(2) what work the Office of National Statistics has done on assessing the likelihood of people completing a (a) four page and (b) three page census form;
(3) whether a question on carers will be included in the 2011 census;
(4) what estimate he has made of the cost of adding an extra page to the 2011 census form.
The information requested falls within the responsibility of the National Statistician, who has been asked to reply.
Letter from Karen Dunnell, dated 23 January 2008:
As National Statistician and Registrar General for England and Wales, I have been asked to reply to your recent questions asking a) out of which Departmental budget the cost of the 2011 Census will be met, b) what estimate has been made of the cost of adding an extra page to the 2011 Census form, c) whether a question on carers will be included in the 2011 Census, and d) what work the Office for National Statistics has done on assessing whether or not people will be less inclined to complete a four page Census rather than a three page Census form. (180520, 180522, 180521, 180523).
I will take this opportunity to answer all the questions together.
Funding for the 2011 Census is provided to ONS by Parliament through the Treasury under the terms of the Comprehensive Spending Review.
Our current funding for the 2011 Census allows for 3 pages of questions per person. We are at present investigating the potential for a fourth page which would allow for the inclusion of more questions. The cost of an adding a fourth page of questions per person has been estimated at £25m.
The value of the information on carers provided by the 2001 Census is well recognised. However, such a question is competing for space on the questionnaire with a number of other questions such as second residence, citizenship, year of entry to the UK, qualifications, industry, end language. Currently ONS is only proposing to include a question on carers if there are 4 pages of questions.
ONS has extensively reviewed and considered the potential impact on response rates of a four page per person census questionnaire for 2011. The 2011 Census Questionnaire Design Team carried out a postal test in April 2007 covering 10,400 households with half of the sample receiving a questionnaire with 3 pages of questions per person and the other half receiving a questionnaire with 4 pages of questions per person. There was no evidence that-the length of a self-completion questionnaire had a significant effect on response rates. However, it should be noted that the postal test was voluntary.
Research into international census experiences was inconclusive on the effect of length of questionnaire but has shown that other factors, including ineffective design and layout, question clarity, and controversial topics have more of an impact on response rates.