Re: st: factor analysis on tetrachoric correlation

[Richard Williams <Richard.A.Williams.5@nd.edu>]

At 11:26 AM 1/23/2004 +0100, jean-benoit.hardouin@club-internet.fr wrote:
> Dear Statalisters,
>
> I search to realize a factor analysis on a specific correlation matrix (as 
> the matrix of the tetrachoric coefficients). The factor program don't 
> accept anothers matrix than the covariance or the correlation matrix. If 
> anyone know a mean to realize this kind of analysis.
It sounds like you have the correlation matrix you want, but you can't get 
Stata to analyze it?  If so, perhaps the -corr2data- command would meet 
your needs?  There was a discussion of this the other day... -corr2data- 
will create a fake data set with a specified means and covariance 
structure.  If the analysis only depends on the means and covariances, you 
are ok.

I caution I didn't know much about factor analysis with tetrachoric 
coefficients.  But, if the only difference between that and factor analysis 
with regular correlation coefficients is in how the correlations get 
computed, I think this might work ok.  If, on the other hand, there are 
other differences in how the factor analysis is supposed to proceed, then 
this may not work.

If this sounds like what you may want, I would suggest first trying to 
replicate a published analysis with known results, where the means, 
correlations and sds were included in the paper.

-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc

*
*   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/