# [R] stratified sampling from known population datafile

Steven Gorlé steven.gorle at telenet.be
Wed Jan 3 22:29:05 CET 2007

```Dear R-wizards,

I have a population from which I want to draw a stratified sample by region.

In  Venables and Ripley "Modern Applied statistics with S" I found some
great procedures for Simple Random Sampling (with and without replacement)
and for Systematic sampling and it works!

For stratified sampling I referred to the manual of the survey package.Are
there any other papers available on this subject?
Is the output correct? And how can I draw a random (stratified by region)
sample from my (population) datafile bmi?

dstrat<-svydesign(id=~1,strata=~REGIONCH, data=bmi)
Warning in svydesign(id = ~1, strata = ~REGIONCH, data = bmi) :
No weights or probabilities supplied, assuming equal probability
> summary(dstrat)
Stratified Independent Sampling design (with replacement)
svydesign(id = ~1, strata = ~REGIONCH, data = bmi)
Probabilities:
Min. 1st Qu.  Median    Mean 3rd Qu.    Max.
1       1       1       1       1       1
Stratum Sizes:
Brussels  Flanders  Walloonia
obs             2571      2987      3006
design.PSU      2571      2987      3006
actual.PSU      2571      2987      3006
Data variables:
[1] "ID"       "WFIN"     "HH"       "REGION"   "EDU3"     "FA3"
[7] "TA2"      "AGE7"     "SEX"      "VOEG"     "BMI"      "LNBMI"
[13] "LNVOEG"   "FLA"      "BRU"      "WAL"      "AGEGR1"   "AGEGR2"
[19] "AGEGR3"   "AGEGR4"   "AGEGR5"   "AGEGR6"   "AGEGR7"   "EDUPRIM"
[25] "EDUSEC"   "EDUHIGH"  "INCLOW"   "INCMED"   "INCHIG"   "REGIONCH"
[31] "PROVINCE" "SGP"      "GHQ12"    "GHQBIN"
> svymean(~BMI, dstrat)
mean SE
BMI   NA NA