There’s a short article in a recent issue of College Teaching on a topic that I don’t see written about very often. I suspect we don’t think about it as often as we should either. It’s the matter of bias in grading.
What sort of “extraneous influences” might consciously or unconsciously influence the grades we assign? The author lists 10: 1) very nice students; 2) ones that have done well in other of our classes; 3) hardworking and conscientious students; 4) those obviously interested in the course; 5) others who are eager (more like desperate) for a high grade; 6) students who need grades for a particular reason, like financial aid, entrance to a major, etc.; 7) students of a particular gender or ethnic background; 8) attractive students; 9) ones who look, act and regularly sound intelligent; or 10) students who are somehow connected to us, say the child of a colleague or friend.
In most cases the influences of characteristics like these are subtle and not easy to admit, even to ourselves. But bias in grading can be prevented—most easily and effectively by grading work without knowing who completed it. Anonymity can be accomplished by assigning number codes to students and having a colleague or staff person record grades. Less cumbersome, students can put names on cover sheets only and instructors can then fold back the cover sheets back before beginning to review and the grade papers.
Bias in grading is also preventable or can be minimized if the instructor uses detailed grading criteria like a rubric. This approach works best when the rubric is shared with students. Although total objectivity is an ideal, teachers do have the responsibility to recognize and work to avoid bias and other subjective criteria.
Reference: Malouff, J. (2008). Grading in bias. College Teaching, 56 (3), 191-192.