Overview

Microdata samples are anonymised individual level records sampled from a single census. They contain a range of individual and household characteristics that can be used to carry out analysis not possible from standard census outputs. Microdata samples range in size from 10% to 1% of households or individuals.

Microdata samples do not contain personal information that could identify any household or individual. We remove information that might identify a person, such as names, addresses and date of birth. We also apply statistical disclosure control (SDC) methods to the microdata samples, for example, record swapping, collapsing variables and restricting detail.

Census 2021 microdata samples

Census 2021 microdata samples for England and Wales will be classified in three ways:

  • public, available to all

  • safeguarded, only available to data analysts under certain terms and conditions

  • secure, only available to approved or accredited researchers

Census 2021 microdata samples for England and Wales will be similar to those available for the 2011 Census. Following user feedback, we also plan to produce two new microdata samples. The first is a safeguarded household microdata sample. The second is a household microdata sample that will contribute to the University of Minnesota's Integrated Public-Use Microdata Series (IPUMS) project. This is an international project that brings census microdata together from over 100 countries.

Public-access sample

This file will contain a sample of data up to 1% of individuals, provided down to a regional level of geography. It will contain around 20 variables and a low level of detail.

Safeguarded individual region sample

This file will contain a sample of data up to 5% of individuals, provided down to a regional level of geography. It will contain around 120 variables and a higher level of detail.

Safeguarded individual grouped local authority sample

This file will contain a sample of data up to 5% of individuals, provided down to a grouped local authority level of geography. It will contain around 120 variables and a low level of detail.

Safeguarded household sample - new for Census 2021

This file will contain a sample of data up to 1% of households, provided down to a regional level of geography. It will contain around 50 variables and a low level of detail.

As we developed this file, we found that individual characteristics posed as an increased risk of disclosure, resulting from individuals being grouped into households. To reduce this risk, we anticipate that we will:

  • provide some characteristics about households, instead of individuals

  • need to combine civil partnerships with marriage

  • not identify same-sex couples

  • not include occupation, but are considering how feasible it is to include the National Statistics Socio-Economic Classification (NS-SEC) and industry instead

IPUMS sample - new for Census 2021

This file will contain a sample of data up to 1% of households, provided down to a regional level of geography. It will contain around 30 variables.

Secure individual sample

This file will contain a sample of data up to 10% of individuals, provided down to a local authority level of geography. It will contain around 200 variables.

Secure household sample

This file will contain a sample of data up to 10% of households, provided down to a local authority level of geography. It will contain around 200 variables.

Accessing microdata samples

Public microdata samples will be available to view from the ONS website.

Safeguarded microdata samples will only be available to data analysts, under certain terms and conditions. These will likely be held by the UK Data Service in line with previous censuses.

The most detailed secure microdata samples will only be available to approved or accredited researchers through the Secure Research Service (SRS). Learn more about accessing secure data.

As census microdata samples do not contain information that can identify individuals, researchers will not be able to use these for data linkage purposes. If an accredited researcher requires this and can demonstrate a wider research benefit beyond their individual project, they can make a request directly to the Office for National Statistics. If the request is approved, we will prepare this work internally and transfer a de-identified version to the SRS. Accredited researchers can then apply to use these data. If a researcher would like to make a request for these data, please email adrcuration@ons.gov.uk.

Release plans

We will begin releasing microdata samples during phase three of the Census 2021 outputs release schedule. We plan to release the public microdata samples first, then follow these with the more complex safeguarded and secure samples.

UK census microdata

We will produce Census 2021 microdata samples for England and Wales. Where possible we will harmonise these with microdata samples from Northern Ireland's census in 2021 and Scotland's census in 2022.

We also plan to harmonise Census 2021 microdata samples with samples from 2011 Census, to better allow comparisons.

Working with others

We are working with a range of internal and external stakeholders through our microdata working group. So that we can better meet our users' needs, this group advises us on the design, creation and dissemination of 2021 Census microdata samples.

Internal members include topic experts on:

  • migration

  • travel to work

  • demography and census transformation

  • population estimates

  • statistical disclosure control

External members include:

  • National Records of Scotland

  • Northern Ireland Statistics and Research Agency

  • Welsh Government

  • UK Data Service

  • local authorities

  • academia

  • market researchers

  • commercial researchers

More information

If you have any questions about Census 2021 microdata samples or are interested in joining our working group, please email census.outputs@ons.gov.uk.