Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: pweights question using gllamm


From   "Wiehe, Sarah Elizabeth" <swiehe@iupui.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: pweights question using gllamm
Date   Wed, 11 Jan 2006 12:12:11 -0500

Hello -

I have a question about pweights and the multilevel model (MLM) command -gllamm-.  I have put together a simple MLM using -gllamm- that involves 2 levels (schools and individuals) which runs and presents reasonable data.  I am having difficulty though when I add in the probability weights.  Sampling was somewhat complex for the data set I am using but for my purposes, I have pweights only for individuals.  Sampling was not done at the level of the schools.  School-level data were collected where each study participant attended.  I have used the -pwigls- command to adjust my survey weights for MLM analyses.  

My questions are:

      Is it possible to do a MLM analysis with a level which was not sampled?  
	If so, is there a way to incorporate weights for a non-sampled level (such as schools in this case) so I am not relying on an unweighted analysis? 

I have tried using a wt = 1 for all schools and "adjusting" the pweights using -pwigls- for the individual level.  I am not sure if it is correct.

I have not been able to find the answer to these questions from the help command for -gllamm- and -pwigls-, www.gllamm.org, and the book on this command entitled "Multilevel and Longitudinal Modeling Using Stata."   I have not heard back from a direct email to the -gllamm- authors.  

Thank you in advance for your time in considering this issue.

Sarah


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index