Stat Computing > Stata > FAQ				Help the Stat Consulting Group by giving a gift
Loading

Stata FAQ
How do I analyze survey data with a one-stage cluster design?

The examples below use Stata 9.  If you are using Stata versions 7 or 8, please see this page.

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 247 simple one-stage cluster sampling

use http://www.ats.ucla.edu/stat/books/sop/tab9_1a.dta, clear

svyset devlpmnt [pweight=wt1], fpc(M)

      pweight: wt1
          VCE: linearized
     Strata 1: <one>
         SU 1: devlpmnt
        FPC 1: M

svy: total NVSTNRS NGE65
(running total on estimation sample)

Survey: Total estimation

Number of strata =       1          Number of obs    =      40
Number of PSUs   =       2          Population size  =     100
                                    Design df        =       1

--------------------------------------------------------------
             |             Linearized
             |      Total   Std. Err.     [95% Conf. Interval]
-------------+------------------------------------------------
     NVSTNRS |       57.5   1.936492      32.89454    82.10546
       NGE65 |      167.5   1.936492      142.8945    192.1055
--------------------------------------------------------------

svy: mean NVSTNRS hhneedvn
(running mean on estimation sample)

Survey: Mean estimation

Number of strata =       1          Number of obs    =      40
Number of PSUs   =       2          Population size  =     100
                                    Design df        =       1

--------------------------------------------------------------
             |             Linearized
             |       Mean   Std. Err.     [95% Conf. Interval]
-------------+------------------------------------------------
     NVSTNRS |       .575   .0193649      .3289454    .8210546
    hhneedvn |       .525   .0193649      .2789454    .7710546
--------------------------------------------------------------

svy: ratio NVSTNRS NGE65
(running ratio on estimation sample)

Survey: Ratio estimation

Number of strata =       1          Number of obs    =      40
Number of PSUs   =       2          Population size  =     100
                                    Design df        =       1

     _ratio_1: NVSTNRS/NGE65

--------------------------------------------------------------
             |             Linearized
             |      Ratio   Std. Err.     [95% Conf. Interval]
-------------+------------------------------------------------
    _ratio_1 |   .3432836   .0075924      .2468131    .4397541
--------------------------------------------------------------

svy: mean nge65dv
(running mean on estimation sample)

Survey: Mean estimation

Number of strata =       1          Number of obs    =      40
Number of PSUs   =       2          Population size  =     100
                                    Design df        =       1

--------------------------------------------------------------
             |             Linearized
             |       Mean   Std. Err.     [95% Conf. Interval]
-------------+------------------------------------------------
     nge65dv |       33.5   .3872983      28.57891    38.42109
--------------------------------------------------------------

This example is taken from Lehtonen and Pahkinen's Practical Methods for Design and Analysis of Complex Surveys.

page 83 Table 3.6  Estimates from a one-stage CLU sample (n = 8); the Province'91 population.

input id str clu wt ue91 lab91 
1 1 2 4 666 6016 
2 1 2 4 528 3818 
3 1 2 4 760 5919 
4 1 2 4 187 1448 
5 1 8 4 129 927 
6 1 8 4 128 819 
7 1 8 4 331 2543 
8 1 8 4 568 4011 
end

gen fpc = 32

svyset clu [pweight=wt], strata(str)

      pweight: wt
          VCE: linearized
     Strata 1: str
         SU 1: clu
        FPC 1: <zero>

svy: total ue91
(running total on estimation sample)

Survey: Total estimation

Number of strata =       1          Number of obs    =       8
Number of PSUs   =       2          Population size  =      32
                                    Design df        =       1

--------------------------------------------------------------
             |             Linearized
             |      Total   Std. Err.     [95% Conf. Interval]
-------------+------------------------------------------------
        ue91 |      13188       3940     -36874.45    63250.45
--------------------------------------------------------------

svy: ratio ue91 lab91
(running ratio on estimation sample)

Survey: Ratio estimation

Number of strata =       1          Number of obs    =       8
Number of PSUs   =       2          Population size  =      32
                                    Design df        =       1

     _ratio_1: ue91/lab91

--------------------------------------------------------------
             |             Linearized
             |      Ratio   Std. Err.     [95% Conf. Interval]
-------------+------------------------------------------------
    _ratio_1 |    .129289   .0065018      .0466761     .211902
--------------------------------------------------------------

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