These were the stats for the CS midterm, and the mean will be curved to a B+:
Median: 104
Mean: 101
Std deviation: 25
When we are talking about the standard deviation, are we saying a 101 is a B+, a (101-25=76) is a B, a (101-25-25=51) is a B-, etc. OR are we saying a 101 is a B+, a (101-25=76) is a C+, a (101-25-25=51) is a D+, etc.?
Thanks!
It is case-by-case for each professor and class.
Some classes don’t curve at all but the exams are designed so that it is reasonable to get 90% and get some sort of A and so on.
Some other examples include lining up the people by the exam scores and apply a distribution. So if too many people did well on the exam, then 96% and above is A, 95% and above is A-, 94% and above is B+, etc. which is on the extreme side, but has happened.
My least favorite one is basing on the highest score and using that as an index - i.e. even if the mean/median is ~60%, if somebody managed to get 98%, then each exam scores are divided by 0.98 which doesn’t give any meaningful boost.
In your case, it seems like 101 will be a B+, but no way that 76 is B but more like C+. 51 is likely no better than C- or D+. This is also dependent on the professor if they want to concentrate as many people as possible on the B range or spread people out.
Many profs are much more lenient on the lower end of the tail. Perhaps you can ask your TA about the Prof’s past history.
About the only thing you can be assured of is that a 127+ (mean + 1 SD) is an A.