My interpretation of the manuals is the same as yours. You don't need to use _robust_ command if you set the weight, psu, and strata using svyset. This is consistent with the results obtained after svy commands, which indicate that the standard errors are robust (even when the _robust_ command was not used).
-----Original Message-----
From: Hoezee, Larry [mailto:[email protected]]
Sent: Tuesday, June 11, 2002 7:10 PM
To: '[email protected]'
Subject: st: Complex, multistage survey (NPSAS:96) and robust
Colleagues,
Does svylogit, using the weight, psu, and strata options calculate a robust
Huber/White variance estimator? The manuals (v.7) seems to indicate that it
does, but I am not certain. Or does one use svylogit, the correct weight,
psu and strata AND program using the _robust command to arrive at the
correct standard errors?
Using NPSAS:96, a database from NCES, necessitates that a weighting and
design effect adjustment is made. I want to be certain to make the correct
adjustments.
I am new to STATA and this listserve. Thanks for the help.
L. Hoezee
*
* 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/
*
* 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/
