# st: how to calculate sampling errors

 From Anna Gueorguieva To statalist@hsphsun2.harvard.edu Subject st: how to calculate sampling errors Date Mon, 12 Feb 2007 19:18:10 -0800 (PST)

Dear Statalist Users,

I have a quick question about calculating sampling error.

I have 2-stage-sampling survey of households using an outdated frame.

I want to show the sampling errors for each estimate.

How do I do that in Stata 9?

Thank you in advance. Best, Anna

Here are some tables if you think this might be useful:
svyset psu [pw=pweight], strata(strata_unique) poststrata(type_unique) postweight(post_str_type) fpc(fpc) ///
>   || hhid_unique
pweight: pweight
VCE: linearized
Poststrata: type_unique
Postweight: post_str_type
Strata 1: strata_unique
SU 1: psu
FPC 1: fpc
Strata 2: <one>
SU 2: hhid_unique
FPC 2: <zero>

. svy: tab poor wave, col
(running tabulate on estimation sample)
Number of strata   =        96                  Number of obs      =      9577
Number of PSUs     =      1198                  Population size    =   7600000
N. of poststrata   =         8                  Design df          =      1102
---------------------------------------------
|               wave
poor |     1      2      3      4  Total
----------+----------------------------------
0 | .6176  .5645    .65  .5453  .5943
1 | .3824  .4355    .35  .4547  .4057
|
Total |     1      1      1      1      1
---------------------------------------------
Key:  column proportions
Pearson:
Uncorrected   chi2(3)         =   68.8616
Design-based  F(., .)         =         .     P =      .

____________________________________________________________________________________
Sucker-punch spam with award-winning protection.
Try the free Yahoo! Mail Beta.