[Rasch] Separability of parameters

Thomas Salzberger Thomas.Salzberger at wu-wien.ac.at
Thu Sep 28 02:46:12 EST 2006

At 18:39 27.09.2006, Snider-Lotz, Tom wrote:
>Hello everybody -
>I recently read a manuscript that said the Rasch model was the only one
>that featured separability of parameters.  One inference I make from
>that statement is that, for 2PL and 3PL models, the IRT item parameters
>are somewhat dependent of the sample on which they are developed.  Is
>that correct, or do I misunderstand this (as I often do).


yes, you are right, insofar as all IRT models with intersecting item 
response functions (IRFs) do not feature separability.

In the Rasch model, the estimation equation of person parameters do 
not contain the item parameters and vice versa (they cancel out). 
That is not the case with multi-parameter IRT models.

However, any model that defines parallel IRFs would, as far as I 
know, yield separability.

Kind regards,
Thomas Salzberger
Dr. Thomas Salzberger
Department of Marketing, Institute for Marketing Management
University of Economics and Business Administration (WU-Wien)
Augasse 2-6 / A-1090 Wien / Austria, Europe
Email: Thomas.Salzberger at wu-wien.ac.at
Tel: (+43-1-) 31336 4406 (Voicemail)
Mobile Mailbox: (+43-) 676 225510933 (Voicemail only)
Personal Fax: (+43-1-) 31336 90 4406
Fax Institute: (+43-1-)  31336 732
WWW Personal: http://www2.wu-wien.ac.at/marketing/user/salzberger/
WWW Institute: http://www2.wu-wien.ac.at/marketing/
WWW The Matilda Bay Club: http://www.matildabayclub.net  

More information about the Rasch mailing list