If your students have been asked to do substantive work in groups or teams, then you are likely to face two challenges:
- how to give students the tools they need to hold one another accountable for their individual contributions and cooperation; and
- if there is a collective product or performance—how to weigh the relative value of individual students’ contributions when assigning it a grade.
Keep in mind that in meeting these two objectives, student perspectives are critical. Students themselves are the ones with the most accurate knowledge of what their peers have done. They have the most relevant perspective on the quality of other students’ contributions and cooperation. Students, themselves, also have the most at stake. They feel the impact of productive or unproductive peer behaviour directly.
Challenge One: Providing tools for peer-to-peer accountability
A system of peer evaluation should provide mechanisms where student voices are factored into the overall evaluation of any individual student’s work.
What are the criteria that your students will use for peer-to-peer evaluation?
It’s important for the students, themselves, to play a role in developing the criteria they will use for this process. One way to do this is through an activity early in the course or team project, where students describe the behaviours they expect from their peers, in order for group activities to be successful.
When asked, students will often cite behaviour such as, “show up every time,” “come prepared,” “do the reading,” “don’t hold back if you know something useful,” “listen,” “be civil when disagreeing,” etc.
Ask students to work individually or in their groups to make short list of preferred criteria. Then, as a whole class, sort these out so that everyone is working from the same set of 3-5 criteria. You don’t want too many criteria, and you don’t want different criteria for every group, as these will become an administrative nightmare to manage if you have several groups.
How will you ensure that students have a chance to adapt to their classmates’ expectations?
Students do not automatically know how their behaviour and attitude are perceived by peers. It’s critical to give them a chance to get anonymous, penalty-free feedback, before they are held fully accountable at the end of a semester or project.
Develop one or two “soft” peer feedback events mid-way through the project or course, where students provide feedback anonymously to one another, using the original criteria developed at the beginning (see above). It can also be useful to set aside time in class for students to discuss with their groups the feedback they received, and reflect on what the group needs to do differently going forward.
Challenge Two: weighing individual student contributions to a group project when assigning individual grades
There are generally two ways of managing peer evaluation as part of a student grade:
- Create a stand-alone category for overall peer evaluation in the course grading scheme (This is especially relevant in courses where the group or team work is extensive and integral to design of the whole course); or
- Use a peer evaluation component imbedded within the grade each student receives for work on a specific team/group assignment.
How does peer evaluation work as part of the whole course grading scheme?
If students have generated criteria for peer evaluation, and if they have had a chance to get anonymous, penalty-free feedback from peers during the course, then it is fair to include a student-generated component of the final course grade, based on overall quality of peer helpfulness. (Be clear when you communicate the plan to students at the beginning of the course. You are not asking them to make judgments about other students’ academic ability or work—you are asking them to evaluate one another’s helpfulness in doing group assignments.)
The peer evaluation component of the final course grade should not be so large as to greatly alter, by itself, a student’s individual grade in the course, but it should be great enough to affect grades at the margins. Typically, 5-10 percent of the total student grade could be assigned to the peer evaluation component in order to meet this objective.
Here are two ways you might accomplish this:
Option One: At the end of the course, ask students to anonymously assess each individual peer in the group, using a Likert-style sliding scale for each criterion. The sum total of points can then be calculated from this scale, and a score derived from the total received from all group members. Students accumulating the highest cumulative score receive full credit toward the 5% peer evaluation component, while students with lesser totals receive a smaller fraction of that credit.
Option Two: At the end of the course, give students a fixed number of points (100, for example), and ask them to distribute these points to their group members, based on the strength of each member’s contribution. As in option one, the raw total received from all group members would translate into a general score, which could then be converted into a fraction of the 5%, based on a comparative, rather than absolute, scale.
iPeer is a software available to VIU faculty for conducting both kinds of evaluation. Download the iPeer user guide for instructors.
How does peer evaluation work when it is embedded within the individual grade assigned to work within a team/group assignment?
First, the peer evaluation component needs to be explained explicitly to students as part of their introduction to the assignment. This explanation needs to represent WHY the peer evaluation will be useful to students, HOW the peer evaluation component will be generated, what the criteria are, and the overall weight it will carry in determining an individual’s grade based on the group assignment.
An imbedded peer evaluation component for a specific assignment can be managed in either of the ways described above (Likert scoring vs. distribution of points), but here it is applied to a subset of the course, such as a specific group project, presentation or team assignment. The objective is to use one of these processes to generate a total score, factor or multiplier that fairly represents each individual student’s relative helpfulness and productivity, as measured by peers in the same group.
At its simplest, the process can be used to generate a score or points that could be added to the group’s score in order to determine an individual’s overall grade for the project.
Or, the process can be used to create a “multiplier” that is applied to the team score, to adjust it up or down, based on how highly peers thought of the student’s contributions.
In addition to the Likert scale and the distribution-of-points approaches, another way to determine the score, “multiplier” or factor for the individual student is a ranking system. Students are asked to rank order the peers in their group according to their overall helpfulness, using the criteria. These rankings are consolidated by the instructor, to generate a correlative number of additional points that are added to each individual score.
Things to consider, whatever you do
- Students may be reluctant to differentiate levels of contribution by their peers, even when done anonymously. For this reason, you may need to create a rule for how points are to be distributed, if you use the point system. The rule can state that no two students can receive the same number of points. The larger the number of points to be distributed, the less stressful this decision will be for them. They will agonize over distributing 20 points over 5 peers, for example, but will feel freer to differentiate if the point total is 100.
- When discussing “peer evaluation” with colleagues, be sure to clarify what you are doing. You are not asking students to evaluate performances or products according to academic standards, which is ultimately the responsibility of the instructor. You are asking students to evaluate behaviour and attitudes within the context of group productivity. These are things that students, and only students, are capable of fairly evaluating.
- Unless a specific goal of the course is “cooperation,” as might occasionally be the case in a management, social welfare or counseling course, be wary of weighting the peer evaluation component of a course or assignment too heavily. Even a modest amount of 5-10 % will get students’ attention. More than that, however, and the assignment or course may fail to be a fair assessment of any individual’s work, as there will be those pragmatic students who take the opportunity to hitchhike on their peers when the peer component is excessive.