Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Afif Naeem <afeef745@hotmail.com> |
To | <statalist@hsphsun2.harvard.edu> |
Subject | st: Factor Analysis with Sampling weights |
Date | Wed, 5 Sep 2012 19:45:55 -0400 |
Hello everyone, I am trying to use exploratory Factor Analysis on 15 variables in the data set. The data set is not a good representative of the general population if used without the Sampling Weights. Hence I have used Sampling Weights in the rest of the empirical analysis. I have two questions in this regard, which are: 1 - Should I use Sampling Weights to construct Factor Scales in exploratory common Factor Analysis. Would the Sampling Weights be counted twice if I use them to construct Factor Scales, and then use the Sampling Weights in regression analysis with these very Factor Scales present in the model? 2 - Is there an easy way to account for Sampling Weights in the exploratory common Factor Analysis. I have been looking for a way to use Factor Analysis while using Sampling Weights, and the one method that works is inputting the Sampling Weight adjusted correlation matrix (which I obtained by regressing each one of the 15 variables on the rest of the variables, one at a time, and taking the square root of the R-squared of the model, which also is Sample Weight adjusted correlation between the two variables). This method takes a very long time. I was wondering if there is more time efficient method to do the exploratory common Factor Analysis using Sampling Weights? Thanks, Afif * * 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/