[Rasch] Negative pt-bis and fit of 1.0? How can this be?
rmt at rasch.org
Wed Mar 7 11:28:48 EST 2012
The HosmerLemeshow test -
http://en.wikipedia.org/wiki/Hosmer%E2%80%93Lemeshow_test - looks like a
fit test for a non-uniform ICC. Is this what you mean or something else?
At 3/6/2012, you wrote:
>Another option we use in our software Kalt is a model similar to the one
>proposed by Hosmer and Lemeshow for ligistic regression. It works fine and
>do not fall in the contradictory situation mentioned by Mark.
Editor, Rasch Measurement Transactions
rmt at rasch.org www.rasch.org/rmt/ Latest RMT: 25:3 Winter 2011
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Rasch