Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: GLLAMM and intraclass correlation


From   Stas Kolenikov <skolenik@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: GLLAMM and intraclass correlation
Date   Thu, 13 Oct 2011 09:06:40 -0500

On Thu, Oct 13, 2011 at 8:07 AM, Gannaway Jack WWG HWWD > I have used
the GLLAMM command for a random intercept ordered logit
> model, which has then obviously produced a variance for the random
> effect.  Can anyone tell me how I get an intraclass correlation from
> this?  I know for a linear ML model with a random intercept it is the
> squared sd of the effect in question over the sum of the squared sds of
> all the error terms, but how do you do it for ordered logit?

Same way. Since the variance of logistic distribution is pi^2/3, your
ICC is [variance of the random intercept]/([variance of the random
intercept]+pi^2/3). You can obtain this by -display- or -nlcom-: the
latter will give you a confidence interval, although you'd probably
want to put rather little trust in it, as the sampling distributions
of ICC are highly skewed.

-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index