Unbelieveable Stats Test Story!!!

[barcotta]

First off--congratulations! Either you knew much more--or maybe much less--then you thought. It has been my experience with the CLEP and DSST tests in particular that there is almost always one ridiculous option per question. That leaves three remaining options on a DSST and four on a CLEP. The majority of the time, there is at least one other highly unlikely option. Now you're at 50/50 for the DSST and you still have to do a little wading for a CLEP--although because they supply five options per question I think they include a greater number of obviously incorrect choices.

So, let's assume that for most questions on most CLEP and DSST exams, you can narrow it down to one of two options. One is the correct answer. The other is the answer-that-looks-like-it-also-could-be-the-correct-answer-to-someone-who-has-studied-but-isn't-sure-about-this-question. I wonder, then, if you're better off flipping a coin sometimes so that you don't fall into too many traps. It seems to me, if you're guessing wildly with a 50/50 chance of hitting the correct answer--you should get about half of those correct. If, however, you take the bait on the close but incorrect answers--maybe your percentage on those 50/50 questions drops.

Since we can't see our answers, there's no way to tell--although I guess you could use some of the practice exams for that--but I never did. Incidentally, on my last exam, which happened to be an ECE, I tallied and gave myself credit as follows:

I know the answer: 100%
50/50: 50%
Guess: 25%

My raw percentage estimate matched the raw percentage reported on the exam results within 2-4%. This does not support my theory but does support my estimation method!

I always felt I was prepared enough that I still would try to select the best answer when testing--but it occurred to me many times to just pick one based on some other criteria. Anyone else have any theories like this?