Grading Schemes and Scales

Our grading is rooted in data. We take data points many times in a given course, typically multiple times during one week. The goal is to even out any hickups and to be able to properly assess the semester-long performance of a student. Learning is a marathon, not a one-time peak performance. To that end, each course defines a grading model which aggregates multiple components. The primitives of these components are individual data points (“mini-grades”) for a particular performance. As we might be communicating those to students during the course, please find below our grading schemes and scales.


The grading scheme “[y/n]” (sometimes just “yn”) is binary: The student passed or not. The domain is either “y” or “n”, or “1” or “0”, with “0” being omitted at times (the empty field in a spreadsheet defaults to 0).


The grading scheme “[0|1|2|3]” is an ordinal scale of measuring performance. It is typically used to assess contribution in class. Here, the value “0” means “student did not show up”, “1” means “student showed up but did not contribute or contributed only minimally”, “2” means “student participated in line with expectations”, and “3” means “student participated and contributed something of significance, e.g. an original thought”. It is not easy to get a “3”.


The grading scheme “[0..10]” is an interval scale with two-digit precision, i.e. it ranges from 0.00 to 10.00. It is calibrated to range from 0 (nothing done/provided) through 5.00 (exactly average performance) all the way to 10 (perfect performance).


The grading scheme “[ABCDE]” is an interval scale of “A” (best) to “E” (worst). It is calibrated with the following meaning: “A” means “excellent”, “B” means “good” (above average), “C” means “average” (as expected), “D” means “poor” (below average) and “E” means “abysmal” (basically failed). While similar to the old German school grading system, this scheme does have an interval scale, so a “failed” in one performance item does not imply failure of the overall course.

Leave a Reply