On Sun, 23 Jan 2000, Brendan Burchell wrote:
> ... I can't help wondering whether [the definition of statistics] is
> only an issue at all
> for a minority of people who consider themselves to be (professional)
> statisticians. Presumably of those who graduate in "statistics", the vast
> majority don't work within the social/political domain at all, but analyse
> data in the realms of biology, physics, signal detection, astronomy, etc
> where these debates would be meaningless?
I agree the debate need not be pushed too far, and there is much in
favour of a postmodern position that words are defined through use.
The confusion over usage does, however, highlight the widespread public
confusion and distrust of statistics. It will be helpful to raise the
profile of the discipline and to expose misuses whenever possible.
One should be as confident that a data analysis has been carried out as
carefully and correctly as a chemical analysis or taxonomic
identification would be; this is blatantly often not the case.
An immediate example came from Any Questions (BBC R4, 21/22 Jan 2000).
I'm open to correction as this is based only on my hearing it once.
In the discussion around the proposed restriction of the right to jury
trial, Gerald Kaufman quoted that "there are 1.8 million cases that come
to court each year and this measure will affect only 18,500 defendants."
I phoned Any Answers to point out that the two figures are not
comparable, but I did not know how many "defendants" came to court each
year - it would be somewhat alarming if about 1 in 35 citizens were taken
to court in a year! I described myself as a statistician. In the event,
I think they filled Any Answers with other topics, so I draw no inference
from not being called back.
R. Allan Reese Email: [log in to unmask]
Associate Manager Direct voice: +44 1482 466845
Graduate Research Institute Voice messages: +44 1482 466844
Hull University, Hull HU6 7RX, UK. Fax: +44 1482 466846
====================================================================
Hull Univ - ranked equal to Oxford for research productivity.
Higher Education Funding Council tables, 3/12/99
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|