Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: effect sizes for mixed effects regression
From
Jessica Nazanine Memarzia <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: effect sizes for mixed effects regression
Date
Wed, 22 May 2013 10:22:10 +0000
Hi,
I have a few questions regarding effect sizes for regressions- I have a rel= atively small sample size so am trying to support findings with effect size= s where possible.
Firstly, is it possible to compute effect sizes for interaction effects in = longitudinal mixed effects regression? I am using xtmixed in Stata 12
xtmixed ghq time ft3_prep_bin || id:, robust
xi: xtmixed ghq i.time*ft3_prep_bin || id:, robust
(data is in long format for analysis over 2 time points)
Secondly, would it also be possible (and perhaps a similar process) to comp= ute effect sizes for a similar mixed effects model interaction effect, whic= h is not longitudinal? (data in wide format again- analysis at t2)
xtmixed fghq_tot group sex || id:, robust
xi: xtmixed fghq_tot i.group*sex || id:, robust (interaction of 2 binary measures on a continuous measure)
Finally, is it possible to compute effect sizes for logistic regressions? U= sing logit and xi:logit for interaction terms? Or are odds ratios here typi= cally enough?
Many thanks,
Jessica Memarzia
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/