Since I read about the experiences of American teachers who had implemented ungrading, I have wanted to test this approach in my own courses.
There are many variants of ungrading, but the basic idea is to grade the student’s work by giving extensive feedback without a grade (whether in the form of letters or numbers). The students are asked to grade themselves.
I tested ungrading in fall 2022, in my Secondary 5 Physics course (weighting 3-2-3) in the Springboard to a DCS pathway. I hereby relate my experience.
What I did
In my course, formative and certification evaluations consist of:
- 2 exams (1 2-hour exam, 1 3-hour exam)
- 2 tests (1h30 each)
- 3 mini-tests (30 mins each)
- 8 lab reports
I asked all my students (19 students for the first 2 weeks, then 18 until midterm, and 17 after) to come and see me at my office for a 1st individual meeting, in order to get to know each other, during the 1st or 2nd week of classes.
Then, I met them every 2 or 3 weeks until the end of the semester, for a total of 5 or 6 meetings per student (the second to last meeting was optional).
The meetings would last between 10 to 20 minutes per person, as needed.
Every time a student handed in an assignment, I would provide feedback (even more than usual!), but I would not give them a grade either on their copy or on Léa. I would instead record the grade I would have given them in an Excel file that I would keep for myself. In the following class, I would hand back the copies without any grades to the students, and I would provide group feedback on the evaluation, as usual. However, the students had now to determine the grade they would have deserved. To do so, the students could grade themselves based on, among other things, the feedback I provided and the solutions to the problems in the exam I had presented on the board.
During the individual meetings, I would go over the student’s evaluations, which they would have completed since our last meeting. I would ask them to grade themselves for each evaluation, based on the criteria described in the course plan, and I would record the grade in my Excel file.
When the estimate of their grade was too far from mine (upward or downward), I would discuss it with the student (without revealing the grade I myself recorded in the Excel file, but by questioning the student in order to make sure they had taken into consideration all the factors). It rarely happened. The students would normally be prompt to reconsider their grade to a much closer estimate to mine, without any pressure on my part.
After each meeting, I would record their self-assigned grade on Léa, so they could monitor their own progress.
On Léa, I had set all these evaluations as being worth 0% of their final grade, since I did not want to adopt a “summative” approach, that is, the final grade being the sum of all the grades. I wanted the final grade of each student to reflect their real progress and understanding of the course material at the end of the semester. Even if a student does not understand a concept on week 3, I believe they should not be penalized if they come to understand it by the end of the semester.
At the end of the semester, during the last meeting, after their final exam, each student first revised and corrected their final exam. Then, the student decided on a final grade for the course (respecting the weighting as defined in the course framework stating that labs are worth 25% of the final grade and theory-based evaluations, for 75%). I then revealed to the students the grades I had given them throughout the semester and the final grade I would have given them if I had had to evaluate them on my own. In most cases, the discrepancies between our final grade estimates were less than 3%. In these cases, the highest grade would be issued to the college.
Why did I want to try ungrading?
I wanted to test ungrading because I believe grades might have a counterproductive effect on learning.
During my university studies in Physics, I would focus more on getting good grades than on actually trying to understand what I was supposed to learn. I think many students in my classes find themselves in a similar situation, and I am trying to find ways to change this. Making assessments more authentic is definitely one way of doing things, but the ungrading approach seems worth exploring. By discussing with students the grades they earned in the course, we can push them in the right direction, and make them understand the importance of “understanding”.
I was hoping the individual meetings would give the opportunity to students to explain, for example, their reasoning behind an unusual answer on an exam. This could make the evaluation “fairer” for let’s say, a student who would have misunderstood the question but would otherwise understand the notion this question was meant to evaluate.
While usually, after an exam, the students pay more attention to their grades that will be posted on Léa, I was hoping ungrading would encourage them to analyze their strengths and weaknesses and allow them to learn from their mistakes (and take pride in their accomplishments).
In addition, I wanted to implement a non-summative assessment approach, so that the student’s final grade represents the best way possible the achievement of the competency at the end of the course. I wanted to offer assessment opportunities throughout the semester but without “freezing” the fact that any evaluation would be worth 5% or 15% of the final grade. Therefore, a student who does not do well on a question about a given concept by the midterm, but does better on a question about the same concept at the end of the semester does not deserve, according to me, to be penalized for his first lower performance.
What were the results in my course?
It happened a few times that I had to redirect students who had been too hard on themselves, but it was never complicated. The conversation was always simple, and the students were systematically willing to adjust their grades. In most cases, they were students presenting a great solution to a problem, but they had made one calculation error in one place, so they would assign themselves a grade below the passing grade.
In the vast majority of cases, the students would assign themselves a grade different from mine by less than 5%. In some cases, the discrepancy was greater. For instance, a student graded herself 50% on a mini-test to which I would have given her 30%. It seems nonetheless acceptable in this case, since what matters is that we both agreed that she did not obtain the passing grade; expectations were not met.
For final evaluations, the students and I were also generally on the same page.
A humble student graded himself 11% below what I would have graded him. I explained to him my point of view and gave him the highest grade.
Two people I estimated as failing tried to assign themselves a 60, while admitting a lack of understanding of the course content. We agreed on lower grades.
At the midterm, an anonymous survey
At the midterm, I asked the students to fill in an anonymous Forms survey in order to give feedback on the method of assessment used in the course. I obtained only 8 responses (out of 18 students).
The responses were mostly positive.
In a multiple-choice list, 4 people answered “pleasant” to describe the method. One person mentioned “unpleasant”. This person and 2 other respondents indicated the method to be “uncomfortable; I don’t like having to do a self-evaluation” and “mysterious; I don’t know what to expect”.
In response to another question:
- 4 people mentioned the method to be “less stressful than a traditional evaluation”.
- The 4 others checked off that it was “no more no less stressful than a traditional evaluation”.
- No one evaluated it as more stressful.
- 4 people indicated the method to be more motivating than a traditional evaluation.
- 2 mentioned it was “no more no less motivating” than a traditional evaluation.
- The 2 others answered it was less motivating.
At the of the semester, an in-person individual reflection
At the end of the semester, at the end of the meeting with each student, I asked each one of them their opinions on the course evaluation method. They all give a positive opinion, except for:
- 1 person simply mentioned preferring a traditional evaluation, without being able to explain why at that moment. This person was failing the course and it was obviously an emotionally-charged discussion. It seems perfectly legitimate that the student was unable to articulate their thoughts. There is no way to know what would have happened with a traditional evaluation, but I do not rule out the possibility that, for this person, a traditional evaluation could have been beneficial.
- 1 person who worked really hard and performed well indicated that they would have preferred being able to see their grades and the weighting for each evaluation, with the group average for each evaluation, etc. This student explained to me that this would have been reassuring in their quest to get good academic results to get admission to a limited enrollment program. I reminded them about the importance of comparing oneself with oneself. Whether or not one knows the group average throughout the semester should not change their grade. However, in some cases (such as this person), it could have had a (positive or negative) impact on motivation… and thus, a (positive or negative) impact on learning and success. I bet it would be more beneficial not to show the group average, but it is quite possible it was not a win for all the people in the group.
- 2 people mentioned having enjoyed the regular meetings with me. They noted it helped them to remain up to date with their studies in the course and that they appreciated having the opportunity to discuss their evaluations. However, they mentioned they would still have preferred knowing their assigned grade in advance, so they could have based their self-evaluation on it.
I am worried such a practice would essentially be counterproductive by exerting too strong an influence on certain people. Moreover, this would turn the spotlight back on the grade when the students would receive my feedback on their evaluations.
All the other people had only positive comments on the method of evaluation. (Obviously, since they were in front of me, there is a possibility some people were too shy to give negative comments.)
In a way I had not anticipated throughout the semester, more than the self-evaluation or the no-grade aspect, it is the individual meetings every 2 or 3 weeks that the students appreciated the most. They liked having a privileged time to ask their questions on the course content (as much on evaluation questions they did not understand, as on homework exercises). At the beginning of the semester, I was apprehensive that some people might be annoyed by the fact that they would have to take time out of class to come to meet me at my office, but in the end, it was not the case.
Due to my work for Eductive (Collecto), I was teaching only one course at the Cégep Limoilou this fall.
When the workloads were distributed for the fall semester, I was supposed to get 12 students in my Secondary 5 Physics course. It was exceptionally suitable to test upgrading. With 12 students, I would have plenty of time to individually monitor each one of them. Finally, I had 18 students in my class. It was still a manageable group size, but it added 50% more time than I had planned to allow for individual meetings with my students (without a reduction in workload).
I felt slightly demoralized at the beginning of the semester when nearly a quarter of the students had forgotten their 1st meeting with me. I wasted time waiting for them, and I had to reschedule all those meetings. However, the situation improved throughout the semester, and no one forgot the last meeting.
Since I only had one course and a flexible work schedule, I was available to meet the students before and after class. This made the meetings more interesting for the students (who usually have a relatively open schedule, in the Springboard program, and could have complained about having to stay at the college 1 or 2 hours after class to wait for the meeting). In addition, I offered video conferencing if it suited them best, but only one student chose this option (for one meeting only) throughout the semester. However, at the end of the semester, 2 students decided to meet via video conferencing for the meeting following the final exam.
Weighing the pros and cons
Creating individual bond
I firmly believe that ungrading helped me build strong bonds with my students. The group was quite calm with many quiet people; they would have certainly stayed more “unknown” to my mind, without our meetings. I was glad to hear at the end of the semester that the student enjoyed these mandatory meetings since I was actually worried that the fact they were mandatory would repulse many.
Nonetheless, this required a great deal of my time. It was manageable, but I could never imagine doing it with, let’s say, 3 groups of 40 students.
With a bigger group or with more groups, I would have had to adapt my method. I could have, for example, saved the ungrading approach for the most important semester evaluations.
The fact remains that I would still have lost something since with my group of students this semester, I was under the impression that having several meetings throughout the semester was beneficial for their learning. In the Springboard program, many students have not necessarily developed the required work methodologies to manage their workload autonomously. I believe our regular meetings help some of them to stay on track (even if some people still failed the course, unfortunately).
Impacts on learning
I cannot effectively measure the impact of this change of evaluation approach on the learning of my students this semester (for lack of proper scientific methodology). I am under the impression, subjectively, that it has caused many students to reflect on their evaluation in retrospect, to benefit from these learning opportunities.
Drawing the attention away from the grade?
As I have mentioned, one of my initial objectives was to repent from my past as a student who was more interested in getting a 100% on her transcript than really understanding the course content. I don’t think ungrading allowed me to achieve this goal this semester.
Most of my students were not really concerned about the grade itself. These people wanted to pass the course, but it did not matter whether it was with 65% or 75%. (They took the evaluation process seriously, but let’s say they did not have a clear objective from the beginning of the semester.) For these people, I believe the implementation of ungrading was mostly beneficial.
For the few people who were aiming for a limited enrollment program, I do not think ungrading truly changed anything. I cannot really blame them: the grade on their transcript can have a major impact on their career choice. It is totally normal to pay attention to the number… Yet, during the individual meetings, I could tell these people were focusing more on “getting the right answer” than “understanding why this is the right answer”. As long as they had the right answers, I could not penalize them.
To try again!
I would like to repeat the experience in other contexts (in Science, for example), to be able to compare.
However, I will not try it this semester (winter 2023), since I work full-time for Eductive (Collecto). In fact, I doubt I will be able to implement the approach again the way I did the last semester, the workload being too heavy to handle. I will have to adapt my approach for the next teaching context.
I would be interested in knowing if some of you have tried ungrading (one way or another) in your courses. Share your experience in the comments! This would provide food for thought on how to adapt the method to make it realistically applicable to a large number of students.