Occupied address (household) estimates: further information

Estimates for the number of households in England and Wales using administrative data have been produced for local authorities for years 2011 and 2015. In future years we aim to produce a back series of household estimates, including 2013 and 2014. Estimates for 2012 will not be produced because of the unavailability of some data sources.

Our methodology uses the same population base that has been used to estimate the size of the population from administrative data. More information about the Statistical Population Dataset (SPD V2.0) we have developed to produce population estimates is available in our size of the population methodology report. The addresses of individuals included in our population estimates have been referenced to a unique property reference number (UPRN). Individuals with the same UPRN are grouped to form “households” on the basis that they are resident at the same address.

These Research Outputs are based on an “occupied address” definition. This is different to the census definition, where a household is defined by shared use of facilities at an address.

It is challenging to meet the census definition of household when using administrative data. This is because administrative data is collected at a person level, whereas survey collections can be designed specifically to collect information about households. One of the consequences of using an “occupied address” definition is that it is difficult to identify multiple households occupying the same address.

We will be researching the potential for combining administrative data with surveys to produce household statistics. While this may offer some potential to align with traditional definitions of “household”, we will also consider alternative definitions that are used in other countries that have register- based censuses. When considering alternative definitions we will have further discussions with users about their information needs for household statistics.

Estimates for the number of households from these Research Outputs are generally lower than household estimates produced from the 2011 Census and from Office for National Statistics (ONS) surveys. Reasons for this can partly be explained by differences in “household” definitions and the quality of address information available from administrative sources. More information is available in the methodology section of our report.

Official estimates of the number of households at local authority level are currently available only from the 10-yearly census. During the intercensal years, survey data are used to produce national household estimates, but there are limitations in providing these for smaller areas. The approach we are developing with administrative data has the potential to deliver estimates of the number of households for small areas more frequently. There is also potential to produce detailed tabulations based on other household characteristics.

These household estimates are produced from the same data sources that were used in our Research Outputs on size of the population. This includes the NHS Patient Register, the Department for Work and Pensions Customer Information System, the English and Welsh School Censuses and data from the Higher Education Statistics Agency. Where addresses are available on these datasets, they have been linked to the Ordnance Survey AddressBase product, which contains a list of residential addresses in England and Wales.

These Research Outputs do not have National Statistics status and are not to be used as a substitute for official household outputs produced across the Government Statistical Service (GSS). Our Research Outputs series has been developed to keep population data users up to date with our assessments of administrative data and, in the longer- term, to show progress towards a possible future census alternative.

These Research Outputs have been published to test the feasibility of a methodology different to that currently used in the production of household statistics.

No. We have been publishing Research Outputs since 2015 to demonstrate the progress we are making with assessing administrative data’s potential as an alternative for producing the information on housing, households and people that is currently provided by a 10-yearly census, after 2021. Our methodology for assessing administrative-data-based outputs will continue to develop each year. On the other hand, household estimates produced from the census and other surveys are produced using an established methodology and remain the official household estimates .

No. We have not yet developed a method for identifying and producing separate estimates for communal establishments. Based on the current methodology a number of communal establishments will also be included in these household estimates. We will be researching the potential to exclude communal establishments from household estimates in future releases.

No. The households included in these outputs have been aggregated to local authority level. Any future Research Outputs on households and their characteristics will be published at a level of detail that is compliant with our disclosure control policy. No individuals can be identified within these Research Outputs or from the text accompanying the outputs.

We are currently working towards the development of an ongoing address indexing service that can support the work of ONS and the Government Statistical Service. Part of this development includes discussions with data suppliers to improve the quality and consistency of address information captured at source.

In future releases we are aiming to produce Research Outputs on household size and household composition. We will also aim to produce household statistics at lower levels of geography with more detailed tabulations about household characteristics.

Currently we have developed our methodology to produce only household estimates at local authority level. In future years we are aiming to increase the level of geographic detail for these outputs.

ONS has recently taken on the responsibility for producing local authority household projections. The decision to bring household projections under ONS’s responsibility was not related to these Research Outputs. A consultation on Proposed Changes to Household Projections in England will run between 31 January to 13 March 2017 to help understand user need and develop the methodology.

A feedback survey is published alongside this publication. In particular we welcome feedback that helps to improve and develop our methodology and definitions. It would be particularly helpful to have feedback relating to this publication by the end of April 2017.