§ Mr. BettsTo ask the Chancellor of the Exchequer how much of Sheffield's 2001 Census information was lost; if he will make a statement on how this information was lost; and what assessment he has made of the effectiveness of the methods used to estimate Sheffield's 2001 Census returns. [126125]
§ John HealeyThe information requested falls within the responsibility of the National Statistician. I have asked him to reply.
Letter from Colin Mowl to Mr. Betts dated 17 July 2003: 457W
The National Statistician and Registrar General for England and Wales has been asked to reply lo your recent question regarding your concerns about how the 2001 Census was conducted in Sheffield. (126125)I can assure you that there were no Census forms lost in Sheffield. It seems likely that your reference to 'lost' information refers to problems with the Census Coverage Survey for Sheffield. This resulted in us not having information for 36 of the sampled postcodes and meant that we could not ise the Census Coverage Survey for Sheffield as we had originally intended.The Census Coverage Survey (CCS) was a key element in the One Number Census process. It was an independent survey conducted shortly after the Census to establish the coverage of the 2001 Census. The information obtained from the Census Coverage Survey was used to enable census population counts to be adjusted for under enumeration at the national, local and small area level. It was therefore important that the Census Coverage Survey was successful in capturing those persons missed in the Census.A quality assurance panel was set up to look at the results of the Census from around the country and also to provide advice as problems arose during the processing. After some consideration, the quality assurance panel felt that the sample for Sheffield from the Census Coverage Survey was not sufficient to make reliable estimates of underenumeration.However, the methodology for the Census included a contingency strategy that could be used for occasions when it was felt that the Census Coverage Survey had failed to find the people missed from the Census. This contingency strategy involved a method known as "borrowing strength". This strategy involves using the adjustments made for similar Local Authorities when compared with aggregate administrative data and applying these to the Authority for which the strategy is invoked. The method for selecting Local Authorities from which to borrow strength involved using the ONS classification of Local Authorities to identify the five most similar areas in terms of socio-economic and demographic characteristics. By applying the borrowing strength method, the population of Sheffield was increased by 6,100 over and above what would have been achieved had the borrowing strength method not been used. This accounts for approximately 1 per cent, of the total population estimate.A consultation exercise was undertaken with all Local Authorities in the winter of 2000/1 inviting each Authority to comment on the acceptability of the list of closest Authorities to their own for borrowing strength purposes. Sheffield was given the opportunity to comment at this stage, before the list was finalised and published.