Census 2011 and the religion question
From the House of Lords' Hansard:
Asked by Lord Laird
“To ask Her Majesty's Government further to the Written Answer by Baroness Crawley on 24 June (WA 282), whether they will review the census question on religion which currently permits one Christian answer without denominational distinction and lists five other religions; and why that question is as it is. [HL4806]
13 July 2009 : Column WA183
Baroness Crawley: The information requested falls within the responsibility of the UK Statistics Authority. I have asked the authority to reply.
Letter from Karen Dunnell, National Statistician, to Lord Laird, dated July 2009.
As National Statistician I have been asked to reply to your recent question asking whether the census question on religion which currently permits one Christian answer without denominational distinction and lists five other religions will be reviewed; and why that question is as it is. (HL4806)
The religion question was introduced in the 2001 census as a voluntary question and has been designed to collect information on religious affiliation, which is required by many users of census statistics for monitoring equality and planning of services.
Question development for the 2011 census began in 2005. A detailed and lengthy process of user consultation, prioritisation of user requirements and qualitative and quantitative question testing has been carried out to inform decisions on the topics, content and design of questions to he included in the 2009 census rehearsal and 2011 census.
As part of this process ONS has already considered the feasibility of an extended list of Christian denominations in the England and Wales census but rejected this approach for a number of reasons.
Testing of a question with Christian denominations indicated that some respondents may interpret and answer the expanded question differently, which would make it difficult to compare data with those from the 2001 census. Trends in religious affiliation over time are required by many census users, primarily for service planning: three-quarters of the respondents to the 2007 consultation regarded comparability with the 2001 census question as essential.
In addition, different and potentially a higher number of categories would be needed in Wales , which would make comparison across England and Wales difficult.
Furthermore, space constraints on the census questionnaire for England and Wales mean that providing detailed breakdowns of the Christian category would result in losing other questions or compromising the questionnaire design, thereby putting the quality of responses and their comparability with 2001 data at risk.
A key reason for including a question on religion in the 2001 census was to provide statistics on minority religions. This helped to provide benchmarks so that employers and public authorities, for example, could fulfil their duties under the Race Relations Act. The proposed 2011 question lists five other religions in addition to Christian: Buddhist, Hindu, Jewish, Muslim and Sikh. These religions are included as they are widely recognised as being the largest of the minority religions within the UK , although it is also proposed that there will be a “write-in” option, where those who wish to record themselves under any other religion may do so.
Full details of this consideration are set out in an information paper relating to the development of the 2011 census religion question, which available on the website:
ONS is currently testing a revised wording to the question on religion that asks “Which of these best describes you?” with no changes to the “No religion” and pre-designated religion tick-boxes."
So there you have it. The extent of the Christian majority was never the issue: the question was to determine figures for religious minorities in order that employers and public authorities could fulfil their duties under the Race Relations Act.
Perhaps Baroness Crawley might learn the difference between race and religion, and appreciate that membership of a particular faith group is not necessarily a pointer to ethnicity.