As end-of-semester papers draw near I find myself facing the age-old question of how to grade them in a way that is consistent and clearly communicable to students. Because I generally give students very detailed guidelines for writing assignments it is relatively easy to assign points to those details and grade papers based on them. This helps me stay consistent from one paper to the next and also allows me to explain grades to students (beyond this, professors who could offer no explanations beyond “you got an A- because you wrote an A- paper” always frustrated me as a student).
In recent semesters, however, I’ve found that grading based on a detailed rubric leads to a few problems. These include a focus on details that can prevent a view of the whole and, as a result, the potential to give a poorly-written paper that has all of the necessary components a higher grade than it would receive using other methods. As a result, for some assignments I’ve moved to a letter-grade system in which I meeting the basic requirements is equal to a B and students need to do particularly insightful or well-written work to earn higher grades.
So far, I’ve been using rubrics for longer papers and letter grades for shorter papers. Eventually, I would like to merge these approaches in a way that goes beyond a “quality of writing and insight” portion of the rubric.