Best strategy for a multiple choice test

S

Sonic_bros

New member
Joined
Apr 5, 2020
Messages
3
100 questions, each with 3 choices. The points are given as follows:
  • 1 point for a correct answer;
  • 0 points for a blank answer;
  • -0.5 points for every wrong answer.
In order to maximize the results of the test, where you need 50 points to pass, in those questions where you are doubting:

Is it profitable to take a guess , supposing you are unable to discard any of the three choices?

Would it still be (or not be) profitable to take a guess if you are able to discard one choice of the three choices presented?

Overall, which would be a good strategy to get at least 50 points?

Many thanks!
 
What are your thoughts?

Have you calculated the expected value of a guess in each case?
 
Dr.Peterson said:
What are your thoughts?

Have you calculated the expected value of a guess in each case?
Click to expand...

I believe its the following:

If guessing between 3 choices: EV= 1 * 1 + 2 * (-0.5)= 0
If guesing between 2 choices: EV= 1 * 1 + 1 * (-0.5)= +0.5

Meaning its profitable to take a guess only when 2 choices? But I am still not sure how can I define a strategy and/or interpret this data.
 
Harry_the_cat said:
Best strategy is to study hard so you don't have to guess any answers!
Click to expand...
No matter how much you study, there are always going to be answers that you wont know how to answer and given the situation it would be great to know how to use probability in our favor.
 
What the probabilities show is that the scoring has been designed so that pure guesses won't help your score, which perhaps is meant to encourage you just to skip questions you have no idea about rather than waste time.

But any knowledge at all will increase your chances of a good score, so unless your answer would be entirely random, you might as well give it a try.

So the best strategy is just to do the best you can, whenever there is any actual thinking for you to do. If you have no thoughts at all, skip it.

Of course, a supplemental strategy might be to do all the questions you are reasonably sure of first, to make sure you get to them, then come back and spend the remaining time on those you have most chance of getting right with some effort.
 
I've never understood the concept of a "blank answer" not being equivalent to a
"wrong answer." To me, a blank answer could be a result of a problem that was skipped, missed, or the person just did not want to put any effort into it. A filled in
reply that is not the correct answer could be a wild guess, an estimated/reduced
guess, one answered by accident meant for another problem, etc. In either case,
you're not supplying the filled in correct answer. I would give zero points for that
question on that hypothetical multiple choice question.
 
You must log in or register to reply here.
Top