R: Examples for 'survey::subset.survey.design'

### ** Examples

data(fpc)
dfpc<-svydesign(id=~psuid,strat=~stratid,weight=~weight,data=fpc,nest=TRUE)
dsub<-subset(dfpc,x>4)
summary(dsub)

Stratified Independent Sampling design (with replacement)
subset(dfpc, x > 4)
Probabilities:
   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
 0.2500  0.2708  0.3333  0.3056  0.3333  0.3333 
Stratum Sizes: 
           1 2
obs        4 2
design.PSU 5 3
actual.PSU 4 2
Data variables:
[1] "stratid" "psuid"   "weight"  "nh"      "Nh"      "x"

svymean(~x,design=dsub)

   mean     SE
x 6.195 0.7555

## These should give the same domain estimates and standard errors
svyby(~x,~I(x>4),design=dfpc, svymean)

      I(x > 4)        x        se
FALSE    FALSE 3.314286 0.3117042
TRUE      TRUE 6.195000 0.7555129

summary(svyglm(x~I(x>4)+0,design=dfpc))

Call:
svyglm(formula = x ~ I(x > 4) + 0, design = dfpc)

Survey design:
svydesign(id = ~psuid, strat = ~stratid, weight = ~weight, data = fpc, 
    nest = TRUE)

Coefficients:
              Estimate Std. Error t value Pr(>|t|)    
I(x > 4)FALSE   3.3143     0.3117   10.63 0.000127 ***
I(x > 4)TRUE    6.1950     0.7555    8.20 0.000439 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for gaussian family taken to be 2.557379)

Number of Fisher Scoring iterations: 2

data(api)
dclus1<-svydesign(id=~dnum, weights=~pw, data=apiclus1, fpc=~fpc)
rclus1<-as.svrepdesign(dclus1)
svymean(~enroll, subset(dclus1, sch.wide=="Yes" & comp.imp=="Yes"))

         mean     SE
enroll 534.56 36.248

svymean(~enroll, subset(rclus1, sch.wide=="Yes" & comp.imp=="Yes"))

         mean     SE
enroll 534.56 40.398

Examples for 'survey::subset.survey.design'

Subset of survey