No pride in numbers



Equal Rights

Written by

Olivia Roney

August 2021

A few weeks ago, I filled out my first census.

I've certainly been captured before. In fact, many times as a resident of my family home.

It's something I took some pride in, akin to turning 18 and being able to vote for the first time. To contribute to something bigger than myself, and to shape outcomes for communities I am part of.

And while it's true that we give our data everyday - by allowing cookies, scrolling Facebook and reading the news, no data collection is quite as deliberate, impactful and widely used as the Australian Bureau of Statistics' (ABS) Census. So when an important part of Australia's identity isn't captured, the consequences are serious and long-lasting.

The way this data collection has been designed, collated and counted has changed in many ways in the 100+ years since the first official Census.

The history of the Census

The first recorded government data collection of the Australian population was in 1828, and the first formal census held in April 1911. The first census was hand-delivered by men on horseback, handwritten and hand-counted.

It wouldn't be until 1921 that Australia would borrow automatic machines to read census punch-cards filled out across the country.

In 1954, for the first time women were invited to process and code the data collected during the Census.

In 1961, the machines used to tabulate Census forms were so large that in order to get them inside the department, a hole was cut in the wall of the building.

In 1966,  the 'head of the household' (read: the man) had to be identified and this person was to fill out the form:

It wouldn't be until 1967, concurrently with significant Constitutional reforms, that the Census captured data on Aboriginal Australians in a meaningful way.

2006 was the first time Australians had the option of completing the census online, and by 2016 it was the default option (in 2021 you could still elect to receive a paper form).

The takeaway is that we have adapted the way we collect Census data to reflect the times - it would be too inefficient not to. And for the most part, the data design too has changed.

The history of Census data

Looking through the Census questions over the years, you get a decent picture of the sociological, economic and political changes in Australia since 1911 - we sent citizens off to multiple wars, the world lived through a global pandemic, women and Aboriginal people fought for their respective rights and technology was on the rise. To reflect some of this, the following topics have been and gone in previous Census':

The Census has a poor history of capturing Aboriginal and Torres Strait Islander communities. At Federation, it was specified that these communities were not to be counted. Communities were counted to varying degrees in other ways, though in extremely insensitive ways - according to assessments of 'blood status', and even included questions about 'half-caste' people prior to 1967.

In 1967, thankfully these practices were scrapped and Aboriginal and Torres Strait Islander people were counted more appropriately.

The point here is that we don't capture the data we used to, and we don't capture it in the ways we used to either. And rightly so - the implications and assumptions behind certain questions are archaic and irrelevant - rubella has been eradicated in Australia and most people have a 'personal computer'.

So what data do we capture in 2021?

The data we do capture

The topics covered by the 2021 Census include, among other aspects and intersections:

Not to be discounted, the Census is a massive undertaking - the greatest logistical effort outside of wartime in Australia. The points above capture many intersections of identity and lived experience. It paints a detailed picture of who Australia is made up of, how we live and love, care for each other, volunteer, study and work. What we cannot see a clear picture of is LGBTIQ+ Australia.

The data we don't

The blaring emission from every the Census since 1911 up to the most recent in 2021 is meaningful data on Lesbian, Gay, Bisexual, Transgender, Intersex and Queer+ (LGBTIQ+) communities. The ABS heard and rejected draft questions that would allow sexuality and gender to be captured in 2021.

On Gender

The ABS defines a person’s sex as 'being based on their sex characteristics, such as their chromosomes, hormones and reproductive organs.' The ABS defines gender as 'social and cultural differences in identity, expression and experience as a man, woman or non-binary person. A person’s gender may differ from their sex and may also differ from what is indicated on their legal documents.' Sex is not the same as gender. I agree with the ABS so far.

But, it was decided early-on in 2020 that data on sex would be captured, but not gender. On sex, the 2021 Census asked:

People who wish to report their sex as other than male or female had the option of a ‘non-binary sex’ response category - despite this also confusing the gender binary with physiology, as typically 'non-binary' is a reference to gender identity (though may capture intersex Australians).

Nevertheless, initially this was welcomed by advocates until it emerged that those who selected 'non-binary sex' on the Census will actually be automatically reassigned as either male or female randomly. I can't think of a more blatant and intentional erasure of identity and good data.

It emerged that those who selected 'non-binary sex' on the Census will actually be automatically reassigned as either male or female randomly. I can't think of a more blatant and intentional erasure of identity and good data.

The ABS' approach to sex and gender demonstrates a fundamental misapplication of these concepts. It also demonstrates a disingenuous approach from the ABS - if they really wanted to capture this data they could.

And that's not all - the 2021 Census also failed to capture the sexuality of Australians.

On Sexuality

Despite going through a consultation process for its inclusion, the ABS ultimately decided that there would be confusion on why such a question was asked, among data privacy concerns. So first and foremost, the decision of whether to include a question on LGBTIQ+ communities was framed from the perspective of non-LGBTIQ+ communities - that the general public wouldn't understand.

LGBTIQ+ Australians have had quite enough of asking heterosexual Australians what they think of us and our relationships (looking at you plebiscite).

Putting that problematic approach to one side, the Census does ask what the relationship of one resident is to another - could be housemate, daughter, partner etc and this, the ABS argues, is how we will get this data.

But what this captures is actions and behaviour and fundamentally, not identity.

The suggestion here is that this is enough to capture LGBTIQ+ communities.

It is not.

A person's identity includes whether they are gay, bisexual, lesbian, queer, straight, pansexual etc. This is part of who you are as a person, and isn't decreed simply because you are in a relationship with a specific person.

What about every LGBTIQ+ Australian who was not in a relationship on the night of August 10, or bisexual people in opposite-sex relationships? These people would not be counted among Australia's LGBTIQ+ diversity.

As a result, Australia's population loses much of its diversity and texture. The Census is reduced in accuracy.

This erasure will have long-lasting and human effects when we look at the outcomes this data informs, and moreover, what outcomes poor data informs.

The importance and use of Census data

By the Australian Bureau of Statistics (ABS) own words, the census:

'provides a reliable basis for estimating the population of each of the states, territories and local government areas, primarily for electoral purposes and for planning the distribution of government funds. Census data are also used by individuals and organisations in the public and private sectors to make informed decisions on policy and planning issues that impact on the lives of all Australians.'

The ABS is proud of the reach and uses of its data - and rightly so. By their own words, census data collections are the largest logistical peacetime operation in Australia's history (in COVID times, I wonder if this still rings true).

Just some of the organisations that use ABS Census data include:

Not to mention the millions of Australians who use this data in our thesis', work research reports, primary school presentations, business pitches and dinner-table arguments.

In the lead-up to the 2016 Census, Queensland Treasurer Curtis Pitt said:

“Queensland businesses, governments and communities use Census data to build a picture of our regions and communities and better respond to changes in our population. The information is valuable in making informed planning decisions about housing, transport, education, industry, hospitals and the environment.”

LGBTIQ+ communities are simply not in these equations—for mental health services, family violence approaches, resources for elders and multicultural communities and the COVID-19 response.

There can be no 'informed planning decisions', and there is no complete picture to be painted from this data.

Put simply, there is no pride in these numbers—and we are all the worse for it.

More writings