To ask Her Majesty's Government further to the Written Answer by Baroness Crawley on 13 July (WA 183–4), why providing options for Christian denominations in the 2011 census would not enable comparability with the 2001 census figures; what duties in relation to religious affiliation need to be fulfilled by public authorities under the Race Relations Act 1976; what are the extra Welsh religious categories that would be required were there a question on Christian denominations; whether religions other than Christianity have difficulty with their single question; and what form of equality monitoring and service planning users of census statistics undertake in relation to Christians and Buddhists. [HL5953]

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

Letter from Jil Matheson, National Statistician, to Lord Laird, dated October 2009.

Further to my predecessor’s letter to you of 14 July, as the new National Statistician I have been asked to reply to your recent Parliamentary questions asking i) why providing options for Christian denominations in the 2011 Census would not enable comparability with the 2001 Census figures; ii) what duties in relation to religious affiliation need to be fulfilled by public authorities under the Race Relations Act 1976; iii) what are the extra Welsh religious categories that would be required were there a question on Christian denominations; iv) whether religions other than Christianity have difficulty with their single question; and v) what form of equality monitoring and service planning users of Census statistics undertake in relation to Christians and Buddhists. (HL5953)

i) An expanded religion question would be likely to affect comparability with the 2001 Census data due to the different way that the question may be interpreted and answered. Testing of a religion question with expanded Christian denominations showed that some people thought of the new question in terms of religious practice rather than religious affiliation (which is the concept we are trying to measure) because of the specific tick-boxes for denominations, and there was evidence that some people chose ‘No religion’ when they would otherwise have picked ‘Christian’.

In addition, if detailed Christian options were provided the lack of space on the questionnaire would require a double-banked format of the question where tick-boxes were listed side by side in two columns rather than in a longer list. Such questions are normally avoided where possible because the tick-boxes on the right, in this case the Christian denominations, frequently get missed by respondents. However, encouraging respondents to notice the Christian breakdown boxes could have the unintended effect of drawing respondents’ attention away from the remaining religion tick-boxes and making them less visible. This could lead to incomplete and incorrect data and a possible undercount for the remaining religions.

These and other concerns about including an expanded religion question are set out in an information paper which details the religion question development which led to the final recommendations for the 2011 Census. This is available on the website:

http://www.ons.gov.uk/census/2011-census/2011-census-questionnaire-content/question-and-content-recommendations-for-2011/index.html.

ii) Respondents to the Office for National Statistics 2007 consultation on user requirements for information on ethnicity, identity, language and religion from the 2011 Census emphasised the use of religion data to enable better understanding of the social and economic position of different religious and ethnic groups, and to help to identify cases of social exclusion. This information would be used to inform policy development and monitoring and enable public bodies to meet their statutory obligations under the Race Relations Act and other equality legislation. The Race Relations Act 1976 (as amended by the Race Relations (Amendment) Act 2000) (the Act) gives public authorities a general duty to promote race equality. The duty means that, in everything they do, public authorities should aim to:

eliminate unlawful racial discrimination;

promote equality of opportunity; and

promote good relations between people of different racial groups.

The Act gives public authorities a general duty to monitor policy and service delivery for different ethnic groups. Definitions of what constitutes an “ethnic group” are subject to much discussion. This is because membership of any ethnic group is something that is subjectively meaningful to the person concerned, and can be based upon a combination of categories such as country of birth, nationality, language, skin colour, national/geographical origin, racial group and religion.

Although the ethnic group question was designed to capture information relevant to the Race Relations Act including colour and ethnic and national origins other concepts or groups relevant to this Act and other pieces of legislation may be captured more easily with other questions creating a suite of ethnicity questions including country of birth, citizenship, national identity, religion and language. For this reason, when prioritising groups for inclusion as new categories in the ethnic group question we considered whether one or more other Census questions provided similar information to an ethnic group tick-box, (in which case there would be less need to include this tick-box in the ethnic group question).

For some ethnic minorities, religion is one of the important defining characteristics and for this reason a question on religion was first included on the 2001 Census. Since the 2001 Census a number of laws to protect religious freedom have been passed, including the freedom to have no religion or belief. The Employment Equality (Religion and Belief) Act 2003 places a duty on all local authorities to monitor employment practices and to address the needs of religious groups. Other laws include the Racial and Religious Hatred Act 2006 and the Equality Act 2006. One of the main needs for religion data expressed by users was to assist legal obligations to prevent discrimination and promote equality.

iii) The different Christian categories which would be required in Wales were there an expanded religion question have not been ascertained from Census users. However they are likely to include non-conformist chapels such as Presbyterians, Independents, Baptists and Methodists since, in Wales, their memberships are higher than those of the Anglican or Catholic Churches.

iv) Some Census users expressed a requirement for information that would break down the Muslim category but there is no evidence of difficulty with a single category for religions other than Christianity.

v) It is not possible to comment on what form of equality monitoring and service planning that users of Census statistics may undertake in relation to Christian and Buddhists—respondents to our consultations generally articulated their need for religion data in general rather than with reference to specific groups. In addition to meeting legal obligations, users emphasised the use of data on religion to enable better understanding of the social and economic position of different religious and ethnic groups, to help identify cases of social exclusion, and to aid decisions relating to the provision of services in areas such as health, housing and education.