help > Effect size for ANCOVA
Showing 1-2 of 2 posts
Display:
Results per page:
Jan 21, 2021  08:01 PM | Marie Jack
Effect size for ANCOVA
Hello Dr. Zalesky,

I've designed an ANCOVA in NBS and have a significant network at a particular F-threshold. I'm aware that Cohen's d effect size can be computed from a t-threshold (d=t/sqrt(n)), and thus from the F-threshold (d=sqrt(f)/sqrt(n)). However, partial eta-squared is more appropriate to use in my case since I am using an ANCOVA. 

Is there a way to extract an appropriate effect size measure for my design?

Thank you,
Marie
Jan 22, 2021  01:01 AM | Andrew Zalesky
RE: Effect size for ANCOVA
Hi Marie, 

Cohen's d is often a reasonable effect size measure. I don't see why it would not be appropriate here, but you may want to consult with a statistician. 

Andrew
Originally posted by Marie Jack:
Hello Dr. Zalesky,

I've designed an ANCOVA in NBS and have a significant network at a particular F-threshold. I'm aware that Cohen's d effect size can be computed from a t-threshold (d=t/sqrt(n)), and thus from the F-threshold (d=sqrt(f)/sqrt(n)). However, partial eta-squared is more appropriate to use in my case since I am using an ANCOVA. 

Is there a way to extract an appropriate effect size measure for my design?

Thank you,
Marie