Let's say a student takes a set of tests, each with different difficulties of multiple choice questions and a different amount of questions.
If I take the percentage of each test, how can I compare them?
E.g. Can I say based on the difference of percentages between "test A" and "test B", the student did a "difference of percentage" better on "test a"?
When comparing the different percentages, what level of measurement would it be? Ordinal, interval, or rational?
IMHO if the two tests have different degrees of difficulty then it is not meaningful to compare the percent correct on test A with the percent correct on test B. 50% could be a great score on test A and a terrible score on test B, yet the difference is zero.