Young Jun Kim <[email protected]> asks:
> I'm conducting survival analysis with stcox command using survey data where
> household is a primary sampling unit (PSU). Several individuals
> participated the study from each household.
> In the Stata, there is no svy command for stcox such as svylogit for
> logistic regression
> Stata manual states cluster option handles clustering in the stcox command
> but it seems not right for my data becasue cluster option treats
> observations in the same cluster as multiple observations from one person.
> I'd appreciate any guidance on how to adjust the clustering effects of
> household with stcox command and also with other survival analysis commands.
-stcox-'s -cluster()- option can be used for both multiple records per subject
and multiple subjects per group, with no problem. It is merely an issue of
specifying a grouping variable in -cluster()- that overrides the default
grouping (subject ID from -stset-).
--Bobby
[email protected]
*
* 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/
