Brilliant. I like that the professor wasn't angry about it but rather, he was impressed with their ability to come together.
I think this is a perfect example why we need anti-cartel and anti-collusion laws. The students have collectively found one way to beat the professor algorithm, managed to create an effective system to prevent anyone from braking the agreement so by my book, as we are talking about programming classes, means that they understand the essential part of what they are supposed to be learning, so I don't see why a good professor would do otherwise.
Quite right collusion is as common a result of the market as anything else.
I actually had a professor rant about this yesterday. But, If all students believe that everyone will boycott with 100 percent certainty, then everyone should boycott (#1). But if anyone suspects that even one person will break the boycott, then at least someone will break the boycott, and everyone else will update their choices and decide to take the exam (#2). While this is sound game theory, it's extremely silly to apply it to this situation. If the students were "blind" (i.e. not in the building) and were actually going on faith, this analysis would be relevant. But since every student was standing next to the door, all that gets thrown out. It's no longer remotely improbable for the students to land on the less likely of the two Nash equilibria -- cooperation.In this one-off final exam, there are at least two Bayesian Nash equilibria (a stable outcome, where no student has an incentive to change his strategy after considering the other students’ strategies). Equilibrium #1 is that no one takes the test, and equilibrium #2 is that everyone takes the test. Both equilibria depend on what all the students believe their peers will do.
Exactly. Although this does demonstrate how an "enforcement" mechanism can move everyone from a shitty Nash Eq to a better Nash Eq. Really, their enforcement changes the payoffs of each student, they could either stay in the hall and get 100% or push through the crowd, incurring social consequences to then have to take the test and maybe receive 100%.
Grading on a curve is retarded to begin with. If the professor only gives an exam about what (s)he taught, then its up to the students to pass or fail. If the professor knows from experience that the mean score on exam will be low, then they can define a score of, say, 40% as minimum passing grade, and, say, 80% as an 'A'. That way it can't be gamed. As long as the target isn't moving it can't be manipulated.
The curve helps control certain problems with test. A problem that was really just too damn hard for almost all the class. For instance a problem where all but one or two students in a high level class can come to grips with in a half hour or so. Material the teacher meant to cover, but didn't. You can say it's all the professors fault and that they should have written a better test, but many professors got where they are by excelling in their field, not by being fantastic test writers. I would rather have a percentage of professors that are respected experts in their field of study then just professors that are good at writing comprehensive exams. Most students look at these things only from a fairness perspective and don't value having talented practitioners with a few teaching deficits, but experts are good for university programs overall. Not that the curve is the best thing ever. I remember an mathematical econ class where there was a student who consistently completed exams in a half hour of a two hour exam who got every problem correct, with no other student scoring 100%. That guy really harshed on our curve.
> That guy really harshed on our curve. Always the problem with curves. Seems that some discarding of outliers could fix that problem.
You can also have to opposite problem I had a professor giving a test so easy that the worse grade was 17 out of 20. I would prefer an almost impossible test and then be graded against my classmates, you just have to provide some rules against gaming (minimum requirements for example).
I had one professor who would pick the best score for each short answer and then then grade everyone else based off of that score. Of course she also had a rule that if the best wasn't good enough she could grade against her own standard. It wasn't probably the most scientific way of grading but it was interesting. She also pretended to be entirely blind to who took each test, so everyone got a random number ahead of time and then you went to talk to her about the test and told her your number, and she'd give you that test. You could then look at everyone's test to compare them. During this conversation she would refuse to admit that it was your test and just pretended like it was something you were interested in looking at.