[Rasch] Not a Fan of Lexiles?
Mike Linacre (RMT)
rmt at rasch.org
Fri Nov 2 11:58:50 EST 2007
Paul writes: "Once you treat data models as "optional", and focus on the
problem to be solved rather than forcing a single data model to be applied
to it, all kinds of analysis and solution options open up."
This is a data-dominated approach - the approach preferred by Fred Lord:
"Building statistical models is just like this. You take a real situation
with real data, messy as this is, and build a model that works to explain
the behavior of real data."
Martha Stocking's summary of Item Response Theory, the statistical
methodology of Frederic M. Lord (1912-2000), New York Times, 2-10-2000
That would seem to be the best approach, but in the search for the Top
Quark, physicists went to expensive extremes in "forcing a single data
model to be applied". There was great rejoicing when they succeeded. They
didn't want "more solutions" that would explain all the messy data they
obtained from the millions of experiments they conducted that didn't
produce a top quark.
Physicists, cooks and carpenters all have a "model" of what they want, and
they do their best to obtain it - even if it means specifying, selecting,
constructing the data that conforms to their models. But social scientists,
in general, reject this approach. But which approach has been more productive?
Editor, Rasch Measurement Transactions
rmt at rasch.org www.rasch.org/rmt/ Latest RMT: 21:1 Summer 2007
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Rasch