Log in

No account? Create an account

Мысли Всмятку

Ad Majorem Annae Gloriam

Previous Entry Share Next Entry
A long time ago I have read a study that said something along the lines of people who change their answer on questions they are not sure about usually do worse on a test than if they left their original answer stand. As far as I can remember, I usually went with my best guess and left it alone, unless I remembered something in the course of working on subsequent questions.

Today I was taking my CPR recertification test, and on two questions I was not sure what the right answer would be. On both I marked the answer different from my initial hunch. I got one right and one wrong. Funnily, my original guesses were also one right, one wrong. Moral of the story: doesn't matter, this way or that, you will get approximately 50% of the questions where you are unsure of the answer if you narrow them down to two possible answers. Pure statistics, regardless of your ability.
Tags: ,

  • 1
I've learned the "leave it well enough alone" in highschool, and I always trust my gut. Unless I am SURE, I don't redo answers. I've tested this several time, and statistically it holds true. You just have a small sample :)

I am sure that if you asked me 100 questions about, say, Korean literature (a subject I know absolutely nothing about), with each question having two possible answers, I would get half of them right and half of them wrong. And I would get a third right if they had three answers each and 25% if they had four answers each. Let's test this theory... I just need to find tests on subjects I am entirely unfamiliar with so some obscure bit of knowledge that is, in fact, in my head, does not confound results. Probably best bet is a test on works of some author I have never read.

Of course I don't take tests on which I know absolutely nothing :)
Oh, and the best test would look like this:
a. b. c. d.


But if it's a subject with which you have even brief familiarity, trust your gut. What's more important, your gut is statistically going to be no less accurate than a random guess anyway, so things can only improve :)

Well, since pretty much all philosophers agree on one point, that too little knowledge is more dangerous than none at all, then gut instinct on a subject you are supposed to know something about but don't know for certain may be wrong more often, than right.

When we had a "bluffing" game in therapeutics class, my theory of a mechanism of action of a particular medication won four times more votes than the actual answer my friend had given. In fact, she told me that as I was speaking, she started doubting whether the written correct answer she held in her hands was, indeed, the right answer. Human mind is easily fooled, fortunately or not.

But again, we are assuming that these tests are actually on something you know more or less. Meaning that if you'd study, you would be able to ace them. Otherwise, of course, it wouldn't work

I took TOEFL about a year after I came to the US. I missed one single question on the entire test. I missed it not because I did not understand the phrase, but because I overthought it. Had I known the language less well, the second interpretaion would not have even appeared to me. Too much knowledge is not necessarily a good thing. And too little knowledge being worse than knowing nothing... even on FER, the Medical subforum is full of the evidence to the truth of this statement.

  • 1