Case Study: The 2021 Census
 

USING INTEGRATED ADMINISTRATIVE DATA TO IMPROVE THE 2021 CENSUS

WHAT IS THE DATA CHALLENGE?

The count of people in Australia on Census night is important because population estimates help set electoral boundaries for all levels of government, underpin the distribution of billions of dollars in public funding, and inform planning for services and infrastructure for every community in Australia.

Although the Census receives high levels of support and participation, a small number of the population do not respond. Where no Census response was received, the challenge is to determine which houses were vacant on Census night and hence should not form part of the Census count.

For households that we have determined were occupied but did not return a Census form, we estimate a count of persons with their age, sex, and marital status by ‘borrowing’ Census counts from a similar, nearby household where forms were received (known as ‘donor’ households). This ensures the Census is more representative of the true number of people in the country on Census night, both nationally and for local regions.

However, a potential risk with this approach is that donor households tend to over-represent older Australians who are generally more likely to have responded to the Census.

HOW CAN INTEGRATED ADMINISTRATIVE DATA BE USED?

We are currently undertaking research into how administrative data can be used to improve the Census count. Administrative data is data collected as part of the day-to-day processes and recordkeeping of governments and other organisations, and are made available to the ABS for statistical purposes.

This research involves assessing the quality of available administrative data sources, and then integrating this data with the 2016 Census to undertake analysis.

Our research is showing that integrated administrative data can help us in the following ways:

1. Choosing more accurate donor houses

Using administrative data (such as counts of people from de-identified Medicare and Centrelink data) to guide our choice of donor houses can help us more accurately reflect the ages of people in houses that don’t provide forms.

2. Improving our decision on whether a house is occupied

Administrative data (such as electricity connections, rentals data, and updates to address information for tax payments) can be used along with the observations we make when we visit houses in person to improve our final decision on whether houses were occupied on Census night.

The ABS takes data security very seriously and the protection of privacy is the ABS’ highest priority. This research is undertaken only by approved users within a secure ABS environment. All administrative data projects comply with legislation, including the Census and Statistics Act 1905 and the Privacy Act 1988.

WHAT ARE THE BENEFITS?

Census data that more accurately represents the Australian population means better decisions across government and community services to improve the lives of Australians.

NEXT STEPS

Your privacy is important to us. Any intentions to use administrative data in the 2021 Census will be considered as part of an independent Privacy Impact Assessment of the 2021 Census, to be conducted in 2020.

The ABS will also work closely with the agencies providing administrative data to ensure the use of administrative data to improve the 2021 Census delivers the most value to the public.

The results of our research and any subsequent plans to use administrative data will be progressively released in the lead up to the 2021 Census.

FURTHER INFORMATION