Census collects LGBTQ+ data - but will we see it?
Australia is counting its LGBTQ+ population for the first time — but the number is only as useful as what the ABS chooses to do with it.
For the first time in its history, the 2026 Australian census will ask respondents how they describe their gender and their sexual orientation. Both questions are optional, both apply to anyone aged 16 and over, and both represent something genuinely useful: a national, self-reported count of a population that has, until now, existed in the official data only as an absence. The questions are well-designed enough. The harder question is what actually happens to the answers.
The case for these questions doesn't require a political opinion
There is a case for these questions that does not require anyone to have a political opinion about gender. Healthcare planners cannot target services toward populations they cannot locate. Housing researchers cannot study overcrowding or homelessness among specific groups without a baseline count. Epidemiologists cannot track health disparities without knowing who is in which group. The absence of this data has not made those disparities disappear — it has just made them harder to quantify and easier to dismiss. Equality Australia's legal director Heather Corkhill put it plainly: "We see gaps in access to healthcare, health disparities, poverty levels, housing and homelessness disparities. But what we don't have is a full picture across Australia." The census is designed precisely to produce that picture.
But here is where the analysis has to be honest about what the data might show, and why that matters as much as whether the data gets collected at all.
A census count will be smaller than some have claimed — and that's fine
Advocacy groups have, for years, cited population estimates that vary enormously depending on methodology, question wording, and sample. Some surveys, particularly those conducted online through convenience sampling, return LGBTQ+ identification rates that sit well above what a rigorous national count is likely to produce. An ABS analysis combining data from nearly 45,000 respondents across four recent health surveys put the proportion of Australians aged 16 and older identifying as LGBTI+ at around 4.5 per cent. Younger cohorts identify at higher rates, and the trajectory is clearly upward across generations, which is itself meaningful data. But the gap between that figure and some of the numbers that circulate in policy debates is wide enough to matter when resources and program design are on the table.
A census count will not resolve every definitional question — the questions are optional, the categories involve self-description, and some people will decline to answer. But the ABS has strong methodology, a large sample, and no financial stake in the result landing in a particular place. Whatever number it produces will be the most defensible baseline Australia has ever had. That is genuinely valuable, for advocates and sceptics alike. It is harder to overclaim a population size when the census has put a number on it. It is also harder to underfund services when the census has confirmed the need.
It is harder to overclaim a population size when the census has put a number on it. It is also harder to underfund services when the census has confirmed the need.
The generational gap in the data will be the most instructive finding
The generational dimension of the data will be particularly instructive. If identification rates among 16-to-25-year-olds are substantially higher than among those aged 50 and over — which preliminary evidence suggests they will be — that raises real questions about whether this reflects genuine population change, shifting social norms around disclosure, or some combination of both. Cross-referencing those figures against socioeconomic indicators and geography will tell researchers things about social mobility, regional disadvantage, and intergenerational change that are currently impossible to see. That is the kind of analysis that actually improves policy.
The data is only as useful as the ABS's willingness to release it fully
There is, however, a legitimate question about whether the ABS will release the data in a form that allows this analysis to be done. The bureau excluded intersex status from the census after testing found the question was not well understood, which was a defensible call. What would be less defensible is releasing gender and sexuality data only in aggregate, or in forms that prevent cross-tabulation with income, age, housing tenure, or location. The census is only as useful as what you can do with the numbers. It is also worth noting that this census has already attracted scrutiny from the Australian National Audit Office over its cyber security arrangements, which is a reminder that data collection comes with data protection obligations that the ABS has not yet fully addressed.
Counting people is a precondition for governing them well. Australia will now count its LGBTQ+ population for the first time. The number will very likely be smaller than some have claimed and larger than others have assumed. Both outcomes are fine. What matters is that the count is real, the methodology is sound, and the data gets released in a form that researchers can actually use.
A census that produces a number and then buries it in a summary table has not solved the problem. It has just made the problem slightly harder to argue about.
Frequently Asked Questions
Will the 2026 Australian census ask about sexual orientation and gender identity?
Yes. For the first time in its history, the 2026 census will include optional questions on sexual orientation and gender identity for all respondents aged 16 and over. Both questions are self-reported and voluntary.
What percentage of Australians identify as gay, lesbian, or bisexual?
A 2021 ABS household survey found around 4.5 per cent of Australians identify as gay, lesbian, or bisexual. Identification rates are higher among younger cohorts, and the trend is clearly upward across generations.
Why does it matter that the census collects LGBTQ+ data?
Without a reliable population count, healthcare planners cannot target services, housing researchers cannot study homelessness among specific groups, and epidemiologists cannot track health disparities. The data has not existed officially, which has made known disparities harder to quantify and easier to dismiss.
What are the risks with how the ABS releases the census LGBTQ+ data?
The main risk is that the data is released only in aggregate form, which would prevent researchers from cross-tabulating sexuality and gender identity against income, age, housing tenure, or location. That kind of cross-referencing is what makes census data useful for policy — a headline number without it tells researchers very little.
Will the census count settle debates about how many LGBTQ+ Australians there are?
It will establish the most defensible baseline Australia has ever had, but it will not resolve every question. The questions are optional, so some people will decline to answer, and the results will reflect self-description at a single point in time. What it will do is make overclaiming or undercounting much harder to sustain in policy debates.