>>13570064>You seriously think it's more likely that there are huge changes in a population from one year to the next than exam questions vary slightly in difficulty between years?Yes. There can be huge changes in the population from year to year.
> Also, you do realize that grading on a curve does not require an exam to be multiple choice right?That was added in addition since it is also something that is done alongside grading on a curve that shouldn't be done to cover all bases. There is nothing in the context of my statement relating it to grading on a curve however. That idea is added by you afterwords.
>Maybe if you have a shitty professor the questions will be slight variations each year No. There is no reason to believe this. Rather it is arguable that a good professor in general will be much more consistent in both lecture and test design than a shitty professor
1. A student can really only solve problems or variations in such a situation that they already practiced.
2. There are already only a limited number of problems and their variations that are even appropriate to the course.
3. Students forget what the professor said in class unless they are made to use it repeatedly through homework problems or by using memorization techniques.
4. What the professor says is written in the textbook.
5. There's only a certain number of a certain type of problems that a student can be expected to be solved in a short time period.
This alone drastically reduces possible variation. Most courses, especially at the ones pertinent to lower division undergrads, have fixed curriculum at as such this further reduces the amount of variation. Moreover, the point of an exam is to be a standardized evaluation of whether or not a student knows a sufficient amount of information to at least say they passed the course. In order to ensure that exams work this way, the variation allowed on an exam must be further reduced.
