Trouble with Testing
Moebius Stripper has a good post, and good comments, over in her weblog about shortcomings of traditional testing, particularly within mathematics. Given an entirely blank slate to work from, I think everyone would agree that testing should, in fact, be a trigger as to whether one is ready to move on to the next set of content. But with multiple students, limited time per week, and a set-length semester, simply changing the testing scheme only tackles part of the problem. When the entire end evaluation of a student is a single letter, there is an awful lot of meaning to be crammed into a single dimension. Two students might get a C, one by doing very well in the first half of the course and then not understanding any of the second half, and another by maintaining a minimal level of understanding across the entire course. Which one is more equipped to move on to the next course in the series? (Answer: it probably depends on the content of the course and its follow-up....)