In this section
- What do you mean by "administrative data"?
- What is an Administrative Data Census?
- What are the advantages of this approach?
- Why is ONS doing this research?
- What are the Research Outputs?
- What are the plans for the future?
- How do you give feedback to data suppliers about the statistical quality of their data sources and to improve data sources?
- How will administrative data be used to make improvements to the 2021 Census production and outputs?
- What about Scotland and Northern Ireland?
- How do you ensure that all this activity and future plans are acceptable to stakeholders?
1. What do you mean by "administrative data"?
Administrative data refers to data that people have already provided to government, for example, in the course of accessing public services. Some of these data could be re-used by us to produce statistics about the population.
We’ve been using administrative data for many years. For example, annual births and deaths statistics are used, as well as NHS patient registrations, to roll forward the population estimates between censuses.Back to table of contents
2. What is an Administrative Data Census?
It’s our ambition to produce the type of information that is collected by a 10-yearly census (on housing, households and people) from an Administrative Data Census. Doing this will require a combination of:
record-level administrative data held by government
a population coverage survey
a population characteristics survey
Read more about our progress towards an Administrative Data Census in our annual assessments.Back to table of contents
3. What are the advantages of this approach?
The reuse of administrative data that are already collected and held within government could lead to a reduction in data collecting costs and the burden placed on the general public providing census and survey data. Administrative data could also increase the frequency, timeliness and range of data outputs, and help make them more responsive to the needs of our users.Back to table of contents
4. Why is ONS doing this research?
Our Research Outputs have been developed to keep you up to date with our assessments of administrative data and, in the long-term, to show progress towards an Administrative Data Census in the future. We've been publishing Research Outputs annually since 2015 and each year we'll continue to expand the range of topics depending on the availability of data and their statistical quality. More detail about the background to this research and our progress towards an Administrative Data Census can be found in our annual assessments.Back to table of contents
5. What are the Research Outputs?
The Research Outputs are part of our research into the potential use of administrative data and surveys to produce population, household and characteristic information currently provided by a census.Back to table of contents
6. What are the plans for the future?
For future releases of our Research Outputs, we plan to produce the following univariate1 Research Outputs by combining administrative data with existing surveys (for example, the Labour Force Survey (LFS)) and the 2011 Census:
the Office for National Statistics, Department for Work and Pensions and HM Revenue and Customs-modelled administrative and survey-based income estimates (2017 release)
qualifications from the Department for Business, Energy and Industrial Strategy and Department for Education for under 30s (2017 release)
number of rooms and floorspace using Valuation Office Agency data (2017 release)
commuter flows from mobile phone data (2018 release)
These outputs will be dependent on access to data and their statistical quality.
In the longer-term, the Research Outputs will be extended to include further topics as data become available and also to introduce multivariate2 outputs. We’ll continue to seek your feedback on these outputs and focus future research in response. A series of independent methodological reviews will be held to give external assurance of our methods.
Notes for: What are the plans for the future?
Outputs using only one variable, for example, estimates of the size of the population by age.
Cross-tabulated outputs using more than one variable, for example, unemployment rates by ethnic group at small areas.
7. How do you give feedback to data suppliers about the statistical quality of their data sources and to improve data sources?
We share information with suppliers on the statistical quality of their data throughout our research because it is mutually beneficial. The information we share with suppliers is provided only at aggregate level to ensure no information is shared about identifiable people, households or addresses. In turn, suppliers advise us on our Research Outputs. This ensures that all information is of the highest statistical quality.Back to table of contents
8. How will administrative data be used to make improvements to the 2021 Census production and outputs?
A significant change from the 2011 Census will be the greater use of administrative data in the design and conduct of the 2021 Census.
We’re pursuing access to further administrative data. These data fall into three types, which include:
activity information such as evidence of contact with government agencies, to help improve our estimates of the size and geographical location of the population from administrative data
information to supplement the address register such as information on occupancy status
information about characteristics of the population such as ethnicity, income or qualifications, to potentially replace or supplement information collected in the census
There are three areas where administrative data will improve the efficiency, effectiveness and quality of the 2021 Census.
We’ll also be seeking to integrate additional administrative data within the 2021 Census outputs as much as possible. More information is provided in the 2021 Census Design Document.Back to table of contents
9. What about Scotland and Northern Ireland?
Scotland is also considering options for producing future census and population statistics using administrative data, and Northern Ireland is considering options for using administrative data to support the 2021 Census. We currently have access to datasets that include only persons registered for services in England and Wales. Both the Northern Ireland Statistics and Research Agency and National Records of Scotland have continuous involvement in our research and will consider any potential for producing Research Outputs separately. Whilst methods may be similar, some of the source datasets used by the devolved administrations may differ from those used for England and Wales.Back to table of contents
10. How do you ensure that all this activity and future plans are acceptable to stakeholders?
We continue to engage with a range of stakeholders about this work to gain their support. As part of this, we've communicated progress through presentations delivered to the Market Research Society, the Royal Statistical Society, the British Society for Population Studies and the Statistics User Forum. Regional working groups have also been set up to discuss findings from Research Outputs and improvements to methods. We also host a research conference to present our current plans and recent findings. Our first Research Conference took place in June 2016 and a report summarising this conference is available.
We also work closely with data suppliers to understand and improve statistical quality issues identified in the data. A Data Suppliers Group has been set up for this, which includes members from important government departments across Whitehall. This group is a forum for discussing issues relating to data sharing, sharing research and building stronger relationships between data suppliers and users.Back to table of contents