If the most raw scores for a test are an F (~40% of the class failed) and the second highest number is an A (~30% aced it) and very few people received a B, C, D - what does that say about the difficulty of a test (if anything?)
The professor is saying the test was too easy - I’m trying to make sense of it.
I’d say the curve represent two populations in the class first the students aren’t taking the class seriously/didn’t study/gave up (the F’s) and second, the rest of the class for which the test was too easy - 50% of that population got A’s
It is either deliberately a weed out class or the instructor is incompetent.
Not inverted, but bimodal distribution. Shows that there was something significantly different between to subsets of the population. It could be the above reasons or it could also be that the test was very difficult and some of the A students cheated.
I don’t think there was cheating, it was essay based with randomized multiple choice sections.
Bimodal distributions are actually rather ordinary results for exams (especially but not only in quite technical and detailed subjects), particularly in cases where the student population includes both a large number of students with a solid preparation for academic work and a large number with poor preparation.
I teach at an effectively open-access university some hundreds of miles away from all but one other baccalaureate-granting institution—it’s simply something you learn to expect, if the exam is at all rigorous.
@dfbdfb - The course is an upper division Political Science class with mostly juniors and seniors. The subject matter is not at all straight forward - it’s very much just analyzing and comparing different arguments about international security, economics, etc. But I also see these results in my Calculus class - my professor explains it by saying that half the class knows what they’re doing really well and half really has no idea.
I also teach at an open-entry school, and many of us experience bimodal distributions. The real problem is not the 40% who fail but the 30% who got As. If the instructor makes the test easier, 40% will still probably fail, and those at the top will learn less. If the instructor makes the test a bit more difficult, 40% will still fail, but grades at the top will fill out the B-C-D spots more naturally; those who excel will get a better education.
This is where I’m confused:
On RMP, multiple (>20) reviews say that his exams are overly convoluted and difficult to properly prepare for. I know RMP is far from a perfect science and doesn’t attract the average student (because the average student isn’t going to bother writing a review about their average experience in a class) but rather students who found the class exceptionally fun/interesting/etc or those who hated it, but whatever - this is a criticism that is apparently unique to this specific Pol Sci professor.
I received a 100% raw score (which was later scaled to a 120 because of the abysmal class average) and my professor told me that he has seen only 5 students pull of 100’s on his midterms in his 15+ years of teaching. This leads me to believe that he consciously tries to make the tests difficult.
The F’s were not just F’s (almost D’s) but rather really, really low F’s - teens, 20’s, thirties.
BUT, almost 1/3 of the class received an A. I was the “only 100” but some of my friends received 97/98/99s so the distinction doesn’t mean too much. Because of this, the professor is claiming that the test was too easy and that he’s going to have to make the next once challenging – despite the huge gulf between the 1/3 that received 90s and the almost 2/3s that received 20s and 30s.
@WasatchWriter - Okay, that makes sense. Whatever caused the 40% of students who failed to fail isn’t going to change if the test is made easier/harder.
“You either get it or you don’t.”