Given below is a report from the Financial Times about the dispute between
the ONS and Westminster council.   The report is self-explanatory, but I
want to draw attention to the final sentence that states that the ONS has
refused to release 'the crude numbers' that it collected for Westminster.

One point to make is that what are called in this report 'crude numbers'
have for at least the past fifty years or so been called 'preliminary
estimates' and have been published within 
a few months of the Census date.  Here the publication of these statistics
is being refused nearly two years later.

The reason for non-release given that 'it would undermine the estimation
process' seems a direct contravention of the National Statistics Code of
Practice.   The Draft Code stated that '...subject to any confidentiality
undertakingings - aggregate or micro-data contributing to analysis or
outputs will be made accessible to users?'   The Approved Code released last
year included the pledge to make 'all record open to scrutiny on request ..'
(p 16).  Many other clauses exude a spirit of openness that seems contrary
to this refusal to  supply basic information.  

Isn't this a body blow to the idea of an statistical service independent of
government?   
Isn't the ONS using its governmental authority to cover up what seem to be a
mistake?

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

FINANCIAL TIMES 8 Jan 2002

The population census figure is an important component of the formula that
determines local authority grants from the government. Westminster feels it
has "compelling evidence of underestimation of its relevant population".
But, despite presenting a portfolio of supporting information to the Office
for National Statistics, it has had no feedback on the ONS numbers. 

Manchester city council, another of the most affected councils, is also
"vigorously pursuing a review of the census figures". 

Westminster says it will lose "hundreds of millions of pounds over the next
decade" if the benchmark population figures are not changed. 

The council points out that, while its allocation is protected in the coming
year by the floor placed under council budgets, it would have fallen by
Pounds 63m if the census figures were relied upon. 

The population is now recorded as being 181,000, a decrease of 26 per cent
on the previous estimate made by the ONS. The council wrote to Mr Cook
yesterday, frustrated at the lack of progress it had made since first
raising concerns following the publication of the census results last
September. 

Nick Raynsford, local government minister, had agreed to reconsid
er the proposed grant allocation if he could be persuaded - before the end
of the consultation period next Tuesday - that the census methodology was
flawed. According to electoral registers and the numbers registered with
doctors in Westminster, numbers rose during the 1990s but the census figures
imply a falling population since 1981. 

The census has recorded fewer households than the council has people paying
council tax. 

Allowing for households avoiding the tax and multiple occupation properties,
in addition to asylum seekers and illegal immigrants who will often choose
to avoid being counted, the council feels there must be even more who have
not shown up in the figures. 

The ONS has refused to release the crude numbers relating to Westminster,
such as how many households were counted in the area, as it feels it would
undermine the estimation process used to get to the "final" figures
published last September.
{Copyright 2003 The Financial Times Limited;   Financial Times (London);
January 8, 2003.
******************************************

******************************************************
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]
*******************************************************