R.E. De Hoyos <[email protected]> asks about generating an indicator variable
that identifies strata with a single PSU:
> I want to run wage regressions for several population groups. Estimations
> have to take care of the survey design (svy). The problem is that for some
> of the groups there are strata with a single psu (primary samplying unit).
> Using -svydes- I can identify does strata, for example:
>
> For wage equation: svyreg wage_1 varlist_1:
>
> svydes wage_1 varlist_1
>
> And Stata will highlight those strata with a single psu. How can I create a
> dummy variable that indicates weather a strata has a single psu given the
> variables that I want to use for the analysis?
> In other words I want a dummy variable that takes the value of 1 if -svydes-
> highlights a strata.
This feature was added to -svydes- in Stata 9. -svydes- now has a
-generate()- option that generates this variable.
In Stata 8, there is no such option; however, the output of -svydes- should
identify the "singleton" strata using the values from the strata variable that
was -svyset-. You can use the output from -svydes- to generate a new
variable.
If you need to do this for a bunch of different groups, you will need to write
a program. You can start with the following:
***** BEGIN: singleton.ado
program singleton, sort
version 8
syntax [varlist(numeric default=none)] [if] [in], ///
STRata(varname) gen(name) [ PSU(varname) ]
confirm new var `gen'
marksample touse
if "`psu'" == "" {
tempvar psu
gen `psu' = _n
}
tempvar u
sort `touse' `strata' `psu'
quietly by `touse' `strata' `psu': gen `u' = _n == 1
quietly by `touse' `strata': replace `u' = sum(`u')
quietly by `touse' `strata': replace `u' = cond(`u'[_N] == 1, 1, 0)
quietly replace `u' = . if !`touse'
rename `u' `gen'
end
***** END: singleton.ado
Here is a log using the above -singleton- program with the auto data:
***** BEGIN:
. clear
. sysuse auto
(1978 Automobile Data)
. * Example with no singleton strata
. singleton, strata(rep) gen(jj)
. tab rep if jj
no observations
. * Example with 2 singleton strata
. singleton, strata(rep) psu(for) gen(kk)
. tab rep if kk
Repair |
Record 1978 | Freq. Percent Cum.
------------+-----------------------------------
1 | 2 20.00 20.00
2 | 8 80.00 100.00
------------+-----------------------------------
Total | 10 100.00
. * Proof that 2 categories of -rep78- have only "Domestic" cars
. tab rep for
Repair |
Record | Car type
1978 | Domestic Foreign | Total
-----------+----------------------+----------
1 | 2 0 | 2
2 | 8 0 | 8
3 | 27 3 | 30
4 | 9 9 | 18
5 | 2 9 | 11
-----------+----------------------+----------
Total | 48 21 | 69
***** END:
--Jeff
[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/
