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]

Re: st: Oblique Rotation returns no correlation between factors?


From   "JVerkuilen (Gmail)" <[email protected]>
To   [email protected]
Subject   Re: st: Oblique Rotation returns no correlation between factors?
Date   Wed, 13 Mar 2013 10:21:45 -0400

On Tue, Mar 12, 2013 at 3:21 PM, I. Kuznetsov <[email protected]> wrote:
> Dear Statalisters!
>
> I doing some research using factor analysis (principle axis factoring) using
> categorical data. I have used polychoric correlations to correct for the
> categorical variables problem. But now when I try to rotate obliquely using
> "oblimin" and use "estat common" to find the correlation between factors,
> Stata returns that there is 0 correlation, which is very strange. I have
> tried different extraction methods and different rotation methods and yet I
> still find no correlation. Any suggestions?

I've used oblimin rotation after factor many times and never had a
problem so short of seeing your data and code I don't really know what
there is for any of us to do.

It is possible to get a 0 value for the correlation from oblique
rotation. An example of the kind of situation that would generate a 0
factor correlation would be a block diagonal correlation matrix, say
two blocks that have substantial correlation with the inter-block
correlations being 0. (In some sense this is the ideal for factor
analysis.) But sampling variability concerns make it likely that the
observed rotated correlation will not be exactly 0.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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