How same-sex couples are captured in the Census
The 2017 Marriage Law Postal Survey paved the way for the historic law allowing same-sex couples to marry in Australia.
Since December 10th, 2017 when the law was passed, over 2,500 same-sex marriages have been registered in Australia.
But did you know that the Census has recognised same-sex relationships for much longer? Since the 2011 Census, same-sex couples in a registered marriage have been included (as these marriages may have been registered overseas). And same-sex defacto relationships have been included since 2001.
To understand how the Census looks at same-sex relationships, you have to look at how the Census constructs household information from the people within it.
The Census doesn’t ask outright what sort of household you belong to. Instead, it asks for each person, how they are related to Person 1 on the Census form (who is normally known as the Household Reference Person. Note that this is not the same as “Head of Household” which is a term that hasn’t been used for about 30 years or so).
Now, some same-sex couples may write their relationship in the “other” box. But the Census will record a same-sex couple if person 2 simply marks the “Husband or wife” or “De facto partner” boxes, and person 1 is the same sex as them.
There has been increasing propensity to identify over time, as LGBTQI issues have gained prominence, and same-sex relationships have become more accepted in society. So there have been large increases in the last few Censuses. Nevertheless, there are likely some same-sex couples who don’t identify, and if they (for instance) marked the “Unrelated flatmate” box, these would not be recorded as same-sex relationships.
So how many same-sex couples did the Census record?
The 2016 Census recorded a total of just under 47,000 same-sex couples in Australia.
The numbers are very even between male and female same-sex couples, with just over 23,000 of each. Overall this makes up 0.9% of all couple families recorded in Australia, with slightly higher rates in NSW and Victoria.
|Male same-sex||Female-same-sex||Opposite-sex couple||Total couples||% same-sex couples|
|New South Wales||9,265||7,494||1,580,124||1,596,884||1.0%|
These numbers are certainly increasing at a faster rate than the total population. The 2011 Census recorded 33,700 same-sex couples Australia-wide, so the 2016 figure represents about a 40% increase. 2006 had just on 25,500.
It’s important to remember that this is certainly an undercount of same-sex couples in the community. As well as some couples not identifying themselves, by marking the “Unrelated” box on the Census form, there is an undercount of all household types, by people being away from home.
If one partner is away on Census night, it should be correctly recorded, as the Census creates household information from “People temporarily absent” which is a page on the back of the form. But if both are away, even if counted together in a Non-Private Dwelling (such as a hotel), the relationship information is lost. This applies to all household types, not just same-sex couples, and arises because people in these establishments on Census night fill in personal forms, with no household info.
Also, it needs to be said that this count is not in any way a measure of “sexual orientation”, which would be an extremely sensitive and difficult topic to collect in the Census (and I suspect such a personal thing will never be part of any Census). It only measures people living in couple relationships and counted at home on Census night.
Same-sex couples can also include children in the household. In the 2016 Census, a total of 4.5% of male same-sex couples and 25.4% of female same-sex couples were families with children (and this is much lower than the 54.6% of opposite-sex couples who had children).
This is important information. Why don’t we have it in our community profiles?
.id’s Local Government Community Profile tool (profile.id) does include same-sex couples as part of the household type counts, under “How do we live?”. However, we don’t separate them out from all other households.
The main reason for this is the very small numbers. About 47,000 couples nationwide breaks down to at most 100 or so in most Local Government Areas. Once you break this down further to suburb/district level, you quickly run into confidentiality issues, and as with all Census data, the ABS makes random adjustments so you can’t identify individuals.
More information on this process (which applies to all topics) is available in our supporting information section in any community profile.
The other thing is that the distribution of same-sex couples is very “clumpy” – some LGAs have fairly high numbers, while many have close to zero.
This means that for most areas subscribing to our community profiles, the numbers of same-sex couples are too small to show, and may actually be misleading, due to the random adjustment, particularly as we break down to the suburb level.
Here are the top 10 LGAs for male and female same-sex couples in Australia. Note that Brisbane is a very large LGA with over a million people, and covers most of the Brisbane metropolitan area. The City of Sydney has by far the largest share of same-sex couples in Australia, where they make up 9.5% of total couples.
|Local Government Area||Male same-sex couples 2016||Local Government Area||Female same-sex couples 2016|
|Sydney (NSW)||2,952||Brisbane (Qld)||1,533|
|Brisbane (Qld)||1,793||Inner West (NSW)||919|
|Inner West (NSW)||983||Sydney (NSW)||740|
|Port Phillip (Vic)||597||Canberra (ACT)||655|
|Melbourne (Vic)||593||Darebin (Vic)||537|
|Yarra (Vic)||580||Gold Coast (Qld)||472|
|Canberra (ACT)||554||Moreton Bay (Qld)||472|
|Stonnington (Vic)||490||Moreland (Vic)||410|
|Gold Coast (Qld)||478||Central Coast (NSW)||334|
|Moreland (Vic)||374||Yarra (Vic)||276|
Most of these LGAs with significant numbers of same-sex couples are in inner metropolitan areas of Sydney and Melbourne.
If you need the number for your area, we can easily extract this for you, but for most areas it’s too small to appear on profile.id, and certainly too small to break down into subcategories. Or, let us know how you would use this data in the comments below!