Skip to main content


Volume 728: debated on Monday 13 June 2011


Asked By

To ask Her Majesty’s Government what proportion of postcodes in England and Wales will be covered by the 2011 Census Coverage Survey, and what are the anticipated benefits and costs of the Survey.

My Lords, the current Office for National Statistics census coverage survey is a validation exercise that will greatly enhance the statistical authority and value of the census. It will cover some 17,000 postcodes across England and Wales. This represents a sample of 1.3 per cent of all postcodes in England and Wales, and will cover some 330,000 addresses. The cost of the field operation of the survey is estimated to be £6.5 million, representing around 1.3 per cent of the total census costs. The cost for the processing of the information collected in the survey is included in the overall cost of the processing operation.

My Lords, I thank the Minister for that very full Answer. However, he will be aware that this is a survey of people who have filled out the census form and about whom no questions have been raised as to accuracy. They are required to stand on a doorstep for 10 minutes, answering personal questions from a complete stranger. If someone refuses to answer, surveyors are instructed to return as many as 10 times to wear them down into answering this survey. As someone said to me, it is fascinating to watch an exercise that not only wastes public money but manages to alienate the public. Will the Minister give us a guarantee that future evaluations of the census will be proportionate, targeted and designed—please—with a dose of common sense?

I am sorry that my noble friend takes that view of this exercise, which is a valuable part of the census. I understand her concern about intrusiveness; these things are not taken in hand lightly. However, the survey is a vital element of the whole of the 2011 census operation. It is necessary in order to assess the extent of any undercount, and to provide information on those persons missed in the census to adjust the final estimates so as to enable all the resulting statistics to relate to the whole population.

If one has to hold an objective survey, is it not necessary to get complete strangers to ask questions? If somebody I knew asked me questions I might not give the right answers, because he might misuse them. The noble Baroness asks why complete strangers ask the questions. Of course strangers have to ask the questions; if they did not, it would not be an objective survey.

A short interview on the doorstep is available and a form can be filled in if people prefer to do that. People may well wonder why they are doing this when they have already completed a census form, but we seek to ensure statistically that the figures which the census is delivering are an accurate representation of the household. This is a valid statistical exercise which complies with quality assurance as defined by the United Nations and is international practice. It is worth investing a little extra effort to ensure that the census really does achieve its objectives.

Will the Minister clarify this for me? I had thought that it was an either/or situation and that if you completed the form you did not have to be interviewed on the doorstep. However, from what he has just said, it sounds as if a certain number of people are still to be interviewed even if they have completed a form. Is that correct? How is it assessed which people who have completed a form should be interviewed?

This is done on the basis of postcodes. The measure is designed specifically to include those postcodes where information has been difficult to obtain and to ensure that the information returned is valid. I thank my noble friend for pointing this out: the actual households may or may not have completed a form in the first place. This measure is designed to ensure that the information that is available is correct.

My Lords, the information in a census is required by law in this country. Does this survey have a penalty attached to it if one refuses to answer any questions?

No, it is entirely voluntary, but I would hope that people would realise why it is important to complete the survey.

My Lords, if it is the purpose of this survey to ensure that the information on the census forms is, broadly speaking, accurate, is it possible to ensure that Members of this House are included in that survey? I wonder whether the Minister is aware that the way that the census form was designed made it extremely difficult for people who are Members of this House to give accurate information about how they spend their working days.

I thank the noble Baroness for that question. I struggled a little to complete my own form and was rather embarrassed by that considering that I have occasionally to answer questions on the subject. So I understand exactly what she is saying. However, I can reassure her that were she part of the postcode lottery—that is, the postcode selected for this interview process—she might indeed find that somebody wanted to interview her about her census form. If so, it would be a very much abbreviated questionnaire compared with the one that she was asked to fill in in the first place.

My Lords, I declare an interest as a former canvasser in an inner London seat. Can my noble friend tell me what happens if the only method of communication with the person being interviewed is through an entry phone?

This is a familiar experience for most noble Lords. It relates in some ways to the difficulties experienced in some inner urban areas in this regard. Noble Lords may remember that I was asked a question on this some time ago and that Kensington and Chelsea had only a 64 per cent return of census forms in 2001. We are on track for a much better return this time, aiming at a 94 per cent return. The early indications are that it is on target, and that is very satisfying. As my noble friend has pointed out, communication is often difficult in these hard-to-reach areas. In the case of non face-to-face communication, it is possible, as I say, for a form to be filled in by the respondee.

My Lords, will the Government have regard to the assessment of population as a result of the census when considering parliamentary boundaries? It is my recollection that the Government closed that issue at the end of last year, and that therefore the new parliamentary boundaries could be drawn up on fallacious figures.

I think that we have discussed this previously, if I may say so, but I thank the noble Baroness for bringing it up. The truth of the matter is that registration for the electoral register is entirely voluntary in this country and not everyone is registered. On the other hand the census is mandatory and designed to make sure that government resources go to where they should. Therefore the two are not compatible; they are drawn up under separate legislation and conducted by totally separate organisations. The Office for National Statistics has nothing to do with local electoral registration.