Thursday 11 September 2003, 2pm, Conference Centre
2001 census - quality assurance issues
Ludi Simpson, Cathie Marsh Centre for Census and Survey Research (CCSR), University of Manchester
The 2001 census set itself hard targets with new methods. How well did it achieve them? It sought to resolve a number of tensions that pull a census in different directions. How well have the solutions worked?
Four tensions or 'trade-offs' which I will address are as follows:
-
The statistics of a complete population versus speedy delivery of results. At the cost of a later delivery programme, users accepted the proposal to improve census output with well-estimated allowances for those missed by a census - using the One Number Census procedures
-
Focus on hard-to-count areas versus information quality. At the cost of some degradation in quality of information through post-back, users accepted a focus of fieldwork effort on hard-to-count areas, to minimise the differential response rates between areas
-
Measures to prevent disclosure versus information which is fit for purpose. At the cost of adding error and inconsistency to the released data, the Office for National Statistics (ONS) removed all 1s and 2s from census tabular output in order to prevent successful claims of breaches of confidentiality
-
Population estimation versus census estimation. The attempt to roll together the population estimates and release of census results means that acceptance of each is dependent on the other
The presentation will draw on interviews with fieldworkers, analysis of 1991 and 2001 census data, ONS documentation, and a review for the Local Government Association of the One Number Census procedures.
CCSR
Dover St Building
University of Manchester
Oxford Road
Manchester M13 9PL
Tel: +44 (0)161 275 4735
Fax: +44 (0)161 275 4722
Email: ludi@man.ac.uk
Census Strategic Development Programme
Peter Benton, Head of Census Strategic Evaluation and Development, Office for National Statistics (ONS)
The Office for National Statistics has announced its plans to consider the future need for census-type information and the options for its collection. This paper sets out the current research programme and progress to date.
The programme aims to answer the following key questions:
-
What information will be required in 2011?
-
How should the ONS meet this need?
-
Where will we be by 2006?
The first phase of the research is a wide-ranging review of future information requirements and possible collection methods. Possibilities included in the review are the use of government administrative records, a rolling or sample census, or modifications to a traditional census such as collection via the internet. Phase one will result in a cost-benefit analysis of the options and a recommended way forward. The second phase will be a more detailed programme of testing and development.
The paper also sets out key interfaces and dependencies and, in particular, the relationship to a study into the feasibility of developing a high-quality common population register, that could be used across the public sector.
Office for National Statistics
Segensworth Road
Titchfield
Fareham
Hants PO15 5RR
Tel: +44 (0)1329 813089
Email: Peter.Benton@ons.gsi.gov.uk