grae313 wrote:Mine had no problems failing people either, it's just very difficult to write test after test across every single class and subject where the difficulty is perfectly normalized such that the standard for a certain grade is always equivalent to the same score on the test. I had very few classes that didn't apply some adjustment factor to the raw grade distribution to go from percentage to letter grades.
Grades as scoring percentages are essentially meaningless to outside evaluators because they don't know how difficult the tests or assignments were. More typically, grades are used to show where you are relative to the rest of your class. "A" means, perhaps, that you were in the top 10% of your class. Outside observers can more easily make conclusions about what it means to be in the top x% relative to your peers than what it means to score x% on a test of essentially arbitrary difficulty...
But even that is not standardized across most universities. Schools don't force the grade distribution into a Bell Curve -- they leave it up to each instructor to determine what level of knowledge constitutes which letter grade. I honestly think MIT is just asking if there is anything special or different they should know about the grading system in cases of other countries or unusual schools. The standard 4.0 scale doesn't have any meaning in terms of percentages at most schools (which is why we have the PGRE).
I see where you're coming from and you may very well be right, huyichen should just ask MIT to be safe and sure. In fact he should definitely do that because we're pointing out exactly how these things vary. Also, I was always told that rec's should address this too. There are obviously differences. Two are clearly shown here, where one is built around the curve relative to your performance compared to your peers and another built entirely around the material taught without recognition of peer performance. Again, this is why it is good to have your rec address anything like this that you feel should be cleared up. For instance in my E&M class I got a 3.3 and that was the highest grade, that definitely needs to be mentioned.... As to the adjustments profs make, I agree they do happen but like grading they vary. The adjustments we had were built into the weight of exams. Cumulative finals always carried a lot of weight, that was because profs would look at exams and see patterns in the questions. If there was a certain question nearly everyone got right a similar one wasn't on the final, but if there was one that everyone seemed to miss a variation of that certainly was. It was sort of like a second chance to prove you knew what you were doing, and it helped that you realized this ahead of time and could prepare for those questions..... Needless to say our attrition rate was horrible, but the math department was even worse. A lot of us hated how unforgiving it all seemed to be.