Many years ago, while I was still a PhD student, a mentor of mine said something very wise about examination. “An exam is an opportunity for the student to demonstrate their knowledge and intelligence to the teacher,” he said, “not for the teacher to expose the student’s stupidity and ignorance.” This is something that both student and teacher do well to remember and, as much as possible, ensure is true. After all, we’re all more stupid and ignorant than we’re intelligent and knowledgeable (on most things, most of the time, we’re not the leading scholar in the field) and an exam is intended to test only what we have most recently (been) taught in a particular discipline. If both teacher and student approach it with the right attitude, it can be one final learning experience in semester that was, hopefully, full of them.
In my last post I described what I believe to be an almost ideal exam at the end of an ordinary one-semester course (e.g., about Hamlet) during which students were asked to read and understand a variety of materials and write a series of essays about them. In my fantasy of the course, they have also been strongly encouraged to compose their paragraphs very deliberately. The learning objectives are therefore quite clear. By the end, they should be able produce an articulate and knowledgeable essay that answers a question that naturally arises from the course content. For the exam, they’re given three hours to write such an essay and are then examined orally on that basis for 20 minutes. In this post I want to say a few words about how to grade the performance.
Ideally, you will have divided your class into groups of eight students to be examined on separate days. (Less ideal conditions can work too.) In the morning they show up as a group and are given a prompt (a different prompt for each group must be designed). After three hours, they hand in their papers. Then, in the afternoon, they’re each given a half-hour slot for an oral exam.
In the first ten minutes (before they enter the room) you read their essay. (They are of course encouraged to do the same.) Since it’s is no longer than 1000 words, you should be able to get through it in that time and even form some thoughts about the competence of the student, both as a writer and as a scholar of the subject. Most importantly, it should give you some ideas for an interesting conversation with the student. If it doesn’t, that’s already a weakness of the student’s performance. Note down whatever reactions of this kind you have, then invite the student in.
Now, talk to the student about the ideas in the paper for twenty minutes. Do not evaluate the paper itself as a piece of writing; don’t talk to them about the paper and the rhetorical choices they made in it. Talk to them about the ideas they expressed. If you think they got something wrong, feel free to explore that with them and give them a chance to correct it. But what you mainly want to do is to poke around looking for the all the things they know and are able to talk intelligently about. If you think they can handle it (which is already promising for their grade) then challenge them directly on points you disagree with (especially if you won’t hold it against them when they stand their ground). Good students should expect to have a spirited (but never heated) debate during this examination. Average students should expect to have a conversation about a topic that both participants are familiar with. Poor students should expect to have a boring conversation about something they know and care little about.
Keep some notes to remind you of anything remarkable that happened during the exam. When the twenty minutes are up, thank them and bid them a good day. Score their perfomance on a scale from 0 to 100. Your criteria are entirely up to you to decide. The point is just that there has to be such a thing as a good, moderate, and bad performance. Now quickly reflect on your first impressions of the paper. Then give the paper a score as well, which may be higher or lower than the oral score. Did the conversation go exactly as the paper got you to expect. Did it go much better? Or did it go worse? The paper score will reflect the discrepancy. Note it down and put the paper, your notes, and the scores, on the “examined” pile. Next student!
When all the exams have been completed (which may take several days), you’ve got a pile of papers, notes from the oral exams, and numerical scores summarizing your impressions. At this point you could simply average the two scores and translate them into letter grades on whatever scale you already use. If you have time, and think it’s a good idea, you can also reread all the papers one more time and give a final grade in the light of the whole experience.
If you’re allowed to do so (in Denmark we are not) I would encourage you to “normalize” the grades — that is, distribute As, Bs, Cs, Ds and Fs on a curve. Let the students know that their challenge for those 3 hours + 20 minutes is to be as articulate and knowledgeable as they perhaps will ever be on the subject. Let them know that any training they do (whether in writing or conversation) will probably benefit them. It’s the students who are best prepared and in best shape for this exam that will get the top grades. All the effort they make, both in the moment and throughout the course, will be rewarded, because it will make them stand out relative to their peers, i.e., the community of those “other knowledgeable people” that their own knowledge is part of.
That’s it. That’s my suggestion for how to grade the ideal exam. Your comments are more than welcome. From the middle of next week, I’m going to take a break from blogging and tweeting until the new year. I will return with a post reflecting on Ian Bogost‘s and Marc Watkin‘s optimism about the, let’s call it, “creative destruction” of the college essay by artificial intelligence. (See also Stephen Marche and my conversation with Charles Knight, Anna Mills, Annette Vee and Marc Watkins.) My approach to writing instruction and examination does, indeed, seem a bit endangered by ongoing technological developments. But I’m not ready to give up yet. Let’s see what 2023 brings!
2 thoughts on “Grading the Moment”
You encourage grading on a curve. One standard counter-argument is that this discourages students from helping each other learn. I’d appreciate your thoughts on this point.
Also, consider this paraphrased quote (I don’t have the source handy): statisticians think that a normal distribution is a law of science, and scientists think it is a law of statistics”. Why should grades follow a particular distribution?
I often see a double-peaked distribution in students’ course average scores. How would you recommend assigning grades in such a situation??
These are good questions. In team sports, athletes compete for positions in the starting lineup while maintaining an objective interest in seeing each other succeed to the best of their abilities. On my approach, students are encouraged to share their work with each other because they are graded on how well they communicate with their peers (not their teachers). The teachers (and graders) have an insight into the level of the cohort because they read, not just the paper they are grading at the moment, but all the other papers too.
As for the naturalness of normal distributions, I agree that it can seem a bit arbitrary. It should be emphasized to the students that it’s an order that is being imposed on their studies, not one that reflects any natural fact. Graduating at “the top of the class” does not mean that you are in any other respect “better” than your peers (or anyone else); it simply means that you succeeded consistently over a range of courses and semesters. It’s a signal that some people find meaningful and others less so.
I don’t know a lot about what is normally taken to explain double-peaked distributions (thanks for the spur to look into it; I’ll report back if I find something interesting). My gut reaction is that I would try to avoid this outcome, either in the design of the assignments or in the specifications of prerequisites for the class. That is, I would want my class to be “challenging” in precisely the sense that would result in most of the students who were “qualified” to take it getting Bs and Cs, and much fewer getting As, Ds, and Fs.