Harvard professor Eric Mazur's Peer Instruction method

In this article, you will find:

• That education is more than transmission of information;
• That peer instruction, a form of active learning, can be experimented with in a university setting;
• And an easy recipe for peer instruction.

"What if I told you that this is not the best way to teach?"

That is what award-winning Prof. Eric Mazur from Harvard University has been advocating since the early nineties. Mazur started teaching physics at Harvard in a most traditional setting. He was convinced he was very successful at his job, and indeed, his students’ feedback was excellent.

All was good until the day he found a test designed by Prof. David Hestenes aimed at evaluating the conceptual understanding of the Newtonian force, a fundamental piece of knowledge when studying physics.

To his horror, Mazur realized that, according to this test, his students had learned almost nothing. Yes, they were certainly able to solve complicated exercises with sophisticated math, but they were unfit to transfer their apparent knowledge into a different context. Moreover, these conceptual questions are supposed to be solved after a few minutes at most, and certainly do not require pages of calculations. He was discouraged, to say the least.

Later on, however, he witnessed a sort of miracle, almost by accident. After having spent half an hour explaining a concept to no avail, he took a break and let the students discuss the subject amongst themselves. Chaos ensued in the Harvard auditorium: people left their desk and migrated to other areas, half of them trying to convince the others, the other half listening and debating them. But the most incredible fact is that after only a few minutes, the students serenely returned to their seats. Then, one of them told Prof. Mazur: "We’ve got this, you can move on".

The Peer Instruction method

This was probably the moment the Peer Instruction method was born. Together with evidence-based considerations about the inefficacy of traditional lectures, Mazur developed the following teaching method:

1. The students are given the course material in advance so that they can read it at home. This way, they can prepare for the lectures. (One can devise strategies to give an incentive for this step).
2. During class, the professor briefly summarises the topic and provides the students with a conceptual multiple choice question. This sort of question focuses on one key notion at a time, and implies for instance to predict the impact of a variable over a system, without using equations or written reasoning.
3. Students think by themselves (silence prevails) and answer. They are also asked about their degree of confidence (for example: “I just tried”, “quite convinced”, “very convinced”). The professor notes the poll’s results and faces three potential outcomes: if less than 30% of students chose the correct answer, that means initial understanding is very poor, hence it is necessary for the teacher to revisit the concept, and they repeat step 3. Between 30 and 70% of the students in attendance have answered correctly. In this case, students are invited to spend some time trying to convince each other of the choice they made. After a while, they must answer the same question and include their confidence level again. The professor can now move to step 4. But if the rate of success is greater than 70%, this result is satisfying: the professor explains the answer and proceeds to the next concept, going back to step 2. What about the people that didn’t get it right the first time? If they still have uncertainties, this step will have given them a red flag, so they have more elements on how to prioritise their study before the exam.
4. According to Mazur this is usually where the magic happens: results show that both the percentage of correct answers and the confidence of students increases. The teacher can safely move to the next topic.

Peer Instruction benefits

Besides boosting the performance of the students, this method has several other advantages:

• When answering the quiz, the students are forced to be committed to an answer, to various degrees of conviction;
• Students have to externalise their answer, turn to their neighbor and convince them of their reasoning;
• By doing this, a student’s answer changes from a simple fact to an articulated reasoning;
• Eventually, students get emotionally invested in the learning process: research in cognitive sciences found that the emotional kind of encoding is among the best to store information.

Information is now easily accessible to students: classroom-taught lessons can no longer be used solely to transmit this information. To allow your students to get the most out of these class sessions, always remember to encourage active participation in the classroom!

For more information:

• Confessions of a converted lecturer, par Eric Mazur
• More info on ConcepTests 
• Hestenes, D., Wells, M., & Swackhamer, G. (1992). Force concept inventory. The Physics Teacher, 30(3), 141–158