you don't say whether you want a one-sample or a two-sample t-test
if two-sample, use -svy: regress- where the only predictors are the
constant and an indicator variable for the group
if one-sample, use -svy: regress- where the only predictor is the
constant; note that if your data are actually in two variables for
pairing, make a new variable that is the difference between the two and
use that for your outcome variable
Rich
On 2/19/13 3:18 PM, Ana wrote:
> Hello,
>
> Does anyone know how to apply a simple ttest using svy command?
> I've tried the 'svy: cnsreg ...' but it did not work out.
> Best,
> Ana.
*
* 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/
