Quote:
Originally Posted by Raev
[You must be logged in to view images. Log in or Register.]
No. I've done the confidence intervals on this kind of stuff before; you need a stupid large number of samples. Like several thousand hits.
|
Huh? Unless the effect of defense is astonishingly small, establishing a 95% confidence interval would definitely not take thousands of hits. Also a paired t-test or regression could make it even easier.