I've made an addition to the Wiki in answer to the question posed by the ONS
"Does this new Code of Practice meet the aim of improving trust in official
statistics, as envisaged in the White paper "Building Trust in Statistics"?
The addition is given below.

You are welcome to respond to this addition directly by posting an amendment
to the Wikipage, at
http://www.seedwiki.com/page.cfm?Doc=RadicalStatisticsGroup&wikiid=353
or by posting a comment on this list.

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

*************************************************
1) Does this new Code of Practice meet the aim of improving trust in
official statistics, as envisaged in the White paper "Building Trust in
Statistics"?

Proposed Radstats response should point out that the Draft Code is not
clearly addressed to the public.

It is suggested that we point out that the draft Code refers to government
uses of statistics and 'users'.  But it does not mention the public except
to emphasise the importance of getting the trust of the public   How can it
Code possibly get the trust of the public if it does not attempt to address
the public?

We could give the 2001 Census as an example. Most members of the public have
forgotten about it. But they will be reminded of it in the Autumn when the
first results are published. They may well ask why the UK seems to setting a
new world record, at least among non-totalitarian societies, in late first
publication of census results. They will also note that the reason given for
the delay is the need for adjustment of the figures for underenumeration.
It is easy to imagine that this delay, and the confession of adjustment of
the figures, will lead to a greater level of mistrust in official statistics
than has ever occured before.

So should not the Code have something specific to say about gaining the
confidence of the public?

There could be, for example, some mention of links to Parliament. Links to
Parliament could be a component of independence from the Government of the
day. Independence from the Government of the day is supposed to be the
objective of the whole exercise, and links to Parliament could be a way of
supporting that independence.

Parliament can also be considered as the body that formally represents those
who are not regular user of statistics, but who expect statistics to serve
the public interest.
*************************************

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