Re: Is there a statistical test for comparing two sets of values (which are statistics)?

Thanks Ray.

I have no further question.

Ray Koopman wrote:
James wrote:
Thanks. Thus in this case am i assuming that the underlying
distribution for r here is normal?

Strictly speaking, yes. However, the operative assumption is only
that the two distributions are close enough to homoscedastic normal
that the nominal p-values from the t-test are not so different from
the actual p-values as to be misleading.

.

Relevant Pages