Dear students,
Here's why we need to move away from marking and grades:
Grades are stressful for students and encourage you to focus on doing well on tests/getting good marks, not on actual learning. With reduced stress (though not no stress at all), students can learn more deeply, feel free to make mistakes—lots of learning happens through productive failure!—, and maintain understanding longer.
Grades are part of larger systemic injustices that privilege some students over others, without taking into account inequities of economic status, gender status, race/ethnicity, religion, neurodiversity, etc. We need more flexibility and compassion in our assessment systems. We also need to move away from grading on a curve or forced distribution of grades, which are just plain inequitable!
Letter/percentage grading in Western education systems has only been around for a little more than a hundred years—if we invented it, we can replace it. It’s not the only way to assess student learning, and it’s nowhere near the best way to assess student learning.
According to Alfie Kohn, “Tests are a deeply flawed way of evaluating students and have been abandoned by many thoughtful educators in favour of more authentic and informative types of assessment” (Ungrading: Why Rating Students Undermines Learning and What to Do Instead, edited by Susan D. Blum, West Virginia UP, 2020, Foreword xix). In other words, we need assessments that model or prepare students for the kinds of work they’ll do in the field/their career (authentic) and that tell instructors about actual student learning (informative), not just that this student can memorize facts.
It's an important step in the process of decolonization and indigenization. Indigenous learning methods don't rely on written tests and percentile grades: this is something that developed in North American, white, Euro-dependent education systems in an effort to impose standards, uniformity, and productivity on young learners. We can improve learning by embracing different methods of showing that students have mastered skills and gained understanding.
Graduate and professional programs are moving away from a focus on applicants’ GPAs and toward pass-fail grades on transcripts: Deng and Wesevich, for example, show that by 2016 eighteen of the top twenty American medical schools only looked at pass-fails on transcripts (“Pass-Fail is Here to Stay in Medical Schools. And That’s a Good Thing”). Apparently, there’s no correlation between med students’ grades in undergrad science courses and their later success as practitioners. And medical schools themselves are trying to reduce student focus on grades as part of efforts to reduce self-harm among med students.
For decades, educational researchers have recommended moving away from grades, but the institutions that have done so across the board are rare (e.g. Montessori schools, Evergreen College in Washington State). Some instructors work within an institutional grading system to amend and/or undermine it. I’ve worked for twenty years to amend SFU’s by giving students more choice, allowing them to weight their assignments to give themselves their best grades, allowing revision and resubmission, and lowering or removing late penalties. Now I feel ready to undermine the system, and I’m going to have students above the first-year level assign their own grades. I will believe you. I will believe in you.
What is Ungrading?
Roughly speaking, ungrading is any assessment system designed to modify/undermine/expand beyond current Western/North American grading practices. You may have taken courses that do a bit (or a lot!) of this, but probably most of your courses still run on the "assign work, do work, hand it in, and get a grade on it" system.
Below, I’ve outlined some different ungrading systems, as they can work in an institution (such as SFU) that still insists on having a final letter grade for a course.
The portfolio model/holistic grading
· Students develop a collection of artifacts (or activities) that demonstrates their understanding of course concepts, critical thinking, creativity, or whatever the major learning goals of the course are. Ususally there's a significant amount of self-reflection built into the portfolio system.
· All students could create the same artifacts/do the same activities, each student could choose assessments from a menu, or each student could self-design their assessment artifacts and activities based on their personal goals, skills, and interests.
· Instructors give feedback/engage in dialogue at various points (formative assessment) and write, receive, or help to develop a holistic narrative assessment of the portfolio (summative assessment).
· The instructor may assign the end-of-course grade, the student may, or they may negotiate it.
· Dr. Dan Laitsch and others in Education here at SFU use variations on this model. [This is the one I’m interested in for my upper-level courses, and I'm using Jesse Stommel as an inspiration.]
The contract model/specifications grading
· Assessment is based on lots of small pieces that are all pass-fail.
· The instructor lays out how many and which assessments students need to successfully complete to get A, B, or C grades. Students can then choose what grade they're aiming for in a specific course and do exactly as much as they need to to get that grade. Students who get A grades need to have successfully done more work from a bundle of more difficult assessments.
· + and – grade increments could be by instructor choice, influenced by attendance and participation, or through self-justification by students.
· Dr. Luke Clossey in History at SFU uses a version of this model in some of his courses.
The mastery model/competency-based grading
· Assessment is tied to learning goals, with students having multiple attempts and multiple means to demonstrate mastery (usually not just retaking multiple-choice tests until they get As, but giving students the opportunity to show their mastery of each skill).
· The number of successful mastery demos equates to A, B, C grades. Sometimes you have to demonstrate mastery in a particular order and if you don't get to the final skills you can't get an A.
· I know that Nursing courses at other institutions use this model…at SFU, Dr. Mark Blair in Psychology has used a retaking multiple-choice tests version.
The effort model/labour-based grading
· Students need to demonstrate regular and long-term activity related to the course in order to get a good grade.
· This system was developed for language learning and literacy classes, and as a way to get beyond the limitations of standardized white academic English.
· Nicole Berry and Lyana Patrick in Health Sciences have done some of this at SFU.
Partial ungrading
· Some assessments in the course may be marked and get grades in the traditional way, while others aren’t.
· Some assignment grades may be self-determined or peer-graded using a rubric.
· Students may be able to drop low grades, choose their assessments, change grade weightings, do an oral exam instead of a written one…lots of possibilities!
· I’ve done two main kinds of partial ungrading systems in the past, which I call bento and buffet: in the bento method everyone does the same assignments and has the same due dates, but students can choose to put different grade weightings on them depending on how they do; in the buffet method, students choose what assignments they will do, when they’re due, and how much of the grade weighting each gets. I'm currently designing a new partial ungrading system I can use in my 100-level classes. Quite a few SFU instructors have done partial ungrading in different ways, including Sheri Fabian in Criminology and Diane Finegood for the Semester in Dialogue.
Yours respectfully and in anticipation of some interesting teaching experiences,
Nicky
[I had a good portion of this drafted before attending an online talk by Firas Moosvi, but I gratefully acknowledge that his talk inspired me to add to and revise this blog post.]