Thanks Nick! Normally that would work. I'm looking at web surveys in
which a fair number of participants simply stopped taking the surveys
at some point through the survey--lots of missing data with no good
way to fill it in. Would still like to use what data I have on early
questions that more respondents answered. Just wish Stata had some
convenient way to automatically recalculate probability weights
reflecting gender based on the observations available in a given
analysis. Looks like I'll have to do this by hand. The for2 command
might help me do this relatively quickly--very handy command.
Hoping all is well w/ you.
Cheers,
Peter
Date: Thu, 24 Sep 2009 15:26:54 -0400
From: Nick Winter <[email protected]>
Subject: Re: st: svyset problem 2: using svy with partially complete surveys
You could consider doing some sort of imputation for the missing data . . .
*
* 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/
