2026 Census topic consultation

AIFS proposes that ABS considers adding a topic on relationship separation in future Censuses – expanding the current registered marital status question to better capture relationship separation for both marriage and cohabitation relationships. Such information is particularly important for understanding the changing patterns of dissolution of couple relationships in order to assist family-related services in local areas for better service delivery.
The patterns in couple formation and dissolution have changed profoundly over the last half century, with the rise in cohabitation and in relationship separations (divorce and or separation of cohabitation) being the main areas of change. The 2021 Census data indicated that one in five couples were living in cohabitation instead of being married to each other and most people in their 20s with a partner were in cohabitation. Cohabitation has also become a setting for having children, with a significant proportion of cohabiting couples having dependent children. According to the 2016 Census, 40% of people aged under 50 years with a cohabiting partner were living with children aged under 15 years, compared to 72% of their counterparts of the same age who were married. While these broader social changes have been largely captured in vital statistics (marriages and divorces) and surveys, it is unknown how many people experienced separation from cohabitation. More importantly, statistics on separation for planning purposes for local areas are lacking.
Currently, the only relevant question in the Census is registered marital status (married, separated, divorced, widowed, never married). This question does not distinguish whether or not people have been remarried and does not fully capture the extent to which people experience divorce, let alone separation from cohabiting relationships. Therefore, this question has become increasingly inadequate to capture the extent to which people experience divorce, and it is a poor proxy for statistics on relationship separation from both marriage and cohabitations.

A good understanding of relationship separation, including from cohabiting relationships, is of national importance. In the context of the rise in cohabitation, current statistics on divorce are increasingly a poor proxy for the statistics on relationship separations. Since 2009, cohabiting couples have had the same rights as married couples under the Family Law Act 1975 (Cth) (where they meet definitional requirements in the case of post-separation property division). They also have the same rights in relation to social services (e.g., government income supports/benefits). However, Census data, a key source of national data, currently capture only a segment of people who have experienced divorce and none of the people who have experienced separation from a cohabiting relationship. There is a strong need to bring Census data in line with social and legislative change by collecting better data on relationship separation.
Relationship separation is often the main or underlying issue for people seeking family-related services in Australia. The ability to identify the extent to which people have experienced separation from married and/or cohabitating relationships, and associated characteristics for local areas (or service catchment areas), is critically important for a better understanding of the needs of potential clients and to enable better service planning and delivery.
While relationship separation data can be obtained from surveys, survey data would not be suitable for making estimates for local areas due to the limitations of sample size. Only Census data would be suited for making local area estimates.
Including this topic in the Census would also enable us to identify rates of separation and divorce among small socio-demographic groups that are not well represented in other national surveys. For example, overseas research suggests that the relationships of military members and veterans are more unstable than those of the civilian population. We do not know if this is true in Australia. The recent inclusion of identifiers for Australian Defence Force members and veterans in the Census make this a valuable source of information on the relationship circumstances and experience of members and veterans, who make up less than three per cent of the Australian population.
The ability to capture separation from both marriage and cohabitation relationships would also enable the provision of national data and assessment of how relationship separations have changed over time. Although surveys collect these data, their methodologies vary and their designs are for different purposes, so they are limited when it comes to assessing trends in relationship separation.
Good data on the experience of relationship separation should be collected in each Census, which would allow assessment of how the patterns of relationship separation have changed.

AIFS agrees with feedback provided for the 2021 Census regarding the need to better capture diverse family structures, and specifically children in shared care arrangements and/or regularly living in more than one household.
For the purposes of this submission, "shared care" is defined as the child spending 35-65% of nights with each parent, as per definitions in child support law; “regular care” is defined as between 14% and 35% with one parent, and primary care is a percentage of more than 65% with a parent. The available data from ABS Family Characteristics Survey (1997-2012) and AIFS’ surveys on recent separated parents suggest that shared care arrangements have been on the rise, but their prevalence is not well established. The rise in shared care also affects the counts of different household types and family types.
As suggested in other submissions (e.g. City of Casey, Response 758505622), a better estimate of the number of children in shared care arrangements could be achieved by allowing a person to identify more than one usual place of residence. E.g. “For persons who live at more than one address, include first and second place of residence.” Further confirmation of shared care arrangements could then be achieved by matching responses from the second place of residence (on other household Census forms) for persons reporting that a person was away on the night of the Census, to support integrity of results.
For children who are in their primary place of residence on Census night (defined as ‘the address at which the person has lived, or intends to live, for a total of six months or more in [year]’), it would allow us to identify those who spend at least 35% of their time living with another parent, to support better estimates of the prevalence of shared care arrangements.
For persons living in multiple households, it would be useful to have information on the approximate share of time (nights, weeks or months per year) they live in their non-primary residence(s). This would support identification of regular and primary care arrangements.

Ongoing national and regional level data on shared care arrangements are important to commonwealth, state and local government stakeholders responsible for providing services and supports for families.
Including this topic in the Census is important as we have insufficient understanding of the prevalence of shared care arrangements for small socio-demographic groups that are not well represented in other national surveys but may have high support needs. For example, overseas research suggests that many children from military and veteran families live in single parent, step or blended families, due to higher rates of marriage, separation and re-partnering among this cohort (when compared to the civilian population). We do not know if this is true in Australia. The recent inclusion of identifiers for ADF members and veterans in the Census make this a valuable source of information on the household and family circumstances of current and former serving members, who make up less than three per cent of the Australian population.
Collecting data on children who live in more than one household will also help capture the diversity of family structures for lesbian and gay guild parent families – those where gay men and lesbian women jointly create a family and co-parent together (Gahan, 2019). Limited data exists on guild parent families in Australia, or their family/household circumstances. Research has shown that children from these families may begin living between 2 households from birth (or an early age) and may later live over 3 or more households if one or both of the original parent couples separate (Gahan, 2019).
This additional information would also allow us to identify adults moving between multiple households for work or other reasons. For example, adults in some occupational groups sometimes live separately from their partners for work related reasons. This is common among Defence personnel, when a member moves to a new posting but the family decides not to relocate with the member (referred to as Members with Dependents Unaccompanied, the Defence member is provided with a separate Defence residence and only returns to the family home on breaks and holidays). This may also be common among other occupation groups e.g. Fly In Fly Out workers. It is not likely these arrangements would be systematically identified with existing Census questions.
For another example, in some separated families, the children live full time in the main family home and parents take it in turns moving between the family home and a separate smaller dwelling. Including additional residence data would make it easier to estimate the prevalence of these varied household and parenting arrangements.
There is a continuing need for these topics in the Census. Capturing this data will allow researchers to identify ongoing trends among these family types, as well as other current and emerging family arrangements. Importantly, we will be able to evaluate the long-term impacts of social, policy, and other changes on the prevalence and outcomes of different family compositions.
(Gahan, L., 2019, Separation and Post-Separation Parenting within Lesbian and Gay Co-parenting (Guild Parented) Families. Aust NZ Journal of Family Therapy, 40: 98-113.)

AIFS agrees with the submissions to the review of 2021 Census topics by the Department of Health (response: 178418291) and Rainbow Families Victoria (response: 443032441).
We would like the ABS to collect data on sex, gender identity, sexual orientation, and intersex variations in the 2026 Census.
We have outlined below the questions we believe will best collect the required data. This includes our preferred phrasing of the questions and answer options. Our questions and answers below are based off the current ABS standards (https://www.abs.gov.au/statistics/standards/standard-sex-gender-variations-sex-characteristics-and-sexual-orientation-variables/latest-release) and the Victorian Family Violence Data Collection Framework (https://www.vic.gov.au/victorian-family-violence-data-collection-framework/data-collection-standards-lesbian-gay-bisexual).

What is your sex?
- Male
- Female
- Another term (please specify)
(Please note: we support asking for ‘sex’ rather than ‘sex assigned at birth’)

What gender do you identify as?
- Man or male
- Woman or female
- Non-Binary
- Another term (please specify)
- Prefer not to answer

Do you identify as Transgender?
- Yes
- No
- Prefer not to say

How would you describe your sexual orientation?
- Straight or heterosexual
- Gay, lesbian or homosexual
- Bisexual or pansexual
- Asexual
- Another sexual orientation (please describe)
- Don’t know
- Prefer not to say

Were you born with an intersex variation (sometimes called ‘DSD’)?
- Yes
- No
- Prefer not to say

This data is of national importance. As with the data that the ABS currently collects through the Census, this data will help organisations to plan and monitor their services and programs, it will support Government policy development, and allow Government agencies and research bodies like AIFS to provide Government and the Australian community with an accurate understanding of the Australian population, Australian families, and Australian relationships.
While the topics of sex, gender, and sexuality feature strongly in current Australian national discourse, there is little to no population data for Government and the community to draw from.
Population data is needed to help researchers to better analyse and report our own findings to shape policy and programs. For example, determining prevalence of intimate partner violence in relationships with an LGBTIQ and/or non-binary person continues to be challenging due to the lack of robust estimates on how many people there are in each of these groups. Research often relies on population estimates such as the Australian Bureau of Statistics Labour Force data to help calculate approximate prevalence rates. This leaves a huge gap in our understanding of a vital national issue and leaves Government at all levels with limited data to enable them to respond adequately. Without these Census data, researchers are also hindered in our analysis of many issues of national importance such as homelessness prevalence, health and wellbeing comparisons, and socioeconomic needs.
The data is needed at all levels. It will allow researchers to provide all levels of government a better understanding of the diversity of their community to allow for better targeting of services and support. Given the small size of these population groups, the Census will also produce national and state estimates that are far more robust than any other available dataset.
These questions will continue to be needed in future Census collections. The research uses noted above will require ongoing evaluation in order to understand the impacts of policy, demographic change, population mobility and other factors.

AIFS agrees with the submission to the review of 2021 Census topics by Rainbow Families Victoria (response: 443032441) to include questions that allow for accurate data on diverse family forms.
We would like the ABS to amend existing questions to allow for diverse family forms.
In Australia, people can have more than one legal de-facto spouse/partner, and/or be married and have additional de facto relationships. However, this data is currently not collected.
We suggest amending the ‘what is the person’s relationship’ question (Question 9 in 2021) to allow for data on people with one or more de-facto partnerships, or who are legally married and have additional de facto relationships to be recorded.
AIFS also recognises that some married people do not use the terms ‘husband or wife’ and suggest changing this to allow for non-binary spousal terminology. E.g. ‘husband, wife or married spouse’.
We would also like data on households where former spouses remain living in the same household. Sometimes people divorce and remain living in the same house due to financial reasons, and/or to continue to parent children in the same household. Therefore, we would also like this question to include the relationship ‘former married or de facto spouse’.
AIFS would also like the ABS Census to collect more data on children who live in more than one household to capture the diversity of family structures. Limited to no data is available on the number of children who live and are co-parented over more than 2 households. This includes lesbian and gay ‘guild parent families’ – those where gay men and lesbian women jointly create a family and co-parent together (Gahan, 2019). Limited data exists on guild parent families in Australia, or their family/household circumstances. Research has shown that children from these families may begin living between 2 households from birth (e.g., with the lesbian parents in one home and the gay male parents in another) and may later live over 3 or more households if one or both of the original parent couples separate (Gahan, 2019).

(Gahan, L., 2019, Separation and Post-Separation Parenting within Lesbian and Gay Co-parenting (Guild Parented) Families. Aust NZ Journal of Family Therapy, 40: 98-113.)

This data is of national importance. As with the data that the ABS currently collects through the Census, this data will help organisations to plan and monitor their services and programs, it will support Government policy development, and allow Government agencies and research bodies like AIFS to provide Government and the Australian community with an accurate understanding of the Australian population, Australian families, and Australian relationships.
Currently we do not know the prevalence of different family forms, where the people in them live, and their wellbeing and economic status. Consequently, some people in some types of families/relationships may not be receiving the services and supports they need. By collecting this data, agencies and research bodies like AIFS will be able to provide this understanding for all levels of government to enable them to provide for the best outcomes of all Australian families.
The data is needed at all levels. It will allow researchers to provide all levels of government a better understanding of the household and family diversity of their community to allow for better targeting of services and support.
These questions will continue to be needed in future Census collections.

AIFS supports the recommendation made in the submission made by Macquarie University (Response 997903127) in response to the Review of 2021 Census Topics that the scope of the children ever born data collection should be expanded to include men. Research on fertility typically focuses on women, however, it is important to understand men’s fertility.
The patterns of forming couple relationships differ for men and women, and patterns and trends in fertility are not necessarily the same for them. Census data over time have shown childlessness has been increasing for women at the end of their reproductive years (e.g., later 40s), as has the prevalence of women with one child. However, it is unknown how common it is for men to have no children, nor how the number of children that men have ever had has changed. Internationally, there has been growing interest in men’s fertility. However, data on men’s fertility in Australia, especially trend data at the national level, is lacking. Making fertility decisions, such as whether and when to start a family and how many children to have, are conscious decisions that people make. Such decisions in a couple can be influenced by both partners’ preferences and circumstances. Each partner’s fertility history can also affect a couple’s fertility decision of whether to have a child or more children. In other words, the fertility preferences and history of the male partner play a role in the timing and number of children a couple achieves. Understanding men’s fertility not only provides insights into the trends in national fertility, but such information would also help better projection of future living arrangements, which is extremely important for policy development in a range of areas such as housing planning and residential care planning.

Collecting information on the number of children men have ever had is of national importance. Understanding men’s fertility not only provides insights into the trends in national fertility, but such information would also help better projection of future living arrangements, which is extremely important for policy development in a round of areas such as housing planning and residential care planning.
Although data of this kind can be obtained from surveys, such surveys typically have a small sample size, often have a lower representation of men, and thus are limited for detailed analysis and examination of subgroups. Moreover, survey data are not well-suited for tracking changes in men’s fertility and Census data are more suitable for this purpose.