[R-sig-ME] glmer

[Steven J. Pierce]

Bence,

There are many different effect size measures, some standardized and some unstandardized (Kelley & Preacher, 2012). Each has its own interpretation, uses, advantages, and disadvantages. Some of them have been adapted for use with mixed models, others haven’t. I think you need to first be clearer about what effect size measure you want to use, then folks on this list can be more helpful about the computational side of estimating the relevant effect size measure from a model fitted by lme4. 

Kelley, K., & Preacher, K. J. (2012). On effect size. Psychological Methods, 17(2), 137-152. doi:10.1037/a0028086

Steven J. Pierce, Ph.D.
Associate Director
Center for Statistical Training & Consulting (CSTAT)
Michigan State University

-----Original Message-----
From: Bence Bagó [mailto:bencebagok at gmail.com] 
Sent: Monday, August 08, 2016 1:06 PM
To: r-sig-mixed-models at r-project.org
Subject: [R-sig-ME] glmer

Dear Community,

I'm working with lme4 package, and have a question regarding glmer
function. I would like to compare effect sizes of a within and a between
subject study - in the within subject study df is twice as big as in the
between subject one. I was wondering whether there is a way to calculate
standardized effect size.

Thank you very much for your help in advance!

Best regards,
Bence Bago

	[[alternative HTML version deleted]]