Sunday, October 17, 2010

Bimodal

Am halfway through grading the Brit Lit midterms, a long and involved process which entails spreading all of the exams out on the floor from best to worst. I was getting depressed, and also a little lightheaded from the extra beer I allow myself on such occasions, so I've decided to knock off for the night.

Scores so far (out of 50, with 45 as the real maximum expected grade): 48, 44, 44, 42, 39, 36, 34, 32.5, 31, 30, 29, 27.

This is a typical distribution for a gen ed literature course at Misnomer U.: two distinct peaks, with only a handful of exams falling in long valley in the middle. The bell curve is upside down. This doesn't make determining grades particularly difficult -- it is obvious where the A and C spikes are, and the fact that there is a long and sparsely populated B-range just means people are less likely to argue about their grades. But it does make teaching the class damned hard. It's kind of nice to have a visual representation of why it's hard, a reminder that it isn't just me. (Maybe I will leave the exams on the floor for the rest of the semester.)

What do y'all's grading distributions look like? Bell curve or bloodbath?

5 comments:

Dame Eleanor Hull said...

I'm halfway through a set of papers where the average mark (so far) is a little under 7 out of 10 points. OTOH, the average is dragged down by one that got 1.5; leaving that one aside, the average is more like 7.2, which makes me feel better. But I haven't done a distribution, just the average.

dance said...

I also spread my papers out on the floor from best to worst, or grouped in piles, anyhow.c

Flavia said...

My exam grades (although not my paper grades) look pretty much like yours--my intention is to allow students who are smart, skilled, and doing the work, but whose writing skills perhaps aren't yet up to snuff, to show their stuff here.

So typically, in a class of 25-30, I'll have at least 10 students whose grades range from B+ to A+, with the biggest cluster around the 90% mark. . . and then I get 8-10 students with Ds or Fs, including students with unbelievably, almost impossibly low scores.

I guess this makes sense, on the principle that they either know it (or get it), or they don't, but it's still disturbing.

(My paper grades are more bell curve-y, with the spike probably around 83%)

Sisyphus said...

Oh thank god. I'm freaking out about the same thing over in my latest post --- with the added problem of angry students sending nasty emails but who turned in almost-blank midterms.

How the hell am I supposed to grade an answer you didn't write anything for? Is this some kind of messed-up magical thinking? Fargh.

hck said...

I call that sort of an inverted bell curve (which is the normal one for most of "my" exams etc. too) a "Homburg curve" (from the shape of the crown of a Homburg hat: |^v^| http://is.gd/g7WJW ).

In my case it also includes papers: there is the group of bright students who work, there is the group of students who want to survive with minimal work and not that much talent, and there are only a few who are brilliant+lazy or industrious+not that talented (for philosophy/renaissance studies).