I plan to estimate a series of hierarchical models -- individuals 
within states, countries, counties, etc.  The data in some cases come 
from complex surveys with design weights and in other cases from 
low-response-rate polls that require post-stratification weights to 
account for differential non-response.

I know that HLM 6.0 allows respondent-level weights as does the suite 
of Supermix programs (MIXREG, MIXOR).  But I'd prefer to use Stata or R.

Best I can tell, neither Stata (xtmixed or glamm) nor R can deal with 
weights.  I am hoping I am wrong.  Can anyone set me straight on this 
(or confirm if you've investigated and are certain)?

Thanks much!
Eric

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Eric Plutzer
Professor of Political Science and
         Academic Director, Survey Research Center
Penn State University
Voice: 814/865-6576
http://www.ssri.psu.edu/survey/

**********************************************************
             Political Methodology E-Mail List
   Editors: Melanie Goodrich, <[log in to unmask]>
            Delia Bailey, <[log in to unmask]>
**********************************************************
        Send messages to [log in to unmask]
  To join the list, cancel your subscription, or modify
           your subscription settings visit:

          http://polmeth.wustl.edu/polmeth.php

**********************************************************