equating of MIRT models created with the mirt package

[Mike Altieri]

Hi all,

I would like to equate/link three forms of a test, each modeled by a three dimensional 2PL model with the mirt package: mirt(mzp_i_tasks,3,SE=TRUE,itemtype = "2PL"), i=1,2,3, by an anchor test design.

Could anybody give me a hint which R package can do this (I know about the package equateIRT but it does not seem to support multidimensional models and the new plink package which seems to be no longer connectable to mirt via read.mirt because this function is missing in the mirt 1.24 release - please correct me if I'm wrong)?

Any help is highly appreciated. Thank's in advance!

Mike 

[Phil Chalmers]

I'm not overly aware of reliable equating techniques for exploratory IRT models, nor would I really recommend them anyway. Your best be will likely be to equate the models using the multipleGroup() function after first selecting some anchor items/parameters and freely estimating the focal group hyper-parameters. But, I would really recommend using a confirmatory IRT model first though, because for EFA models you cannot freely estimate the latent means and variances required to do the equating. 

Phil

--
You received this message because you are subscribed to the Google Groups "mirt-package" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mirt-package+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.