Page 2 of 2 [ 18 posts ]  Go to page Previous  1, 2

A350XWB
Snowy Owl
Snowy Owl

User avatar

Joined: 5 Dec 2007
Age: 29
Gender: Male
Posts: 172

23 Dec 2008, 3:45 pm

DNForrest wrote:
At Oregon State University that would be the three Physical Chemistry Exams, where the class average is usually 20-40%, with just one or two people achieving an A- percentage. I somehow managed to get a 133/150 on the third class' final. I was most hated for that one.


How large are the classes? For one to get a 133/150 when the average is 20-40% implies a high casino index, at or near 1. And also a few others who failed miserably or were disadvantaged by the high CI of the tests. :oops:

As ValMikeSmith said, a test is like a lottery ticket "for winning a degree."

Quote:
Yeah, my neuroscience research methods class was nothing but casino exams the entire semester. The average grade in the class was a 60 and no one made an A in the class (though a couple people, including me, made an A-). The professor who taught the class is widely regarded in the department as being one of the worst teachers at the university, and his tests are notorious for having vague wording, a lot of typos, bad structure, and on his answer key the answers that he considers right he often admits were wrong when people question him about them (though he never gives back points for those incorrectly graded questions).


This can contribute to high casino indexes. The casinoest exams usually are not very well done because of exactly these factors, for one. Also, making questions too vague or too precise would uselessly increase the CI of an exam.

And "adaptive" exams have CIs directly proportional to the difficulty of the hardest questions: the higher the difficulty of the last questions, the harder the exam, and thus the CI gets raised by dozens of %. Because relative difficulty is also a factor that gets in the way of calculating a casino index. A given exam that would be hard for one at a high school level but a medium one for a first year of college would not have the same Casino Index. In the former case, the casino index would be in the 0.800-0.900 range while the college class, when faced with the same exam, has a CI of 0.400-0.500.


_________________
My favorite emoticon :D


DNForrest
Veteran
Veteran

User avatar

Joined: 18 Jan 2008
Age: 35
Gender: Male
Posts: 1,290
Location: Oregon

23 Dec 2008, 5:28 pm

A350XWB wrote:
DNForrest wrote:
At Oregon State University that would be the three Physical Chemistry Exams, where the class average is usually 20-40%, with just one or two people achieving an A- percentage. I somehow managed to get a 133/150 on the third class' final. I was most hated for that one.


How large are the classes? For one to get a 133/150 when the average is 20-40% implies a high casino index, at or near 1. And also a few others who failed miserably or were disadvantaged by the high CI of the tests. :oops:

As ValMikeSmith said, a test is like a lottery ticket "for winning a degree."



It's a class required to be taken by all Chemical Engineers and Chemistry majors, so it was about 100-200 people in the class, depending on the year. The problem was the way the professor taught, his tests were over materials almost never covered in the homework or in the lectures. Admittedly my final's high score was due more to me actually studying from the book rather than luck, but it was still a pain in the ass.



A350XWB
Snowy Owl
Snowy Owl

User avatar

Joined: 5 Dec 2007
Age: 29
Gender: Male
Posts: 172

27 Dec 2008, 11:16 pm

Some subjects, like creative writing or sociology, are subjects where grading is subject to the luck of the draw. Where one could almost gamble his/her grade on the roll of the dice and make no difference. The base casino index is often very high.

However, science is known not to be a bunch of casino subjects (although some teachers can give exams with high CIs).


_________________
My favorite emoticon :D