Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I was discussing a small experiment with a friend of mine this week. He said, "we can just do 3 trials."

I said, "sure, but the subject will have to get all 3 trials right to be better than chance."

We discussed it a bit and verified that you would indeed have a 50/50 chance of correctly guessing at least 2 out of 3 coin flips.

I started wondering how many trials you'd have to have to get statistical significance. I figured out that the probability of guessing at least $n-1$ correct trials is this formula:

$$\frac{(n + 1)}{ 2^n}$$

If I set that equal to $.05$, how do I out how many trials you'd have to have to get a 95% confidence if the subject misses no more than one trial?

id est, if $$.05 = \frac{(n + 1)}{ 2^n}\,,$$ what is $n$?

More importantly, how did you solve for $n$?

Thanks a lot!

Patrick

share|improve this question
add comment

1 Answer

up vote 1 down vote accepted

At $7$ rolls the probability is $0.0625$, at $8$ it is $0.035156$. I just calculated them from 2 to 27 in a spreadsheet. There is no algebraic solution without invoking Lambert's W function.

share|improve this answer
    
Thanks for the response. I'm pleased there wasn't something really simple I was overlooking. Edit: Evidently, hitting enter submits the form. Because I don't know any better, I'm going to leave it open for just a little longer to see if there are any conflicting opinions and then give credit where credit is due. Thanks Ross. –  D. Patrick Jun 27 '11 at 20:38
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.