Peer Assessment: The challenge of receiving feedback

A number of instructors at McGill have implemented peer assessment (PA) in their courses and have generously shared some of their reflections on the experience.

 Dr. Claire Trottier teaches Introductory Immunology (MIMM 214), a required U1 course for 300+ students majoring in microbiology, immunology, anatomy and cell biology. During our conversation, Claire shared how she implemented PA in this course, reflected on the challenges of receiving feedback from peers, and offered advice to other instructors considering implementing PA in their courses.

What motivated you to try PA in this course?

First, I wanted to fulfill one of the Microbiology and Immunology program outcomes, which is that students develop their ability to communicate to a non-scientific audience. Since it’s difficult to put together an instructor-graded communication assignment for 300 students with limited TA support hours, PA seemed like a good alternative. Second, peer review is such an important part of science, and it has a difficult learning curve. PA allows students to practice peer review as they learn how to give and receive feedback. Third, PA sensitizes students to how much work professors do when they’re grading: Students say, “I spent four hours on just a few reviews” and I’m like, “Yes.”

Can you describe the assignment?

The students write a newspaper article—two pages, double spaced—on a topic of their choice related to immunology. They also fill in a questionnaire (pp. 1-2) where they describe their topic, including how it connects to the course content and why they chose it, a specific audience for their article, and where they would publish their article. Their newspaper articles end up being about a huge variety of topics, aimed at different non-scientific audiences. Each article is peer assessed in two rounds. Based on feedback from the first round, students revise their assignments and then submit the revised versions, which are also peer assessed.

I have a detailed rubric with specific prompts that I ask students to answer when giving written feedback. They see the prompts in advance so they know what to expect. I provide examples of good feedback, medium-quality feedback and bad feedback in an online document, and this year I plan to record a short video to post online that will discuss both how to give feedback, and the process of receiving feedback.

How much is the assignment worth?

The assignment is worth 25% of the course grade and is based on the feedback from their peers. This grade is calculated by the software and incorporates a number of factors, including accuracy of the student’s grading and timely submission of reviews. Students spend a lot of time on the assignment, including the PA component, so I feel like that should be reflected in how much the assignment is worth.

 How is the assignment graded?

In 2019, this 25% is divided up as follows:

Assignment task	% of assignment	% of course grade
Round 1: draft 65% for writing the newspaper article 35% for reviewing peers’ writing (PA)	40% 26% 14%	10%: 6.5% 3.5%
Round 2: revision 65% for writing the newspaper article 35% for reviewing peers’ writing (PA)	60% 39% 21%	15%: 9.75% 5.25%

In 2018, after I explained the assignment and grading, I used a polling question to have the students choose between several grade weighting options for writing the newspaper article vs. reviewing other students’ newspaper articles. The option that got the highest number of votes was then implemented. I had done modelling with my grades from the previous year to get an idea of what impact each option would have on their assignment grade, so I gave options that I would be okay with. In 2018, the greatest number of students voted to have writing the paper be worth 65% and to have the reviewing (PA) portion be worth 35% of the assignment. That was interesting because in 2017, I got complaints from students about the distribution of the grade. They felt that at 25%, the PA portion was worth too much. Yet the 2018 students chose to have more weight on the PA portion. I did the same polling exercise at the end of the semester, and the majority of students chose the 65/35 option, so I’ve decided to stick with that in 2019.

How did students do in terms of giving or receiving feedback?

While we may think more about students learning to give constructive feedback, learning to receive feedback can be very challenging, too. Receiving feedback involves dealing with the emotions about the feedback. Sometimes the feedback is good and you’re going to make a change based on the feedback; other times you’re going to disagree with the feedback and that’s okay, too. Either way, going through that process is important.

This year I used Peergrade software to manage the logistics of having 300+ students do PA. [Editor’s note: To learn more about Peergrade, fill out this form.] Peergrade includes a flagging system: If a student doesn’t like the feedback they received for whatever reason, they can flag it. Also, if a reviewer gets a bad back-evaluation—that’s feedback on the feedback they received—they can flag it. I had about 150 flags for the first round of PA! I reviewed the flags one by one to determine which ones required action, which was time consuming. Action might involve adjusting a student’s grade or revisiting with the class what constructive feedback looks like, for example. It’s true that there were a few students who graded really harshly. But in some cases, the flags showed that students were working their way through the emotions of getting the peer review: they felt upset by something someone had written. I understand that! Sometimes when I get feedback, my initial reaction is pretty strong, too. Then you sleep on it, and work through it.

I model how to cope with feedback in class by doing a mid-semester survey to invite students’ feedback on the course. I tell them that there are a couple of reasons I do the survey. First, I want their feedback so I can see if I can improve, and I want to be responsive to their needs. Second, it’s a way for me to show them that it can be difficult to get feedback, and that’s okay. I explain in class that here’s a piece of feedback and, you know what, this is a really good idea and I’m going to do it. And then here’s another piece of feedback that’s an interesting idea, but I’m not going to do it, and this is why. So I try to model it that way.

Did you see a difference in students’ feedback between the first and second rounds?

The quality of the written feedback improved substantially from the draft to the final version. There were more detailed examples, much more specific and constructive feedback. In 2018 I gave them prompts like “how did the author communicate enthusiasm for the subject to their chosen audience?” and then, “how could the author make this paper more engaging for the audience?” So some people said, “this sentence was confusing”, and then they quoted a sentence and said, “Maybe this is how you should phrase it…” with really specific examples, rather than “Just try to make it more peppy.” As authors in the first round they got feedback from three reviewers, so they also saw what others were doing and they could learn from it. This year I have modified the prompts so that they are more closely associated with the numerical criteria on the rubric; this was a suggestion from several students last year to improve the feedback process.

What advice do you have for instructors interested in trying PA?

• Include some students in designing the assignment. I asked a couple of students, “Are the instructions clear?” “Is the rubric clear?”
• Be explicit in class about your expectations and create a rubric to accompany the assignment. You might base it on a rubric that worked well for you or a colleague in the past.
• The prompts for students to provide written feedback on one another’s work should be specific and designed to elicit what you want the students to get out of the assignment. Students may find it more difficult to respond to general prompts.
• Don’t underestimate the amount of time PA can take in terms of how to guide the students through it and deal with technical logistics.
• Make sure the students know that it can be difficult to accept feedback, and that’s okay. That was really important for my students to hear.

Readers: How do you help your students learn to be receptive to feedback from peers?

Check out Teaching and Learning Services’ other peer assessment resources.

Original publication date: January 31, 2019