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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: R: Re: Factor Analysis and Multiple Imputation


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: R: Re: Factor Analysis and Multiple Imputation
Date   Mon, 10 Oct 2011 12:53:26 +0200

On Mon, Oct 10, 2011 at 12:30 PM, Mihir wrote:
> Any rational for doing so? Does 5 times bigger sample size will affect the
> factor score/bartlett score?

No, should not get a five times bigger sample size, you should get
five samples of the same size. The difference between these samples
are the imputed values, which are random draws from the posterior
distribution. It is a different matter how you store these different
samples, in some (correct) ways of doing so make it appear that you
have a 5 (or 6) times larger sample size. Important is that Stata
knows that that is not the case, and treats these cases accordingly.
If you want/have to do it manually (i.e. cannot use -mi- or -mim-),
you really need to be careful in order to do this right. Anyhow, the
logic is the same as the logic underlying multiple imputation, see any
introductory text on that.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany


http://www.maartenbuis.nl
--------------------------
*
*   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