Skip to content

Zhang 2011 (re)analysis

2 messages · Reinhold Kliegl, Ben Bolker

#
I compared the respiratory data with the original publication; Davis,
C.S. (1991, Statistics in Medicine, 10, 1959-1980). From this it
appears that
data("respiratory", package = "geepack")
contains exactly the data reported in Appendix I of Davis (1991).

Note that for work with lme4, it is recommended that unique id numbers
are assigned to subjects of the second center, e.g.:
respiratory$id <- ifelse(respiratory$center==1, respiratory$id,
respiratory$id+56)

The second version of these data:
data("respiratory", package = "HSAUR")
differs from the original ones in two respects:
(1) male and female labels are exchanged
(2) for record 428, the outcome/status should be "poor" (0), not "good" (1)

So now we should be ready to replicate Zhang et al. (2011) :)

Reinhold

PS: The page numbers for Zhang et al. (2011) are 2562-2572.
PPS:  The Everitt & Hothorn-Book was published 2006 (not 2002, as I
wrote earlier); the HSAUR package goes with this book.
On Mon, Oct 31, 2011 at 6:44 AM, Ben Bolker <bbolker at gmail.com> wrote:
#
On 11-11-01 06:21 AM, Reinhold Kliegl wrote:
Another way to do this is to make a new "id nested within center"
variable, i.e. idctr <- interaction(id,center) ...
thanks!

  Ben Bolker