> The UK is one of the few countries in Europe to still manage a
> successful census. Some countries have been forced to rely on
> population registers, administrative data and large surveys; after
> political opposition to the census (on privacy grounds to particular
> questions which are not asked in the UK) and lessening public co-
> operation.
This is optimistic. Surely the position is that we don't know how accurate
is the UK census? We know that 1991 wasn't. And it is not clear whether
we shall ever know about the accuracy of this year's census. There will
always be anarchists, dropouts, and illegal immigrants who don't want to be
counted. No system will easily estimate these groups.
One of the advantages of using counts based on administrative systems is
that they can be expected to be comprehensive for the coverage of the
administrative systems. But it is expected that the 2001 census will miss
people who are recorded in existing administrative systems. It is not
clear that the methods used in 2001 will make it possible to distinguish
between those missed who are part of existing administrative systems and the
anarchists/dropouts/illegal immigrants most of whom are not recorded in any
administrative system.
There is a robust defence of administratively based counts by Philip
Redfern, former Deputy Director of the OPCS, in RSS News for June 1999.
Ray Thomas, Social Sciences, Open University
Tel: 01908 679081 Fax 01908 550401
Email: [log in to unmask]
35 Passmore, Milton Keynes MK6 3DY
**********************************8
problem with reliance on a census that is independent of the administrative
system is that it also misses another group - those who are counted in
existing administrative systems but
Ray Thomas, Social Sciences, Open University
Tel: 01908 679081 Fax 01908 550401
Email: [log in to unmask]
35 Passmore, Milton Keynes MK6 3DY
> -----Original Message-----
> From: Ludi Simpson [mailto:[log in to unmask]]
> Sent: Wednesday, February 14, 2001 9:01 AM
> To: [log in to unmask]
> Subject: Re: UK Census
>
>
> John Whittington wrote:
>
> > I can't help but wonder how justifiable it is, in statistical terms,
> > to undertake a full Census, at great cost etc.
>
> The arguments for the full census include those for accurate data for
> sub-national areas. The census provides profiles for areas
> that are very
> small (tables will be published for output areas with average
> populations of around 200 in 2001); it allows measurement of change
> for local authority and smaller areas; it provides broad social and
> employment indicators and cross-classifications of these, again for
> small sub-national areas. As a public data source (most libraries
> provide access to local tabulations) it can be a democratic instrument
> giving evidence to local communities and campaigns that has not only
> official status but includes comparative data for regional
> and national
> areas.
>
> An attempt to use the major national surveys to measure change in
> selected indicators since the census for each local authority
> (populations 20,000 to one million) found that the surveys even when
> combined did not have sufficient sample size to measure change
> reliably, even for those relatively large areas. The failed
> attempt was
> commissioned by the DETR who wished to update the indicators of
> local authority need during the decade between censuses. The study
> was carried out by Harvey Goldstein, Patrick Heady and others, but I
> don't have a published reference.
>
> The census procedure does recognise and incorporate the value of a
> survey, by holding a large-sample follow up survey, in order to
> measure the number and nature of people missed by the census. If
> successful (it wasn't successful in 1991, but many more resources have
> been committed to it for this April's census) the census
> results even for
> local areas can be adjusted to make them more likely to be reliable
> estimates of the 2001 population.
>
> The UK is one of the few countries in Europe to still manage a
> successful census. Some countries have been forced to rely on
> population registers, administrative data and large surveys; after
> political opposition to the census (on privacy grounds to particular
> questions which are not asked in the UK) and lessening public co-
> operation.
>
> Lack of a census denies governments and non-government
> organisations a valuable democratic resource.
>
> Ludi
>
> ------
> Ludi Stephen Simpson, [log in to unmask]
> This email is used for my work with the Cathie Marsh
> Centre for Census and Survey Research, University of Manchester.
>
> Most days in 2000 I work at the Policy and Research Unit,
> Bradford Metropolitan Council, Jacobs Well 4th floor,
> Bradford BD1 1HY, UK,
> Tel: +44(0)1274-754252, Fax: +44(0)1274-752004. Email:
> [log in to unmask]
>
> If you want to send letters to me that are not Council
> business, the quickest route is:
> 41 Park Crescent, Bradford BD3 0JZ, UK.
>
> ******************************************************
> Please think before you press the 'Reply' button! Note that
> if you press
> the 'Reply' button your message will go the individual who
> posted this message
> not to the list. With many mailers you will have a
> 'Reply-to-All' button that
> will send automatically to the list address of
> <[log in to unmask]>. The
> Radstats list is set up for public discussion so please be
> generous with your
> thoughts and share them us all.
> *******************************************************
>
******************************************************
Please think before you press the 'Reply' button! Note that if you press
the 'Reply' button your message will go the individual who posted this message
not to the list. With many mailers you will have a 'Reply-to-All' button that
will send automatically to the list address of <[log in to unmask]>. The
Radstats list is set up for public discussion so please be generous with your
thoughts and share them us all.
*******************************************************
|