I obviously royally screwed up the formatting on my last reply!
Anyway, it's fairly obvious why using a true Bell curve wouldn't work very well in academia - a student could potentially get 98% of the material correct and still fail.
My wife is a college science professor who worked very hard for her PhD. When she grades, she will frequently throw out a question if most or all of the students missed it, because it points to a deficiency on her end of the equation, not on the students'.
Anyone can write a test in which the students are guaranteed to pass. Equally, anyone can write a test in which the students are guaranteed to fail. When 90% of the students in a freshman level general Intro to Biology class are failing, it would appear that the teacher is either failing to teach or intentionally writing the tests so that students will fail.
“When she grades, she will frequently throw out a question if most or all of the students missed it, because it points to a deficiency on her end of the equation, not on the students’.”
That’s what I do too — if 70% get it wrong, the question goes. But there have been times that it just means that 70% of the students didn’t read the text, because it was something in the text but not highlighted in a lecture.