 |
      |
||||
|
Help the Stat Consulting Group by
giving a gift
| |||||
|
Loading
|
|||||
Stata FAQ
How do I analyze survey data with a stratified random sampling design?
The examples below use Stata 8. If you are using Stata version 9, please see this page.This example is taken from Levy and Lemeshow's Sampling of Populations.
page 136 stratified random sampling
use http://www.ats.ucla.edu/stat/books/sop/hospsamp.dta, clear svyset [pweight=weighta], strata(oblevel) fpc(tothosp) svytotal births Survey total estimation pweight: weighta Number of obs = 15 Strata: oblevel Number of strata = 3 PSU: <observations> Number of PSUs = 15 FPC: tothosp Population size = 157.99993 ------------------------------------------------------------------------------ Total | Estimate Std. Err. [95% Conf. Interval] Deff ---------+-------------------------------------------------------------------- births | 183982.9 34014.33 109872 258093.8 .7035474 ------------------------------------------------------------------------------ Finite population correction (FPC) assumes simple random sampling without replacement of PSUs within each stratum with no subsampling within PSUs. Weights must represent population totals for deff to be correct when using an FPC. Note: deft is invariant to the scale of weights.
svytotal births, by (oblevel)
Survey total estimation
pweight: weighta Number of obs = 15
Strata: oblevel Number of strata = 3
PSU: <observations> Number of PSUs = 15
FPC: tothosp Population size = 157.99993
------------------------------------------------------------------------------
Total Subpop. | Estimate Std. Err. [95% Conf. Interval] Deff
---------------+--------------------------------------------------------------
births |
oblevel==1 | 14931 2669.857 9113.882 20748.12 .15648
oblevel==2 | 117116.9 33067.66 45068.68 189165.2 1.089405
oblevel==3 | 51934.98 7508.399 35575.58 68294.37 .0330073
------------------------------------------------------------------------------
Finite population correction (FPC) assumes simple random sampling without
replacement of PSUs within each stratum with no subsampling within PSUs.
Weights must represent population totals for deff to be correct when
using an FPC. Note: deft is invariant to the scale of weights.
This example is taken from Lehtonen and Pahkinen's Practical Methods for Design and Analysis of Complex Surveys.
page 74 Table 3.3 Estimates from an optimally allocated stratified simple random sample (n = 8); the Province'91 population.
NOTE: In this data set, the fpc changes with the strata. This is different from all of the previous examples.
input id str clu wt ue91 lab91 fpc 1 1 1 1.75 4123 33786 7 2 1 2 1.75 666 6016 7 3 1 4 1.75 760 5919 7 4 1 6 1.75 457 3022 7 5 2 21 6.25 61 573 25 6 2 25 6.25 262 1737 25 7 2 26 6.25 331 2543 25 8 2 27 6.25 98 545 25 end svyset [pweight=wt], psu(clu) fpc(fpc) strata(str) svytotal ue91 svyratio ue91 lab91
Survey total estimation
pweight: wt Number of obs = 8
Strata: str Number of strata = 2
PSU: clu Number of PSUs = 8
FPC: fpc Population size = 32
------------------------------------------------------------------------------
Total | Estimate Std. Err. [95% Conf. Interval] Deff
---------+--------------------------------------------------------------------
ue91 | 15210.5 4279.452 4739.059 25681.94 .20649
------------------------------------------------------------------------------
Finite population correction (FPC) assumes simple random sampling without
replacement of PSUs within each stratum with no subsampling within PSUs.
Weights must represent population totals for deff to be correct when
using an FPC. Note: deft is invariant to the scale of weights.
Survey ratio estimation
pweight: wt Number of obs = 8
Strata: str Number of strata = 2
PSU: clu Number of PSUs = 8
FPC: fpc Population size = 32
------------------------------------------------------------------------------
Ratio | Estimate Std. Err. [95% Conf. Interval] Deff
------------------+-----------------------------------------------------------
ue91/lab91 | .1277788 .0031736 .1200134 .1355442 .3803409
------------------------------------------------------------------------------
Finite population correction (FPC) assumes simple random sampling without
replacement of PSUs within each stratum with no subsampling within PSUs.
Weights must represent population totals for deff to be correct when
using an FPC. Note: deft is invariant to the scale of weights.
UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services
The content of this web site should not be construed as an endorsement of any particular web site, book, or software product by the University of California