Hi,
re: Your reply (Idea being..with a 7% SD ... the 3% difference in the test sample
would be expected to be consistent with the average of 20%)
I. Yes, .07/sqrt(100)
II. With Excel function: P(z <4.29)) = NORMDIST(4.29) = .999
Ho: p = 0.2
Ha: p > 0.2 (claim)
sample proportion = .23
z(.23) = (.23-.2)/.07/sqrt(100) = .03/.007 = 4.29
p-value = P(z < 4.29) = .999 OR 99.9% (10% significance level)
p-value greater than 10%, accept Ho
The attrition rate is same as 20%