A different kind of representation

On the necessity of including questions about gender and sexual orientation in Australia's census, and the importance of representation in data.

In August 2021, the Australian Bureau of Statistics (ABS) will conduct Australia’s 18th National Census. The census is run every five years by the ABS and is a chance to collect crucial data about the entire population. The census questions cover topics including marital status, occupation, languages spoken, and family size. The data gathered is used to inform policy development, funding and resource allocation, upcoming research, and evaluate government programs. Notably absent from past censuses, and the upcoming one, is any question regarding the sexuality or gender identification of those surveyed.

Every census cycle, the ABS consults with the public, various organisations and government bodies to consider changing the questions to better reflect the needs of current society. All the same, the topics and questions have remained the same since the 2006 census. This cycle, the ABS has undertaken extensive consultation regarding new topics for consideration.

From April to June 2018, public consultation saw over 400 submissions received from different stakeholders. A broad range of questions and topics were suggested, including gender and sexuality, disability and career status, education and training, and unpaid work. These topics were then analysed against criteria including whether the topic can be efficiently and accurately completed by households, if there was no other suitable alternative for the data in question, and if the topics would be considered acceptable by census respondents. Gender and sexual orientation, amongst seven other topics, were found to warrant further consideration, and were then analysed with regards to their potential inclusion, focusing on the burden to respondents and the cost of collecting and analysing information.

As a result of their consultations, the ABS found that data on gender and sexuality would be useful to all levels of government, and that the Department of Health and the Department of Social Services require this information. Despite this, and the lack of large-scale sources on data and sexuality, the ABS recommended that the government not include questions on sexual orientation, however non-binary will be added as a response category to the question regarding gender. While this is a definite improvement, there still remains no question or response option which allows for respondents to indicate that they are intersex, transgender, or genderqueer.

What makes the census so powerful as a data collection tool is obvious; unlike surveys, it can collect data from the entire population, rather than relying on a subset of respondents, which often poorly represents the entire country. Censuses remove many of the biases which often underlie surveys. In particular, selection bias is eliminated, which generally arises in surveys when the selection of individuals or groups for data collection isn’t properly randomised, meaning the data will not be representative of the entire population. This is also compounded by the existence of respondent bias, in particular when the questions asked are of a sensitive nature, as they would be in this case. When dealing with a sizable, variable, and historically disadvantaged subset of the population, it is necessary to collect the most accurate data possible.

Inclusion in the national census would be symbolically significant with the government acknowledging the existence of LGBTI people as an important part of Australia’s population. Far more importantly however, this data has enormous potential to inform government policy and services, including housing and health, as well as services provided by advocacy groups or within the queer community. Without widespread and reliable data, organisations struggle to provide resources to support LGBTI people in specific and sensitive ways. The LGBTI community is one which has been made vulnerable by the purposeful disengagement of government; a lack of data only serves this purpose and reduces pressure on the government to provide necessary services.

The most obvious use of this data would be gaining accurate knowledge on the proportion of Australia’s population that are LGBTI. Current estimates put this percentage from as low as 2% to over 10%. The accuracy of current statistics is particularly poor for trans and gender diverse people, with survey techniques frequently failing to distinguish between non-cisgender identities, forcing people to respond inaccurately and obscuring the true spread. Small scale surveys with a limited sample size often require the entire set of data to be coupled together, further obscuring demographic information such as age distribution, which is necessary to know because LGBTI people, like the general population, have different needs at different ages.

Another extremely important use of this kind of data is for allocating funding for social services. The LGBTI community is disproportionately likely to need and use these programs, especially services such as health and housing. LGBTI people are at least twice as likely to experience homelessness than their non-queer counterparts, which is compounded by the fact that homeless shelters are often faith based, and discriminate against queer people. Information on housing and homelessness would prompt the government to address this issue, and ensure that there are homeless shelters and housing which are safe and inclusive for LGBTI people.

In many areas, LGBTI people have poorer health outcomes than the general population. This includes higher incidence of mental health problems, smoking, HIV positive status, disordered eating, substance abuse, and a higher likelihood of experiencing violence. LGBTI people may also be at higher risk of cancer, including breast, prostate, and liver cancer. This disparity in health is likely due to actual or perceived queerphobia, which may result in inferior care or an avoidance of health professionals. Census data on LGBTI people would demonstrate the gap that already exists, and may be useful in pushing back against the Religious Discrimination Bill, which will exacerbate queerphobia in healthcare. There are multiple national programs such as the Aged Care Diversity Framework, and the National Mental Health and Suicide Prevention Plan, which consider LGBTI communities to be vulnerable groups who must be prioritised for action. Such data would allow these programs’ effectiveness to be evaluated over time.

If included in the 2021 census, these questions would likely be accepted in Australia’s future censuses. This means we could see trends over time in both societal makeup and social attitudes. Questions about same-sex couples have been in the Australian census since 1996, and in this time, the number of same-sex couples living together, according to census data, has quadrupled. This is likely due to both an increase in partners living together, and an increased willingness to self-identify, reflecting improved societal attitudes towards queer relationships. Census data would also be analysed to reveal the intersections between gender and sexual orientation, and education, work, religious affiliation, occupation, and income, as well as where queer communities are living. These trends could be tracked over time to see how they change.

The Australian Government appears to have taken on board the ABS’ recommendation not to include these questions, despite the clear importance of this data, and its potential use in all levels of government. Why, then, did the ABS recommend non-inclusion of questions regarding sexual orientation? The ABS cited a desire to keep the “burden we place on responding households to a minimum”, likely referring to potential confusion arising from the questions. Another reason cited was that respondents may have a lack of trust in the government, leading to privacy concerns which could reduce the accuracy of the data. This potential decrease in accuracy, due to mistrust or misunderstanding, would likely be able to be adjusted, as other areas of data are, to account for the expected undercalculation. In addition, in the small-scale testing which was undertaken with the relevant questions, it was found that the adverse reaction wasn’t significant enough to impact either the quality of data or response rates. Even in the case of partial undercounting, this data’s benefits extend beyond finding the net amount of queer people, to their patterns of location, demographics, education and occupation. Ultimately, even if the ABS’ concerns are correct, this data would still far surpass current data about LGBTI Australians, and would be of immense value to the community.

At this point, public consultation is over, and the questions for the 2021 census are being finalised. They will be released later this year, but we do know that there will be no questions regarding sexual orientation, and no potential response to indicate being a binary transgender person, or intersex. Including these questions should not be unthinkable for the government. Both New Zealand and the United Kingdom, countries we see as politically and socially comparable to Australia, are including questions regarding sexual orientation and gender identity, including transness, in their 2021 census. Australia’s reluctance on this matter will see sexuality and gender diverse people waiting until at least 2026 to see ourselves represented where it counts, in the data which informs policy and social services.

Filed under: