Grading

From EvaluationWiki

One of the fundamental evaluative processes, to be sharply distinguished from ranking. Grading requires the assignment of evaluands to one of a set of categories that are ranked by merit (or worth or significance, etc.) and labeled by reference to a corresponding set of ranked general evaluative terms, e.g., Excellent, Good, Satisfactory, Weak, Unacceptable, often abbreviated as A, B, C, D and F (or E). While one can rank evaluands that are pulled from different categories, not more than one from each, since the categories themselves are ranked, one cannot rank evaluands from the same category without applying some further evaluative operation. So one cannot in general infer the rank order of any two items pulled out of a graded group: they must have come from different grades. Note that scoring is even less a guide to grading than it is to ranking, since if the test is easy enough, a score in the 90% region may only indicate weak competence; and if hard enough, a score of 20% may be correctly graded excellent. Grading is the most basic example of criterion-referenced evaluation.