Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: RE: RE: coefficients in factor analysis


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: RE: RE: coefficients in factor analysis
Date   Wed, 5 Apr 2006 22:48:48 +0100

With factor analysis you need 
a miniature dictionary, as more-or-less
standard things get renamed and 
there seems to be a conspiracy to
hide the fact that factor analysis
is not that extraordinary, just a 
bit weird. 

In line with this, Stata 9 expects -- nay, 
requires -- you to use -predict- after
-factor-, not -score-, as it is telling you. 

Crudely, scores are really (predicted) values, 
loading are really correlations, 
and coefficients are really coefficients. 
There is lots of ah but to make that crude, 
if not wrong, but it is a start. 

If you are having difficulty with factor
analysis you might be better off with 
principal components, which requires 
fewer religious commitments. Either way, 
it sounds as if you really do need a good
textbook and a very close study of the manual. 

Nick 
n.j.cox@durham.ac.uk 

Paley, Irina
 
> Regarding the score command: using stata 9, when I type in: 
> score f1 right after factor, stata tells me: score is not 
> valid after factor.
> 
> In the help section, it mentions: predict f1 f2: score first 
> two, rotated factors
> 
> 1) Am I doing something wrong with the score command?
> 2) Do you think that the (rotated or unrotated) coefficients 
> that are displayed after factor command are the ones that 
> should be used to actually create the factors? Would you be 
> able to refer me to how these are related to their 
> "regressions scores"? I am sorry I can't seem to find what a 
> regression score is...

*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index