[Rasch] Fwd: Re: Fwd: CCA query

Mike Linacre rmt at rasch.org
Tue Apr 24 04:55:58 EST 2012


Thank you for asking for our suggestions, Michael.

How about investigating DPF (Differential Person Functioning). We expect 
that random behavior will cause a considerable drop in person measure? Also 
compute a person-by-stage misfit statistic. We would expect misfit to 
considerably increase in the presence of random behavior.

Mike Linacre





More information about the Rasch mailing list