Message
I
would welcome correction if I am wrong, but don't Census tables give a
false impression of accuracy? Do the Census tables
give response rates for individual questions? Don't they
just 'interpolate' figures instead?
I
would be pleased to be shown that I am wrong but my experience as an enumerator
and as a respondent is that enumerators are not paid/motivated enough to get
100% responses on all questions from all households. Have response
rates for individual questions been given for the 2001 Census?
It may be that the quality is generally good, but
post-enumeration checks cast a lot of doubt on the matter. As I
recall the matter a table showing the distribution of rooms by
household provided a rich contrast
between the Census count and the post-enumeration survey in 1991. I
havn't seen any such tables for 2001. Do they
exist? The post-enumeration
surveys I have seen reported focus wholly on overall response
rates.
It would be useful to know more about the data used in
biomedical researech and how that is elicited from the patient.. But
in the case of the Census the issue of growing importance is the coverage rather
that the quality of the categorised data. The
'don't-want-to-be-counted' includes a number of illegals and
semi-legals. Failed asylum seekers, illegal immigrants, people
working without permits, etc. And people who believe that they might
fall into one or other of these
categories
One of the most interesting
groups don't-want-to-be-counted groups are the million young men
who said were abroad at the time of the 2001 Census. That must
have left a lot of young women lonely. But one reason for
scepticism was that there were no reports of men missing abroad in the
agony-hearts columns. These columns who would surely have picked up the
issue had they receive a just a few letters from deserted girl
friends. The probable explanation was that the boy friends were
living with their girl friends who as single mums were in receipt of
housing benefit. If the boyfriend confessed to cohabiting his
girl friend would lose her benefit. A good reason to wish not
to be counted.
It appears to be increasingly
difficult to maintain the dual function of the Census - as a basis for
estimating the total population and as a source of detailed information on the
population. The Government has already been obliged to use
administrative sources for estimates of the geographical distribution of the
population. Surely that trend will continue?
Surely the Census will
increasingly be seen as a large scale social survey? Wasn't
the 1966 Census conducted on a 10% sample basis? Might not such a
solution be on the cards for 2011?
Ray
Thomas
*****************************
At
05:04 01/11/2007 +0000, R.Thomas wrote In part):
Sadly nothing seems have been learned by the Government from this
experience. I asked the senior statisticians at a meeting last
week what plans the ONS are making to try and count people who don't want to
be counted. His reply was that he did not see making such
estimates as belonging to the ONS.
I have to say that I
have quite a bit of sympathy with that view. It's a bit like asking the
ONS to make estimates of how many MPs have committed undetected criminal acts
- it requires other government agencies to attempt to investigate such
matters! Of course one expects them to take 'all reasonable steps' to
reach (and count) as many people as possible - but as for those who are
'deliberately hiding' .....!
A further fall in response rate in 2011 can be expected to lead
to calls for the abandonment of the census. It will be diffiult
to resist these calls in terms of data needs. So the
abandonment it will be regrettable only in terms of loss of the the
one opportunity that all ciizens have of participating in the production of
statistics.
That all sounds a bit 'OTT' and
melodramatic. The Census does, and probably always will, provide a
wealth of valuable data, even if it is incomplete and slightly biased by the
absence of data from the self-selected "don't want to be counted" group.
To put things in perspective, I would imagine that the Censuses (or do I have
to try to construct the Latin plural?!), even the 2011 one, produce a far
higher 'quality' of data than is possible in most biomedical
research.
Kind Regards,
John
----------------------------------------------------------------
Dr John Whittington,
Voice: +44 (0) 1296 730225
Mediscience Services
Fax: +44 (0) 1296 738893
Buckingham MK18 4EL,
UK
----------------------------------------------------------------
? ????????? ?????????? ?????? SPAMfighter ??? ??????? ?????????????.
????????? ??????? 180 ??????????? ?????? ????? ?? ?????????? ???????.
???????????? ??????? ?????? ?? ????? ????? ????????? ? ?? ??????????? ???????.
??????????? ?????????? ??????! SPAMfighter
******************************************************
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.
*******************************************************