Created 5 July 2017
By National Research Manager, Jodie McNair
The Australian Bureau of Statistics (ABS) published the first comprehensive release of 2016 Census data in June 2017. Data relating to employment qualifications and population mobility is slated for release in October 2017.
The big question in the minds of many Australians involved in policy, economic, planning and other fields that rely heavily on the Census is whether the new data will be reliable. The official line from the ABS is that “On Census night, 9 August 2016, the online form suffered a series of outages due to distributed denial of services (DDoS) attacks. Australians accessing the online form did not cause a system failure, submission rates were within expectations and load capacity. An attempt to restore the system during the fourth DDoS attack led to the failure of one of our supplier’s routers, compounding network issues.” In addition, in the lead up to the 2016 Census, there was significant public debate regarding the intention of the ABS to retain all the names and addresses collected “to enable a richer and dynamic statistical picture of Australia.” This second issue has largely been overshadowed by #censusfail in the months since August 2016.
The ABS has been active in their efforts to assure Census users that the 2016 collection is reliable, citing high participation, rigorous quality checks and establishment of an independent assurance panel.
“I’m a bit of a freak for evidence-based analysis. I strongly believe in data.” Gus O’Donnell, economist
What are the options for Census users? Disregard the 2016 collection? Qualify every use of Census data as being ‘less reliable than optimal’?
The Census is “unique in that it is a total survey of the population, covering a range of social and economic variables. At present, it is the only way such data is obtained in Australia.” Other surveys are representative, meaning that broader implications are drawn from a sample collection. For example the 2015 SDAC included “around 63,500 people from over 25,500 private dwellings, and a further 11,700 people from 1,000 establishments.” In short, there is no equivalent data source in Australia that compares to the Census, and currently no way to cobble together ‘virtual census’ from existing administrative data collections. Interestingly, international efforts along these lines “Ironically…need to be checked against a census.”
Social scientist Dr Nicholas Biddle presents an important reminder: “The most important thing to keep in mind throughout this period is that no Census has ever been perfect. There are no halcyon days where everyone filled out their Census on the allocated night, every form was filled out completely, honestly and accurately, and it was collected by ABS staff seamlessly and with no fuss.”
For more information, give Jodie McNair a call on 1800 983 776 or email firstname.lastname@example.org
 2016 Census of Population and Housing: Details of Overcount and Undercount, Australia 2016