[Rasch] Rasch person reliability and coefficient alpha

Mike Linacre (RMT) rmt at rasch.org
Sat May 31 08:02:38 EST 2008


They are reporting Cronbach Alpha. They probably have many extreme scores 
on their 4 items. This would produce very high reliability 
(reproducibility) for raw scores, but very low reliability 
(reproducibility) for Rasch measures.

In physical measurement terms, we are very sure the basketball-players are 
taller than the tape measure (Alpha), but we are very unsure what their 
heights actually are (Rasch).

Mike L.

At 5/30/2008, you wrote:
>they say the items have an alpha of .84. Yet when I Rasch analyze them, I 
>get a reliability of 0.00.

Mike Linacre
Editor, Rasch Measurement Transactions
rmt at rasch.org www.rasch.org/rmt/ Latest RMT:  21:4 Spring 2008
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20080530/e53eb3ea/attachment.html 

More information about the Rasch mailing list