[Rasch] Use of e in the Rasch model

Anthony James luckyantonio2003 at yahoo.com
Fri Sep 9 18:57:09 EST 2011



Mike,

Can't we write the model as P=a-d/1+(a-d). This is additive. Why do we need
log?

I know it will be awkward when a=d but the multiplicative form has this problem
too when d=0.



Anthony



--- On Wed, 9/7/11, Mike Linacre <rmt at rasch.org> wrote:

From: Mike Linacre <rmt at rasch.org>
Subject: Re: [Rasch] Use of e in the Rasch model
To: rasch at acer.edu.au
Date: Wednesday, September 7, 2011, 7:22 AM




Anthony writes: p = (a/d)/(1+(a/d)) is within 0 to 1.


Comment: Yes, that is the multiplicative form of the Rasch model. Georg
Rasch uses it in his book, "Probabilistic Models".


If we want to transform multiplicative numbers, like a and d, into
additive numbers, then we take logarithms .... as Georg Rasch himself
does in his book.


Mike L. 

 
Mike Linacre

Editor, Rasch Measurement Transactions

rmt at rasch.org
www.rasch.org/rmt/
Latest RMT: 25:1 Summer 2011
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20110909/434937a6/attachment.html 


More information about the Rasch mailing list