UCLA professor gamed his game theory midterm

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

syoung

MS-3
10+ Year Member
Joined
Jan 3, 2011
Messages
1,026
Reaction score
76
What do you think? Could science and medicine progress even further if exams in college were like this?

http://blogs.kcrw.com/whichwayla/20...-a-ucla-professor-gamed-a-game-theory-midterm

tl;dr
Professor told his class they could use whatever notes, call whomever, bend most if not all the normal rules of "cheating" that governed a typical exam for this midterm. They could not however do anything that broke state or federal laws like blackmail, extortion, kidnapping...

They lived the week as evolutionary ecologists... And he found that the scores were 20% higher than in the past.
 
Meh. My guess is that the smart kids still got the high grades, except this time the dumb kids were able to "work with" the smart kids which leads to your 20% increase.

Also, who couldn't score higher when you could use the internet?
 
I kind of think students should be allowed to use notes, books, and internet during tests because in real life most people have access to these tools. And in order to make sure students aren't just passing because of fact checking the internet, you could make the questions very hard conceptual questions to test true understanding.

It would be hard to do this though as most people would abuse the system.
 
By creating a higher bar for students who want to be top 10 in the class will in theory result in "smarter" students. The students who only had to get a 92-95% now need to get 100% to be top 10 which means they know more so not only does this help the kids who don't give a **** but helps the ones on the top too. However, the negative side is that if ~30 kids get 100%s then this method was a major flop and didn't prove anything or than the fact that given enough time anyone in the 21st century can find an answer to things.
 
Modern science/medicine isn't about the vast amount of information you can recall, but rather how you integrate said information, which can't be taught as easily as straight facts.

Secondly, I think GPA/class rank is useless for understanding what a student knows and how he thinks. This board is replete with examples of gaming the system for high GPAs, but a good standardized test (whether or not the MCAT is this is debatable) is unparalleled to gauge a student's talents.
 
The question the professor posed was not something that could be just "found" on the internet.
 
What do you think? Could science and medicine progress even further if exams in college were like this?

http://blogs.kcrw.com/whichwayla/20...-a-ucla-professor-gamed-a-game-theory-midterm

tl;dr
Professor told his class they could use whatever notes, call whomever, bend most if not all the normal rules of "cheating" that governed a typical exam for this midterm. They could not however do anything that broke state or federal laws like blackmail, extortion, kidnapping...

They lived the week as evolutionary ecologists... And he found that the scores were 20% higher than in the past.

Fairly common practice in upper level math classes. These tests are usually insanely hard and not something that could be googled.
 
The question he asked was extremely loaded, so in that case the group scenario is interesting.

If it was a regular exam with MCs and short answers, this would have been a dumb idea.
 
Easiest way to ace the test: everyone brings $5 and pays a poor grad student the money to complete the test.
 
We have "tests" like this in organic chemistry. They are graduate level organic chemistry problems, and we are given a week to solve 5-10 of them. Depending on the semester, you can either only work in a group or you can work with the entire class. They are extremely challenging and time consuming problems, but you can use all of your notes and book and talk to people and such. We're technically not allowed to use the internet, but it's not like it matters because we can't just google the answers.

We have two normal tests and one of these in organic chem. I think it's a really cool and unique system, as it challenges how much we know in a way that they could never do on a test just due to sheer lack of time.
 
Top