[R] how to prepare my data to be used in the "survey" package

David Winsemius dwinsemius at comcast.net
Mon Aug 29 20:29:48 CEST 2016

> On Aug 29, 2016, at 9:48 AM, David Remotti <david at agros.it> wrote:
> Looking for some help in using the "survey" package
> I have gone through the documentation but still find problems in understanding how to prepare my data to be analyzed wih this package.
> In particular I dont understand how to prepare the fpc data.
> The documentation provides a "small survey example" where 6 variables are given:
> stratid,psuids,weight,nh,Nh and x
> My survey is a two stage stratified sampling so I have for each stratum the number of primary
> and secondary units in the sample and in the population, and using these data I can prepare such
> data frame, except for the last variable "x".
> I cant understand what does it means and which information I have to put there.
> A second (minor) problem is that my stratification is based on two variables, but it looks like it is only possible to give 1 variable as strata parameter.
> Is it possible to give two ?

This tutorial by Lumley illustrates both data preparation and the use of a two-stage clustered design:


Found with a google search on "primary sampling unit secondary sampling unit r survey package lumley"


David Winsemius
Alameda, CA, USA

More information about the R-help mailing list