June 01, 2002
Grade Deflation?

Eugene Volokh, ringleader of The Volokh Conspiracy, has an interesting post on grading on a curve. We can start with the objection he quotes:

Reader Agim Zabeli writes: "If 20% of your students must get an A, another 60% must get a B, and the last 20% must get a C, how the hell does somebody fail your class?? Is it the policy of UCLA Law School that failure is impossible?"

After explaining that he occasionally gives a D or F, Volokh goes on to justify this rather lax-sounding policy as follows:

I've never thought hard about whether this is a good policy, but my tentative sense is that it probably is. Top 20 law schools preselect their students quite well; the median incoming GPA at UCLA is about 3.6, which is to say just below an A-. I suspect that even the worst of our students are actually pretty sharp, and that even the worst of their exams are not abysmal -- just bad compared to the other students.

. . . . my (possibly Pollyannish) guess is that virtually no students at UCLA are really D or F students in any absolute sense; and only when I see a grade that is way off the left side of the curve do I give a grade that low.

I have heard this argument before, not least when I taught at one of the best prep schools in New York City last fall. My department head told me that since all of the students at the Name-Deleted School were very good to superb, grades of B- to A+ were virtually the only ones needed. That seemed reasonable, and it certainly made my job easier: I got very few angry telephone calls from parents. However, it bothers me a bit that the same argument is never applied the other way around. An example will show what I mean.

There are said to be forty universities in the Boston area. I don't know which one is the least selective of these, and don't really want to know, but it must have quite low standards for admission, if it takes students turned down by the other 39. If Harvard and (I suppose) M.I.T. give out very few grades below a B+, shouldn't Dregs University, or Craphole College, or the College of St. Simeon the Holy Fool, or whoever is at the bottom of Boston's scholastic food chain be giving out virtually no As or Bs, on the grounds that even the best student in any given class would be only a fair-to-poor student at most of the other universities in town, and would flunk out of Harvard or M.I.T.? Shouldn't the curve at such a school include, say, 1% As, 2% Bs, and all the rest Cs, Ds, and Fs? Maybe that's how it works, but I doubt it.

As with central banks and the money supply, so with professors and the grade supply, it seems likely that inflation will always be far more popular, and far more prevalent, than deflation.

Posted by Dr. Weevil at June 01, 2002 09:49 PM