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

Following my question here, I am also looking at the difference between males and females and I have conducted linear regression in a general linear model setup for this purpose.

My effect size for this part of the project is eta squared (which is 0.25).

Eta squared is equivalent to r squared (which is the appeal for me in this case).

I am wondering how do I calculate the confidence interval of eta squared. Does anyone know of any online calculators that can I can use. I am totally lost!

share|improve this question

1 Answer

up vote 4 down vote accepted

I could well be wrong about this, but...I think I would take the square root, yielding eta, which is akin to multiple R. Then I would use the standard method for finding a confidence interval for a correlation, which involves transforming to a Fisher's Z, obtaining upper and lower confidence limits based on Z and N, and transforming back. Then I'd square these limits to get my upper and lower limits for eta-squared. (One online calculator that could help is Vassar's.)

share|improve this answer
There is only one word that I can say to you: thank you! (two words, in fact) – Amarald May 11 '12 at 19:45
This is really quite a clever solution, +1. I went looking for my meta-analysis book this morning, but I couldn't find it... – gung May 11 '12 at 21:16
And intuitive too! – Amarald May 11 '12 at 21:25
Thanks for your kind words, but be ready for someone to come along and poke holes in my method! – rolando2 May 12 '12 at 17:56

Your Answer

 
discard

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.