On Thu, 26 Jan 2012, James Holland wrote:
I'm trying to kick off some of the rust and learn some of the R packages
for Rasch modeling. When I tried using the eRm package, I get item
difficulty estimates for all my items accept the first (in terms of order)
item.
#Begin code
library(eRm)
r.simulation - sim.rasch(20,100)
r.data - r.simulation$items
#eRm results
erm.rasch - RM(r.data)
names(erm.rasch)
erm.items - erm.rasch$etapar
erm.items
#end code
All items accept V1 (the first item in the list) show up.
As explained on the manual page ?RM, one of the parameters is not
identified (as the Rasch scale is latent with arbitrary 0). Either you can
restrict the sum of all item parameters to zero (default) or the first
item parameter. In both cases, the print output of the model omits the
first item parameter. But coef(erm.rasch) will give you the appropriate
expanded parameter vector.
See also vignette(eRm, package = eRm) for more details.
Z
James
[[alternative HTML version deleted]]
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
__
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
