survey (version 0.9-1)

svyglm: Survey-weighted generalised linear models.

Description

Fit a generalised linear model to data from a complex survey design, with inverse-probability weighting and with standard errors corrected for cluster sampling.

Usage

svyglm(formula, design, ...)

Arguments

formula

Model formula

design

survey.design object. Must contain all variables in the formula

...

Other arguments passed to glm

Value

• An object of class svyglm.

Details

Standard errors for cluster-sampled designs are computed using a linearisation estimate (equivalent to the Huber/White sandwich formula in GEEs). This does not incorporate the finite-population correction, making standard errors potentially conservative.

References

~put references to the literature/web site here ~

See Also

svydesign, glm

Examples

## Independent sampling
  df<-data.frame(x=rnorm(1000),z=rep(0:4,200))
  df$y<-with(df, 3+3*x*z)
  df$p<-with(df, exp(x)/(1+exp(x)))
  mpop<-lm(y~x+z, data=df) 
  xi<-rbinom(1000,1,df$p)
  sdf<-df[xi==1,]

  dxi<-svydesign(~0,~p,data=sdf)
  m<-svyglm(y~x+z,data=sdf,family="gaussian",design=dxi)
  m1<-lm(y~x+z, data=sdf)

  summary(m1) ##wrong
  summary(m)  ##right
  summary(mpop) ##whole population

  ##cluster sampling
  df$id<-rep(1:250,each=4)
  df$clustp<-by(df,list(df$id),function(d) min(exp(d$x*d$z)/(1+exp(d$x*d$z))))[df$id]
  mpop<-lm(y~x+z, data=df) 

  xi<-rbinom(250,1,df$clustp[4*(1:250)])
  sdf<-df[xi[df$id]==1,]

  dxi<-svydesign(~id,~clustp,data=sdf)
  m<-svyglm(y~x+z,data=sdf,family="gaussian",design=dxi)
  m1<-lm(y~x+z,data=sdf)
 
  summary(m1) ##wrong
  summary(m)  ##right
  summary(mpop) ##whole population