It is simple, providing you already have a variable in your data that
identifies the stratum for each observation and have correctly svyset
your data. From what you've said perhaps this isn't true.
I'd double check the svyset before getting deep into recoding.
If your svyset is OK, in Stata 9 you can use svydes to generate a new
variable that identifies the offending strata. The trick is to include
in the svydes command the variables you are using when the singleton PSU
problem arises.
Cheers,
Emma
>>> <[email protected]> 01/06/07 17:25 >>>
This does not seem like a very difficult problem to solve, but
I am unsure how to go about fixing it. I have a bunch of
strata with a single psu in STRATA 9. I want to combine those
single psu strata into a new strata. I assume it would involve
giving each strata some sort of identification number and
combining those whose psu==1, but I do not know what commands
to use. Any suggestions on how best to accomplish this? (I
know there are multiple previous postings on singleton psu,
but I am still confused, sorry!)
Thank you
Eleanor Friedman
*
* 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/
