Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

st: RE: confa


From   Nick Cox <n.j.cox@durham.ac.uk>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   st: RE: confa
Date   Thu, 7 Jul 2011 14:28:08 +0100

Working backwards, 

1. "Browne (1984b)" fails a longstanding request on this list to use full references. It may be one of Stas' references in the article cited but it is still poor practice to expect people to go somewhere else to decode your references. 

2. Any principle that maximum likelihood not be used with categorical data would invalidate a rather large slice of Stata and of much current statistical practice. 

3. You are asking us to comment on an argument, but you do not say what it is. 

Poor posting, which wastes your time most of all by asking an unanswerable question. 

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

Aggie Chidlow

I hope somebody can advice me..as I am quite new to "confa".
I would like to use confa with my survey data.

I have to say I am familiar with the Stanislav Kolenikow's work :
Confirmatory Factor Analysis using Confa (2009) Stata Journal, 9(3),
329 - 373. I use is as a guide.

Today I have read that "ml" should not be used with categorical data.
Instead it is recommended that "wls" (i.e. weighted least squares)
should be used as proposed by Browne(1984b).

Is that true?
Many thanks in advance, Aggie

*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index