2071.0 - Census of Population and Housing: Reflecting Australia - Stories from the Census, 2016  
Latest ISSUE Released at 11:30 AM (CANBERRA TIME) 28/06/2017   
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product


A question on Sex has been included in every Australian Census, since the first one in 1911. The Census is authorised to collect information specifically on Sex through the Census Regulations. Self-reported sex (or, in many cases, gender) is used for a range of purposes including population projections, estimates of life expectancy, family structure and gender comparisons.

Until the Census 2016, there were few or no ways of reporting anything other than male or female.


There has been increasing recognition of sex and gender social issues in Australia. In November 2015, the Australian Attorney General's Department updated the Australian Government Guidelines on the Recognition of Sex and Gender, which required Australian Government departments and agencies to progressively align their existing and future business practices with the Guidelines. Accordingly, the aim of 2016 Census was to make it possible for all Australians to report their sex in a way not limited to ‘male’ or ‘female’.

For the 2016 Census, the Australian Bureau of Statistics (ABS):

  • recognised that individuals may identify as a sex or gender other than the sex they were assigned at birth or may not identify as exclusively male or female
  • provided methods for all form types so that people could record their sex in the way each thought most appropriate.

In addition, the ABS aimed for the 2016 Census to gather information that would assist the design of questions, expand the Standard Classifications of Sex and Gender, and support more informed collection of information on sex and gender in future ABS collections.


Collecting accurate information on either sex or gender in the Australian Census has many challenges. Most households complete the Census without contact with an interviewer or any interaction with ABS staff. This means that the ABS cannot answer any questions from the respondent or explain each question – for example, explaining the difference between sex and gender.

In households, it is common for one person to complete responses on behalf of other household members, who may not know or respect how these others would report their sex or gender. Household forms may also introduce privacy concerns for some individuals. While the ABS had arrangements to provide private forms on request, some respondents may not have been aware of these or may have chosen not to take this additional step.

The Census forms have limited space, due both to the physical limitations of the paper form and also the risk that completion rates would decrease if households felt the Census form was too intrusive or too long. This space limitation means that often only one question can be asked on a topic, even when multiple questions would be clearer.

There is personal choice in relation to reporting as male or female. For example, a person that is intersex may not identify themselves as other than male or female due to issues such as acceptance of their assigned sex or previous discrimination.

The ABS did not expect the 2016 Census to be able to provide an accurate measure of the number of people with other than strictly male or female sex and gender.


The ABS made a number of changes to the approach to collecting information about sex in the 2016 Census. Procedures were developed and refined through engagement with a number of stakeholders. Their contribution is gratefully acknowledged.

The 2016 Census aimed to have most people complete the Census online. Thus, the ABS focussed on ensuring that the online form's development supported the ABS’ aims to support people in responding as other than as male or female.

A special online form was developed that had the categories of Male, Female and Other (please specify). This online form was available to households or individual respondents on an opt-in basis by contacting the Census Inquiry Service or submitting an online assistance request.

Nearly two in every three persons (63%) in occupied private dwellings were counted on the online form, and in every online form there was advice on how to report as other than male or female. In addition, the ABS developed 'How to Answer' instructions which were on the ABS website and in a fact sheet distributed to stakeholders.

The paper form, unfortunately, did not provide any instructions about the special procedures. However, the 'How to Answer' instructions provided directions on how to report other than male or female on the paper form.

The ABS tried to ensure that Census field staff, call centre staff and office staff were all aware of the special procedures and able to sensitively support members of the public.

In addition, a small pilot test was conducted during the 2016 Census. Approximately 29,900 addresses were given the special online form as their default form to test attitudes and responses among the wider population.

The ABS changed data capture and processing procedures to ensure that responses to the Sex question would be accurately captured and could be analysed.

For the first time, Census data will be made public on the number and characteristics of people who report they are neither male nor female. This is being done via this overview and several analytical articles to be released in the upcoming months.


The special procedures relied on contact with the Census call centre. A significantly larger than expected volume of calls was received before and after Census night, which meant that some people were unable to get their calls answered in a timely fashion. This may have deterred some or many from following through on the special procedures to receive an access code for the special online form or to seek advice on procedures for the paper form.

There were also some reported incidents of call centre staff being unaware of the special procedures, possibly due to additional staff being brought on quickly to assist with the larger than expected peaks. Call centre management addressed reported issues as quickly as possible.

The withdrawal of the online form for 43 hours also may have led to some people reporting on paper rather than online, and they may have been unaware of how to report as other than male or female on the paper form.


The ABS does not consider the count of people reporting as other than male or female in the 2016 Census to be an accurate measure of the number of people with other than male or female sex or gender. The ABS has made this assessment in consideration of the inherent limitations of the Australian Census for this topic, the limitations of the approach used for the 2016 Census and the operational challenges experienced.

There were 1,300 people who provided a sex or gender response other than male or female that could be validated, a rate of 5 per 100,000 people. Approximately one-third of these simply answered with no text or no meaningful text. The remaining two-thirds gave descriptions that will be helpful for understanding the complexity of sex and gender.

A further 2,400 people provided a response which may have been meant to indicate another sex or gender, by marking both the Male and the Female responses on the paper form (multimark) without writing any text. These marks could have been intentional, for those people not aware of the special procedures. The marks may also have been mistakes. Where both sex boxes were marked, but one was clearly crossed out, the records were excluded from this analysis.

There was a dramatic increase in the proportion of this type of reporting from the 60 years and over age groups. This could indicate that the older age groups found it difficult to or chose not to follow the special procedures, or alternatively they could have simply made an error and not crossed it out. In addition, multimarks were only possible on paper forms, and the proportion who used paper forms also increased with age.


(a) Sex or gender response other than male or female that could be validated.
(b) Marked both Male and Female responses on the paper form without writing any text.
Source: Census of Population and Housing, 2016

In addition to the groups discussed so far, there were an additional 6,500 people whose initial response might have indicated a sex other than male or female (e.g. they selected the Other box on the online form or there was text next to the Male and Female boxes on the paper form). When these responses were manually reviewed, however, it was clear that these were not intended as valid responses of a sex/gender other than male or female.

Of these 6,500 people:
  • a small number of people provided an apparently intentionally invalid response
  • an even smaller number gave a response exclusively based on their sexual orientation
  • the vast majority of this group wrote unrelated text, such as information about other household members, linking the male and female boxes to different household members, people's names or symbols.

The highest possible count, therefore, for sex/gender diverse responses as measured in the 2016 Census is 3,700, representing a rate of 16 per 100,000 people. Around one- third (35%) of this group (1,300) were able to be validated as intentional. It is not clear how many of the remaining 65% (2,400) intended to indicate that their sex or gender was other than male or female.

Response to the Sex question
Rate per 100,000 people

Valid and intentional sex/gender other than male or female
1 300
Possible sex/gender other than male or female(a)
2 400

Wider group of possible sex/gender diverse responses
3 700

Other responses - clearly not an intended 'other' response(b)
6 500

23 717 400

(a) Includes people who marked both the Male and the Female responses without writing any text.
(b) Includes people who clearly did not intend an 'other' response (e.g. information about other household members, the names of people or various marks on paper forms), those giving a deliberately invalid response and those whose response was exclusively about their sexual orientation.
(c) Includes all responses to the Sex question.
(d) Includes overseas visitors.
Source: Census of Population and Housing, 2016

Results of the pilot test

The ABS conducted a small pilot test during the operations of the 2016 Census to test attitudes and responses among the wider population, and to gain insights to guide future data collection of sex or gender.

Approximately 29,900 addresses were selected in the pilot test and sent a letter with an access code that took them directly to a special online form. Dwelling residents were not informed that they had been selected in the pilot test. The initial sample of addresses included some out of scope dwellings (e.g. commercial premises and under construction), private dwellings which were unoccupied and dwellings where the occupants did not respond to the Census. There was a mix of forms returned from responding dwellings - paper forms, standard online forms and special online forms. All households received codes for the special online form in their initial letter and in a reminder letter after Census night. However, households that had not responded to the Census within approximately two weeks of Census night had access codes to the standard online form delivered to them by field officers.

Preliminary analysis indicated that the conduct of the pilot test did not have a significant impact on response rate for the selected households. The proportion of people in the pilot test sample who returned an online form was higher than the national average, however further analysis needs to be done to understand whether it was consistent with other similar areas.

Preliminary results indicate that the pilot test had a lower level of invalid responses to the Sex question than the rest of Australia, probably due to the higher use of the online form in the pilot test. Invalid responses included those which were apparently intentionally invalid, exclusively based on sexual orientation or unrelated text.

There were almost no people in the pilot test who marked both the Male and Female boxes on the paper form without writing any text (multimarks), reflecting the low number of paper forms in the pilot test sample. There were approximately 200 people who provided a sex or gender response other than male or female that could be validated. This indicates that other sex/gender responses were about 50 times as likely in the pilot test sample as across the rest of Australia.

However, there was a difference between the pilot test and the rest of Australia in the provision of further meaningful text from those people who gave a sex or gender response of other than male or female that could be validated. Less than one-fifth of this group gave further, meaningful text in the pilot test, with almost all of the remainder leaving the text box blank. This was significantly lower than the level of meaningful text descriptions received outside the pilot test, which was approximately three-quarters of people with a sex/gender other than male or female. The lower level of text responses in the pilot test may be due to accidental reporting or other invalid responses. It could also indicate there was a proportion of the population who wished to report as 'Other' without any further details but were less willing to request the special online form than other people. If all reports of 'Other' with no meaningful description are excluded from the analysis, the reporting of sex/gender other than male or female was more than 10 times as likely in the pilot test than across the rest of Australia.


This overview is the first release of results from the 2016 Census's collection of non-male and non-female responses to the Sex question. Over the upcoming months, the ABS will publish several articles on this topic:
  • Final results of sex and gender reporting in the 2016 Census, including analysis of descriptive terms
  • Characteristics of people whose sex/gender is other than male or female
  • Lessons for future collections.

The ABS will seek advice on articles from peak groups and undertake targeted peer review, and welcomes feedback from the community.

The ABS is committed to implementing the Australian Government Guidelines on the Recognition of Sex and Gender and will continue to work with the community on approaches for future collections including the 2021 Census.


For more information about 2016 Census data release and products, go to www.abs.gov.au/census.

Overseas visitors who were intending to be in Australia for less than one year are included in this analysis.