I agree with Dietrich. One reason for my agreement is that the use of the
word statists as a singular noun (meaning statistical method) does not
correspond with the everyday meaning of the word.
In everyday use statistics is a plural noun, and it means the figures that
millions of people read about or work with. They would not recognise the
term 'data' as applying to, say, the latest increase reported in the RPI, or
in the level of unemployment, or in checking their earnings, or monitoring
the activities of their employer in terms of sales, output, etc.
But this reason is symptomatic of a rather more fundamental problem.
Statisticians in using statistics to mean statistical method tend to be
blind to the fact that statistics, e.g. official statistics, have qualities
other than numerical values. Many, if not most statisticians are rather
like 'cheerful robots' (to use Wright Mills' classic phrase) in their
handling of statistics. They are uncritical of the origins and purposes of
the statistics they use or help produce. They expect to be able to learn
something just by manipulating the numerical values given.
The restriction of the meaning of statistics to statistical method has also
been fairly disastrous for the development of the social sciences in Britain
in recent decades. At present we have a government that has declared that
it expects to be assessed on the basis of statistics relating to its
performance. That is a historically unprecedented declaration. But there
has been little reaction from academic social scientists - except for a few
who have jumped on to the quantification bandwagon.
Ray Thomas, Social Sciences, Open University
Tel: 01908 679081 Fax 01908 550401
Email: [log in to unmask]
35 Passmore, Milton Keynes MK6 3DY
> -----Original Message-----
> From: Dietrich Alte [mailto:[log in to unmask]]
> Sent: 04 January 2000 11:59
> Cc: [log in to unmask]
> Subject: statistics and data
>
>
> julian.p.wells wrote
>
> > Incidentally, it has often struck me that, given the
> > contents of the journal, Radical Statistics is something of
> a misnomer:
> > "Radical Data" would capture its nature more accurately.
> >
>
> the word "statistics" has surely more than one meaning. one of it -
> and probably older than most others - is one that comes from the latin
> "status" = state (political unit, conditional, physical etc.) and has
> got something from "description of the state of the state" in it -
> which is usually done with data on social, economic, political etc.
> constructs.
> as i understood radstats, it is _this_ meaning of data (and
> despriptive+inferential statistics) that the "stats" in the list
> focusses on.
>
> -------------------------------------------------------------
> Dietrich Alte, Diplom-Statistiker
>
> Universitaet Greifswald
> Institut fuer Epidemiologie und Sozialmedizin
> Organisationszentrum Community Medicine
> Walther-Rathenau-Str. 48, D-17487 Greifswald, Germany
> phone +49 (0) 3834 - 86 77 13, fax +49 (0) 3834 - 86 66 84
> email [log in to unmask]
> -------------------------------------------------------------
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|