I'm confused by the semantic turn in this discussion. It is not, for
example, within normal grammatical usage to suggest the use of a capital
letter should specify the meaning of a word.
It would be more reasonable to suggest that the use of statistics as a
singular noun to mean statistical method should be regarded as
ungrammatical! Economics and econonomical methods have quite different
meanings. Why should we treat the meaning of statistics and statistical
method as synonymous?
I'm even more confused by the suggestion that we should use the term data
instead of statistics in its plural sense. But the point of substance in
Julian Wells' message is to say:
> My original motive was to point up what strikes me as an
> omission, from both
> this list and from RadStats journal, of an important task, namely the
> critique of statistical method.
I would argue that use of 'statistical method' is optional and easily
challengable. If we disagree with a method being used we can challenge by
using a different method on the same material. But it is much more
difficult to challenge statistics (used as a plural noun here and later in
this message). There aren't usually any alternatives. Only one
organization can produce census statistics, for example.
I would applaud the work of Radstats in appraising statistics in the Journal
and in books like Statistics and Society. But I think this work is fairly
minuscule in relation to the unjustified influence of statistics.
Economic statistics, for example, dominate our individual lives and our
economy and our culture. But economic and associated statistics like
statistics of 'real income', or statistics that appear under the heading
'standards of living' are highly misleading. The ONS publishes such
material without qualification. And I don't see many serious appraisals -
inside or outside Radstats publications.
Ray Thomas, Social Sciences, Open University
Tel: 01908 679081 Fax 01908 550401
Email: [log in to unmask]
35 Passmore, Milton Keynes MK6 3DY
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|