These data are from Section 12.2 of Levy and Lemeshow. They describe
(a possibly apocryphal) study of survival in out-of-hospital cardiac
arrest. Two out of five ambulance stations were sampled from each of
three emergency service areas.
Usage
data(scd)
Format
This data frame contains the following columns:
ESA
Emergency Service Area (strata)
ambulance
Ambulance station (PSU)
arrests
estimated number of cardiac arrests
alive
number reaching hospital alive
Source
Levy and Lemeshow. "Sampling of Populations" (3rd edition). Wiley.
Examples
data(scd)
## survey design objects
scddes<-svydesign(data=scd, prob=~1, id=~ambulance, strata=~ESA,
nest=TRUE, fpc=rep(5,6))
scdnofpc<-svydesign(data=scd, prob=~1, id=~ambulance, strata=~ESA,
nest=TRUE)
# convert to BRR replicate weights
scd2brr <- as.svrepdesign(scdnofpc, type="BRR")
# or to Rao-Wu bootstrap
scd2boot <- as.svrepdesign(scdnofpc, type="subboot")
# use BRR replicate weights from Levy and Lemeshow
repweights<-2*cbind(c(1,0,1,0,1,0), c(1,0,0,1,0,1), c(0,1,1,0,0,1),
c(0,1,0,1,1,0))
scdrep<-svrepdesign(data=scd, type="BRR", repweights=repweights)
# ratio estimates
svyratio(~alive, ~arrests, design=scddes)
svyratio(~alive, ~arrests, design=scdnofpc)
svyratio(~alive, ~arrests, design=scd2brr)
svyratio(~alive, ~arrests, design=scd2boot)
svyratio(~alive, ~arrests, design=scdrep)
# or a logistic regression
summary(svyglm(cbind(alive,arrests-alive)~1, family=quasibinomial, design=scdnofpc))
summary(svyglm(cbind(alive,arrests-alive)~1, family=quasibinomial, design=scdrep))
# Because no sampling weights are given, can't compute design effects
# without replacement: use deff="replace"
svymean(~alive+arrests, scddes, deff=TRUE)
svymean(~alive+arrests, scddes, deff="replace")