Hi all,
I am running sem on a small dataset while correcting for sampling design, i.e. using svy sem, and I would like to report gof indices. Apparently the svy option reduces the number of available indices radically, so only SRMR and CD remain. Would it be an acceptable solution to specify the model without the svy option, just to obtain the full set of gof indices (Chi2/df, NFI, CFI, RMSEA, SRMR and CD), while retaining the svy option in the final model to estimate the structural effects and covariances? Are there any other options for estimating gof in svy sem?
Most grateful for help!
Jan Hultgren
*
* 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/
