Un texte encapsul? et encod? dans un jeu de caract?res inconnu a ?t? nettoy?... Nom : non disponible Url : https://stat.ethz.ch/pipermail/r-help/attachments/20060924/fd6c0fc0/attachment.pl
On Sun, 24 Sep 2006, justin bem wrote:> Hi dear all, > > I have a complex survey data to analyse. I have a stratified survey with three > stages level. The first stage is a PPS, but all other are SRSWOR. The solution >propose in survey package is to considere PPS as SRS with repetion.No, the solution proposed is to consider it as PPS with replacement.> I have try this > > E3stage<-svydesign(ids=~stage1id+stage2id+stage3id, fpc=~fpc1+fpc2+fpc3, ...) > > This is the value for WOR sampling at all stage. What value should I give to >fpc1 to have the same value of variance that in sampling with repetion ? I try >fpc1<-rep(0,length(data)) I get a error message.For sampling with replacement you just omit the fpc and specify the first-stage id and the sampling weights E3stage <-svydesign(ids=~stageid1, weights=~weights) -thomas Thomas Lumley Assoc. Professor, Biostatistics tlumley at u.washington.edu University of Washington, Seattle
Un texte encapsul? et encod? dans un jeu de caract?res inconnu a ?t? nettoy?... Nom : non disponible Url : https://stat.ethz.ch/pipermail/r-help/attachments/20061001/dab1b2f2/attachment.pl
On Sun, 1 Oct 2006, justin bem wrote:> Dear Thomas ! > > Thank you for you help ! I need and explaination. If I suppress second > and third stage what become the variance du to second and third stage ? > > I have try this > >E3stage<-svydesign(ids=~stage1id, weight=~wei ...) > and > >E3stage<-svydesign(ids=~stage1id+stage2id+stage3id, weight=~weights...) > > And I get the same resultsYes, just as in the example on the help page for svydesign. When the first stage is sampling with replacement the variance estimator is based on the p-expanded cluster totals being a simple random sample from a population distribution. This is conservative, and better variance estimators are possible in principle. I am hoping to include better estimators in a future version of the package, but I need to do some more reading first -- there does not seem to be agreement on the best variance estimators (except for the impractical one that requires pairwise sampling probabilities). I would welcome suggested references (especially if they are in English). -thomas Thomas Lumley Assoc. Professor, Biostatistics tlumley at u.washington.edu University of Washington, Seattle