Population Statistics (Hansard, 26 November 2002)

§ Ruth Kelly

The information requested falls within the responsibility of the National Statistician. I have asked him to reply.

Letter from Len Cook to Sir Sydney Chapman, dated 26 November 2002: As National Statistician and Registrar General for England and Wales I have been asked to reply to your recent question to the Chancellor of the Exchequer asking if he will institute an inquiry into the accuracy of(a) the 2001 Census population figures and(b) the 2001 mid-year estimate population figures. (82946) I am confident that the 2001 Census figures provide the most accurate estimate of the 2001 population, both nationally and locally. The 2001 Census results are the first to represent the entire population. This was achieved through a new strategy known as the 'One Number Census'. One of the key elements was an independent follow-up survey— the Census Coverage Survey (CCS), which involved an intensive face-to-face survey of a sample of over 16,000 postcodes containing 320,000 households covering every local authority in England and Wales. By combining the results of the Census and the CCS, it was possible to estimate the total resident population in 2001— the 'one number'— to a high 157W level of precision, plus or minus 0.2 per cent, and to provide information on the characteristics of those persons missed by the Census. Thus in 2001, the Census results represent 100 per cent of the population. Census response, that is the proportion of people returning a form in England and Wales, was 94 per cent. The 6 per cent of people estimated to have been missed by the Census have been added to the counts using a statistical process that was subject to extensive consultation, and based on the results of the CCS. In addition, the 2001 Census has undergone rigorous testing and quality checking throughout both the fieldwork and processing. Census tests were carried out in 1997 and 1999 and a survey designed to measure the accuracy of response to each question has been completed. Furthermore, a comprehensive set of verification and validation checks were built in to the processing of the data, and data quality played a key role in all aspects of processing the forms. All information on accuracy will be contained in a Census Quality report to be published in 2003. The population base for the 2001 Census was defined so as to be entirely consistent with that used in the calculation of the mid-year estimates, and the 2001 Census figures will form the basis of the mid-year population estimates throughout the next decade. Thus, the 2001 mid-year estimate was calculated by adjusting the census figures to take account of births and deaths and an estimate of the net migration in the interval between Census day (29 April) and the 30 June. However, in the light of the evidence, shown by the 2001 Census figures, that the previous (pre-2001) mid-year estimate series had been over-estimating the size of the population and growth in the population over the last few years, the Office for National Statistics will be conducting an assessment of accuracy of the mid-year population estimates that is aimed at learning lessons from the 2001 Census results and making improvements to the method and data sources used to produce mid-year estimates in the future. The results of this assessment are due to be published on the National Statistics website in February 2003. A National Statistics Quality Review on international migration is also currently underway. Some local authorities have experienced an unexpectedly large shift between their population estimates for mid 2001, and the Census values. We are working with several of these to examine our explanation of the differences. For the above reasons, I therefore do not feel it necessary to instigate any further inquiry into the accuracy of the 2001 Census and 2001 mid-year estimate figures.