any sound, but they look beautiful. So I ask my students to design a beautiful-sounding musical instrument from recycled parts for El Sistema. They use physics to improve their instrument. This encourages them to read the textbook, because the content – rather than a goal – becomes a way of achieving something that’s more meaningful in their mind.
How do you grade your students?
At the end of each month, we hold a fair in a public space where they exhibit their projects, and they have to take an oral exam. I appoint a panel of judges, since instructors who grade their own students have a conflict of interest – we can’t switch from coach to judge. So I ask my colleagues to come and query my students. I give them the brief and tell them what physics the students should know. My colleagues then select one person from the team and ask him or her questions. I tell the team that the student’s answers will determine the team’s grade. So if there’s a weak student, the others work very hard to bring him or her up to speed. It works beautifully, and it validates what I do in my class in the eyes of my colleagues. What’s more, I get to stay the good guy. But switching over to this project-based approach takes a lot of work and requires you to think differently.