By Trevor Hedberg
Anyone who has taught even a single philosophy course has probably had to deal with borderline cases at the end of the term. A borderline case occurs when a student is right on the edge of receiving a higher letter grade in the course. Perhaps a 90% in the course is required for an A-, and the student has finished the course with an 89.8%. These cases seem common. In my own experience as an instructor, roughly 20-25% of students find themselves in this situation when the semester concludes. But here’s something that’s a little puzzling: beyond my own courses, very few syllabi that I have seen explicitly address how borderline cases are handled. In fact, I cannot recall seeing one course syllabus that addressed this issue during my entire undergraduate career.
From both first-hand and second-hand experience, I gather that what usually happens is the professor evaluates these cases on an ad hoc basis by looking at (1) the student’s perceived level of effort in the course, (2) the student’s body of work as a whole, or some combination of these factors. If the professor is appealing to (1) and the student appeared to put forth a lot of effort in the course, then that student gets bumped up to the higher grade; otherwise, the student gets the lower letter grade. If the professor is appealing to (2), then they examine the major assessments and see if the student’s body of work is indicative of a higher level of course mastery than the raw numbers indicate. Typically, if the student demonstrated improvement over the term, then they will get bumped up to the high letter grade; otherwise, they will receive the lower one.
The problem is that neither (1) or (2) will produce consistent or fair verdicts about these cases. Student effort is absurdly difficult to measure. Because we see our students for such a small portion of each week, we have little insight into how many hours they are putting into our course. Moreover, effort cannot be equated to quantity of in-class participation: students who frequently participate in class with half-baked ideas are not obviously putting in more effort than students who contribute rarely but with insightful thoughts and observations. Poor contributions to discussion may be evidence of a failure to complete the assigned reading or a failure to spend time reflecting on the material. But they also might just be evidence that the student is struggling to grasp the course concepts despite her best efforts. How are we to know how much effort the student is really putting into the course?
(In the prior paragraph, I'm conceding for the sake of argument that it is permissible to factor effort into final grades for the term, but I should not that this position is controversial. See, for example, section 3 of Daryl Close’s “Fair Grades”.)
Assessing a student’s body of work is similarly ineffective. You already know that the student’s body of work produced an unclear result: that’s how the borderline case arose in the first place. If later assignments in the course are supposed to be weightier in determining a student's level of mastery, then that fact should be reflected in the actual weight of the assignments (i.e., their percentage of the final course grade) and affect all students rather than being applied selectively to certain students at the end of the term.
Moreover, higher grades on later exams and papers will not always be a reflection of improvement in knowledge of course material. A student may, for example, have bombed an early test because she had three other exams that week and was therefore unable to adequately prep for all of them. In such a case, her higher grades later in the term will be a reflection of having fewer commitments during the weeks those assessments are due rather than any meaningful improvement in her knowledge of course content. (Cases like this are quite common because college instructors have a tendency to schedule their assessments at similar points in the academic calendar.)
So what should we do about borderline cases? Well, here’s one easy solution: just don’t bump anyone up. Full stop. Such a policy treats all students equally and can be added to your syllabus with a single sentence. It also saves you work because you don’t have to puzzle over borderline cases at the end of the term. Sounds great, doesn’t it?
There’s just one problem: when you get that student who made an 89.8%, are you really that confident that a B+ is the appropriate grade? We all know that the difference between, say, an 84% on a paper and an 87% on a paper is difficult to judge. I feel that almost any fine-grained judgment I make on a paper or essay has a margin of error of +/-2%. As an illustration, if I gave a paper an 86%, I suspect that one could reasonably make a case that it should receive an 84%, and that one could reasonably make a case that it should receive an 88%. I’m skeptical of anyone who claims that they can make consistent judgments on philosophy papers or essay answers to the exact percentile, particularly since we often grade a lot of these and under significant time pressure.
What this means is that you cannot be fully confident of your judgments in borderline cases. You may have been overly harsh on the student on a previous assessment or simply made an error when grading under conditions of fatigue or stress. And a small error or two like this could be what makes the difference in a student’s course grade. If we want students’ grades to accurately reflect their mastery of the course material, then this is a problem.
I’m still actively considering the best way to adjudicate borderline cases, but I have tentatively settled on using a system of Borderline Points to resolve them. Here’s an except from a recent syllabus that explains the system:
There is no extra credit in this course, but there are opportunities to attain Borderline Points. These points will be the determining factor in Borderline Cases (i.e., cases in which a student is within 5 points of reaching a higher grade in the course). If a student has accumulated (for instance) 936 total points on all the course assessments, then I will look at their total number of borderline points and add it to the point total. Suppose that after adding the borderline points, the student has 941 total points; in that case, the student would receive an A for the course (rather than an A-). If, on the other hand, the student did not have enough borderline points earned to reach 940 total points, then the student would receive an A-. Details on acquiring these points can be found on Blackboard in the handout “Borderline Point Challenges.”
My courses have 1000 total points possible, and these borderline point challenges allow students to accumulate a few points that will be added to their grade if they find themselves on the edge of receiving a higher letter grade. These challenges are usually short writing assignments that are graded on adequate completion. Each is worth 1-2 points.
One of the major advantages of this system is that the students control their own fates in borderline cases. I don’t have to make difficult judgments or worry about being biased in favor of students I have grown to like over the term. I just look at their borderline point totals. Suppose a student accumulated 896 points: if she got 4 borderline points or more, then she gets an A-; if she got 3 borderline points or fewer, then she stays at a B+. End of story.
It may not be an optimal system, but it’s the best one that I’ve devised thus far. I’m curious, however, what others think about this issue. How do you all assess borderline cases at the end of the term? And do you include that information in your course syllabus?