[Rasch] Re: Rasch-Poisson

Anthony James luckyantonio2003 at yahoo.com
Sat May 23 02:58:16 EST 2009

Thanks for your reply Mike,
What's the equation that I wrote in my previous email then?
Is there any way to get ratio scale measures with winsteps (or  by hand) according to that equation?
How can one convert an interval scale logit measure of say, 1.5 to a ratio measure according to the equation in my email? Will  it be again in logits? 
Is Poisson a mathematician?

--- On Thu, 5/21/09, Mike Linacre (RMT) <rmt at rasch.org> wrote:

From: Mike Linacre (RMT) <rmt at rasch.org>
Subject: Re: [Rasch] Rasch-Poisson
To: rasch at acer.edu.au
Date: Thursday, May 21, 2009, 7:23 PM

Thank you for asking us about Poisson, Anthony.

The Rasch-dichomous model can be written:

Log (Pni1 / Pni0) = Bn - Di

A Rasch-Poisson model is:

Log (Pnix / Pni(x-1)) = Bn - Di - log(x)

where x is a category on a rating scale from 0 to infinity. Since log(x) linearizes the x's, the x's must be on a ratio scale.

You can see that the Rasch-Poisson model becomes the Rasch-dichotomous model if the rating scale is limited to 0,1.

For more technical material, see Georg Rasch at www.rasch.org/memo18.htm

Mike L.

At 5/21/2009, you wrote:

A while ago I heard about Rasch-Poisson model which supports a ratio scale.
Mike Linacre
Editor, Rasch Measurement Transactions
rmt at rasch.org www.rasch.org/rmt/ Latest RMT:  22:4 Spring 2009 
-----Inline Attachment Follows-----

Rasch mailing list
Rasch at acer.edu.au


Please consider the environment before you print

-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20090522/46484a5d/attachment.html 

More information about the Rasch mailing list