[Rasch] The Rasch Rating Model and the Disordered Threshold Controversy

Jeremy Hobart jeremy.hobart at pms.ac.uk
Sat Jul 7 20:48:17 EST 2012

Hi Mike,

Doesn't  the note you quote from David Andrich relate to a somewhat different issue than is being implied here


On 7 Jul 2012, at 10:33, "Mike Linacre" <rmt at rasch.org<mailto:rmt at rasch.org>> wrote:

Jeremy and Folks:

Let's cut the Gordian knot ....

Disordered thresholds signify disordered modal probabilities for the categories, not substantively disordered categories nor disordered raw scores nor disordered ability, difficulty, leniency, ..., measures.

Please see my note at http://www.rasch.org/rmt/rmt131a.htm (do try that experiment yourself), and also David Andrich's note at http://www.rasch.org/rmt/rmt251c.htm (do try that experiment for yourself).

This is proved mathematically in the cited paper (in the current issue of Educational and Psychological Measurement) .


Mike L.

At 7/7/2012, Jeremy Hobart wrote:
Is it really controversial that a higher "score" on a rating scale ought to mean more of the problem being measured than a lower score? If so 4cm is actually shorter than 3cm.

Please can you circulate the full reference as well as the naked abstract, (I didnt get the full ref). Can the authors please make the PDF available so that we can make a judgement?

Also, I'm sure David Andrich has some comments to make. It would be be good to see those.

Mike Linacre
rmt at rasch.org<mailto:rmt at rasch.org> www.rasch.org/rmt/<http://www.rasch.org/rmt/> Latest RMT: 25:4 Spring 2012

More information about the Rasch mailing list