berkeman
Admin
- 69,406
- 24,778
Was the distribution of scores Gaussian? In a couple of my undergraduate physics classes with lots of students (required frosh-sophomore physics classes for physics and engineering students with several hundred students per class), the average score on midterms and finals was around 50%-60%, the distribution of scores were very Gaussian, with a number of the best students scoring in the 90% to high 90% range. The instructors were excellent and very accessible, with several TAs per class who were also very accessible. And in one of those classes, the instructor even issued a handout at the start of the class of the most likely 200 questions that could show up on exams.HAYAO said:Average scores for test in such courses were around mid 40s to 50s. However, due to the scoring system of the college, most people passes.
The tests and questions were hard, but the instruction was great. The scores followed a Gaussian distribution, and the passing grade was whatever the standard was for the University. It was a bit like it is in college sports -- the students/athletes with the best natural abilities who were willing to work the hardest did the best. As long as I can see that Gaussian distribution of test scores and can interview the best students to see how they felt about the instructor, I feel okay with how the instruction is going. If the distribution is skewed one way or the other, that seems like a red flag for grade inflation or extremely clueless instruction, IMO.