chapter 7 presented a ci for the variance s2 of a normal population distribution. the key result there was that the rv x2 5 (n 2 1)s2ys2 has a chi-squared distribution with n 2 1 df. consider the null hypothesis h0: s2 5 s20 (equivalently, s 5 s0). then when h0 is true, the test statistic x2 5 (n 2 1)s2ys20 has a chi-squared distribution with n 2 1 df. if the relevant alternative is ha: s2 . s20