Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

I have a situation that is the opposite of the usual two-sample t-test for equal means: I'm trying to show that it's statistically unlikely that two samples are different.

Is there a statistical test that has the null hypothesis being that the two samples are unequal, and the alternate hypothesis being that the two samples are equal? The number of samples in these tests is small (5 samples each), so I need an equivalent to the t-test, not to the z-test.

Thank you for your help!

share|improve this question
Try stats.stackexchange.com - it is more statistics oriented than MO. –  David Roberts Jun 24 '11 at 4:16
Thanks, David. I didn't know about that site. –  user1998 Jun 24 '11 at 4:27
This question is quite unclear. Why would one not simply let the null hypothesis be equality of population means? That's what you'd normally do if you're hoping the data will justify the conclusion that they're not equal. –  Michael Hardy Jun 25 '11 at 20:42

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.