Page 1 of 1

Guessing and eliminating answer choises

Posted: Mon Dec 05, 2005 4:22 pm
by jax
If I can eliminate one answer choice, should I guess on that problem? I was just wondering what is the best way to approach a problem when you can eliminate one answer choice. I know they subtract some percent of the questions you get wrong from your score, so how many do you have to eliminate in order to safely guess?

Posted: Tue Dec 06, 2005 3:57 am
by Grant
"As a correction for haphazard guessing, one fourth of the number of questions you answer incorrectly is subtracted from the number of questions you answer correctly." - ETS
I wrote something about this on the following page:
http://www.physicsgre.com/testing-answers.shtml
Feel free to check my math, but if you eliminate 1, 2, or 3 of the answer choices then that statistically corresponds respectively to 1/16, 1/6, or 3/8 points added to your raw score (assuming you guess randomly from the remaining answers).

Posted: Tue Dec 06, 2005 10:21 am
by jax
So even if you eliminate one it's still worth guessing?

If I eliminate one, I would have a 1 in 4 chance of getting it right and they take away 1/4 of the ones you get wrong, so it doesn't seem like that would be productive...

Posted: Tue Dec 06, 2005 12:30 pm
by Grant
It makes more sense if you consider 4 different questions where you can eliminate one answer choice in each question. The odds say you will guess correctly on 1 out of the 4 questions (i.e. one correct and three incorrect). For the correct guess you get +1 on your raw score and for the three incorrect ones you get -3/4 from your raw score (i.e. minus 1/4 for each incorrect answer).