Social Learning: Peer Instruction and Assessment Insights with Harvard's Dr. Eric Mazur

Can questioning transform teaching more effectively than lecturing? Join us on The Brighter Side of Education as we welcome Dr. Eric Mazur, the Balkanski Professor of Physics and Applied Physics at Harvard University, to discuss social learning and assessment.  Dr. Mazur is the innovator of Peer Instruction and Perusall. His motivational lectures on interactive teaching, educational technology, and assessment have inspired people around the world to change their approach to teaching. Our conversation focus on social learning strategies, as they align with the flipped classroom model, will be broken down into three parts: Peer Instruction, Just in Time Teaching and Perusall. We end with a discussion on rethinking assessment in the age of AI.

We start with Peer Instruction, its workflow, and its alignment with the flipped- classroom model. Dr. Mazur recounts his eye-opening journey from traditional lectures to an approach that prioritizes student engagement through questioning. This episode highlights how shifting from information transfer to active learning can elevate comprehension and retention, proving that social learning dynamics are key to successful education.

Then, we dive into the Just-In-Time Teaching methodology and its evolution into the Perusall platform. Dr. Mazur explains how these tools have transformed pre-class assignments into engaging, interactive experiences, leading to improved classroom performance. This segment delves into the principles and benefits of moving content interaction outside of class, setting the stage for more meaningful in-class learning.

We end with concerns over the shortcomings of high-stakes assessments and explore how the rise of generative AI necessitates a shift in our teaching and evaluation strategies. 

Transcript

Dr. Lisa Hassler: 0:10

Welcome to the Brighter Side of Education. I'm your host, dr Lisa Hassler, here to enlighten and brighten the classrooms in America through focused conversation on important topics in education. In each episode, I discuss problems we as teachers and parents are facing and what people are doing in their communities to fix it. What are the variables and how can we duplicate it to maximize student outcomes? Today, we explore peer instruction as a method to increase student comprehension and engagement, particularly through social learning styles for teaching and assessment.

Dr. Lisa Hassler: 0:43

Peer instruction, a transformative teaching methodology pioneered by Dr Eric Mazur, sparked a paradigm shift in education since its emergence in the 1990s. This innovative approach leverages social learning dynamics, resonating with educators striving to engage and motivate students effectively. Dr Mazur is the Bolkansky Professor of Physics and Applied Physics at Harvard University and Academic Dean for Biological, chemical and Physical Science and Engineering. Renowned for his groundbreaking work in nanophotonics and his dedication to transforming teaching practices, dr Mazur has authored numerous scientific publications, secured 52 patents and penned multiple influential books, including the acclaimed Peer Instruction, a User's Manual. With his unparalleled expertise, dr Mazur joins us today to delve into the power of peer instruction and share insights on transforming classroom learning. Welcome to the show, eric.

Dr. Eric Mazur: 1:41

Well, thank you, Lisa. Thank you for having me.

Dr. Lisa Hassler: 1:43

Now, your innovative approach to teaching has earned you the title of a catalyst for change in scientific communication and education. What inspired the development of peer instruction and how does it differ from traditional teaching methods?

Dr. Eric Mazur: 1:57

Well, first of all, this was never planned. I never aimed at becoming a catalyst for change in education. I simply had a problem to solve in my classroom and that's what gave rise to everything else. So I don't want to take credit for becoming a catalyst for change. That was an accidental byproduct, but essentially sort of recapping my educational trajectory, or my trajectory as an educator, I should say.

Dr. Eric Mazur: 2:24

I've been teaching at Harvard now for 40 years, and when I joined the faculty as an assistant professor in 1984, I simply did to my students what my instructors had done to me to lecture. I didn't know any better, and even though I knew from my own education that lectures were not necessarily always engaging and certainly not the place where one actually learns, I guess I forgot all of that when I was told okay, eric, you have to teach us introductory physics course for non-physics majors and I simply mimicked my own professors which you know, I think we all tend to do right, we project our own experiences I sort of naively thought that's how I learned my discipline and therefore that's how my students are going to learn it. And you know, to make matters worse, what happened was that in the beginning, all the signs that I was receiving were telling me that I was doing the right thing. My evaluations were very high, even though for most of my colleagues they were very low when they taught non-majors. My students did well on my exams and, you know, I pretty quickly started to think I was the department's best lecturer, which turned out to be a complete illusion. Department's best lecturer, which turned out to be a complete illusion.

Dr. Eric Mazur: 3:47

And it was not until, you know, 1990, when I taught six years, that sort of through an accident, I discovered that my students, yes, they liked my lectures and, yes, they were able to solve what I consider difficult problems on the exams, but they hadn't really mastered the most simple and basic principles. If I took a question out of context and presented that to them in a way that was different from the context they had learned in, they were totally stumped. And I think we all agree if you can't transfer what you've learned from the context in which you've learned into a different context, you haven't really learned. To a different context, you haven't really learned. So you know there were three possibilities. A my students were dumb, but it's hard to maintain that at my institution. Two, I was doing a really poor job, but that didn't make much sense either, because I got these high evaluations and they were not in exams or two. My method of testing them out of context was not really good. It took a long while for me to convince myself that it was actually me and my teaching, and then, sort of totally serendipitously, I stumbled into peer structure. I didn't call it that way because it didn't have a name, but I started shifting from teaching by telling to teaching by questioning.

Dr. Eric Mazur: 5:09

I realized that, you know, education is not simply the transfer of information, and that's what lectures focus on, right, they transfer information just as a book does, or the internet or whatever. There are so many ways now of transferring information. A lecture is precisely what it is. It comes from the Latin verb to read, it's the instructor reading the textbook, in a sense, to the students, and what needs to happen in order to learn is that, as a learner, you need to build the mental models that underpin the information that is being transferred. So you have information transfer and then you have assimilation of that knowledge, building the mental models that permit you to transfer things to another context, having the aha moments, if you want, and for most of my students obviously also in part given because they knew they were not going to continue in the discipline did not happen. So in a sense, peer instruction, which I introduced in 1991, was sort of the first implementation of the flipped classroom. I call it invert the sequence. So rather than putting the emphasis in class on information transfer and then have students on their own work on making sense of the information I said you know what, rather than reading my notes, on which I work so hard, why don't I just give the notes to my students, have them read it before class and then in class I teach by questioning. And that shift took off in part because I collected data that showed that it worked and also in part because it's so easy to implement.

Dr. Eric Mazur: 6:48

You just take a number of questions, you go to class, you ask a question, you ask students to commit to an answer. Initially we use scanning sheets and putting hands on their chest, indicating with the fingers you know the number of the choice, that these were multiple choice questions initially, that they selected. So I could immediately see it was sort of a low tech clicker, if you want. And then later we had a wired network of clickers and you know it gave rise to the whole clicker movement too.

Dr. Eric Mazur: 7:19

And then you tell students okay, now, after committing yourself to an answer, find a neighbor near you who has chosen a different answer and try to convince that person that you're right and he or she is wrong. And the first time that I did that, I was totally, you know, shocked by the engagement. I had 250 students in class that they all talked. They were completely oblivious to me in front of the classroom. I'd never seen this, and what happened was that very quickly I noticed that they would zoom in on the right answer and initially it was a shock. You know why can't I, as the expert, convince them as effectively as they can?

Dr. Lisa Hassler: 8:00

Yes.

Dr. Eric Mazur: 8:01

The reason is that I suffer from the expert blind spot. I don't know what it is to begin to learn, whereas if the student next to me has the right answer and the right reasoning and talks to me, he or she is more likely to have gone through the conceptual blocks.

Dr. Lisa Hassler: 8:20

Yeah.

Dr. Eric Mazur: 8:20

Whereas for me it was such a long time ago I forgotten about all of that right and, and so then I have them vote a second time, we wrap up and the cycle repeats until the class time is up, and you know, the rest, in a sense, is is history. I mean, after a year I submitted a grant, I had to brand it, so I called it p instruction, and then, years later, somebody came up with the idea of calling it the flipped classroom.

Dr. Lisa Hassler: 8:46

And yeah, I don't know, it was a total surprise to me because I was really focused on solving my problem in my classroom Right. The peer instruction workflow and its alignment to the flipped classroom model can be applied not only to a college classroom, but in high school and in elementary school as well, so its transfer is not just for large classrooms in universities. This model can be applied to any classroom to increase the engagement.

Dr. Eric Mazur: 9:13

Yeah, you know. To get back to what you said about it being applied even in elementary schools and middle and high schools all around the globe actually, and even in corporate training, you know, surprised me enormously at first. And the first time I was invited to talk for an audience of elementary school teachers, I thought what in the world do I, as a Harvard faculty member, have to tell people in elementary school who are doing a much better job than anybody in higher education in terms of educating young learners? So I was actually nervous. The first time that happened I thought, you know, I'm going to be sort of talking down from my ivory tower. And then I realized that, you know, my message actually did resonate with them. And even, you know, for corporate trainers oh my God, that was. You know, I kind of always talk for corporate trainers. I thought what I'm doing is irrelevant to them. But it wasn't.

Dr. Eric Mazur: 10:08

And the reason is learning. Is learning is learning right? We don't learn by watching, we learn by doing, and that's sort of the essence of the method. So just-in-time teaching teachers are curious about what this is. Can you shed some light on the just-in-time teaching method, its principles and the benefits? And what I'd done in the early 90s was throw the information transfer out of the classroom and focus on the, you know, sense-making in the classroom. When I'm there to help them, because I consider that the harder part of it- yes.

Dr. Eric Mazur: 10:56

And you know, initially I just thought, okay, the information transfer out of the classroom is simple, you know, let's just tell students read this. And of course I would discover that not all students did it. I tried different ways of incentivizing it. And then in the late 90s I found out about this method, just-in-time teaching, which was introduced at Indiana Purdue University, iupui and Davidson College and the US Air Force Academy called just-in-Time Teaching, which was actually the perfect complement to what I was doing, because it was sort of finding a way to incentivize the students to do the pre-class information transfer, mostly reading, in a way that would actually connect much better to the in-class activity. And essentially what just-in-time teaching does is it assigns a reading, it has a reading prompt and then asks students to answer a number of questions.

Dr. Eric Mazur: 11:54

The two or three the first two are typically content-based, but they're not something that you can just fish out of the reading. You have to think, and they're evaluated on effort. And then the third question is always the same please tell me one thing you found difficult or confusing in the reading. And if you did not find anything difficult or confusing, please tell me what you found most interesting. Okay, so that second part of that last question is just so they can't get away by saying everything was clear and then I have to say something.

Dr. Eric Mazur: 12:28

But what happens is that by having students express I don't quite understand why, blah, blah, blah, as an instructor you can, before class, skim the answers to that third question and have a good sense as to what is happening in the brains of the students, and then you can tailor your questions in class to address those questions. You know, in a pre-class assignment one of you asked this following question and then you project the question that one of the students had the student go wow, he's actually reading the questions that they have. You know, let's think about that question and rather than just me answering the question, I would bounce the question at the students and have them think about it, commit to an answer, talk to each other to try to convince each other of the answer, vote again and you know, that sort of motivates the students more intrinsically to put in the effort before class, because by doing that and by communicating what they don't understand, there's a bigger chance that the instructor will actually help them address the question in class.

Dr. Lisa Hassler: 13:35

Yes, and that is like in elementary school. We have the exit ticket. This is almost like the intro ticket, and so it's just it's flipped. It's flipped, yes, but that reminds me of the exit ticket strategy. Yes, absolutely. And so now you have been instrumental in the development of Perusall, and that is the interactive social learning platform. How does Perusall enhance student engagement and learning outcomes, and how can educators incorporate it into their classrooms?

Dr. Eric Mazur: 14:05

Well, it's actually a very nice segue to your last question because, in a sense, perusal is just in time, teaching on steroids. You know, I realized sometime I would say about 15 years ago, and I don't know why it took me so long that I'd worked very hard on making the classroom a more social activity. Students were constantly socially engaged, talking to each other, interacting with each other, but the just-in-time teaching part was still a completely isolated individual experience, right? I mean, they were doing the reading and then they were communicating to me, all the students to me, the questions that they have, and I would take some of those questions and then put them in the class, but it's still, you know, all these students individually pointing at me, so to speak, with their questions, not interacting with each other. And so I started experimenting.

Dr. Eric Mazur: 15:03

Probably around 2014 or so, I started to experiment with annotation systems and then I thought, you know, maybe we can design an annotation system which is a little bit like a social platform elements of Facebook and so on where students essentially can discuss content either in the form of printed material or in the form of video with each other. Let's say, you know, highlight something, or mark a moment in a video and say I don't quite understand this. I was thinking that, blah, blah, blah, is there somebody who can help me get this? And then, essentially, somebody else can click on that marker or that highlight and it opens a chat window which looks very much like the chat window on your phone or in Facebook or whatever, and they can have this conversation, which is asynchronous but still interactive with each other. And in order to incentivize the students, we correlated online behavior, which means not just what they write but also how they engage with the platform, like each other's comments, illicit responses from other students, time spent on tasks and so on to better performance in the classroom, and then reverse engineered that and built into the platform ways of incentivizing the students to contribute in a meaningful way.

Dr. Eric Mazur: 16:29

So if a student, for example, says I don't understand this, then the feedback from the system is you know why don't you understand this? You should essentially express your thinking and your thought process. Express your thinking and your thought process, right, so it sort of stimulates students to expose their thinking, not just ask questions. I mean, they're fine to to just chit chat and and that's perfectly fine, right, but if they want to show that they've actually done the pre-class assignment. They need to engage a little bit more than just chit-chatting.

Dr. Eric Mazur: 17:03

And then we saw that it actually improved classroom performance significantly. And of course we added machine learning right from the beginning, 10 years ago, when we implemented Perusall of really making this pre-class assignment much more engaging for the students but also an avenue for the instructor to have sort of a window into the brains of the students. I can get a report which is compiled by a large language model that just tells me. Here are some of the points that your students are struggling with. So even before I step into the classroom, I I've already interacted with them in a sense and I can tailor my classroom activity.

Dr. Lisa Hassler: 17:48

So when you talk about the types of engagement that they're having, is there any sort of like voice? Are there videos on this?

Dr. Eric Mazur: 17:56

Absolutely yes. So as an instructor, you can either use videos as a medium, as a content, or you can use text. I mean, I've had music instructors upload musical scores for students to annotate, believe it or not. Yeah, I was totally surprised. I've seen instructors who teach art uploading art. You know you can upload a podcast. Yes, yes, have people discuss it and annotate it. Yeah, because I mean anything, so you can put a marker and say I have a question about this or I find this interesting because, or, hey, this makes me think about what I learned in this other class and engage with each other. And in terms of Perus, I mean, it's grown significantly. I think there are ways of doing voice notes as well for the students so that they can do that. I'm not 100% sure. I mean, you know it started in my class and now it became it's just growing. It's just growing and we have sort of you know, over 4.4 million students or something like that using Perusal. It's pretty amazing.

Dr. Lisa Hassler: 19:07

So how would teachers access Perusall? How would they be able to like, get into it and use it?

Dr. Eric Mazur: 19:13

Well, I mean, you can just go to perusallcom. So Perusall, there's two L's, perusallcom, and the basic platform is free, although we hope that people will contribute. Because, as the number of students has grown, the cost of supporting the platform has gone up. So if you adopt a textbook, for example, rather than going to the bookstore have students go to Amazon, they can buy the book through perusal for the same price as on Amazon or anywhere else, and instead of Amazon getting the commission, we get the commission and we use that to support the platform. But I would say probably 80% of our users do not use textbook. They upload their own materials, which is perfectly fine. In that case, you know the institution can sign a license, but we don't obligate anybody to do that, which provides the institution additional support. Or people can use a course pay model, but again, in principle, the platform is free. We want to be sure, however, that we can keep supporting all of our millions of users.

Dr. Lisa Hassler: 20:20

This is wonderful. I love free resources, things to help in engagement, things to help teachers.

Dr. Eric Mazur: 20:26

Absolutely.

Dr. Lisa Hassler: 20:28

In one of your lectures you emphasize the need to rethink assessment practices. Can you elaborate on why traditional assessment methods may be out of sync with societal needs?

Dr. Eric Mazur: 20:38

I was already thinking that before the generative AI revolution, but I think it's becoming even more important now. I think that over the years, as I was going from lecturing to peer instruction to team and project-based learning, I sort of realized that if you put yourself in the shoes of a student, the assessment especially in higher education and especially at a competitive institution like mine is sort of the tail that wags the dog. You know you can do all you want to change the pedagogy in the classroom and to motivate the students. Ultimately, they want to get a good grade and they'll do anything they need to do in order to get that good grade and therefore, in a sense, the assessment really is sort of the hidden curriculum. That is what determines how students are going to engage with the material study. And also, I think the assessment in a sense robs the standard. High stakes assessment in particular robs students of their ownership of learning right, so it's no longer that they're learning for them, they're learning in order to pass the assessment. And please, the teacher Also, just think about this.

Dr. Eric Mazur: 21:53

Most assessments are closed book. Yeah, and are people separated from each other? Yeah, I heard that for the SATs. Now you can't even wear a smart watch anymore? Yeah, because it could give you information and you're separated from each other. You're only allowed to use pencils and no other writing utensil that could permit getting information. People separated from each other, cut off from each other. Think about it. That's not how we work.

Dr. Lisa Hassler: 22:24

No, it's not.

Dr. Eric Mazur: 22:25

By the standards of what we call teaching in traditional education. I cheat every day, and I would say probably most people cheat every day, because we look things up, we talk to other people. So how in the world can we think that we're measuring something that has any bearing on real life with the traditional high stakes assessment, right, with the traditional high-stakes assessment. So I've completely modified my assessment, which is a long conversation, but I would like whoever is listening here to ponder how does what we measure in our traditional forms of assessment really relate to what the societal needs are that we have and we know? If our assessments were really good, then people would not interview anybody for jobs, but they would just look at the results of the assessment and hire.

Dr. Eric Mazur: 23:15

So I think, in a sense, most of our assessment practices are I'll use a strong word here misguided and probably a social injustice, because they sort of favor people who have traditionally succeeded and you know who are coming from a particular socioeconomic class. But now it's even worse. Right, because you can get generative AI to do the work for you. You'll find that a lot of things that we're assessing can be done by generative AI, which means that they will be done by generative AI be done by generative AI, which means that they will be done by generative AI. So, luckily, there is now a very strong outside force that is going to push us in the right direction and force us to rethink our assessment, because if it can be done by generative AI, then it's probably not a worthwhile task to teach our students.

Dr. Lisa Hassler: 24:06

I like that, and you also had said something about the silent killer of learning. Exactly, but I'm glad that you also had said something about the silent killer of learning.

Dr. Eric Mazur: 24:10

Exactly.

Dr. Lisa Hassler: 24:10

But I'm glad that you're talking about a high stakes test. At what point do we look at those and say is it really worthwhile and could there be a different way? And I do agree that it isn't in line with society when collaboration is such a big emphasis in life. You're not the expert in every field and everything, so you have to go out and source for different perspectives to be able to work together. So, amidst the shift to remote learning, you mentioned experiencing your best teaching ever, and I would agree with that, where the best teaching I ever did was remote. So what advice do you have for educators striving to excel in this educational landscape?

Dr. Eric Mazur: 24:46

You know, everything that was published widely after the pandemic or even still during the pandemic, was that remote teaching was just a bad joke and didn't work. But I think the reason for that was bad pedagogy. Right, if you mostly teach for lecturing and then now you have to lecture remotely, I mean it can already be deadening. It's even more deadening when it's online. So, since I wasn't doing any lecturing, I mean it can already be deadening.

Dr. Eric Mazur: 25:09

It's even more deadening when it's online, right? So, since I wasn't doing any lecturing, I was, you know, engaging my students, having them interact with each other. Yes, that translated to online learning very, very easily, and the students had at least a sense of connection and interaction in my classes, whereas for most other classes they were watching the recorded class rather than the live class, because it's so much better to watch it at your convenience than the teacher's convenience. And then you're totally alone. I mean, you're already alone at home and now you're alone in the learning process, which is very different from what was happening in my class. So, to me, all the negative press about remote learning was simply due to bad pedagogy. If it didn't work, it was simply because people did a lousy job. Pardon my strong language here. I agree it was bad pedagogy. There are plenty of examples of very successful online learning environments. For example, western Governors University is one of the largest institutions. It's completely online and it's one that creates the biggest satisfaction and future career success among its learners. And it's certainly not a very highly selective institution, but it's been rated many times as an absolutely top institution in terms of satisfaction and career success. And then there's the Minerva schools, who are completely online. So there are plenty of examples of it working.

Dr. Eric Mazur: 26:36

But one of the things I realized when I was teaching fully remote during that one middle year of the pandemic was that it created an incredible amount of flexibility that was free, sort of, from the constraints of a physical classroom, where you can only be when everybody is together there at the same time. In a sense, the physical classroom imposes a huge constraint on when the learning takes place, but Zoom or Teams or whatever platform you use, is always available to anybody. You don't need to book it, you know, you could just hop on a call and meet somebody. And I realized also that many things that I had done synchronously could equally well be done asynchronously. So at the beginning of the pandemic, at the beginning of that year, I went fully online. I made a list of all the activities in my classroom and I asked myself does this have to be synchronous and does it have to be led by me or is there something that could be done self-paced? And I noticed that Just about everything could be pushed into an asynchronous self-paced, especially if it's done in a team environment where students hold each other accountable it could be moved into that category freeing up my time to be there for the students when they needed me and essentially support them with individual difficulties, rather than just talking to them, thinking this is something they need to hear from me. And so I think that permitted me to really address student difficulties in a way that I've never been able to do before, and I've started to restructure my teaching in the physical space the same way.

Dr. Eric Mazur: 28:22

So my class right now it has 160 students. We don't have a set classroom anymore. Students meet all over the buildings. They can sit in the coffee shop, they can sit in the atrium, they can go to the maker space. As long as they find a table that has four or five chairs, they can sit around that table and, through technology, they call us in whenever they're ready to present something that they have done, which they need to do six times a week. They need to, you know, present a little exercise that they've done to show their growing mastery of the subject. Yes, so I've really started to completely rethink. So I started by thinking about pedagogy with spear instruction. Then I have been rethinking the assessment and sort of the last step which I've engaged in in the last few years was rethinking the physical learning space, where learning space could be construed very broadly, from a physical classroom to a maker space, to a building, to Zoom or Teams or anywhere where people learn.

Dr. Lisa Hassler: 29:31

I have some quotes from students, and one was Eliza saying that she was very pleased with the way the course was taught and in particular, with Professor Mazur. He was extremely dedicated and passionate about teaching. He held long office hours each week and was always more than happy to answer my questions. So I thought that that supported what you were just saying about having those hours and the availability, and it was obviously very appreciated by your students that were very happy to be able to have that time with you. When you rethink what needs to be done, when and how, it's a free time to address the students' needs and, with engaged students and mastery, with convincing one another what's right, that's real learning and it's happening in your classroom. So you have a wonderful model.

Dr. Eric Mazur: 30:19

That's wonderful, and I invite any of your listeners, if they're ever in the Boston area, to come and visit my class. I just promise one thing After visiting my class, you will come out convinced that I teach kindergarten, because that's very much what it looks like.

Dr. Lisa Hassler: 30:36

Herding cats and a lot of chaos, but it's actually organized chaos, I'm sure. Well, thank you so much, Eric, for sharing your expertise on peer instruction and offering valuable insights into transforming classroom learning through interactive social learning and alternative assessments.

Dr. Eric Mazur: 30:52

Thank you so much. I had a great time. Thank you for having me.

Dr. Lisa Hassler: 30:56

As we wrap up, I encourage you to embrace social learning styles in their teaching and assessments, transcending the confines of traditional classrooms. If you have a story about what's working in your schools that you'd like to share, you can email me at lisa at drlisahasslercom, or visit my website at wwwdrlisahasslercom and send me a message. If you like this podcast, subscribe and tell a friend. The more people that know, the bigger impact it will have. And if you find value to the content in this podcast, consider becoming a supporter by clicking on the supporter link in the show notes. It is the mission of this podcast to shine light on the good in education so that it spreads, affecting positive change. So let's keep working together to find solutions that focus on our children's success.