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 174.
This example uses the dogcats data set.
gen n_pop = 1300
gen n_type = .
(50 missing values generated)
replace n_type = 850 if type == "dog":type
(32 real changes made)
replace n_type = 450 if type == "cat":type
(18 real changes made)
svyset _n, fpc(n_pop) poststrata(type) postweight(n_type)
pweight: <none>
VCE: linearized
Poststrata: type
Postweight: n_type
Strata 1: <one>
SU 1: <observations>
FPC 1: n_pop
svy: mean totexp
(running mean on estimation sample)
Survey: Mean estimation
Number of strata = 1 Number of obs = 50
Number of PSUs = 50 Population size = 1300
N. of poststrata = 2 Design df = 49
--------------------------------------------------------------
| Linearized
| Mean Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
totexp | 40.11513 1.163498 37.77699 42.45327
--------------------------------------------------------------
svy: mean totexp, over(type)
(running mean on estimation sample)
Survey: Mean estimation
Number of strata = 1 Number of obs = 50
Number of PSUs = 50 Population size = 1300
N. of poststrata = 2 Design df = 49
dog: type = dog
cat: type = cat
--------------------------------------------------------------
| Linearized
Over | Mean Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
totexp |
dog | 49.85844 1.44369 46.95723 52.75964
cat | 21.71111 1.96505 17.7622 25.66003
--------------------------------------------------------------
svy: total totexp
(running total on estimation sample)
Survey: Total estimation
Number of strata = 1 Number of obs = 50
Number of PSUs = 50 Population size = 1300
N. of poststrata = 2 Design df = 49
--------------------------------------------------------------
| Linearized
| Total Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
totexp | 52149.67 1512.548 49110.09 55189.25
--------------------------------------------------------------
svy: total totexp, over(type)
(running total on estimation sample)
Survey: Total estimation
Number of strata = 1 Number of obs = 50
Number of PSUs = 50 Population size = 1300
N. of poststrata = 2 Design df = 49
dog: type = dog
cat: type = cat
--------------------------------------------------------------
| Linearized
Over | Total Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
totexp |
dog | 42379.67 1227.136 39913.65 44845.69
cat | 9770 884.2723 7992.988 11547.01
--------------------------------------------------------------
This example is taken from Lehtonen and Pahkinen’s Practical Methods for Design and Analysis of Complex Surveys.
page 97 Table 3.10 A simple random sample drawn without replacement from the Province’91 population with poststratum weights.
input id str clu wt ue91 lab91 poststr gwt postwt sruv srcvs 1 1 1 4 4123 33786 1 .5833 2.333 .25 .43 2 1 4 4 760 5919 1 .5833 2.333 .25 .43 3 1 5 4 721 4930 1 .5833 2.333 .25 .43 4 1 15 4 142 675 2 1.2500 5.0000 .25 .20 5 1 18 4 187 1448 2 1.2500 5.0000 .25 .20 6 1 26 4 331 2543 2 1.2500 5.0000 .25 .20 7 1 30 4 127 1084 2 1.2500 5.0000 .25 .20 8 1 31 4 219 1330 2 1.2500 5.0000 .25 .20 end
poststratified conditional estimates
Note that you cannot get the deff with the postvar/postwgt options. The numbers in the postwgt variable must be integers (i.e., whole numbers) and are the population totals.
gen fpc = 32
gen postw = .
(8 missing values generated)
replace postw = 7 if poststr == 1
(3 real changes made)
replace postw = 25 if poststr == 2
(5 real changes made)
svyset [pw=wt], fpc(fpc) poststrata(poststr) postweight(postw)
pweight: wt
VCE: linearized
Poststrata: poststr
Postweight: postw
Strata 1: <one>
SU 1: <observations>
FPC 1: fpc
svy: total ue91
(running total on estimation sample)
Survey: Total estimation
Number of strata = 1 Number of obs = 8
Number of PSUs = 8 Population size = 32
N. of poststrata = 2 Design df = 7
--------------------------------------------------------------
| Linearized
| Total Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
ue91 | 18106 6013.646 3885.986 32326.01
--------------------------------------------------------------
svy: ratio ue91/lab91
(running ratio on estimation sample)
Survey: Ratio estimation
Number of strata = 1 Number of obs = 8
Number of PSUs = 8 Population size = 32
N. of poststrata = 2 Design df = 7
_ratio_1: ue91/lab91
--------------------------------------------------------------
| Linearized
| Ratio Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
_ratio_1 | .1297472 .004386 .119376 .1401184
--------------------------------------------------------------
poststratified unconditional estimates This has been skipped for now.
pure design-based estimated under srs
svyset [pw=wt], fpc(fpc)
pweight: wt
VCE: linearized
Strata 1: <one>
SU 1: <observations>
FPC 1: fpc
svy: total ue91
(running total on estimation sample)
Survey: Total estimation
Number of strata = 1 Number of obs = 8
Number of PSUs = 8 Population size = 32
Design df = 7
--------------------------------------------------------------
| Linearized
| Total Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
ue91 | 26440 13282.26 -4967.551 57847.55
--------------------------------------------------------------
svy: ratio ue91/lab91
(running ratio on estimation sample)
Survey: Ratio estimation
Number of strata = 1 Number of obs = 8
Number of PSUs = 8 Population size = 32
Design df = 7
_ratio_1: ue91/lab91
--------------------------------------------------------------
| Linearized
| Ratio Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
_ratio_1 | .1278159 .0040873 .1181511 .1374808
--------------------------------------------------------------