Design of a census coverage survey and its use in the estimation and adjustment of census underenumeration

Abstract

Many countries in the world conduct censuses of their populations. The UK is no exception, and every ten years there is a census undertaken in England and Wales, Scotland, and Northern Ireland. Although in theory it is three separate censuses, in practice these are planned and conducted as a joint project. The 1991 Censuses suffered from an increased level of underenumeration relative to the 1981 Censuses. This underenumeration was not detected by the 1991 follow-up survey that was designed to estimate this underenumeration.

The work presented in thesis develops the design of a new follow-up survey for the 2001 Censuses that will be able to measure the level of census underenumeration. Much of the work presented in this thesis deals with the development of an effective estimation strategy for this follow-up survey that utilises all the available data. In addition the thesis contains an analysis that adjusts this design and estimation strategy to facilitate its implementation within Northern Ireland. The ultimate goal for the 2001 Censuses will be to create a 'One-Number Census'. This is a census database where the estimated underenumeration has been fully integrated into the output database so that all tabulations are consistent with the agreed national population adjusted for underenumeration. This thesis considers the development of an imputation system for this purpose. The system utilises a donor imputation approach but also makes use of the estimates already available to ensure that the resulting database is consistent with agreed population estimates.

More information

Identifiers

Catalogue record

Export record

ASCII CitationAtomBibTeXData Cite XMLDublin CoreDublin CoreEP3 XMLEndNoteHTML CitationHTML CitationHTML ListJSONMETSMODSMPEG-21 DIDLOpenURL ContextObjectOpenURL ContextObject in SpanRDF+N-TriplesRDF+N3RDF+XMLRIOXX2 XMLReferReference ManagerSimple Metadata