Skip to main content

Census

Volume 508: debated on Tuesday 23 March 2010

To ask the Minister for the Cabinet Office pursuant to the answer of 27 January 2010, Official Report, column 795, on the 2011 Census, what the most recent estimate is of the (a) absolute response rate in terms of number of respondents and (b) percentage response rate in each of the pilot local authority areas for the 2011 Census rehearsals; in which (i) wards and (ii) postcode areas rehearsals were held in each of the pilot areas; and what criteria were used to select the sub-areas. (318330)

The information requested falls within the responsibility of the UK Statistics Authority. I have asked the authority to reply.

Letter from Stephen Penneck, dated March 2010:

As the Director General for the Office for National Statistics (ONS) I have been asked to reply to your recent question asking pursuant to the Answer of 27 January 2010, Official Report, column 795, on the 2011 Census, what the most recent estimate is of the (a) absolute response rate in terms of number of respondents and (b) percentage response rate in each of the pilot local authority areas for the 2011 Census rehearsals; in which (i) wards and (ii) postcode areas rehearsals were held in each of the pilot areas; and what the criteria were by which the sub-areas were selected. (318330)

The rehearsal was unlike the Census in that it was voluntary and publicity was necessarily limited. It included 133,890 addresses (61,725 in Lancaster, 38,618 in Newham and 33,547 in Anglesey); 49,300 returns were received.

In order to rehearse our procedures appropriately, 51% of addresses were pre-selected to be subject to the full follow-up process; the remainder were not followed up if they did not respond. In the Census itself, it is the intention that all non responding households will be followed up.

(a) Measuring return rates from those 68,000 households subject to the full process, we had 27,500 returns (41%).

(b) The percentage return rate in each of the rehearsal local authority areas is:

Percentage

Lancaster

48

Newham

28

Anglesey

49

(i) and (ii) All wards in Lancaster and Anglesey were included in the rehearsal.

Areas in Newham were selected by lower super output areas, a statistical geography, which do not always align with ward boundaries. All or part of the wards and postcode sectors shown at Annex A were part of the Newham rehearsal.

The particular criteria for choosing sub-areas in Newham were that they should contain around 40,000 households, a number of communal establishments and a traveller site; that they should have high concentrations of multiple occupancy and students; that they should be contiguous. The areas were chosen in order to reflect some of the particular challenges representative of hard to enumerate areas.

Annex A

(i) Wards in which the rehearsal was held

All wards in Lancaster and Anglesey and the following in Newham:

Ward reference

Ward name

LA name

Level of ward inclusion

OOBBGU

Stratford and New Town

Newham

Whole

OOBBGG

East Ham Central

Newham

Partial

OOBBGK

Forest Gate North

Newham

Partial

OOBBGD

Canning Town North

Newham

Partial

OOBBGF

Custom House

Newham

Partial

OOBBGE

Canning Town South

Newham

Partial

OOBBGS

Plaistow South

Newham

Partial

OOBBGB

Beckton

Newham

Partial

OOBBGJ

East Ham South

Newham

Partial

OOBBGX

West Ham

Newham

Partial

OOBBGT

Royal Docks

Newham

Partial

OOBBGL

Forest Gate South

Newham

Partial

OOBBGH

East Ham North

Newham

Partial

OOBBGC

Boleyn

Newham

Partial

OOBBGW

Wall End

Newham

Partial

(ii) Postcode areas

Newham

E1

E12

E13

E15

E16

E3

E6

E7

Lancaster

CH3

LA1

LA2

LA3

LA4

LA5

LA6

LA7

PR3

Anglesey

LL00

LL58

LL59

LL60

LL61

LL62

LL63

LL64

LL65

LL66

LL67

LL68

LL69

LL70

LL71

LL72

LL73

LL74

LL75

LL76

LL77

LL78

To ask the Minister for the Cabinet Office pursuant to the answer of 3 December 2009, Official Report, columns 953-6W, on the Census, (1) what estimate she has made of the cost of printing the Census forms; how many Census forms of each type will be printed; and what estimate has been made of average cost of printing a form of each type; (322892)

(2) what estimate has been made of the cost of storing the Census forms;

(3) for what reasons the forms for the Census are being printed in March 2010;

(4) how many Census forms of each type will be printed in March 2010.

The information requested falls within the responsibility of the UK Statistics Authority. I have asked the authority to reply.

Letter from Stephen Penneck, dated March 2010:

As the Director General for the Office for National Statistics (ONS) I have been asked to reply to your recent questions asking:

pursuant to the Answer of 3 December 2009, Official Report, columns 953-6W, on the Census, what estimate she has made of the cost of printing the Census forms; how many Census forms of each type will be printed; and what estimate has been made of average cost of printing a form of each type. (322892)

pursuant to the Answer of 3 December 2009, Official Report, columns 953-6W, on the Census, what estimate has been made of the cost of storing the Census forms. (322893)

pursuant to the Answer of 3 December 2009, Official Report, column 953W, on the Census, for what reasons the forms for the Census are being printed in March 2010. (322894)

pursuant to the Answer of 3 December 2009, Official Report, columns 953-6W, on the Census, how many Census forms of each type will be printed in March 2010. (322908)

The contract for census processing was awarded to Lockheed Martin UK in August 2008. This is the largest of the census contracts with a value of around f 150m.

It is estimated that the cost for the printing of the 2011 Census questionnaires is in the region of £20 million; which would equate to an average cost of around £0.53 per questionnaire.

Lockheed Martin UK acts as the prime contractor with several sub contractors undertaking aspects of the work.

There are a number of types of census questionnaires. England, Wales and Northern Ireland each have a different question set, and in Wales everything needs to be printed twice, once in English and once in Welsh.

Due to the differences in questions and language used across these areas, there will be 17 separate types of questionnaires printed in total. These can be broadly grouped into four categories, and will be printed in the corresponding quantities:

Household questionnaires: around 34 million

Individual questionnaires: around 3 million

Communal Establishment questionnaires: around 144,000

Census Coverage Survey questionnaires: around 600,000

ONS docs not have the costs of printing each specific form type.

Storage costs are factored into the fixed price service contract so it is not possible to extract specific costs for the storage of any questionnaires printed in the period up to census day.

In addition to the questionnaires there are information guides, translations of the questions into various languages, leaflets, envelopes and other necessary material to accompany the questionnaires; as well as the various administrative forms used by census staff to undertake the census. This amounts to around 50 separate types of document to be printed; and is estimated to total approximately 107 million items, on top of the roughly 1 billion pages of questionnaires.

Initial printing of the questionnaires is running from March 2010 to July 2010; once this is complete a second process of overprinting will begin in July 2010, during which individual barcodes, unique identifiers and internet access codes will be printed onto each questionnaire.

For this process to be a success the scanning and processing software needs to be developed and rigorously tested. This is a large and involved project and must start in April 2010 for the systems to be ready in time for census day on 27 March 2011.

In addition to being needed for the printing timetable, a finalised print-tested questionnaire is needed for the programming of the scanning and printing software and for creating test data for the various stages of testing necessary to ensure that the final processes meet the high standard demanded by the census.