S4xE6: Peer Instruction (Teaching Practice Byte)
Episode Summary
Peer instruction is a way to move the easy-to-learn content to before lecture, so you can spend more time during lecture on developing understanding. In this teaching practice byte, we talk to Dr. Beth Simon from UC San Diego about peer instruction, what context she’s used it in, how she does it, and nuanced details that aren’t always discussed when reading about this active learning technique.
You can also download this episode directly.
Episode Notes
https://peerinstruction4cs.com/
Transcript
[00:03] Kristin: Hello and welcome to the CS-Ed podcast, a podcast where we talk about teaching computer science with computer science educators. I am your host, Kristin Stephens-Martinez, an Associate Professor of the Practice at Duke University. Joining me today is Dr. Beth Simon, teaching professor in Education Studies at UC San Diego, formerly teaching professor in Computer Science and Engineering. Beth, thank you so much for coming on the podcast.
[00:28] Beth: Hi, Kristin. Thanks so much for the invitation and so excited to be sharing my experiences with peer instruction with you and your audience.
[00:34] Kristin: Awesome. So today’s episode is a TPB, teaching practice byte. In a single sentence, what would you say is your practice?
[00:43] Beth: I use peer instruction to move the easy part of learning before lecture so that I can spend my time together in class with students working with them interactively to help them develop their understanding of the hard part of learning whatever concept we are on.
[00:58] Kristin: Great. So before we dive deep into the teaching practice byte, how about you tell us the context, like where does this make the most, the most sense to use? What kind of classes? How many students can this kind of practice use?
[01:13] Beth: Right. So, peer instruction has actually been shown to be effective in a pretty large range of undergraduate STEM classes. I’ve used it for 10 plus years in a CS0 course. A CS1 course. And I’ll say that my classes generally are bigger, I would say 50 plus up to 400 ish. But I also have close colleagues that have found success even in very small classes, like five students.
[01:39] Kristin: Really? I, I’m a little surprised at five. I’ll be honest, though, my range is more in the 100-something to 200-something. So I always appreciate when I hear it’s like this works in a very large class. No worries. Though, I have no idea the sizes of my audience’s classes. I suspect it’s, it’s the whole gamut, though.
[02:00] Kristin: So. How about you go into the details of the tip itself because I do use peer instruction. I suspect mine might be a little different than yours. But we’ll see.
[02:07] Beth: Absolutely. Peer instruction is one way to implement that infamous thing we always talk about a flipped classroom. Again, where you sort of move that easier, more passive learning to before the lecture, and then you cognitively engage students during lecture to cover the most challenging content. The peer instruction approach is focused on having students attempt very challenging multiple choice questions and voting on their answer in some way, whether that’s clickers or some other techno technology for it getting the votes recorded. Well, that might seem simple. We found there’s a surprising number of ways faculty can fail to implement this technique with fidelity as shown in the research literature. Meaning that the techniques that research has used to be found effective aren’t always the things that that faculty pick up on and remember to implement in the classroom. Over time, I really realized that there’s actually a surprising number of nuances to implementing it. But that, I think it’s this combination of those nuances that is probably really what makes it so generally successful in so many undergraduate STEM classrooms. Because if it just worked in computer science, you know, maybe we’d have some easier way of doing it.
[03:18] Beth: So I’ll just briefly list a few, although not exhausted, sort of parts of the peer instruction process that I think are really important for a faculty member to implement. So there’s this vote, individual vote, discuss, group vote approach where you just put this question in front of students, and you have to give them time to think about it individually. Don’t skip it. Don’t skip any part of that. Don’t skip the individual vote. Don’t skip the discussion. Don’t skip the group vote. Regardless of how many students get it. Right. Even say in the individual vote. A lot of people are like, well, if 90% of them get it right. Can I just skip it? Honestly? It’s not about picking the right answer. And you have no idea if people voted for the correct answer, what their thinking was around that. So, it’s the discussion is really important because it gives students the space to develop their analysis and explanation skills on our technical topics, and again, they often even develop a better understanding even if they got the answer right on the first vote.
[04:18] Beth: Don’t show the results of the first vote, which, by the way, I did the entire first year that I taught with it. It, it just drives the discussion in wrong in the wrong way. People just like, oh, B was the tallest bar. It must be B. Let’s all just discuss B. It just limits discussion. Definitely compared to our usual instincts in developing a multiple-choice question. Do not create a question with five possible answer options. At most pick two, maybe three numbers of wrong answers or distractors. And then, and the next section, I’ll tell you about always train your class that you’re going to be asking them to not only tell you why the right answer is right but why each of those wrong answer is wrong. Because you should pick those distractors to be common misunderstandings that students have about that particular topic. It’s not just, again, we’re not just trying to get like, oh, well, we don’t want them to be able to just guess the right answer. We’re asking them to engage in analysis about interesting options.
[05:19] Beth: Always lead the classwide discussion, and after the, the group vote to and have different groups share with the whole class what did they talk about. This is really important because maybe an individual group of students doesn’t necessarily have the best discussion or explanation skills. But as they hear other people in the class explaining, that helps them sort of develop their explanatory skills to be able to understand what is it should we be really be doing when we’re, when we’re discussing this with each other.
[05:53] Beth: One more thing, not always but frequently early in the term, show the students how much the class has improved their understanding from that initial vote until the group vote, and feel free to borrow my absolute favorite sentence. I’ve ever gotten to say in a classroom. “Look, we have documented evidence that you learned something in class today.” How often do you get to say that in other classes and just be prepared for everybody to laugh a lot. So it’s, it’s very good in large classes, especially for all of you who have those. That there’ll be this one point in the term, just one, probably where 100% of people will get it right on the group vote. And that’s not because they couldn’t have done it other times. But, you know, technical things, stuff happens, but show that when it happens because I’d be like, look, we got 100%, and the class will burst out its own applause for, for each other. And I’ve had so many students tell me that that is one of those most memorable moments that they have. I feel like the alumni association should pay me for that. But anyway, it’s definitely something that really, really brings some magic to the class and really is, is memorable for students, and it’s a great experience for me as well.
[07:07] Kristin: Yeah, I think I, I have had that happen. I think a couple of times when I’ve taught the CS1 using peer instruction. I have so many questions. I think I want to start with. when you say group vote? What does that mean? Can you also be detailed in some of the like mechanics of how you do it, like I get clickers. But if we don’t have clickers, I use, for example, Microsoft Forms because that’s what we’re Microsoft campus. So that’s what I use. But could you provide some more concrete of each of these pieces?
[07:44] Beth: Yes. So the point of the individual vote and the group vote and group vote just means I tell students, everybody in your group needs to decide on an answer together, and then each of you will vote again. All of you get to vote again. The whole thing of telling them they need to come to a, you know, a consensus is just to really again, encourage that group discussion and help make that the normal. So, but it’s really everybody votes individually, they discuss it, and then everybody votes again with their new idea of what they think. The rationale for actually registering it is that we, we want to give students the impetus. They’re like, oh yeah, you know, just think about it. What do you think it is? We can just sit there, but you know, your brain gets lazy. That’s very natural human brain. Don’t burn any energy, don’t need to. So you need to have some way of recording but just so people can get points for it. So I give 5 to 10% of the grade in the class comes from voting, not from getting the answer correct. That’s very important if you require correctness, students hate you because they’re like, you’re testing me before you even taught me. I’m not testing you. This is an incentive to engage.
[08:55] Kristin: I call it class participation points.
[08:57] Beth: It is. It really is. And it’s the funnest ones they’ll ever have. So, that’s the key with the registering things. It doesn’t, the only key about it is you need to be able to register accurately and for a large number of people. And that’s one of the reasons I like clickers and it needs to be anonymous. Don’t like freaking holding your hands up, people. It’s very interesting. I was actually working with a faculty member who was trying this very hard. And I had him using the clickers. I was like, anonymity is important. And then the first thing he does, you know, trying it out the first time in class. He is like, and after the, the group vote, he’s like, OK, so raise your hand if you voted B, I’m like, no, no, no, no, no, you’re missing the point. You just made everybody feel very, very uncomfortable. And it’s just a very natural reaction. It was like, personally, we’re attached to like seeing our students and being like who, you know, who did it. And it’s like, just generally put it out there, especially after the, the group vote. Somebody want to tell me what you talked about in your group. Not like who can explain why B is correct. It’s not the worst thing to say. Who wants to tell me why they thought B was the right answer? But it can be better and honestly easier for us as instructors to just get into that mode to just always say, “somebody tell me, what did you talk about in your group? And somebody else tell me, what did you talk about in your group?” And that just leaves it open there. The best answer that I ever got is we talked about the fact that we think this question is really confusing. And I’m like, oh, that’s really interesting. Why don’t you tell me about that? And so that just makes it much easier for people to feel free that they could, you know, volunteer and say something.
[10:29] Kristin: So I have a specific question because I’ve been using peer instructions since I started at Duke, basically my whole teaching career. How do you manage the stragglers to fill out the form? Because like, you don’t really know, how many are there during the first round? I call it round one during the individual vote. But, you know, by then you do have it like, from that number, you know, how many are mostly present for the group vote or round two? But there’s always those stragglers that like, take forever or basically, they’ve checked out, especially once they realize that I don’t, I only give points for filling out one of the forms. I’m getting closer and closer to saying like every form is half a point. So like if you want the whole point for this peer instruction, you need to fill out both forms or something because I like it’s just getting more and more frustrating that I think a good a chunk of the students have checked out, and they’re like, I’ve already filled out the form once I’m only here to get the points and not to do any actual engagement. So, I’m not even going to fill out round two. So, what do you do?
[11:33] Beth: So I think the key thing is try to change your policy so that you can avoid that because when people come just to get that one point and then check out, it makes for a worse experience for the people around them. And that’s, that’s a bummer for those students. So I say instead, I say you have to, you have to have voted for at least 75% of all of the questions I asked today, and given that they are really only like six votes in a lecture period or something like that. That sort of means all. But it also means like you can miss one, you could be at the bathroom, you could not make it in time, you know, get, get it done in time. But what I do is when I’m especially actually either on both group vote or the individual vote, I’ll be like, I’ll start with the Clicker software. It’s also shown up on the, the screen, how many votes have come in so the students can see that, which I think is a good mental pressure. I see it, and when I get close to that number, like if I have like, I think I have 200 votes today, I start getting to like 170. I go OK, let’s get, let’s everybody vote 5, 4, 3, 2, 1. I spend all this time counting down in my classes, I count down from five, and I use the microphone and big class, and I move it closer to my mouth and same way on the, the the when you’re calling for the group vote because they’re discussing right? So it can be very hard to get their attention. But also that’s where I open up the clicker software again to get the group vote. The thing pops up. So students know, oh she’s taking a vote, and people like elbow each other and things like that. And then I do say I do a countdown, and if they didn’t get it, I’m like, hey, it’s ok if you didn’t get it, but you know, you need to push something, so push something.
[13:10] Kristin: Like what I do. So I, I also lower the pressure by saying you need 80% of the possible points from the entire semester to get full credit for that 5%. And that way also, I’m not dealing with absences and having to like, do anything with that red tape. But in the way I do it is, I’m like, ok, so I, we will start discussing when we hit 120. I know that like 135 or something is from round one, but sometimes it just drags on like it takes another couple of minutes and I’m like, I, I don’t know if I can’t go back on my word because then that like sets a different kind of precedence. So I’m like, this is annoying.
[13:50] Beth: Yah, I feel like being, setting a particular number is putting yourself in a bad situation that is probably going to be hard to manage the whole paper trail thing, I understand. And this clicker, the iClicker software is explicitly set up for this. It’s like based on however many votes you have in a period, you know, how many times do students need to vote to get a point? So it literally has that setting already set up for. And then I also have, you can miss, you know, usually 10% of the classes and, and still get your full points participation, and it’s not an all or nothing game, it’s tiered blah, blah, blah. But again, that’s one of the things that the I click software actually has set up. But I do, I totally understand the, the, the, the dragging on is like, it really grates on you because, you know, here you, we’re all built to feel like we’re trying to optimize our time in the lecture hall and there’s so much to do. So I, I don’t know exactly what I would do in your setting, but I think I would, I would probably lean on the, the counting like you guys just need to listen because when I start counting backwards, like you need to have to vote in by the time I get to one. And that just gives you a lot more flexibility.
[14:55] Kristin: Yeah, so it sounds like you actually have two aspects. You have like overall you need 90% of the possible points, and then, but in the class, you also need a certain number for any of the points to count?
[15:07] Beth: Every class period is either a zero or one in terms of for the day. And it’s like what does it take to get that to get your point? You need to have clicked in at least 75% of the available clicks that day.
[15:18] Kristin: Oh, ok. I get it. Interesting. I’ll have to think about how I would use something like that, but I feel like we’re getting too far into the weeds. And I will, I will hint now that Beth has agreed to do a second episode where we more dive deep into how I do things just so I can find out how I’m doing it wrong and can do better. So let’s now go to the next question in the teaching practice byte of where else would this work or where would it not work? So, in this case, I probably, maybe the discussion would be more interesting about where would it not work?
[15:53] Beth: Right, and I’ve gotten this a number of times and I do think there are cases where this will not work. I think it’s very difficult to use it in a class that doesn’t have any sort of textbook or anything to support the pre-learning of the easy stuff that said, I know in the psych department in my university they frequently use it in classes without a textbook. And they just, I think these more sparingly and intersperse a little bit of lecture. I think in, you know, my classes just, just, I don’t feel like there’s enough time to cover the easy stuff and have peer instruction questions to the difficult things. And one of the challenging things for people implementing it the first time is that they’ll back down and be like, oh, I’m not sure they read that. So maybe I should lecture on it just quickly, really, in case. And, and it’s like you got to hold the line and be like, nope, nope, you guys got to read it before you come. You’re expected to know that part. People often be like, oh, well, you wouldn’t do this in graduate student classes, which is my favorite question because I’m like, what people’s brains, the way they work magically changed between undergrad and grad school? No, your brain still learns the same way. This is based in brain science. So, it makes sense. And definitely, there have actually been research studies on using it in physics Ph.D. courses. So that’s it.
[17:03] Beth: I will say I have been for the last 15 years focusing more and more on working with K 12 classroom settings and specifically for a long time high school and AP classes. And I have found that it can be very, very challenging to implement in the high school setting for a number of different reasons. So first off, even in an AP class, some high school students are just way less invested in the class, and having a really significant number of students who are unwilling to engage, makes it really uncomfortable for everyone. Additionally, at least with the CS people, teachers, I’ve been supporting, they are often coming to teach computer science out of a number of different reasons, but without any real content knowledge, and they’re the most brave people you have ever met, but it can be very difficult for them to lead the discussion around this question when they’re not exactly sure, like often, I will create the, the peer instruction questions and give them to the teachers, but they didn’t create them themselves, they’re not really sure why the distractors are common student misunderstandings, and it can just be very hard to even in our classrooms to deal with students give you an explanation, and you’re like, what was that? So, I’ll just say, I think, you know, absolutely can work in some high school settings. But it’s a little bit more difficult, I think to work with.
[18:22] Kristin: Yeah, that’s an interesting thought. I, I especially think your comment about the pedagogical knowledge of like why the wrong answers are wrong. That definitely I can see why peer instructions might be a little bit more uncomfortable initially because you can’t, you can’t riff off of a student’s answer to quickly help the students learn from the student’s response without having all of that extra contextual understanding of everything behind the question, which makes a lot of sense. We are almost out of time, especially since we want to keep this one short. So let’s do TL;DL. Too long. Didn’t listen. Can you give me a rundown of your practice in three minutes or less?
[19:04] Beth: Sure. These are more like the steps that you need to go through. Stop lecturing. Start having all students in your classroom, not just the ones who raise their hand, discuss and learn where they, you direct the learning through the creation of challenging questions on your class content. You need to tell students why you’re doing this weird thing and why things look very different in the classroom and reinforce the things that we think really support students focusing them on this. Hey, growth mindset. We just all learned something. Developing professional analysis and speaking skills, and helping students honestly get to know each other. I think that’s especially important in lower-division classes. Last thing is so many people say, ok, ok, I’ll try this for like I can commit to trying it for two weeks. But I just, you know, I, I can’t. It, it doesn’t work to do that. I’m just gonna tell you, we’ve seen it happen plenty. It doesn’t work. I promise you have to explain to your students why you’re not lecturing. And then when you go back to lecturing all that stuff, you just said about why you’re not lecturing now you’re lecturing and it’s torture for you. It’s torture for your students. It just it, you, you’ll find that you can’t. So you just have to jump in with both feet and plan some time during the term before or the summer before that you can go ahead and start creating questions. I also would recommend that you start by looking at some of our vetted examples at our website which we will link down below, but it’s https://peerinstruction4cs.com/ at a four in there. And just use those as models for the kinds of questions you do want to ask, which are hard questions with not too many distractors and and not skip all those easy ones.
[20:44] Kristin: Yeah. So for the website, it’s peer instruction, the number four, cs.com definitely we’ll link in the show notes as well. But just to make sure that we heard what, what the four was because there are so many different fours in this world. Oh, all right. Well, thank you so much for joining us, Beth.
[20:59] Beth: I’m so glad to have had a chance to share. I want to encourage faculty to reach out if you have questions and consider for yourself or a new colleague looking for a CS new faculty teaching workshop that covers using peer instruction because it’s great to get some support from people who have expertise.
[21:16] Kristin: Awesome. And I will tease that the next episode will be Beth doing more of a deep dive on trying to figure out how to make my peer instruction better. That will be the next episode on this podcast feed. And this was the CS-Ed podcast brought to you by our patrons! You all help keep the mic running. I’d like to especially thank our three randomly chosen patrons Yesenia Velasco, Jacqueline Smith, and Mark Guzdial. You all are awesome! If you’d like to support the podcast and keep it commercial and sponsor-free, consider becoming a patron on Patreon, link on our website. Otherwise, share the podcast far and wide and give us a five-star rating. Help us get the word out! This episode was hosted by me, Kristin Stephens-Martinez, and produced by Chris Martinez. And remember teaching computer science is more than just knowing computer science. And I hope you found something useful for your teaching today.