[Rasch] Rasch applied rashly to quash innovation
Steven L. Kramer
skramer1958 at verizon.net
Fri Dec 14 03:59:30 EST 2007
Theo: Yet another and perhaps more important question.
" I agree with your concerns. First of all, the folks who developed these tests have erred if they think that identifying a typical learning pathway with the Rasch model implies that the pathway that has been isolated is the only legitimate pathway to learning. In fact, Rasch analysis can help us identify alternative pathways by providing detailed information about individuals who "misfit" the model
Yesterday, Andrew Kyngdon of MetaMetrics announced on the Rasch list-serve that the next MetaMetrics conference will deal with their new "Quantile" scale. The Quantile scale is a formative assessment tool, used to place students on a vertical scale and tell what math they know, what might be fruitful to review, and what they are ready to learn next. They are trying to link state tests to their quantile scale, and may also have other formative tests, perhaps with some adaptive testing (I'm not sure what their plans are.)
In any case, I suspect that harm can be prevented if developers of a Quantile scale explicitly look for multiple pathways through mathematics, and place students not only at a position on a pathway, but also on the pathway that applies to them. (As a curriculum researcher, I suspect that often the cause of differing paths may not reside in the student, but rather in the instruction/curriculum to which the student is exposed.)
I have never much used or understood infit, outfit, etc. Can you perhaps provide references, or else provide a practical description of how "multiple pathways" might be developed and used?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Rasch