Hypothesis test for population variance

[naivesincerity]

I have a sample of 8 values, and need to test the null hypothesis sigma=25 against the alternative, sigma>25, where sigma is the population S.D. What's the correct test and DF to use, given that I only know the sample SD, and will probably be estimating sigma?

 

[vinayan]

You can use the chi-square test (n*s^2)/sigma^2 follows chi suqare distribution with n - 1 degrees of freedom.

S^2 is the sample variance,
sigma^2 is the population variance and n is the sample size.

 

[naivesincerity]

Thanks vinayan. Strange that that doesn't seem to be in my notes