Final course grading is a sticky subject with both faculty and students. I have heard plenty of instructors (my own as a student and my colleagues) proclaim that their letter grade cutoffs, so clearly stated in syllabus are infallibly written in stone such that a student 0.001 points below the cut is still below the cut. As a student, this did not seem fair, and as a graduate student I realized that this was likely a simple time-saving device for final grades. I certainly relied on this mechanism, particularly when I had my own finals to complete and grant proposals past due. But at the Ellbogen Center for Teaching and Learning, two pedagogy faculty asked a simple question: What do final grades truly represent?
We often think of final grades as some representation of what a student "earns" through content knowledge and hard work, which leads to the accumulation of points. Once these points reach a particular threshold, we assert that the student has adequate knowledge to claim they know (something) about the topic of the course and pass the class. But therefore does the accumulation of more points allow us to claim that the student knows more content or that we are more confident that they know enough? I prescribe to the latter.
The majority of my students are pre-professional, ultimately planning on becoming athletic trainer, physical therapists, doctors, etc. Some of my students inspire confidence through their hard work ethic, attention to detail, and mastery of both content knowledge and application. These inspirational students are often the ones that I would trust as my therapist or doctor. As such, my courses are designed from the ground-up to provide all students the opportunity to convey these components necessary to inspire that confidence. Grading ultimately reflects that.
But in the end, when I need to ultimately decide if a student is assigned an A- or a B+, this places me, and any instructor, in a quandary. If the classic point cutoff for an A- is a 90% and a hypothetical student earned 89.91%, does this mean that I have significantly less confidence in their ability compared to a student that earned only 0.09 more percentage points? More concerningly, are my assignments and grading strategies sensitive enough to illuminate such a difference? Interestingly, I have polled students in my classes across multiple years, and get divergent answers!
Statistically, the answer to these questions is no. Each assessment carries some level of confidence about its accuracy; the lack of perfect confidence accumulates and leads to a need to carefully re-assess final grades at the end of a course. If the effective confidence interval (which may not be possible to calculate) includes the next higher letter grade cutoff, the student may likely be performing at that higher level. More concerningly, as many of my courses function as gatekeepers to subsequent courses and programs in higher education, not identifying this phenomena may eliminate the student's ability to demonstrate that they do have the reliable capability to operate at the higher confidence level!