Methods of Coverage Assessment and Adjustment in the UK 2011 Census of the Population
Alan Taylor, Owen Abbott, Christins Sexton, Mark Baillie, James Brown
Methodology Directorate, Office for National Statistics, Titchfield, United Kingdom; Southampton Statistical Sciences Research Institute (S3RI), University of Southampton, Southampton, United Kingdom

The UK decennial population Census took place on March 27th 2011. Every effort was made to ensure that every member of the population was counted. However, no census is perfect and some people will have been missed. In terms of resource allocation, this can be an important issue as the people that are missed are often those who attract higher levels of funding. A smaller number will also have been counted twice, usually in two different locations. It is expected that this will be more of an issue in 2011 due to nature of population changes from 2001. This Census builds on the successful methodology used in the 2001 Census developed to address the issue of undercount. The 2001 Census was the first UK Census to have a fully integrated coverage assessment and adjustment process. The aim was to provide population estimates that took account of the undercount estimated using the census coverage survey through a dual system estimation methodology. In this talk I will describe the components of the 2011 methodology and how they have been development from those used in 2001. These components include the design of the coverage survey, use of dual system estimation techniques and how this is integrated to estimate population totals by administrative area. In additional, I will discuss supporting evidence from simulations carried out by the ONS Census Coverage Team.

Keywords: Census; Coverage assessment; Coverage adjustment; Dual-system estimation

Biography: Dr Alan Taylor, Statistician working in the Methodology Directorate of the Office for National Statistics. Academic background in statistics and applied medial statistics in the area of psychology and psychiatry.