[Rasch] qualitatively described, quantitatively responded

Trevor Bond trevor.bond at jcu.edu.au
Wed Nov 7 11:52:10 EST 2007

Shirin's problem is now much clearer and the advice much more applicable

Augustin's advice is important:
Once you have a meaning of the measures, proceed on reverse 
direction, try to identify the construct involved with the positions 
on the scale and define from less to more the best suitable names, 
but not only: poor, medium, good, excellent and so forth, but include 
a description, such as:
a) excellent is able to write phrases with a correct syntax, make 
interpretacions, propose conclusions, etc.
b) poor is able to write incoherent phrases, bad use of construction 
rules, etc.
hope this helps

BUT, I would be not be surprised at all to find that list members 
could provide Shirin with exemplars of scalable scoring systems for 
essay writing. Even if those scoring rubrics do not solve the 
problem, they will certainly help....references, pls colleagues?

and remember:
Even if some clever professor can develop a 9-level hierarchy of 
Grammar usage (for example);
it is unlikely that all those fine distinctions will be applicable in 
the essays you are grading and
unlikely examiners will  be able to judge reliably between the 9 
categories, anyhow.

Another note: although Andrich's original work set out to solve the 
essay scoring problem in particular (if my memory serves me well), I 
would venture that the PCM rather than the RSM is now used more 
frequently with essay scoring. (I am not referring to the thresholds 


Trevor G BOND Ph D
Professor and Head of Dept
Educational Psychology, Counselling & Learning Needs
D2-2F-01A EPCL Dept.
Hong Kong Institute of Education
10 Lo Ping Rd, Tai Po
New Territories HONG KONG

Voice: (852) 2948 8473
Fax:  (852) 2948 7983

More information about the Rasch mailing list