2089.0 - Planning the 2021 Census, Australia  
Latest ISSUE Released at 11:30 AM (CANBERRA TIME) 16/09/2019  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product

Data quality assurance

Among national statistical agencies, data quality is generally assessed on whether the data is fit-for-purpose. This means that data quality is assessed not only on the accuracy of statistics, but their timeliness and relevance.

ABS has a quality framework for all its statistics, which it will be applying to Census data. More information on the ABS Data Quality Framework can be found on the website.

In order to produce high quality data from the 2021 Census, we are focussing on the design of the Census form and the procedures for data collection and processing.

Quality assurance

The quality assurance process makes sure the Census data is fit-for-purpose and meets the requirements of our data quality framework. The quality of Census data will be managed and assured through several activities.

Statistical risk management

Statistical risk is the chance or likelihood of something going wrong in a process that affects the quality or integrity of data. Statistical risks can occur at any time during design, execution and processing of the Census. Statistical risk arises for various reasons, including inadequate information, changes to existing processes, human error or changes in the external environment. Statistical risk is managed through the data quality controls discussed below.

Data quality controls

Data quality controls are business processes that ensure a part of the Census runs smoothly, for example management of field staff or distribution of Census materials to all dwellings. There are also data quality controls that explicitly manage the quality of Census data.

Data quality controls for the 2021 Census include:

  • use of quality checks at key stages during Census collection and processing to ensure that any potential causes of data errors are detected early and rectified
  • testing of processes, systems, field operations and new content will be undertaken in the lead up to the Census:
    • Questions have been tested through focus groups, individual interviews using cognitive testing techniques, and field tests. This will ensure high quality data can be obtained for the proposed new and amended topics.
    • Field tests will help refine the wording of questions and ensure new and changed questions do not impact the quality of responses.

Census Post Enumeration Survey

The Census is a large and complex exercise. There are always going to be a small number of people missed or who are counted more than once.

People may be missed for several reasons. For example, they may have been travelling and were difficult to contact, the person completing the form may have mistakenly thought that they shouldn't be counted, or the household was difficult to find. People can also be counted when they should not have been. This can occur when a person was overseas on Census night, but they were still included on a Census form at their home.

The Post Enumeration Survey is a short survey run the month after the Census to determine how many people were missed, counted more than once, or counted in error.

It is run separately from the Census to provide an independent assessment of the completeness of the Census.
From the Post Enumeration Survey, it is possible to determine the number of people who should have been counted in the Census. This is then compared to the number of people who were counted in the Census. The difference is referred to as ‘net undercount’ or ‘net overcount’.

For more information

For further information on the 2016 Post Enumeration Survey and net undercount, refer to the Census of Population and Housing: Details of Overcount and Undercount, Australia, 2016 (Cat.no. 2940.0).

2021 Census Statistical Independent Assurance Panel

The ABS will establish a 2021 Census Statistical Independent Assurance Panel. The 2021 Panel will independently assure the 2021 Census data and publicly release a report on their findings.

This initiative follows the establishment of the 2016 Census Independent Assurance Panel. The 2016 Panel published their findings in the Report on the Quality of 2016 Census data, which is available on the ABS website. The 2016 Panel recommended that the ABS continue the practice of an Independent Assurance Panel reviewing the quality of the Census data to provide greater transparency and accountability.