Tell me more ×
Cross Validated is a question and answer site for statisticians, data analysts, data miners and data visualization experts. It's 100% free, no registration required.
up vote 1 down vote favorite

I have conducted a one-way ANOVA on 8 groups. I understand that the degrees of freedom for that test is 7, but I don't understand how to ascertain the degrees of freedom for the post hoc tests:

I conducted a series of t-tests as the multiple comparison post-hocs using the Holm-sidak method of familywise error rate reduction. How does one figure out what the degrees of freedom are for such tests? Is it simply n1 + n1 - 2, as it would be for regular t-tests?

If so, since my samples sizes are 893 and 169 respectively for the post-hoc t-test in question, are the sample sizes getting a little high, forcing my Type I error rate up?

Thanks!

share|improve this question
1  
Sidak-Holm is a step-down test. I believe df should be N-k where N is total sample size and k is number of groups. – ttnphns Aug 10 '11 at 1:35
Thanks for your comment! To clarify, in this case, N would be the sample size (n1 + n2), and k = 2, since only 2 groups are being compared in the post-hoc test? Or would k = 8, the original number of groups? – Mog Aug 10 '11 at 1:39
1  
If you run ANOVA with 8 groups and and ordered to perform multiple comparisons on them it will be all 8 groups. – ttnphns Aug 10 '11 at 6:54
Excellent! Thanks for your help! – Mog Aug 10 '11 at 15:28

Know someone who can answer? Share a link to this question via email, Google+, Twitter, or Facebook.

Your Answer

 
discard

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

Browse other questions tagged or ask your own question.