CivPro Exam Forum

[ph14]

Add me to the magic hypo list?

Also, OP your prof does sound a little...tense...Is he a jerk in class?

haha, no he is an excellent professor, brilliant guy. but he is certainly enthusiastic.

he uses this: http://en.wikipedia.org/wiki/Point-bise ... oefficient

to determine when to drop out a multiple choice question. Yaleies...go figure.

Can we get the lay person's explanation? that was too complicated for me lol.

[bdubs]

Add me to the magic hypo list?

Also, OP your prof does sound a little...tense...Is he a jerk in class?

haha, no he is an excellent professor, brilliant guy. but he is certainly enthusiastic.

he uses this: http://en.wikipedia.org/wiki/Point-bise ... oefficient

to determine when to drop out a multiple choice question. Yaleies...go figure.

Can we get the lay person's explanation? that was too complicated for me lol.

Less technical explanation:
http://www.apexdissertations.com/articl ... ation.html

Basic idea: The professor can't determine how well the multiple choice question is measuring understanding of the topic with "standard" statistical measures, so he's using something specifically tailored to categorical variables.

It's a measure of "fit" like r, or r-squared.

Why this is relevant to whether he's a good professor or not is completely unclear to me.

[ph14]

Add me to the magic hypo list?

Also, OP your prof does sound a little...tense...Is he a jerk in class?

haha, no he is an excellent professor, brilliant guy. but he is certainly enthusiastic.

he uses this: http://en.wikipedia.org/wiki/Point-bise ... oefficient

to determine when to drop out a multiple choice question. Yaleies...go figure.

Can we get the lay person's explanation? that was too complicated for me lol.

Less technical explanation:
http://www.apexdissertations.com/articl ... ation.html

Basic idea: The professor can't determine how well the multiple choice question is measuring understanding of the topic with "standard" statistical measures, so he's using something specifically tailored to categorical variables.

It's a measure of "fit" like r, or r-squared.

Why this is relevant to whether he's a good professor or not is completely unclear to me.

How does this play out? closer to 0 the hardest the question, closer to 1 the easier the question?

[bdubs]

I'm assuming that he's dropping questions that are too easy so that the curve disperses more when the essay questions are factored in. Since it's supposed to be a variant of the Pearson moment, I think that you're right that the higher the degree of correlation the more likely it is that the question is easy.

Although an extremely hard question in which the majority of people are misled may also have a high correlation measure value. He might still drop it though.

[brickman]

sorry to have derailed, was just trying to show how much thought and technicality he puts into the class...absurd.

In any case, anyone know of good outlines with PJ hypos?

I'm just going to rip through my schools library tomorrow and try to find something but suggestions would be helpful.