It is time to abolish grading on a curve
I taught my first big class at the University of Alberta in the winter semester of 2008. After a long and arduous semester, it was time to convert the cumulative marks to grades. I had been provided a smooth ideal curve that the distributions of the grades within the class are expected to follow — a grading practice termed: “Norm-referenced”. I decided on the grades by finding natural gaps in the marks to differentiate between the groups of students, which precipitated a curve with two pointy peeks that stretched higher than the “ideal” curve and precipitously dropped forming a gorge that swallowed any hope that my curve could ever look normal. The “ideal” curve recommended at least one student getting an “F” and a few getting “D-“; none of my students had failed or done so poorly to warrant a D-. “This is ridiculous, am I supposed to arbitrarily fail a student?” I thought to myself. It was then that I abandoned the curve and never looked back.
Twelve years later, as the Associate Chair of Undergraduate Studies in Civil and Environmental Engineering, I am responsible for approving the undergraduate course grades in our department. The grades provided by the instructor of one of these courses perfectly and suspiciously followed the “ideal” curve. When I inquired, the instructor was utterly perplexed; they had assigned the grades to match the curve, isn’t this how grades are assigned? They did, however, acknowledge that grades cutoff had to be arbitrary and rather extraneous to how students actually performed in relation to the objectives of the course. The incident left me with the nagging questions: What about other instructors in the department and the university and what is the institutional policy on curving the grades? What about the students? Does “norm-referenced” grading of the students help them in their quest for knowledge or is it time to abolish it?
Whether you should “curve” or not depends on who you talk to; some instructors religiously follow the curve, others use it as a guide, and a small group, to which I belong, act as if it doesn’t exist. This varied level of adoption is a natural consequence of a historically confounding policy. Robert Luth, a provost fellow at the University of Alberta, reported that in 2010, the University of Alberta adopted a “flexible” policy while the General Faculty Council (GFC) policy was written in a manner suggesting that “norm-referenced” grading is recommended to ensure “consistency” across fields [1]. Based on campus wide consultation, Robert noticed a similar varied level of adoption citing an example of an instructor being “surprised to discover that GFC policy does not force the use of the curve”[1]. While the report fell short of recommending abandonment of the curve, the University of Alberta online news publication, Folio, reported in 2012 that the University revised its policies to proclaim that the University of Alberta does not and did not ever have a “bell curve”[2]. The policy[3] which lives till today declares:
“Grades in any course, examination or other academic assessment shall not be mandated on the basis of a curve or historic distribution of student grades. The distribution of grades shall not be predetermined by any system of quotas that requires a certain number or percentage of grades at a particular level. However, a faculty, department or unit may develop and provide guidelines to instructors setting out a reasonable distribution of grades in the faculty or department.”
When it comes to Engineering, the 2017 version of the Academic and Engineering Course Policies for instructors[3] is silent about any adopted “curves”.
One of the major drawbacks of curving the students is that the process arbitrarily limits “the number of students that can excel” [4] which is in stark contrast to the ultimate aspiration of a university education: to shrink the gap of knowledge among the students. Professor Adam Grant at the Wharton Business School of the University of Pennsylvania contends that those who adopt grading on a curve follow the view that “the world is a zero-sum game”[4]; there is a limited number of good grades that can be imparted. However, such process is ultimately detrimental to both the morale and the well-being of the students. It creates an artificial hyper-competitive environment requiring the students to compete against each other for the scarcely available “good” grades, and eventually leaving behind a few students who do not believe that they can compete with the high achievers within a class. The same view is eloquently articulated by Professor Timothy Burke, a professor in the department of History at Swarthmore College. In his article: “Grading on the curve is always a bad idea”, Timothy [5] argues that the curve acts as a mechanism to decide that even “under the best circumstances” some will fail, and most are average. The practice, he argues, shifts the focus from the desired outcome of a class to the actual curve itself. In fact, the only function the curve serves is obfuscating an instructor’s inability to properly designate achievable course outcomes with concomitant appropriate and relevant assessments.
So, why do many professors still follow this practice? Eugene Volokh who teaches at UCLA School of Law argues that grading on a curve is needed since an instructor is not able to assess the level of difficulty of their exam [6]. In other words, the curve guarantees the same outcome independent of the level of difficulty of the assessment. The curve, he argues, is also needed to guarantee the students’ outcome for new professors and unusually harsh professors [6]. This view, in my opinion, is designed to maintain an ostensibly appropriate standard of education while purposefully disregarding any abnormalities or aberrations by shrouding them behind a curve. This needs to stop. Instead of using a curve to maintain an artificial façade of appropriate standards, institutions should invest in educating their new and seasoned instructors on assessment methods that focus on student’s learning rather than arbitrary sorting.
Some proponents of following “the curve” cite “grade inflation” as the alternate repercussion; more students are going to get higher grades. I have been teaching at least three relatively large classes per year since joining the University of 2007 and more often than not, I have been assessing the students solely based on the expectations that are properly communicated beforehand with the grades divided among many components including assignments, quizzes, multiple midterms and a final exam. This practice combined with grading that is based on predefined overall marks relieves the anxiety and stresses associated with norm-referenced assessments. Ultimately, I have found that in all my undergraduate classes, students end up “curving” themselves. In other words, anxiety and stress are relieved without grade inflation.
It is 2020 and the literature on modern assessment and grading is ubiquitous with the majority pointing out the deficiencies of grading on a curve. However, approving the grades in our department made me realize that the “curve” is alive and well, seeping as a silent, pretentiously salient, insidious threat to students learning and well-being while obscuring any indications of assessment’s inadequacy. It is time to focus on educating the students rather than arbitrarily aligning them on an artificial curve. It is time to abolish the mere existence of the grading curve.
References:
[2] https://www.folio.ca/university-clarifies-grading-curve-policy/
[5] https://blogs.swarthmore.edu/burke/blog/2012/08/23/grading-on-the-curve-is-always-a-bad-idea/
[6] https://www.washingtonpost.com/news/volokh-conspiracy/wp/2015/02/09/in-praise-of-grading-on-a-curve/
