On 23-May-10 15:06:47, Ray Thomas wrote:
> The President of the Royal Statistical Society has described
> statistics as THE MIRROR through which we view society.
>
> Isn't this particular analogy misleading to the point of stupidity?
> A mirror gives a full picture. The main point of statistics is to
> summarise.
>
> A mirror reflects. And what has reflection got to do with statistics?
> Ray Thomas
Of course there are also distorting mirrors, and mirrors which are
smeared and fuzzy or cracked ... and in any case a mirror gets
everything back to front!
However, I would see Statistics as more than mere summarisation.
It embraces the whole of what is involved in the selection, filtering,
transformation, enhancement and suppression of information; and the
professional statistician should be accountable for exactly what
happens in every aspect of this all the way along the chain.
For many years I have been pushing what I think is a better analogy:
between Statistics and Photography.
Consider a professional photographer, familiar with the scientific
basis of photography in all its aspects. And from the pre-digital
era.
You have a scene, you can capture the light emanating from it, and
make a picture. Along the way, you can fiddle with the light before
it hits the camera plate. Then you can fiddle with what happens to
the film when you develop it, and to the paper on which you print it.
Depth of focus, aperture, exposure time, the intensity and colour
and direction of illumination of the scene, the interposition of
colour filters, the choice of aspect from which to take the photograph,
the framing of the scene (i.e. choice of what you explicitly include
and what you leave to the viewer's imagination), the choice of film
(speed, sensitivity, grain, colour response -- or black&white),
development technique, printing technique including fancy tricks
like "insolation", and so on and so on, are all parameters of the
process by which the photographer can select and highlight what he
wants the viewer to see and suppress what he does not.
The end result is marks on a sheet of paper, which the viewer
will look at and perceive something, highly influenced by the
photographer's technical and artistic choices.
There are very direct analogies of all the above elements in Statistics,
The arranagement and illumination of the scene etc. correspond to the
design of the investigation. The light beams on their way from the
scene to the camera lends are the raw data. Subsequent elements
correspond to choice of statistical analysis and presentation.
And the end result is marks on a sheet of paper, which the viewer
will look at and perceive something, highly influenced by the
statistician's technical choices.
Of course, nowadays a naive amateur photographer can go into a shop,
and buy a consumer camera equipped with a host of pre-programmed
features and sophisticated software. This photographer can then go
out and complacently indulge in point-&-click Photography.
And a statistician can similarly indugle in point-&-click Statistics.
Best wishes to all,
Ted.
--------------------------------------------------------------------
E-Mail: (Ted Harding) <[log in to unmask]>
Fax-to-email: +44 (0)870 094 0861
Date: 23-May-10 Time: 17:51:30
------------------------------ XFMail ------------------------------
******************************************************
Please note that if you press the 'Reply' button your
message will go only to the sender of this message.
If you want to reply to the whole list, use your mailer's
'Reply-to-All' button to send your message automatically
to [log in to unmask]
Disclaimer: The messages sent to this list are the views of the sender and cannot be assumed to be representative of the range of views held by subscribers to the Radical Statistics Group. To find out more about Radical Statistics and its aims and activities and read current and past issues of our newsletter you are invited to visit our web site www.radstats.org.uk.
*******************************************************
|