I’ve been reflecting on the concept of grade drift recently. My brain usually turns to grade drifting at the end of each semester as I evaluate how objectively I have scored the students. If you’ve never taught before, what I call “grade drift” is the tendency for an instructor to allow their grading standards to drift as they grade an assignment. I’m not sure how prevalent this is for assessments that are concerned with exact answers, such as math, science, engineering, programming, etc. But in assessments that deal with qualitative or creative responses, it’s common for you as an instructor to change your evaluation thresholds as you make your way through the pile of papers.
There are a few reasons why this happens. In my first courses, my grades would drift if I didn’t have a sufficiently robust marking guide to help me narrow down the kinds of answers I was looking for in a student’s response. Now that I’ve taught the course a few times, I know generally what I’m looking for and can go through the assignments with a checklist of items. This form of grade drift is usually related to lacking experience in teaching and grading, especially if your course delivery isn’t tied to your course learning outcomes.
However, there is another reason why your grades might drift. On the one hand, you want to approach your evaluations as if they are objective – that there are clear right and wrong answers. After all, your grading should be defensible, and one way to defend the grade you grant is by pointing to an objective standard. However, on the other hand, you want your evaluations of your student’s work to allow for imagination, creativity, and novel connections between ideas. You also want your evaluations to acknowledge that your classes have unique compositions of people, with their own experiences and their own progressions through your course. Just because you teach the same material each semester doesn’t mean your students progress through it at the same rate. Different students will take the material differently, which means the average response you read through will be different for each cohort of students.
This means that some groups will be “better” than others in how they perform. You can choose to penalize the students for not meeting an abstract standard that you’ve set, or you can meet the students where they are and make it your goal in the course to improve their performance as you go.
I know that some instructors will take the first approach. They will believe their teaching, evaluations, and courses are reflective of an external, eternal, objective standard; they don’t think they are being arbitrary. I’m willing to bet that these are also instructors who believe that a 70% in their course is “like a 50% in other courses” (a literal thing I’ve heard said by a peer). I think this approach is wrong and it reflects a misunderstanding of what it means to pass a course. When a student passes a course, it means they have met the minimum requirements of the course learning objectives. Anything above that should reflect varying degrees of competence and mastery. If your course is designed so that a student is only “just” meeting the standards with anything above a 70%, you haven’t calibrated the course or your expectations very well. At an undergraduate level, your job is to elevate your students, and improve their abilities while meeting the program learning outcome objectives.
I find myself on the other side of this issue. I understand what the goals are of the course I teach, and my aim is less concerned with ensuring my students can adequately explain all the theory they are exposed to in the course. Instead, my goals is to make my students better thinkers and writers from where they enter my course to their exit. If they can explain Aristotle’s ideas of tragedy by the end, great! But I won’t lose sleep if they can’t. My course is a general elective, and that means I’m supposed to round them out as students and people. I care less about them absorbing my esoteric knowledge and instead I care more about them learning how to think, reason, and communicate their ideas clearly.
And that’s why sometimes my grades drift as I score the student’s rubrics. I start off with some ideas of where they could be, then I calibrate my grading to meet them at their level. It’s a little more work for me, but I think it better captures the student’s performance.
As indicated above, there are a few strategies for overcoming this, such as having a clear rubric and clear notes on what you are looking for. Also, marking all the students on one question, one at a time, rather than marking an entire submission will make it easy to compare student fairly.
While this is an unfortunate thing (the students like to think they are being objectively evaluated) this will happen as you come to understand the natural curve of your class. You can either grade your students to one standard, or you can reflexively respond to your class’s own knowledge and aptitudes. If your course objectives are clear, then you can feel free to adapt your grading to match the student’s progress.