Grade Distribution

What is the appropriate distribution of grades in a university class? Is there a standard that is readily accepted? These kinds of questions often motivate grade discussions amongst university instructors. 

Some general points of consensus: larger classes more often fit normal distributions (i.e. grade curves) than do smaller classes. Larger lower-level classes tend to have a lower average than smaller upper-level classes. There should be some kind of consistency across equivalent classes so that students don’t think grades are totally subjective and potentially unfairly distributed. 

Up until a few years ago the Faculty of Arts had a grading guideline that recommended a typical grade distribution in which no more than 75% grades could be in the A and B range; that A grades should be between 5 and 25% of the total grades distributed, and; that no more than 20% of the grades should be a fail. This guideline was maintained (as a policy recommendation) in the anthropology department up until the present. Some faculty feel constrained by the above guidelines while other consider it too lenient. 

It is helpful to ask what grades are supposed to measure. If they are a measure of relative rank that leads us in one direction. If they are a measure of quality/quantity of learning that might take us in a different direction. It would seem that instructors generally consider them some measure of learning (but that’s where consensus starts to break down). Students often seem to consider that grades are correlated to the amount of time they invest in completing an assignment or preparing for an exam. What we do know is how grades are used: to allocate access to scarce resources like awards and admission to graduate and professional programs. 

There is a big literature on grading. There is a lot of talk about finding the best way to grade authentically. To give formative feedback to learners. There’s even a critical literature focused on “ungrading.” Both standard beliefs and radical critiques often ignore the reality that capitalist society is graded all the time. Of course we can be compassionate, we can care about our students (and trust they extend that same care to us), but that doesn’t belie the reality of our ranked, hierarchical, scarce resource driven society. Grades are just one more part of that bigger system. 

So. Why not resist the power and refuse to grade? Or, if not total refusal, change the means of grading more in line with so-called ungrading (which still results in a final grade). Why not indeed. Radical ungrading would be best, but for most of us the reality is that ungrading ends up being more work (arguably better feedback for students) and the same normative grading as before. So ungrading becomes another plank in the ubiquitous student-centric learning model in which grades are seen as rewards for showing up and trying. 

There is another augment used against grading – the social economic discrimination thesis. Here the augment is that grading norms white social norms (more accurate would be the claim they replicate specific class values, not racial values). Here students achieve differentially (and their grades are thought to reflect this) according to their demographic marginalization from whitestream middle class society. They are kneecapped before they start the course. Here an argument is made to use grades to counter inequities of history. Ignore obvious errors in composition, for example, and focus on ‘content’ mastery instead. 

Then there is the weakest (in my opinion) argument against grading: “the other guys are skewing their grade distributions higher than we are and thus our students lose out. This isn’t so much a critique of grading as an accommodation to grade inflation. 

I personally hate grading, and this is something that I think many colleagues would agree with. It can be tiresome, anxiety producing, and leave one feeling a deep sense of futility. As a consequence, I have shifted the types of assignments I give my students (more weight on completion graded assignments – if you complete at standard defined, you get full points- this does tend to skew class averages positively). I have also tried to lay out the grading criteria clearly so that there is no ambiguity in what I am grading. I try to use strength-based assessments (I identity where an essay, for example could be improved as opposed to explaining everything that is wrong with it). All this is still grading and at the end of the day I am responsible for submitting a number on a grade sheet. 

At the same time that I hate grading I also value the opportunity to facilitate learning, revision, and transformation through discussion and collaborative reading and reviewing of student work. Some quick witted observer will probably realize that’s really marking without numbers. Ideally, in my field of work, the formative part is guiding learners toward improvements, marking (assigning a value to that process) is simply one small piece. But, it’s the piece we end up focussing on. 

At the end of the day grade distributions have no natural level. They are indeed arbitrary. But because we use them for allocating access to scarce resources we have an obligation and a responsibility to be consistent with how we distribute them. Ideally, we should be consistent across a university, not just a department. My personal preference is toward criteria referenced (or standards-based) grading were there is some clear standard that we expect learners to obtain. It would be based on a combination of knowledge growth within the individual and in comparison to how that individual learned relative to a pre-set criteria. We aren’t in that world though and thus must be content with some variation of normative ranking.