Do we need a census, when most of the interesting things you want to know about me aren’t in it? 

There’s been quite a lot of public discussion about the security and privacy of individual data within the Census both before and after the 2016 iteration.

If we look at the rationale for the Census, the Australian Bureau of Statistics (ABS) says:

The Census of Population and Housing (Census) is Australia’s largest statistical collection undertaken by the Australian Bureau of Statistics (ABS). For more than 100 years, the Census has provided a snapshot of Australia, showing how our nation has changed over time, allowing us to plan for the future.

The aim of the Census is to accurately collect data on the key characteristics of people in Australia on Census night and the dwellings in which they live. … The information you provide in the Census helps estimate Australia’s population, which is used to distribute government funds and plan services for your community – housing, transport, education, industry, hospitals and the environment. Census data is also used by individuals and organisations in the public and private sectors to make informed decisions on policy and planning issues that impact the lives of all Australians. 

But for the first time in 2016, the ABS would keep names and addresses of respondents on file for four years, rather than the earlier 18 months. It also emerged that the ABS had been using names and addresses from Census data to cross-reference records kept by other Commonwealth and State government agency departments since 2006. In addition, a randomly selected 5% of respondents had been tracked from Census to Census since the same year.

It’s ironic that many of the same people who are calling out security and privacy concerns about the Census, and indeed many of the Australian population, make even finer grained and more current information available online through social media, with little thought for what might be done with that data if it was collected and analysed systematically. That’s over and above the information that Government collects about me and my family through numerous other means.

It begs the question – if Government want to know where I live, how much money I earn, what kind of education I have, how much time I spend online, whether I own or rent a house, how many kids I have and where they go to school, or where I was born, how many hours a week I work, why use a Census at all?

For each Census that is run, there’s generally a public consultation process that takes place to inform design of the Census. But to be equally fair, most of these consultations are predicated on the idea that a Census is necessary.

In 2011, the UK National Statistician opened Pandora’s box by asking in public consultation what the future held for population statistics, rather than the more traditional focus on “which questions should be asked”.  The lid was firmly nailed shut quickly, and the (next) 2021 UK census is proceeding on very traditional lines.

Meanwhile, there are numerous academic research activities which are focused on deriving those same key characteristics of Census respondents (and more) from data which is openly available both from public sources and social media. The results suggest that highly comparable data can be obtained both at small scale (individual to suburb/postcode), and even at acceptable precision at large scale (state-wide). Commercial enterprises have already worked this out, and there’s a whole group of companies who have made a lucrative business of sourcing and supplying high quality individual, group and small area data.

The Census is a useful reference for the long term, but it’s increasingly questioned as a medium term strategic input, and almost entirely irrelevant for short term / high precision targeting. Perhaps it’s time to think about whether or not a high proportion of Census information is already known to Government (and funding spent sourcing/integrating it rather than a broad scale population survey). Or even, if we’re courageous, like the US Bureau of the Census did in 2013, to open Pandora’s box again and ask if someone else already has better information on us that could be more effectively leveraged for medium and long-range planning.

Baden Hughes
Associate Consulting Director

Dr Vanessa Douglas-Savage
Consulting Director