[Rasch] Not a Fan of Lexiles?

Stephen Humphry stephen.humphry at uwa.edu.au
Fri Nov 2 15:52:17 EST 2007


Mike:
 
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?
 
Paul:
 
And physicists, cooks, and carpenters can all manipulate/equate magnitudes
of their variables empirically - which makes a big difference to evaluating
the veracity of a "measurement" application.
 
But this kind of discussion "look, all things Rasch is wonderful" -vs- my
"is it really" response  is going nowhere. 
 
****************************************************************************
*********************************************
 
Hi Paul.
 
I wouldn't characterize the discussion that way -- I think some nice points
have been made by Mike Linacre, Andrew Kyngdon and others. Although I
disagree with what you say, it is great to have discussions like this one,
and I resepct that you speak your mind.
 
Here is a statement by Joel Michell about the importance of understanding
measurement, in psychology ...
 
"Psychology might be on the way to becoming a successful quantitative
science, but as a body of workable, quantitative theories and laws, it is so
far short of the example set by physics that no one yet has a clear idea of
what a successful quantitative psychology would look like.  The history of
science teaches us many things, but I do not think that one of them is that
we can expect to make progress by ignoring pertinent matters." Michell
(1999, p. 217).
 
There are some questions we might ask about the role of measurement in
science. Can measurement be successfully achieved in physics, in general,
without experimental work? -- i.e. the measurement of all of the quantities,
not just one any particular one. Is it a given that an instrument will
measure, or is this is something that needs to be experimentally established
and verified? Can the measurement of all kinds of quantitative attributes be
achieved without a theory, or at least a clear definition? For example, what
about the measurement of temperature, or even force?
 
We might also debate what is required to establish whether something has
quantitative structure. However, that aside, the question is: should these
things be treated as hypotheses? One of the most important things Rasch did,
in my view, was to clarify the sense in which it is a scientific hypothesis
that measurements are actually such, by identifying a fundamental
requirement of measurement compatible with physical definitions and laws;
i.e. invariance. He analysed and described this requirement in terms of
Newton's second law (e.g. Rasch, 1960, pp. 110-114), as well as other
physical law. I agree that scientific work will not be accomplished by
supposing that the application of a measurement model will solve all
problems in and of itself. However, I think Rasch developed a general
probabilistic framework and basis for experimental work that needs to occur
to measure quantitative attributes.
 
There is no doubt substantive considerations are equally important as
measurement requirements, and in practice, it is possible to lose sight of
substantive considerations if there is too narrow a focus on measurement
models to the exclusion of substantive phenomena. As made clear in Kuhn's
(1661) analysis, however, I also think it is a serious mistake to think that
all experimental work happens after successful measurement. I also believe
this is central to debates like this one.
 
Kind regards,

Steve
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20071102/30ea56f1/attachment.html 


More information about the Rasch mailing list