Thursday, April 06, 2017

Grading

Before I get into the more interesting matter of curriculum design I want to propose something that I believe would improve grading practices in college. Let us imagine that students take three courses per semester and there are 600 students in a cohort. That means that 1800 grades would be given out to each cohort every semester. My idea, now, is to draw these grades from fixed pool of available As, Bs, Cs, and Ds.

The distribution could vary slightly from school to school. The simplest is 20% As, 30% Bs, 30% Cs and 20% Ds. Teachers would "draw" their grades from this pool in rounds that allow for a maximum of 360 As, 540 Bs, and 540 Cs, with the remaining 360 going to Ds and Fs. If a teacher wants to give a student an A after the pool has run out, then they would have to square them off against the picks of other teachers in that round, before a panel of disinterested [faculty] peers who would look only at the general quality of the submitted work.

This would give teachers the task of designing assignments that demonstrate mastery in a relatively objective way, beyond the "subjective" (or specialist) judgment that the course teacher is able to give. Students, too, would need to make sure that they are not just impressing their teacher, but doing work to an inter-subjectively enforced standard. Teachers at each level should of course reach some agreement about what sorts of assignments to give their students—essays of what length, etc.

Oral examination could also be done this way, with some students being called in to perform before the judges to determine who gets the As and who must settle for Bs. The teachers would, of course, not send their best students into these battles because they would be at risk of getting a B (and that would be rather unfair). They would send those that deserve an A only if they outperform their [student] peers in general academic skills (presentation, coherence, reasoning, etc.).

After some time, I imagine the teachers would get a good sense of each other's standards and would be able to trust each other's judgment. The point is that an "easy A" would, on closer examination, not be so easy. Teachers would in a sense be nominating their students for grades, not merely assigning them unilaterally.

PS I apologize for the breezy style, but it's having a cheering effect on me just to dash these ideas off the top of my head like this. If this system sounds like a lot of work, I think you're overestimating how many grades would actually end up in the "danger zone" between pools and how difficult it would be to decide between them.

No comments: