Just to share something I posted on SMU LMS Micro discussion forum

I have a proposition that midterms should only be held after week 7. From the perspective of current students, it might be a little too late to even bother thinking about this. However it is precisely because the midterms are over that it is appropriate to post this in case this issue somehow snowballs.

So before I go any further, let's have some common understanding about the significance of the midterms. Usually it is a significant component % of the overall grade> Hence it is important to do well, if not don't screw up, or the consequence could be disastrous when you get back the final grade.

However, do think for a moment. Can you avoid the bad consequence, or at least significant avoid it, if the midterms were to be held in week7 or before?

The answer is yes. Knowing you did very badly for the exam (e.g. through discussing questions with friends after the test), you can choose to withdraw from the module to avoid getting C or below.

Assuming perfect knowledge(not too far off from reality since in many instances you can roughly guage how well you did relatively), those who form the bottom bell curve will have a huge incentive to drop the module if they are concerned about their GPA. If they do, adverse selection occurs. Those who remain in the class are the more competitive; The bell curve shifts adversely. If the prof does not have a min. cap of grades for once decent but now relatively poor performers (e.g. meaning the curve is really belled from C to A, with the comparatively lowest scoring student guaranteed a C no matter how well his raw score is), the new group of people who form the bottom bell curve will have an increased incentive to drop the class. And this creates an issue of fairness for the remaining students.

Also, with people dropping classes, it necessarily means resources were not allocated optimally. The class size could have been larger if there was a benefit of foresight to know the exact number of people who drop the class. And for the people who dropped the class, they had wasted their first 7 weeks of time studying for the module.

Of course, probably you don't see many of such instances in SMU. Because most of the time not everyone knows that a) The bottom bell curve has dropped the class. And so consequently b) Few people dropped the class, the bell curve does not shift as much for you to care and you will not have the incentive to drop anyway.

But from my personal experience, it has happened on a significant scale before. Nearly 10 people from my MFE class dropped the mod after the midterms. The size of the class shrinkage is probably because econs students are well... more economical. I was one of the unfortunate ones who do not know about point a) during that time to act on it.

Then you could argue, didn't the students who dropped the class on week 7 benefit? It could be a zero-sum. But I disagree. Having midterms on week 7 or before creates a moral hazard. One can choose not to study for the midterms and hope for the best. In the event that one does well, he/she had not learnt as much as should have been. And if the event that one does badly, he/she will drop the class. However, by having midterms after week 7, the incentive to study harder becomes stronger and the educational system will produce students who are more knowledgeable.

I know I have a somewhat extreme take on this issue but any thoughts?