User Consultation: Advisory Groups Meetings - October/November 2004
Summary of issues and topics covered at the Census Advisory Group meetings held during October/November 2004
Summary of issues and topics covered at the round of Census Advisory Group meetings held in October/November 2004. The groups involved were:
Business and Professional Interests Advisory Group;
Central and Local Government Information Partnership (CLIP) Census Sub-Group;
Health Service Advisory Group;
Academic Advisory Group;
Departmental Advisory Group; and
Special Populations Advisory Group.
These Groups cover the interests of the business sector, central government, local government, health authorities, the academic community, and special needs and minority populations respectively.
Corporate user consultation arrangements/role of Advisory Groups
The discussion paper AG (04)08 had been prepared as part of a wider project on National Statistics consultation. The paper noted that reorganisation of the structure of ONS might necessitate change to the approach established for previous user consultation on the Census. Advisory Group members were asked to give their views on the form of consultation which would most effectively combine the various user community and topic interests in a future Census whilst aligning with the corporate approach being taken within National Statistics.
Academics in particular tended to be interested in a number of themes and said that they
would find it difficult and fragmented if consultation was theme-based. Other views were that consultation on the 2001 Census had been effective and had been one of the better examples of how ONS conducted user consultation. It was felt that ONS needed to be responsive to the demands and needs of users rather than users having to fit into a rigid ONS organisational structure.
Users felt that it was also important for ONS to understand that an approach of 'one size
fits all' would not necessarily work and that it would be a great pity to lose the expertise,
experience and good will that currently exists among census users. User interest was not confined to single topics since there was a widespread need to cross-tabulate a range of census characteristics. The important aspect was to ensure that users were consulted, and in good time.
The ONS Office Management Board subsequently considered proposals for improved user consultation in December 2004 and decided that further work was necessary before any wider circulation of the recommendations. Advisory Groups will be kept informed of further developments.
A statistical framework for the 2011 Census
Paper AG (04)05 highlighted the key statistical aims and high priority areas of research. The paper noted that recent censuses and surveys have suffered from falling response and in particular differential response. It is likely that contacting and obtaining a response from some sectors of society would continue to be increasingly difficult; in addition the population is becoming more mobile and patterns of living more complex. These sorts of issues provide challenges for both data collection and outputs.
There was general concern expressed at all the meetings that the development of an Address Register was a high risk and could take a very long time to come to fruition. ONS noted that pre-listing of addresses was not being ruled out and that assessment of post-out would be part of the 2007 Test.
ONS reported on plans to work in partnership with Local Authorities to use their local knowledge for targeting enumeration and for assistance with creating address lists. Close
involvement with Local Authorities on the production and quality assurance of address lists would give LAs an opportunity to sign up to the enumeration lists in advance of the Census.
The issues on which ONS wish to consult with Local Authorities as partners will be those
where LAs are best placed to assist ONS in achieving a number of its mission-critical
aims, in particular to:
maximise response and minimising differential non-response;
ensure that key minority groups are well-defined, located and enumerated;
build, through a Community Liaison Programme, effective partnerships in planning and executing the enumeration;
inspire public confidence, and encourage participation, in the census operation; and
improve user confidence in the final results.
A Local Authority Liaison Steering Group has been set up to take these issues forward; its next meeting has been scheduled for 10 February 2005.
Summary of responses to Consultations on Population Definitions
Earlier consultation on the paper on population base definitions for outputs had been web-based and the current paper AG (04)06 was tabled to fulfil ONS' commitment to report back on user responses to consultation. The consensus at all the meetings was that that the concept of usual residence was becoming increasingly blurred, and that this had been recognised as an international issue.
ONS reported that a Population Definitions Working Group had recently been set up, which included representatives from central government, local government, academics as well as colleagues within ONS. Discussions at the first meeting had focused on the hard-to-count groups such as 'gap year' students, medical students away from medical school, migrants, asylum seekers and children of divorced parents. Accommodation such as second homes and holiday homes were also recognised as being hard to identify. The aim of the Working Group was to come up with harmonised definitions.
A second meeting of the Working Group has been scheduled for 12 January 2005.
Population base for 2011 Census Enumeration
The discussion paper AG (04)07 highlighted how important the decision on the choice of population base was for census enumeration purposes. Changes in society impact on living arrangements and family structures and call into question the traditional concept of usual residence. The paper outlined various options for the enumeration base for the 2011 Census; Annex A drew comparisons of the options, and Annex B, covered possible states for residential households on Census night.
There was a general consensus that either option 3 ('population present + questions about
usual residents temporarily absent') or option 4 ('usually resident + questions about visitors') would offer users the most relevant base, although there was some concern that the necessity for more residence questions might mean a reduction in the number of topic questions.
A 'person present' base would mean a simpler definition but would require more questions to establish residency, whilst a 'usual resident' base would involve a more complex definition but fewer'definitional' questions.
ONS are currently researching the amount of space which would be taken up on the census form by additional residence questions and will report back to Advisory Groups when more work has been done.
Many members were concerned at the reduction in face-to-face contact that would result from using post-out as the primary method of delivery of forms, especially in areas where there were language difficulties. ONS reiterated that there would be multiple enumeration strategies. It was planned to identify in advance those areas where traditional doorstep contact offered a better method of delivery than post-out.
Strategic issues surrounding 2011 Census questionnaire content
A discussion paper AG (04)04 sought views on a number of issues affecting the length and content of the census questionnaire. Many members were doubtful that a shorterquestionnaire would result in improved response. It was generally felt that the complexity of questions and how formidable the form looked were more significant
factors.
There was a widespread view that small area output geography, a wide range of topics and full processing of all data were all fundamental requirements for users. It was suggested that Internet or telephone completion offered the scope to have more tick-boxes with drop-down categories. Market research users noted that the removal of the age 75 or over cut-off for employment questions adopted for the 2001 Census had made it impossible to create social classifications of the elderly population, and they hoped that this limitation would not be applied for 2011.
The issue of Disclosure Control was raised at each meeting and it was agreed that this needed to be considered at the same time as topics and census geography. ONS reported that it planned to engage with the user community on disclosure control methodologies in general in the spring 2005.
ONS reported that as part of the research into census content, the 2001 Census business
cases would be reviewed to see whether the purposes for which people had said they wanted the data had actually been effected. The Statistical Team would be contacting those who had made the 2001 business cases and an update would be given at the next Advisory Group meeting.
Community Liaison Programme
ONS reported, through paper AG (04)03, that the Community Liaison Programme was being initiated much earlier in the planning than had been the case prior to the 2001 Census. The different enumeration approach which envisaged post-out as the prime means of form delivery, and the resulting loss of contact with the public, meant an
increased need to target minority and hard-to-count population groups.
The earlier start to the Programme was welcomed and some useful experiences from the 2001 Census were reported by members who had been involved in field activities. Various suggestions were offered such as drop-in centres and multi-lingual help desks. It was hoped to build partnerships with organisations represented at the Special Populations Advisory Group and others.
Community Liaison would be initiated in the areas of the Census Test in 2007 and should
give useful feedback and information with which to develop plans for wider implementation in the 2011 Census itself.
As part of the research programme the 2011 Census Team are consulting with the Central Office of Information to investigate the most effective means of engaging with local communities and hard-to-count populations.
UK governance
There were clear messages from members about the need for consistent UK outputs. ONS
reported that a UK Design and Methodology Committee has been formed to provide a forum for users to advise on the statistical and methodological aspects of the 2011 Census design with a view to achieving a common approach across the three UK Census Offices, and to achieve consistent and comparable UK-wide outputs from 2011 Census by adopting common methodologies, definitions and questions.
In addition a UK Census Harmonisation Committee has been created to identify and agree, at the working level, areas of research into those issues where there is scope for, and benefit in, following a common approach across the three UK Census Office, with the aim of achieving a high level UK Census.
It has also been proposed to establish a UK-wide Census Users Advisory Committee, through which views of the several Census Advisory Groups could be directly channelled through to the Census Offices, resolving where necessary any conflicting views and requirements expressed by the different user communities across the UK. It is planned that this Committee will initially meet after the next round of Advisory Group meetings.
Next round of meetings
Planned for April/May 2005.
Census News
