[rkward-users] Variable Class and IRT

[Thomas Friedrichsmeier]

Hi,

On Tuesday 28 December 2010, Craig Johnson wrote:
> New user to R and RkWard so I'm trying to work through things.  One of the
> things I'm trying to do is play around with the IRT features.  The problem
> I'm having is that I don't have the proper "class" to run the statistics. 
> I can change the variable type (integer, factor, string, etc) but I can't
> figure out how to change the class.  Any suggestions would be helpful.

could you give a bit more detail, exactly which feature you are trying to use, 
and exactly what the symptoms are?

The model fitting features should accept objects of class "matrix" or 
"data.frame". Chances are that you really do have a data.frame. (Note: A 
data.frame is an entire "set" of data, not a single variable). Otherwise you 
could try convering to one on the R Console:
  some.name <- as.data.frame (x)

The tests work on fitted models. Here you'd select an object as produced by one 
of the model fitting dialogs (e.g. "estimates.rasch").

Regards
Thomas
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/rkward-users/attachments/20101229/8aa748bc/attachment.sig>