Highlights from a Conversation with Eric Mazur

This article was originally published in the Fall 2002 issue of the CFT’s newsletter, Teaching Forum.

As this year’s Department of Physics and Astronomy Forman Lecturer, Eric Mazur of Harvard University gave a talk titled “Understanding or Memorization: Are We Teaching the Right Thing?” on September 26, 2002. Professor Mazur holds a triple appointment as Harvard College Professor, Gordon McKay Professor of Applied Physics, and Professor of Physics at Harvard University. An internationally recognized teacher and scientist, he leads a vigorous research program in physics education as well as in optical physics. In 2001 he was honored with one of the first National Science Foundation (NSF) Director’s Distinguished Teaching Scholar awards, the highest distinction conferred by the NSF.

One reason Professor Mazur’s teaching model receives so much attention is that he focuses not on “coverage,” but on “uncoverage.” The term “coverage” refers to the familiar process of covering the contents of a course. “Uncoverage,” by contrast, refers to the process of surfacing common misconceptions, and enabling students to see how complex ideas in a discipline or course fit together. (See Understanding by Design by Grant Wiggins, available in the Center for Teaching library, for more about these terms.)

Professor Mazur pursues “uncoverage” in various ways, including a process called “Just-In-Time Teaching.” He spoke about these strategies and the role technology plays in his physics courses at a luncheon hosted by the Center for Teaching during his visit to Vanderbilt. The following are selected excerpts from those remarks.

What do you see the role of technology playing in improving teaching?

Technology is not a magic bullet. One of the greatest IT inventions is the book, and we haven’t really used it effectively for 500 years. I think that if you look at the way many people are using information technology in education right now, it’s not very interesting. It’s used as a new way of presenting old material, whereas really the way you want to use technology is to facilitate new modes of learning. So I think the technology is nice when it can do something for you that you couldn’t do before.

What is special about the way you teach Introductory Physics at Harvard?

In a nutshell, I move all of the information transfer out of my classroom. Education is basically two things. One is transfer of information; the other is assimilating that information. Just one of the two is not good enough–you need both. But if you step into a random classroom, especially in the sciences, and you look at the teacher and the students and you ask yourself what is happening here, in most of the classes today, it’s mostly transfer of information-the instructor presenting material to the students. The assimilation of that information is supposed to happen outside of the classroom, later – some of it perhaps in the class, but you can’t assimilate that much material that quickly. It takes time. You need to take the information and adapt it to your own experiences, resolve misconceptions you may have, and so on. If you ask yourself, what is the more difficult of those two steps–getting the information or assimilating it–I think we’d all agree that it is assimilating the information that is really the hard work.

Before there was information technology (and I’m not talking about computers here or even VCRs, I’m talking about information technology that’s 500 years old-the printing press) the only way to transfer information was to put the sage on the stage and disseminate the knowledge to others, whether they were humanists or scientists or whatever. It was the only way to transfer information to the next generation of scholars. Now, we have books and many other means of disseminating information-but somehow we’ve stuck to this very old method of transferring information, the lecture. In fact, lecture comes from the Latin word for “to read,” and students used to accuse me of reading from my notes to them, so it was sort of ironic that I had not started thinking about this much earlier in my teaching career.

So in my classes I have moved this information gathering stage out of the classroom. Students have to come to class prepared. In many business and law schools, courses are taught using the case study method. You better come to class prepared because the faculty member might point at you and say, “You, explain your point of view.” You are put on the spot. Therefore, there is a good motivator for coming to class prepared. If you haven’t read the case before coming to class, you are likely to or you might well make a fool out of yourself in front of your peers. That’s a high-pressure motivator, which I don’t want to use in my classrooms because I want to really use the classroom for discussion and to foster this interaction between students.

So for a long time I struggled with finding a good model, a good way to get the students to prepare themselves for class. I would assign a chapter to read, but if I didn’t offer a carrot and a stick, half of the class would not read it, even though I told them, “Look, you have to read that chapter anyway. You can read it before class. You can read it after class to understand what happened in class. Or you can read it two weeks before the exam in order to pass the exam. Obviously, you’re going to get the most mileage if you read it before class.” That was not good enough.

Then I came across this teaching technique called “Just-in-Time Teaching” that is the icing on the cake. It was developed at the Air Force Academy and a number of other institutions.

How does Just-in-Time Teaching work?

The basic technique is the following (and I think it works in any field-it’s not discipline specific). In addition to a reading assignment, the students have to complete a web-based form that has three questions. Two deal with the content they have read and are very hard questions; they are not questions that you can lift out of the text. In fact, not many students will get them right, but the students are graded on effort, not on correctness. The most important part is the third question, which is always the same: “Please tell us what you found difficult or confusing in this reading assignment. If you did not find anything difficult or confusing, tell us what you found most interesting.”

Normally when I prepare to teach a class, I sit down and look at the material and decide, based on my experience and other things, what I think the students need to be lectured on. But that process is based on my experience, which might be very different from that of the students I am facing because I am teaching pre-meds and engineers. Students are going to have very different careers from my own, and I’ve noticed that I’ve often made the mistake of projecting my own experience onto students who actually are very different and will be very different from me.

With the Just-in-Time system, I have an opportunity to actually look at what the students tell me. They’ll say “I found topic A difficult” or “I found topic C difficult.” The night before I teach, I press on a button on my Web site and out come all the answers to that third question. And I can quickly scan them and decide what needs to be covered-or uncovered-in class.

What type of feedback do you receive from this third question?

Basically the feedback I get from students falls into three categories. One is “useless”-responses like “everything is clear” or “everything is unclear.” There’s not much you can do with that information.

Then there are responses that fall into a category I call “real important things that you would like to address in class.” In fact, sometimes students will have questions that as an instructor you might not have thought of. And I use them the next day in class by just bouncing them back to the students. Instead of explaining to them what the answer to the question is, I just pose their question as a ConceptTest, as a question to the class. I’ve had some students come to me at the end of the lecture saying, “Professor Mazur, that question you asked me, that was exactly the question I had.” And I look at the student and say, “That was your question!”

And then there’s this middle category of students who have small difficulties, small semantic difficulties. They’ve misunderstood a word or there’s something minor that is an obstacle to understanding that would take too much time in class but would fairly easily be resolved one-on-one if you could say, “Look, this word, you’ve used it out of context.” Or, “Look, you’ve misunderstood the meaning of that part of the equation.”

How do you respond to these students?

My fingers were itching just to send emails to these students. So I made a link so that you could click on the name and it would start my email program and I could type in an answer. And then I fairly quickly found myself sending forty emails, many duplicates of one another. Because you read the feedback from one student, you answer that, and ten students further down there’s one that has exactly the same problem. You go into your sent mail folder, you cut and paste.

Then I said, you know, I should really build this on a database. Take a computer database. The assignment is a record in the database. The student is a record in the database. The answer of the student to the assignment is a record in the database, and so should my answer to that student be. This way I can just use it again. The program knows what the assignment is, so each time I press the button that says “answer the student,” it can pull all the answers I previously sent on that particular subject, show them to me, I click, and it creates a personalized email to the student with that student’s feedback, using my answer to another student’s feedback, but personalized: “Dear David,” blah, blah, blah, and so on, “I hope this helps, Eric Mazur.” And the students gets an email responding to his or her feedback, never knowing that I might have sent that ten years ago to another student, because I recycle it from year to year.

This has actually allowed me to interact a lot more with students in very little time. I want to emphasize here I don’t spend more than an hour per lecture preparing myself that way. And, of course, it gets faster as you do more, because the database gets cleverer. So we have now implemented it in a way that other instructors can use it. Yale, MIT, Brown, and the University of New Hampshire are now beta test sites, and we’re slowly adding more and more people. 

How do you respond to students or even other faculty members who seem doubtful of new methods of instruction?

Phil Sadler is an astronomer who works at the Harvard-Smithsonian Center for Astrophysics. He told me an interesting anecdote, which I think applies here. Suppose you are a tennis coach, and you are coaching a student, and you find out she is not holding her racket correctly. You say, “You should really be holding your racket like this, instead.” So she changes her grip, and all of a sudden, while she was playing a decent game before, now with her changed grip, the balls are going in the net and out of bounds, and she gets annoyed. She tells you, “Why don’t you let me play the way I played before? I played better.” As a coach, you know that she’ll improve very soon and that she’ll play better than she played before.

The same is true in a physics class. If you plot performance as a function of time when you’re making a change, it doesn’t immediately get better. It often gets worse before it gets better. It sometimes takes a while, one or two months, or maybe only just before the exam that you start to really improve. In between, there is that period of big discomfort.

If you are one month into your term and you have fifty percent of your students who are unhappy, the logical thing to do, or to think at least, is, “I must be doing something wrong. Let’s go back to what I was doing before.” Unless we all realize that we have to go through a period of pain and change, we are never going to get to that place to where we are actually performing better than we before. We just have to face that collectively.