qIf we draw two samples from one population and obtain two different means, how do we estimate the
significance of the difference between these two sample means?
qIf we repeat these pair-sampling many many times, the probability distribution for the mean
difference will be a normal distribution with a mean of zero and a standard deviation of:
q
q
q
qTest the null
hypothesis that the samples come from the same population
( m1=
m2
as well as s1=
s2)
use the t-distribution with
q v = N1 + N2 - 1
q