st: Using sem for CFA taking into account couples correlation (dyadic analysis)

[Marta Garcia-Granero <mgarciagranero@gmail.com>]

Hi everybody:

Is it possible to add dyad information (husband-wife correlation) to a 
CFA using sem on Stata 12.1? We have used -sem- for a 4-factor CFA on 
the 32 ordinal items of Dyadic Agreement Scale (DAS). We have 415 
married couples and our initial analysis did not include the couples id:

sem (CONSENSUS-> das1-das3 das5 das7-das15)(SATISFACTION-> das16-das23 
das31 das32)(COHESION-> das24-das28)(AFFECTEXP-> das4 das6 das29 das3

One of the referees mentioned "Dyadic Analysis", by Kenny, Kashy and 
Cook (2006) as reference for the effect that the lack of independence 
had on the variances, standardized statistics and degrees of freedom of 
our model.

I already have a couple id variable, as well as gender, and I would like 
to incorporate that information (within couple correlation between wife 
and husband) into my sem model.

Thanks a lot,
Marta
*
*   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/