Presentations from this session have been added and, where available, can be accessed from the hyperlink in the title line. Copyright in all papers and presentations remains with the author(s), to whom permissions requests should be addressed. All presentations present the views of the author(s), not those of BSPS.

Beyond 2011: The future for population statistics?
Jonathan Swan, Office for National Statistics

The UK Statistics Authority set up the Beyond 2011 Programme to establish and test models to meet future user needs for census type statistics in the UK, including wider socio-demographic statistics. It is internationally recognised that it is becoming increasingly challenging to conduct traditional censuses, partly because of the complexity of doing so with increasingly mobile populations. There are three broad categories of models under consideration within which are numerous variations; census or survey based models, aggregate administrative data based models, and record-level administrative data based models. It is unlikely a single source will deliver the whole solution therefore combinations of these options, as well as variations dependant on geographical location, are also being investigated. The feasibility of each option will be evaluated against criteria such as cost, statistical quality of outputs, and breadth of statistical outputs available. The Beyond 2011 Programme will be conducting a public consultation in Autumn 2011 for twelve weeks to engage with and receive views of stakeholders. The aims of the public consultation are to create a wider awareness of the programme, to foster a broad public debate about the future user need for population and socio-demographic statistics, and to provide an opportunity for statistical users to feed in their specific requirements. This will then inform the work on the statistical options. The presentation will introduce the Beyond 2011 Programme and the development of the options under consideration, as well as outlining the public consultation and how stakeholders at all levels can get involved.

Email: jonathan.swan@ons.gsi.gov.uk

Beyond 2011: Administrative data sources and low-level aggregate models for producing population statistics
Sophie Larcombe, Pamela Dent, Office for National Statistics

The dynamic nature of the modern UK population and demand for more frequent and in-depth statistics is driving a change in methods for the collection of population and socio-demographic data, a trend which can be seen across western countries. The Beyond 2011 Programme was set up to establish and test alternative models for meeting user needs for census type statistics in the UK, including socio-demographic statistics. A wide variety of solutions are being considered which can be grouped into three broad categories; Census and survey based models, aggregate administrative data based models, and record based administrative data based models. Administrative data sources collected by a host of departments and organisations have become more available with data sharing opportunities established with the Statistics and Registration Service Act 2007. The suitability and statistical quality of a range of currently available administrative data sources are being assessed in relation to their use as part of both aggregate and unit-level solutions for the collection and derivation of population statistics. The presentation will outline the range of administrative data based options that are being considered. We will focus particularly on the aggregate administrative data model options and present the results of some early research.

Email: sophie.larcombe@ons.gsi.gov.uk

A conceptual framework for UK population and migration statistics
James Raymer, University of Southampton; Ann Blake, Office for National Statistics

The Office for National Statistics is considering the use of administrative data to improve the estimation international migration flows and population totals at the subnational level. In this presentation, we introduce the issues that arise when combining information from multiple data sources and provide an overall framework for how the pieces of information fit together. The work forms part of the Migration Statistics Improvement Programme.

Email: j.raymer@soton.ac.uk

The development of 2011 Census outputs
Joe Traynor, Office for National Statistics

The 2011 Census aims to provide users with a greater range of outputs than ever before; allowing users to make the best possible use of the data. Some outputs have been designed for comparability with the previous Census and will change little from those produced in 2001. Others outputs will be entirely new taking advantage of the new information gathered. Current plans include outputs based on alternative populations in addition to the main residential base, and a desire to provide more detailed outputs for minority groups. The presentation will give a clear picture of the proposed outputs from the 2011 Census, how these outputs were formulated and some of the key decisions made along the way.

Email: joe.traynor@ons.gsi.gov.uk

Proposed dissemination methods for the 2011 Census
Chris Ashford, Office for National Statistics

Dissemination plans for the 2011 Census will primarily be web-based, with the intention of providing users greater flexibility and easier access to 2011 Census data. The presentation will show how using user market research and segmentation to develop user personas has informed internal design and testing of 2011 Census dissemination mechanisms and products. The presentation will, as a result, update users on how census data will be accessed and the functionality available to support users.

Email: chris.ashford@ons.gsi.gov.uk

2011 Census: the London Consortium
Eileen Howes, Greater London Authority

This paper is about the London Census Consortium through which the GLA, all London boroughs, Transport for London and the London Fire and Emergency Planning Authority will work together to make best use of the 2011 Census data. It will explain how funding for a ten-year programme has been raised and the work it is planned to cover. This includes acquiring the standard data; making use of the opportunity to define additional data in a number of different ways; commissioning a bespoke set of London tables or datasets; providing advice, training and support to all members of the consortium; and a programme of analyses relevant to London. The analysis is of particular interest here as we are currently planning a list of reports to be produced very quickly when the statistics are published and a further list of longer term work which we hope will involve some collaboration with academic experts as well as some commissioned research that there are not enough resources or expertise within the GLA to produce for ourselves.

Email: eileen.howes@sky.net

Census 2011: A question of confidentiality
Carole Abrahams, Keith Spicer, Paul Groom, Joe Frend, Martyn Hill, Office for National Statistics

The 2011 Census poses new challenges for data confidentiality. There is a legal obligation for published outputs to protect the information provided by respondents. The big question is how to do this while maintaining a high level of data utility for users. The Statistical Disclosure Control team at the ONS has developed innovative solutions which have been welcomed at public consultation events. This presentation will look at some of the confidentiality issues and outline the methodology which will be used to deal with them.

Email: carole.abrahams@ons.gsi.gov.uk

Administrative data linkage to support 2011 Census data quality assurance
Louisa Blackwell, Nicky Rogers, Office for National Statistics

Record-level administrative data are being used to help validate Census population estimates. The innovative data matching approach described here is just one component of a comprehensive data quality assurance strategy.for the 2011 Census. The focus here is the validation of Census estimates of undercount. A matrix of linked information on addresses and the people living in them, drawn from administrative data, can shed new light on households and individuals that Census processes may have missed. The administrative sources will be described, together with the methods being used to clean and link them. Analyses of these sources at aggregate level, intended to inform data matching, are also described. New Census information on second residences and visitors, together with addresses one year ago, will support the interpretation of linked records and of unmatched residuals. These results and methods draw on a series of data matching pilots designed to strengthen our understanding of available administrative sources and their quality.
Keywords: Census, Administrative Sources, record matching

Email: louisa.blackwell@ons.gsi.gov.uk

Using administrative data to count local populations
Gillian Harper, Mayhew Harper Associates Ltd

There is growing evidence that official population statistics based on the decennial census are inaccurate at the local authority level—the fundamental administrative unit of the UK. The research reported in this paper was partly funded by the ESRC UPTAP programme. It investigates the use of locally available administrative data sets for counting populations. The method uses truth tables for combining different data sources with different population coverage according to a defined and therefore replicable set of rules. The result is timelier and geographically more flexible data which is more cost-effective to produce than a survey-based census. Associated techniques for linking diverse data sources at individual and household level are briefly discussed. The methodology is then applied to administrative data from a London borough with about 170,000 people. The results are evaluated and compared with other population sources at different geographical scales. The paper concludes by discussing potential improvements including scaling up the work to cover multiple local authorities. The practicalities of using alternative central government data sets are briefly considered.

Email: harpergill@gmail.com

Population and Household Projections for Wales
Tony Whiffen, Welsh Assembly Government

The 2006-based set of local authority-level population and household projections produced by the Welsh Assembly Government Statistical Directorate were published over the period 2008-2009 and were the first to be produced separately for Wales. Since then a 2008-based round of projections has been completed. For those working in a local planning context household projections are more useful than population projections as they provide a more meaningful indication of housing demand. As a result the demographic projection outputs for Wales now include separate projections for National Park authority areas as these also have a statutory planning function. This paper considers the methodological and data issues arising from the work to produce projections for National Park areas. It also looks at development work to produce population and household projections for residual areas of unitary authorities where part of the authority is inside a National Park. It will also provide a more general comparison of recent household estimates with Annual Population Survey data for Wales and outlines plans for retrospective analysis of recent projection results after the results of the 2011 Census are known.

Email: tony.whiffen@wales.gsi.gov.uk

How will ethnic diversity change in the UK between 2001 and 2051?
Philip Rees, Pia Wohland, Paul Norman, Peter Boden, University of Leeds

A team of researchers at the University of Leeds, funded by the Economic and Social Research Council, has developed a suite of population projections for ethnic groups living in the local authorities (LAs) of England (Wohland et al 2010, Rees et al. 2011a, 2011b). The results of these projections will be released in the summer of 2011 for general use. This paper reports on one analysis of the projections. We examine the ways in which ethnic diversity may change across LAs to mid-century. First, we develop a conceptual framework that integrates competing measures of diversity, building on a tradition that has seen such a review written once a decade since the 1950s. Second, we compute selected indexes and diversity profiles for total and age-specific LA ethnic group populations in 2001, 2031 and 2051. Third, we analyse the variation and dynamics of changing ethnic diversity across LAs, showing how the ethnic compositions of the England LA population are likely to change. Because the database on which this analysis is based are future projections rather than past censuses, we assess the uncertainty range of outcomes by computing all measures based on a projection (TRENDEF) that replicates the National Population Projection (2008-based), a projection that uses different assumptions but the same model for international migration (UPTAPEF) and a projection that changes the international migration model (UPTAPER). We identify the degree to which diversity changes vary between these projections.

Email: p.h.rees@leeds.ac.uk

Measuring uncertainty in the Local Authority Population Estimates
Dorothee Schneider, Ruth Fulton, Office for National Statistics

This paper presents an update on a project initiated by ONS in 2007 to improve the understanding, measurement and reporting of the quality of the mid-year population estimates. Potential error measures are not straightforward to produce because the population estimates, calculated using the cohort component method, are derived from multiple data sources (Census, administrative and survey sources) and use a variety of estimation procedures often within a single component. There is also a lack of independent data with which to corroborate the estimates, because data sources that provide good comparisons for population statistics or related components are used in the estimation process itself, and remaining sources have associated quality issues with no total error measures. Measures of potential error can be achieved by initially mapping out and describing the sources of potential error (quality issues) associated with each of the data sources and methods used, estimating the potential error for each quality issue and developing methods for combining these into a total quality measure of potential error. This paper will summarise the overall theoretical approach, the simulation methodology adopted for measuring total potential error and the composite quality measures considered. It will describe in more detail, • the new methodology adopted for measuring uncertainty in the internal migration estimates, and • the methodology for measuring uncertainty due to statistical error in the International Passenger Survey (IPS) estimates of immigration and emigration. It will also include a short discussion of some of the issues and challenges encountered in this work.

Email: dorothee.schneider@ons.gsi.gov.uk