full mirt equating

165 views
Skip to first unread message

somayeh Bahmanabadi

unread,
Sep 15, 2019, 9:21:47 AM9/15/19
to mirt-package
Hi dear all
Can I run multidimensional IRT equating true score with mirt package?

Phil Chalmers

unread,
Sep 15, 2019, 9:53:14 AM9/15/19
to somayeh Bahmanabadi, mirt-package
Yes. You would need to freely estimate the variance components and mean structure in the focal group using the multipleGroup() function, which would require at least D suitably selected anchor items (where D is the dimensionality of the model). 
 
Phil


On Sun, Sep 15, 2019 at 9:21 AM somayeh Bahmanabadi <s.bamna...@gmail.com> wrote:
Hi dear all
Can I run multidimensional IRT equating true score with mirt package?

--
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...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mirt-package/103e86d1-c055-4cce-9728-a1d4589ce979%40googlegroups.com.

somayeh Bahmanabadi

unread,
Sep 16, 2019, 5:28:24 AM9/16/19
to mirt-package
Hi dear Phil 
thanks for your response.
my data design is equivalent groups, and I dont have any anchor test.
can I run FULL MIRT observe score equating with multipleGroup() function, too?
can you send me an guideline for FULL MIRT observe score equating, please. 
thank you

Phil Chalmers

unread,
Sep 16, 2019, 7:05:27 AM9/16/19
to somayeh Bahmanabadi, mirt-package
The mirt package does not include observed score equating methods, you'll have to look elsewhere. Though note that using observed scores for equating is effectively the same as treating all the items as anchors, so the previously described approach would be just as valid in this setting. 

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...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages