Quote:
Originally Posted by seped
[You must be logged in to view images. Log in or Register.]
It answers the question of if your sample is large enough at the same time.
|
Well, here's the thing: if you look at this page
http://en.wikipedia.org/wiki/Binomia..._approximation
Then it seems that if we set n=100 and p=0.01, it fails all the given tests for whether 'n' is high enough for the normal distribution to be a good approximation (on the other hand, if n=100 and p=0.3, then we seem to get a good approximation).
That Wilson score interval looks pretty good, though. Maybe I'll use that. Thanks for the link!