[R] eRm - Rasch modeling - First item missing from estimation

[Achim Zeileis]

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 at 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.
>