Some weeks ago, I blogged about a new way of curving exam grades which I’m trying in my principles course. Monday I handed back the second exam, which is the most difficult of the year, primarily because it covers the most analytical material. The average grade was 63%, which is normal for this exam. This week the students have been coming by my office for the ‘do-over.’ Their charge is to explain orally why the correct answer on the problems they got wrong is correct. This shouldn’t be that difficult, since they have several days to re-view the problems and re-think the answer. For each problem which they correctly explain, I give them half credit.
I’ve found this to be an interesting (and different) sort of assessment. Not only do students have a chance to explain their reasoning (which they don’t on the multiple-choice exam), but I give them more than one chance by prompting for more information. When students still don’t understand the reasoning behind the correct answer, it is very clear. Whether that is due to lack of preparation, or a true lack of understanding I’m not sure, but I have no trouble denying them credit.
I think I like this was of curving exam grades because it reinforces that exams should be opportunity for learning as well as assessment. Additionally, I think it has the incentives right, unlike my past practice of using a traditional curve. Here, they get more credit if they get the problem right the first time, so there’s a disincentive to just blowing off the in-class exam. After all, 62.5% (assuming 25% randomly correct) is not a very good grade for someone able to get all the questions right orally.