Danger: Curve Ahead
In order to make the case for going gradeless in our classrooms, we also have to shine a light on the ways in which traditional grading practices are inequitable.
Grades serve as mirrors for the structural inequities that are woven into the fabric of our educational systems, and they magnify these disparities due to their impact on the futures of the students who are affected by them. Often used for the twin purposes of comparison and competition, grades are drivers of injustice. Students do not need to compete with each other to learn effectively. In fact, any kind of competition will immediately privilege those who have played the game before and those who know the rules of the game. Many traditional grading practices not only embrace this kind of competition, but thrive on it.
In this post, I want to share an excerpt from my forthcoming book about grades that deals with one of the biggest perpetrators of both competition and inequity in our schools today: grading curves.
In order to make the case for going gradeless in our classrooms, we also have to shine a light on the ways in which traditional grading practices are inequitable.
One of the most frequently used curving models is called norm-based grading. In this framework, the teacher adds up all of the test scores and divides them by the number of students in the class in order to determine the average of all the scores. Those students who then score at or near the average get a C, those whose scores fall at increments above the average (statisticians refer to these increments as “standard deviations”) get Bs and As, and those whose scores fall at increments below the average get Ds and Fs. This is true even if the average grade was, let’s say, a 27%, which would practically mean that many people actually failed the exam, but with the curve some folks who may have scored in the 40s or 50s would get much higher grades on the exam.
This is ludicrous, and there is a lot to unpack. First, let me be clear that if a test has an average score of 27%, then it means there is a problem with the test, not the students. The goal of teaching should be learning, so it follows that the tests and other assessments we design should be opportunities to reflect that learning. If the test is so hard that most students cannot pass it, then a) they have not been taught very well, b) the test is poorly designed, or c) all of the above. Instructors who design exams like this often do so to create an illusion of rigor. But exams that are so difficult that many students fail them are not rigorous. They are flawed, and this is a really important distinction to keep in mind.
More importantly, though, many who implement norm-based grading do so because they believe, or have been seduced by, claims that intelligence falls in a normal distribution in populations. This faulty, pseudoscientific framework essentially suggests that some people will always have average intelligence and ability to learn, some will have a level of intelligence classified as being above average, and some will have a level of intelligence classified as being below average. This kind of normal distribution is commonly illustrated as a “bell curve.” We call it a bell curve because, well, it looks like a bell when you see it on a graph, but also because of an infamous book by the same name published in 1994 by Richard J. Hernnstein and Charles Murray. The book, its findings about intelligence and race, and its methods are rooted in racism and deficient, partially formed beliefs about the ways in which our biology determines our destiny.
I do not have enough space in this single post to outline all the ways The Bell Curve is wrong, but noted scientist and author Stephen J. Gould provides a decimating critique of the book’s arguments in his revised and expanded edition of The Mismeasure of Man, which he wrote originally to put to rest any theories that have biological determinism as their foundation. Suffice to say that a grading model using a bell curve as its statistical and philosophical foundation is inequitable from the outset because it takes as its starting point the faulty notion that some students by virtue of their biology will always fall below average on the curve.
The other type of grading curve commonly in use is a simpler one. In this model, which is typically applied (again) if students on the whole do rather poorly, an instructor looks at all of the scores on a test or a paper and moves the highest grade—whether it be an 88 or a 72 or a 64—up to 100 and then all of the other grades are bumped up proportionately in relation to their distance from that initial grade. So if the highest grade was, in fact, an 88, then an original grade of 86 would be moved up to a 98 after the curve was implemented.
A curve like this, which I have experienced myself as a student, is often implemented under the guise of helping students. The problem, though, is that students still know that they did not do well on the assessment and that they still need to learn more. A curved, inflated grade does not change their inward feelings of failure, because they may believe they are being unjustly rewarded, nor does it help them develop the conceptual understanding they will need for subsequent courses where their grades are now likely to suffer. This only serves to extend opportunity gaps that students may already be experiencing rather than helping them to succeed.
A grading curve is perhaps the most egregious example of a traditional practice that furthers inequities in our schools, but it is far from the only one. As we move toward going gradeless, we must use evidence like the kind I have presented here to demonstrate that a more progressive approach to assessment and evaluation is not just a good idea or the next new edu-fad. Indeed, when you look at the larger issues with injustice and inequality in our educational systems today, going gradeless is becoming a moral imperative.
Josh Eyler is Director of the Center for Excellence in Teaching & Learning and Director of the Think Forward Quality Enhancement Plan at the University of Mississippi, where he is also Clinical Assistant Professor of Teacher Education. Eyler is the author of the book How Humans Learn: The Science and Stories behind Effective College Teaching, which Book Authority named one of the “100 Best Education Books of All Time.” His forthcoming book, Scarlet Letters: How Grades are Harming Children and Young Adults, and What We Can Do about It (Johns Hopkins University Press, 2024), is about one of the most urgent issues in education today—grading and alternative assessment. You can email him at jreyler@olemiss.edu, or find him on Twitter at @joshua_r_eyler, on Blue Sky @josheyler.bsky.social, and on LinkedIn at Joshua Eyler.