Following on from this question on reporting eta squared, I would like to report 95% CIs for my effect size point estimates (partial eta and eta squared values).

  • What statistics do I need to know and is there a straight forward formula for calculation?
  • $\begingroup$ Very interesting. Normally I care alot about such things, but this one i have never thought of before. $\endgroup$ – rolando2 Sep 24 '11 at 12:35
  • $\begingroup$ stats.stackexchange.com/questions/15958/… is a post with a good long answer providing some alternatives to eta (it's the answer that mentions Cohen several times). It's possible that these alternatives have a straightforward confidence interval. Like rolando2, I find the question interesting, but don't know the answer. $\endgroup$ – zbicyclist Sep 25 '11 at 1:28

As a preliminary answer:

| cite | improve this answer | |
  • $\begingroup$ I'm impressed. I've seen some of Karl Weunsch's work before, on teaching statistics, and liked that too. Thanks for pointing this out, Jeremy. $\endgroup$ – rolando2 Sep 26 '11 at 23:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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