> -----Original Message-----
> From: Coomaren Vencatasawmy

> ..... please find a list ordered by decreasing interest.
>
> 1) Chatfield, C. 1991. Avoiding statistical pitfalls.
> Statistical Science,
> 6, 240-268.
> 2) Nelder, J.A. 1999. Statistics for the millennium. The
> Statistician, 48
> (2), 257-269.
> 3) Lindley, D.V. 2000. The philosophy of statistics. The
> Statistician, 49
> (3), 293-337.
> 4) Hand, D.J. 1998. Breaking misconceptions - statistics and its
> relationship to mathematics, The Statistician, 47(2), 245-250.

A very useful selection.   Most of these seem to focus on the relationship
of statistics to mathematics.  And that focus helps to clarify how
statistics as a social science might be defined.

Nelder adopts a hard line mathematical approach and wants to redefine
statistics as statistical science being mainly concerned with the results of
experiments.  That is of course the opposite of statistics as a social
science.   Experiments have very little place in mainstream social science
that is concerned with the study of society.

Nelder does put the matter very clearly in pointing out that mathematics is
deductive system with no necessary relationship to the real world.    But he
asserts that statistics is by contrast 'concerned with inferences from data
from the real world'.    For Nelder the real world is the results of
experiments.   But for statistics as a social science would recognise that
real world data comprises artefacts that are the products of human
activities and the way we think about them.

The most important single example is economic statistics.   Economic
activity is defined artefactually (actually by the System of National
Accounts) in terms of activities where there is a contractual relationship
between buyer and seller.    And that economic reality is what we all
suppose makes the world go round!

But of course the world of reality displayed by SNA economic statistics is
only part of the real world.    So should not statistics as a social science
have special concern with data that is related to economic activity but
which is not part of economic activity as conventionally measured?    Easy
to think of examples - unpaid domestic work, education, depletion of natural
resources, creation of environmental problems, etc. etc.

I think this example demonstrates that statistics as a social science would
go in a very different direction from statistics as the analysis of data.
Statistics as social science would be primarily concerned with matters such
as the coverage and meaningfulness of data in giving a picture of human
activities.

Ray Thomas, Social Sciences, Open University
Tel: 01908 679081 Fax 01908 550401
Email: [log in to unmask]
35 Passmore, Milton Keynes MK6 3DY