sample size needed for power of a study
View Single Post
Nov28-04, 08:30 PM
Math Is Hard
Actually z = (x-m)/sd
doh! sorry - typo!
and if you are using m=200 you will need to find the z value corresponding to
Pr(observation < z) = 0.80
and then your x (observation) would be
x > m + z*sd
This will give you 80% confidence if x really is big enough -- for 48/sqrt(N) to be less than or equal to 8 sqrt(N) will have to be > 6, N>36
Thanks for your help! I am not familiar with Pr() and "observation" - but I think I see what you're saying.