NOTE: If you want to see the design effect or the misspecification effect, use estat effects after the command.
This example is taken from Levy and Lemeshow’s Sampling of Populations.
page 168 stratified random sampling
use https://stats.idre.ucla.edu/stat/books/sop/jacktwn2.dta, clear
svyset [pweight=sampwt], strata(stratum) fpc(npop)
pweight: sampwt
VCE: linearized
Strata 1: stratum
SU 1: <observations>
FPC 1: npop
svy: total twin
(running total on estimation sample)
Survey: Total estimation
Number of strata = 18 Number of obs = 831
Number of PSUs = 831 Population size = 256998
Design df = 813
--------------------------------------------------------------
| Linearized
| Total Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
twin | 26055.4 3791.044 18614.01 33496.78
--------------------------------------------------------------
svy: total twin, over(quart1)
(running total on estimation sample)
Survey: Total estimation
Number of strata = 18 Number of obs = 831
Number of PSUs = 831 Population size = 256998
Design df = 813
1: quart1 = 1
2: quart1 = 2
3: quart1 = 3
--------------------------------------------------------------
| Linearized
Over | Total Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
twin |
1 | 19183.8 2661.629 13959.33 24408.28
2 | 6737.907 2696.605 1444.778 12031.04
3 | 133.687 126.7443 -115.0976 382.4715
--------------------------------------------------------------