The Australian Census and Changing Family Forms

It is now more than forty years since the first births of babies born through in vitro fertilization (IVF) in Australia. How have changing forms of reproduction and family-making been reflected – or not – in Census documents?

It is now more than forty years since the first births of babies born through in vitro fertilization (IVF) in Australia. In the wake of the 2021 Australian Census, I was curious to see how changing forms of reproduction and family-making are reflected – or not – in Census documents.

Australian Census form, 2022. Image courtesy of the Australian Bureau of Statistics.

In the last four decades, there have been extensive changes to the ways in which Australians make families due to developments in assisted reproduction. The use of artificial insemination by donor (AID) has broken the link between biological parenthood and social parenthood. More recent development of procedures in third party provision of ova allows for separation between the biological provider of ova, the birth mother, and the social mother. These developments have also made possible the use of commercial or altruistic surrogacy, with complex configurations of gamete providers, birth mothers and social mothers.  

The first tranche of  the results of the 2021 Census were announced in June 2022, with further reports expected in late 2022 and early 2023. The Census no longer demonstrates assumptions that all people live in nuclear families with two married heterosexual parents and their children. Complex and blended family forms are recognised in various ways. Even so, it has long been argued that the Census only reflects what governments choose to count. For example, in earlier decades the Census did not recognise the unpaid work of married women, who were largely defined in terms of their relationship to a male partner. In the most recent census, there are several questions about paid and unpaid work, caring responsibilities and volunteer activities, with no assumptions as to the gender of the person who carried out such work (Questions 55 to 57).

The Census includes explicit recognition that there are single person households and various relationships including marriage, de facto relationships, and the co-residence of adults who may not be in a marital or de facto relationship (Question 9, Question 10). In questions about parenting, same sex relationships are explicitly mentioned. In general, the language is gender-neutral: ‘Person 1’, ‘Person 2’, ‘spouse’, partner’. The number of couples in Australia who formalise their relationship as marriage, though, has been steadily decreasing, with the COVID-19 crisis leading to a steeper drop in the number of marriages.

There is one question (Question 7) where individuals are asked to identify as ‘male’, ‘female’, or ‘non-binary’. Since the census was administered, this question has been criticised for its imprecision. Would transgender or intersex individuals recognise themselves in this characterisation?  The Australian Bureau of Statistics (ABS) explains that it was only required by the government to collect data on (biological) sex and not (social and cultural) gender identity. The 2021 census did not include any specific questions on whether people identify as heterosexual, gay, lesbian, bisexual or transgender. The ABS has undertaken to conduct further consultation on questions of sex, sexual orientation and gender identity. Community groups have lodged a formal complaint about the lack of questions on sexual orientation.

The Australian Census can be completed via an online form, 2022. Image courtesy of the Australian Bureau of Statistics.

In questions about who was away from the residence on Census Night, it is recognised that some children may have been absent from the house due to ‘shared care arrangements’ (Question 4). Questions 12 to 14 ask about the person’s current and prior addresses. It recognises that some individuals may have ‘no fixed address’ because of ‘family conflict’ or ‘eviction’.

Several questions invite responses on children in a household. Question 37 asks women over the age of fifteen whether they have ever given birth to a baby and how many babies they have borne. It is explicitly mentioned that this does not include adopted children, foster children or stepchildren. (Stepchildren can be mentioned in Question 9, but there does not seem to be an explicit question about adopted children.) The ABS is presumably interested in compiling national statistics on fertility rates. Parents are asked whether any child is ‘the child of Person 1’, ‘the child of Person 2’, or the child of ‘both person 1 and person 2’. A transman who has borne a child may also find it difficult to answer this question.

What about children born through various forms of assisted reproduction? The first thing to note is that the Census does not distinguish between biological and social parenthood. This is not to suggest that either biological parenthood or social parenthood should be privileged. For many adoptees and children born of conception through the third party provision of gametes, questions of heritage can be extremely important in an emotional and visceral sense.

If a woman has borne a child conceived through artificial insemination by donor (AID) sperm, then there could be scope to distinguish between the provider of sperm and the social father/parent. This has become controversial in recent years, with children conceived through third party provision of sperm campaigning for the right to access information about their sperm provider fathers. Because of the earlier convention of anonymity of sperm providers, someone born through third party provision of sperm would be unlikely to be able to answer relevant questions about their biological father or his ancestry or his nationality. 

In 2016, the State of Victoria passed legislation which requires that children be able to access information about their donor fathers. For those born in the earliest days of AID, however, when records were often destroyed, there has been little way of gaining information about their heritage. One could imagine that such people might be stopped in their tracks by such questions as Question 18: ‘In which country was the person’s father born?’ 

The campaign for children to have access to information about their sperm provider fathers has been spearheaded by such people as Lauren Burns. Her situation was portrayed in ABC Television’s Australian Story in 2014, updated in 2021, and she has recently published a memoir. The campaign was given urgency by the story of Narelle Grech, who passed away from bowel cancer, and only managed to meet her biological father six weeks before her passing. Without information about her genetic heritage, Narelle had no way of knowing that she had a genetic susceptibility to this disease.

In the Census questions, there is, however, a recognition that not all individuals know where their mother or father was born. In those cases, they are advised to provide the country of birth of their ‘second parent’. In the case of ‘same sex parents’, they can provide the country of birth of one of their parents (Question 19). Similarly, Question 22, ‘What is the person’s ancestry’, might be difficult to answer for someone conceived through anonymous sperm provision. In addition, the instruction to only provide two forms of ancestry cannot capture the complex reality of many ethnically diverse families. Nevertheless, the most recent census does provide a snapshot of an ethnically diverse Australia.

Because the question about birth explicitly refers to how many children an individual woman has borne (Question 37), this would not capture any information about the genetic heritage of her offspring. In some cases, children have been born through both third party ova and third party sperm provision. In such cases, they would have two biological parents and a birth mother and a social mother.

In the case of children born through surrogacy, the social mother and the birth mother are distinct. This would mean that the social mother would not be able to answer Question 37, but that the children would appear in answer to other questions about residence in the household. Where a woman has borne a child for altruistic or commercial surrogacy and relinquished the child to a friend or relative, the relinquishing mother would be able to answer the question of how many children she had borne, but her household structure would not reflect this. The same would be true of those who had relinquished children for adoption.

In other cases, AID and/or IVF may have been undertaken with gamete providers who are known to the parent(s). The Census might not be able to capture relationships where, for example, lesbian parents and their children maintain a relationship with the sperm provider father, although he may not be resident in the household. Similarly, in cases where gay male parents are in contact with the provider of gestational surrogacy, this may not be captured in questions about household structure. As far as the Census is concerned, the basic unit of society is the household.

These issues are compounded when gametes are obtained or purchased from overseas, or where international surrogacy arrangements have been entered into. The questions about children’s ancestry or heritage will be difficult to answer in these cases. Do they simply refer to the ancestry of the social parents, or should they recognise the heritage of the providers of gametes or gestational labour? Different parents have differing answers to this dilemma, particularly where the child or children do not ‘look like’ their parents. As Jaya Keaney has explored, some parents take an assimilative approach, while others actively encourage their children to explore the culture of their birth mothers or providers of gametes. 

In recent years, we have seen the campaigns of Australians born through adoption or third party gamete provision to be able to trace their heritage. This will be even more challenging for those conceived from often anonymous providers of gametes in other countries. We can expect another generation to start the search for their transnational heritage, as has been carried out by those who were the subject of transnational adoption. It will be some time before the Australian Census captures the complexities of families formed through the use of assisted reproduction, particularly when that has been carried out across national borders. 


Vera Mackie is Emeritus Professor in the School of Humanities and Social Inquiry at the University of Wollongong.  She has recently published a chapter on Australian popular cultural representations of assisted reproduction: ‘Google Earth and Google Babies: Nation, Transnation and the Australian Reproscape’, in Paul Sharrad and Deb Narayan Bandyopadhyay (eds) Transnational Spaces of India and Australia (Palgrave Macmillan, 2022).

Vera tweets as @veramackie.

This entry was posted in Family, Reproduction, Research blogs. Bookmark the permalink.