Heather Ridolfo <evd7@CDC.GOV> is using -svy: mean- with options -subpop()-
and -over()- and noticed that it is reporting a smaller estimation sample than
the number of observations in her dataset:
> I recently posted a question regarding losing observations and PSUs when
> using svy regress and the subpop command. Jeff Pitblado recommended I
> update the version of stata I was using to correct this problem.
>
> This did in fact fix the problem. However, I am still having a problem
> when I run svy, mean.
>
> svyset psuscid [pweight = gswgt1], strata(region)
> svy, subpop(bhgsp): mean efficacy esteem child_sex age meduc new_inc2
> dm_support dm_college m_indep MDCOLLEG md_spprt MDACHIEV, over(racebh)
>
>
> Survey: Mean estimation
>
> Number of strata = 4 Number of obs = 4657
> Number of PSUs = 121 Population size = 3921354
> Subpop. no. obs = 2126
> Subpop. size = 1736680
> Design df = 117
>
> If STATA is dropping observations from my original dataset due to
> incomplete data, is the survey design information from these
> observations retained in the calculation of the standard errors? Is
> there any way to fix this or another way to get means while using the
> survey commands and subpop?
We have verified that -svy: mean- is incorrectly dropping out-of-subpop
observations that contain missing values in the variables of the varlist. The
only other affected commands are -svy: proportion-, -svy: ratio-, and
-svy: total-. We hope to have this fixed in the next Stata update (within the
next few weeks).
--Jeff
jpitblado@stata.com
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/