# Thread: Best Stat Question Ever?

1. ## Best Stat Question Ever?

Got this off Twitter (@galka_max) and found it interesting:

If you choose an answer to this question at random, what is the chance you will get it correct?

A) 25%
B) 50%
C) 60%
D) 25%

2. ## Re: Best Stat Question Ever?

I meant to post this in the Stats Forum, not the "research" one.

My apologies.

3. ## Re: Best Stat Question Ever?

How do you define "correct"?

4. ## Re: Best Stat Question Ever?

Originally Posted by rogojel
How do you define "correct"?
I didn't know that was important. Why do you ask?

It is my understanding that there actually isn't an answer to this question so I don't know how to define, "correct."

5. ## Re: Best Stat Question Ever?

Originally Posted by The Game
I didn't know that was important. Why do you ask?

It is my understanding that there actually isn't an answer to this question so I don't know how to define, "correct."
If an answer is chosen at random, then each option has a 25% chance of being selected. If, then, there is a single option of 25% that is keyed as correct, the question is consistent: There is an answer such that the probability of selecting it randomly is the same as the displayed percentage, and that answer is keyed as correct.

But in the displayed question, A and D are both 25%, which screws things up. If A was keyed as correct, would D be too?

6. ## Re: Best Stat Question Ever?

It doesn't actually specify what probability distribution would be used to choose "at random". One could make a case for almost any answer.

7. ## Re: Best Stat Question Ever?

Good point, assumed uniform.

8. ## Re: Best Stat Question Ever?

People need to stop with the assumption of random = uniform! Stop this madness!

9. ## Re: Best Stat Question Ever?

Maybe fifty percent.

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts