Episode 8a

Part 1: Group Work: What Could Go Wrong with Peggy Brickman

[MUSIC PLAYING] STEVEN ROBINOW: Welcome to Teaching for Student Success. I'm Steven Robinow. In this episode, I have a conversation with Peggy Brickman of the University of Georgia about group work. But this is a different perspective of group work. Instead of trying to sell you on how great it is, we're going to talk about the problems of group work.

I encourage anyone interested in doing group work or frustrated by doing group work or intimidated about doing group work or have done group work and is sick of it and won't do it again, I encourage any and all of you to listen to these episodes to talk about the problems of group work and possible solutions.

We're going to talk about two papers today, a 2018 paper entitled, "When Group Work Doesn't Work: Insights From Students," and a 2021 follow-up paper entitled, "What to Expect with Group Work: Problems, Frequency, and Success of Mediation." Because this was such an extended interview, I'm going to break this into two parts. In part 1, we're going to start with the first paper, "When Group Work Doesn't Work: Insights From Students." In part 2 we'll talk about the second paper, "What to Expect with Group Work: Problems, Frequency, and Success of Mediation."

I'm very excited to introduce Dr. Peggy Brickman. Thank you so much for joining us on Teaching for Student Success. Welcome, Peggy.

PEGGY BRICKMAN: Thanks for inviting me, Steve.

STEVEN ROBINOW: Absolutely. Happy to have you. Dr. Brickman is a professor in the Department of Plant Biology at the University of Georgia in Athens, Georgia, and also holds an appointment in the Department of Math and Science Education. In addition to her teaching responsibilities, Dr. Brickman serves as the director of an NSF-funded REU site, Research Experiences for Undergraduates, titled ominously, Undergraduate Biology Education Research Version 3.

Dr. Brickman—Dr. Brickman runs an active education research program. Dr. Brickman publishes extensively in a variety of areas impacting the student experience. She has published on student perceptions and motivation about learning, TA training for improved student engagement, instructional techniques, and institutional change. Dr. Brickman has won numerous teaching awards.

I actually don't know if there is a teaching award at the University of Georgia that she hasn't received. She received what is likely to be one of the most prestigious education awards in the system. She received the Regents Award for the Scholarship of Teaching and Learning from the State of Georgia. Congratulations on that particular award and the many, many others.

PEGGY BRICKMAN: Thank you.

STEVEN ROBINOW: In this conversation, Dr. Brickman and I will talk about two recent publications, a 2021 paper by Brickman, Lannen, and Beyette entitled, "What to Expect With Group Work: Problems, Frequency, and Success of Mediation," published in the Journal of College Science Teaching, and an earlier paper from 2018 by Chang and Brickman entitled, "When Group Work Doesn't Work: Insights from Students," published in CBE—Life Sciences Education.

OK. Well let's talk about your research on group work. Perhaps we should start with your 2018 paper, "When Group Work Doesn't Work: Insights from Students." Maybe you could set up the problem for us. Maybe first by talking about why we think student work is—group work is a good thing or some of those data, and then set up the problem that you've chosen to address and describe how you set up your research protocol.

PEGGY BRICKMAN: OK. So, I started this because I kept going to these great workshops like the Summer Institute or National Center for Case Study Teaching in Science. They just preached group work as an amazing solution to lots of problems. I would say, but I have 300 students. Do—no. Do not use it with 300 students.

So that was frustrating. The more I read about it, the more—for example, there's so many of these meta-analyses where they've looked at hundreds of studies that have compared students who used some kind of group or cooperative learning. There's lots of names for it. COGAL, scale-up in the physics world. And compared the learning gains for those students who worked collaboratively with students who are in a more traditional competitive classroom where each student is on their own.

And the data is very strong that students perform better, they have better attitudes. And I think one of the things that was very convincing to me was that things like the jigsaw that we use at the Summer Institute and others was actually created in Texas by collaborators that were interested in solving the diversity issue that they had with newly integrated schools.

So they would bring in—in Texas in the 1970s, now we have integration. We have Black students in with white students. And so the jigsaw and those other kind of collaborative learning techniques were actually introduced to reduce some of the barriers between students of different racial groups interacting with each other successfully in the classroom.

So, being in a predominantly white institution, I was like, yep, I don't like the idea that students are segregated within a classroom, and you can really see it if you're teaching in a 300-student lecture hall that's tiered. Just does not seem like students are getting to know each other.

For example, just to break my heart, the final exam, students had to come down and show their ID when they turned in their paper. I would see two students standing next to each other and they would look at each other and say, you're in this class? I can't believe you're in this class! An entire semester went by and they didn't know that one of their friends was in the class because it's so anonymous.

So, collaborative learning and group work could help alleviate some of those problems. So I just was like, all right, we're going to do this. And I also tell people when I'm doing a workshop on group work or collaborative learning, I have done everything wrong. Because I think group work is touted as this panacea that's going to fix all your problems, but there's lots of cognitive research science that's looked into the kinds of tasks that are beneficial when you have collaboration and things that are actually harmful.

So for example, memorizing tasks or just trying to understand something is really difficult if you have other people talking to you at the same time. So I'll often tell people who are in a French class, yeah, you want to learn that vocab on your own quietly. What you want to practice with a group is the speaking, the complex kinds of things where it benefits you to have an interaction with someone.

The paper in 2018 with Eunice Chang, she was amazing. She's in our College of Ed Technologies. She heard about me because I was trying to do these kinds of things using technologies, and so she was like, I would love to just talk to you about what you're doing and how you're doing it.

And so she spent an entire semester sitting in on class and made a list of—like these are the things—this is what you call what you're doing and these are the things that I'm noticing that students are not doing when you ask them to do it.

So it was very humbling to have someone sort of pick apart your classroom, but that evaluation is so helpful. So she did interviews and she was like, this is what they're saying, they're not actually doing that. You're setting up all of these best practices for students to collaborate with each other and they're just not doing it.

So I was like, huh, we should probably collect some data on this, because other people are following this prescription. Like you should do this, this, this, this. If you read Johnson and Johnson, they're very famous talking about collaborative learning, cooperative learning. We were doing all those things, but the students were not doing what they needed to do. And so that was the data that we collected for that paper.

STEVEN ROBINOW: Oh, so you had some preview into this before you actually collected the data.

PEGGY BRICKMAN: Yeah. I already knew that there were some issues. It wasn't like—as an instructor you're like, ah, this is frustrating. Most of the time as an instructor you see it at the end of the semester, the peer evaluations that they're doing for each other are really nasty. There's a lot of complaining, there's a lot of just actually anger toward me that I was like, I don't understand what's happening.

But I was setting up a situation which they were resentful about and wasn't really monitoring the interactions to the degree that I really needed to. I thought, oh, I'm doing it. I'm definitely monitoring this, but it's like, nah, you're not monitoring and you're not intervening, really.

STEVEN ROBINOW: All right. So can you describe how you set up your study and what you actually did? And how you collected your data?

PEGGY BRICKMAN: Yeah. So at the time I was teaching multiple sections of a class. And so it was really easy to have hundreds of students that would participate in this. I would have 70 or 80 groups per class, three sections of it, lots of places to collect data on it. And then we were also using, like I said, some of these techniques that are recommended like assigning students to roles within their groups.

Having them do a midsemester peer evaluation and a final end-of-semester peer evaluation where they actually rated each other. How well is this student doing on all these different facets? And then if you had to divvy up points, let's say you had 100 points to give out to each of your group members, how would you divvy it up?

Then the data that we collected were to find out how students were performing in the course, whether they were doing well like a high performer versus a medium versus someone who's like really bombing on most of the tests. And then we look to see, could we interview maybe a dozen students? And interview some of the top performers, some of the medium performers, and some of the lower-performing students.

And asked them a series of questions about how they were divvying up roles in their group, what kind of problems they felt like they were finding, and then in the end, we looked at all the quantitative data as well, like the peer-evaluation scores they gave each other. What their grades look like, what the grades look like for all the students in the group.

We were having them do outside-of-class projects with each other. And then also this activity called a group test, just like a tiered test. You take your test individually and then you get to retake it as a group. And so we could see how a high-performing group looked and how students within that group, like how many of them were also higher-performing and how many of them were maybe not doing so well within that group?

And teased apart not only that data, but also what the kinds of comments they gave each other on the peer evaluations looked like. So we basically just threw a big net and collected all of the data that I was collecting in order to try to monitor how the groups were doing and then we teased it apart. Like statistically just separate out high performers and low performers, and also really dig into the interviews and see what students are saying.

STEVEN ROBINOW: The students self-assigned groups. So you didn't assign groups in this study.

PEGGY BRICKMAN: Yes, at the time.

STEVEN ROBINOW: At the time they self-assigned?

PEGGY BRICKMAN: I was letting them choose their own groups because there was so much like contentious issues, and I felt like there wasn't—and there isn't—really good data on how to set up your groups perfectly. And so with 300 people, there's lots of different surveys you can give them, but I thought, it's so much work. I'm just going to let them choose their own groups and then I'll see what happens.

STEVEN ROBINOW: If I understood correctly, among the high-performing groups, for example, high-performing groups seem to support the lowest performers in that group better than in low-performing groups.

PEGGY BRICKMAN: Yeah. So we can follow them how they did individually and then how the group did on group tests, and we found that over the course of the semester, a low performer would make greater gains if they were in a high-performance group than a low performer in a low-performance group, which is great. It means that they were really getting that kind of interaction that was helping them learn the material.

STEVEN ROBINOW: It's great as long as you're in a high-performing group.

PEGGY BRICKMAN: Exactly. And what was probably the worst that made me realize like I can't let them self-select, definitely that. But also that when they do their peer evaluations, you take two identical low performers and one is in a high-performance group and one is in a low-performance group, there was extra penalties. That low performer was getting even worse peer evaluations than the same student doing the same exact work in a high-performing group. So it was like double indemnity. I don't know if that's the right term.

STEVEN ROBINOW: For a better understanding of the term double indemnity, see the 1944 movie by the same name with Barbara Stanwyck, Fred MacMurray, and Edward G. Robinson.

PEGGY BRICKMAN: That low performer was getting even worse peer evaluations than the same student doing the same exact work in a high-performing group. So it was like double indemnity. But already they're in a group that's not doing well, so their grades are not good. But then they get double-dinged because now when their end of semester comes and it says like, how much of the group work are they responsible for? Their peers would say, well not very much. And so they would end up—it was really bad.

So at that point, I knew that I was like, OK now, no, we're not going to let them self-select groups because what happens is high performers are self-selecting with other high performers.

And also this—I stopped doing peer evaluations because I realize that students aren't very good at doing peer evaluations, and as a result of writing the paper up, which I find a lot—like I don't know if other researchers experience this, but you'll do a certain amount of reading into the literature, figure out what your question is, and zone in. But it's until you find the results where you're like, huh, I gotta look more into that.

And so I really started looking into the research that has been done on evaluating each other, like how students evaluate each other, and it's just really not good. They are not good at evaluating because of that exact thing. Like if the group is doing well, they're kind to each other. If the group's not doing well, they're extra negative to each other. So I stopped doing that, too. That leads to the second paper that I worked on with Austin.

STEVEN ROBINOW: OK. Well, before we get to that, there's a lot of interesting things here to think about. One of the things that concerned me in this paper was this notion of students grading each other. Students are there to learn the content, they're not there to rate their colleagues. And so that seems like that's a concerning issue or situation, an uncomfortable situation in which to put students. And it's almost an unfair thing to do.

PEGGY BRICKMAN: Where did that come from? So I'll tell you where it came from, which is engineering. So engineers, they work in groups the whole time. It's one of their accreditation standards, is to work—because that's how engineers work. And so a lot of these peer evaluations came about in engineering and they would have these—you'd ask them six or seven questions and they'd be able to rank the students on a scale from 1 to 5.

And so when I started, I was totally convinced by these engineers who were like, we've got this great setup and it doesn't matter how many students you have and they can evaluate each other. And I cannot tell you how many issues I would have. I mean, even if they did it every single time, they would rate each other.

At the end of the semester, students were basically digging their colleagues, their fellow students just because they didn't like them and then they would do that weird like your best defense is a good offense. So you'd have a student who had been terrible all semester. And so he would offensively go in, when it was time, and he would ding all the other students.

And so then I'd have this whole team of students in my office just screaming at each other. I would say, wait a minute, wait a minute. So I understand what he did. Why did y'all all give him 100s if you're saying right now after the fact that he wasn't doing any work? Well, I just felt bad. I felt really bad doing it.

So I was like anecdotally I'm seeing that this isn't working, but it wasn't until I really dug through the research and found papers that I discovered, oh no, this is so real. Yes, there's a lot of problems with doing that kind of peer evaluation.

STEVEN ROBINOW: OK. So we're moving into the second paper here now where you do mediation among student groups. So before we go there, I want to—in the discussion of your first paper I'm going to read two sentences at the end of the paper. And the topic here is, "Are time-consuming strategies that facilitate group work worth all the effort?"

And the second sentence in that section says, "In student interviews, we observe that only about one out of seven lower-scoring students perceive group work and instructional-facilitating activities to be beneficial for their performance compared with three of seven in high-scoring groups."

So whether you're a low scorer or a high scorer, less than half of them think group work is any good. But then at the top of the next page it says, "We found that regardless of group composition or group performance level, students were likely to report positive experiences with group members."

I'm going to read those two sentences again. "In student interviews, we observe that only about one out of seven lower-scoring students perceive group work and instructional-facilitating activities to be beneficial for their learning performance compared with three of seven high-scoring students."

The next sentence. "We found that regardless of group composition or group performance level, students were likely to report positive experiences with group members."

So those seem at odds with each other. Can you talk about that?

PEGGY BRICKMAN: Students mix their comfort and sense of belonging with learning. They keep those as totally separate things even though we know that when you have a greater comfort and sense of belonging, students actually learn more. So this is a work that was done by Lee Theobald, for example, at the University of Washington.

I would have to—and having redone those interviews again, I would try to dig into what they really meant by useful. Because when you tease that apart, I think what students are saying is, I love to be able to come into the room and know somebody that I could talk to and ask questions to if I missed class, I could catch up on stuff.

Just kind of—you can sense it in the room as the semester goes on. If you teach a class that doesn't have groups, like everybody is sitting there quiet, no one's talking, whereas in a class that has group work, there's so much chatter and they're all hugging each other on the last day of class. And I think that that is real and important.

But I think in their perception they are thinking, is this group going to be like a tutor for me? Are they going to identify things that I have trouble with? And without experiencing it without groups and with groups, I don't think they really can understand there are gains.

You can see, over the course of the semester, students who are doing poorly are doing better because of their group. They don't realize that necessarily. Also, it sucks to be in a bad group. It really does. And if you're in a bad group, it's miserable. At the time, although there are all these resources for identifying problems in a group, the remedy was to do the peer evaluation and ding each other on their grades.

There was no—like, what am I supposed to do with these people? Nothing. And so I think that's why we went into the second paper, which is, OK, this is ridiculous. You can't tell people that group work is going to work and then not help them mediate when there's problems because we know there's going to be problems.

STEVEN ROBINOW: So before we get to the next paper, just the final question, I guess, about this paper. There's lots of things to talk about in here that are fascinating. I recommend people read it. So at the end of the day, did the group work—did your efforts at group work actually improve student learning? Did it improve student retention, for example? Did you look at equity gaps? Did it decrease the equity gap?

You have these good groups, these bad groups, but overall of course, were more students passing and were more minoritized students doing better?

PEGGY BRICKMAN: I can't say about the minority students because the numbers were not very high, not high enough to statistically say if you've got less than 10%, it's problematic. And that was frustrating to both Eunice and I. In particular because there have been a lot of reports that for minoritized students, having collaborative learning is really beneficial.

Students did better on the same exact kinds of tests. For me, I was able to actually incorporate more higher-order problems on tests that I had never been able to use in the past because they could troubleshoot and talk through things together.

And so for me, teaching a course like that that's meant to help develop science literacy, science literacy to me is not memorizing the bases in DNA. It's really more about, well, should I get this genetic test or shouldn't I? What do I do about climate change? Which of these solutions is showing the best gains in terms of reducing carbon? And those kinds of questions are higher order. So it was really important for me that I could add more of those kinds of questions to my test, but data shows that they're—it's improving.

STEVEN ROBINOW: The rates at which you were giving these F's and students withdrawing—your DFW rate was dropping.

PEGGY BRICKMAN: Yeah. I think, too, though, when you have group work, it changes the kinds of assessments that you're doing. And so it isn't just a bunch of multiple choice tests but you have students doing writing, they're turning things in for a homework grade, because once you have groups, it's more easy to grade things like that.

So I would say it's difficult to tease that apart because I'd literally changed the kinds of assessments that I could offer to students because groups allowed me to introduce those kinds of things.

STEVEN ROBINOW: OK. So your assessment strategy changed, and the opportunities for students, then, to show their learning became more diverse in the course as well.

PEGGY BRICKMAN: Yeah. So many different metrics. So we created a test of science literacy skills, showed great gains in those for our students who are working through these projects where they had to tease apart some skills that they need to make decisions in their real life. And all of that—I could never have done that without group work. It's not with 600 people or 700 people I'm teaching.

STEVEN ROBINOW: So despite the title, "When Group Work Doesn't Work," there were still sufficient benefits for group work.

PEGGY BRICKMAN: But when it doesn't work, it's bad. And I don't think there had been enough work done on the negative side of it as if we were just whitewashing it.

Like oh, it's going to be fine. And then when I would go and talk to faculty, oh, I hate group work, I never use it my classes. It's horrible, it's a disaster. I'm like, OK, where's the mismatch here? Between all these papers that say it's great and then if I go and talk to a campus, a faculty, they're just like, oh, I hate group work. I hated it when I did it, I hate doing it. So I was like, OK, this is a serious problem.

STEVEN ROBINOW: So these are implementation problems. Design, design and implementation, yeah.

PEGGY BRICKMAN: Yeah. And a lack of real-world experience. If it doesn't—like maybe they were right saying it doesn't work with 300 people. It's fine if you've got 12 people. Do not try it if you've got 300 people. You just don't have a way to moderate what's going on, which, I think, is a problem because for a big R1 like me, they're not changing our class sizes.

STEVEN ROBINOW: But you do group work in a class of 300?

PEGGY BRICKMAN: I do, absolutely. In 280 people.

STEVEN ROBINOW: And you see gains from group work?

PEGGY BRICKMAN: Yep.

STEVEN ROBINOW: Whatever the reason, whether it's because you change assessments in addition to groups, there are so many factors. But at the end of the day, this structure has improved your student performance in your course?

PEGGY BRICKMAN: Yes.

STEVEN ROBINOW: In spite of all the problems?

PEGGY BRICKMAN: Yeah, which is why I keep doing it, even though I know it's going to be a hassle.

STEVEN ROBINOW: OK. So I'm going to end episode 1 here. So group work is worth it. In spite of all the hassles, if you're here for your students, group work can work in large classes and in small. It's not easy, it's not simple, but it can be helpful to your students. In part 2 of this series, Peggy Brickman will talk about what to expect with group work. She'll talk about problems, she'll talk about the frequency of problems, and she'll talk about the success of mediation. So please go to part 2 of this episode.

For more information about Peggy Brickman, some of her favorite readings, please go to our website at teachingforstudentsuccess.org. Thanks again for spending time with us today. I hope this discussion on group work has been helpful to you, and I hope you tune in for part 2. Thank you for caring about your teaching and your students. Thank you also to the education researchers out there who are working hard to improve the learning experiences and learning environments for all of our students.

Those of us at Teaching for Student Success would love your feedback. I've broken this episode up into a couple of pieces. Let me know what you think about that. If you like it, great. If you don't, great, let me know. Please contact us through our website at teachingforstudentsuccess.org.

Teaching for Student Success is a production of Teaching for Student Success Media. Let's end this podcast as I always do with some music by JuliusH. Julius's music or some of Julius's music can be found on Pixabay.

[MUSIC PLAYING]