2049.0.55.001 - Information Paper - Methodology for Estimating Homelessness from the Census of Population and Housing, 2012  
Latest ISSUE Released at 11:30 AM (CANBERRA TIME) 05/09/2012  First Issue
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product  
Contents >> Future Directions >> FUTURE DIRECTIONS


On 11 September 2012 the ABS will publish its official homelessness estimates from both the 2001 and 2006 Censuses, consistent with the methodology outlined in this publication. They will be published in the publication Census of Population and Housing: Estimating Homelessness, 2006 (cat. no. 2049.0). Official estimates of homelessness from the 2011 Census, using the same methodology, will be published on 12 November 2012 under the same catalogue number, after Census second release variables are published on 30 October 2012.

This ABS methodology will be applied in producing homelessness estimates from future Censuses. However, improvements are expected to be made to both questions and field procedures which will provide for new and better estimates for tracking future changes in homelessness. The transparency and repeatability in the methodology will allow for an alternate estimate to be made that is consistent with 2011 to provide a link in monitoring changes over time.

The Census is not the only source of data from which the ABS publishes estimates of homelessness and the ABS has developed a statistical program on homelessness for the next few years. Some of these developments are outlined below.

The ABS will publish an Information Paper: Guide to Homelessness Statistics (cat. no. 4923.0) in November 2012 to assist users with analysing not just estimates compiled from the Census but also from other available data sources to obtain a more complete picture of homelessness. The guide will outline which parts of the homeless definition ABS collections can, or cannot capture.


ABS will investigate using its 5% sample of the Australian population captured in its Statistical Longitudinal Census Dataset (SLCD) to undertake longitudinal analysis of the circumstances of those who have been classified as likely to be homeless on Census night. The circumstances of people classified as likely to be homeless in 2011 can then be compared with their circumstances in 2006, and vice versa to look at both antecedents for and situations after an episode of homelessness. Into the future it should be possible to report on aspects of repeat periods of homelessness and long term outcomes as seen in the SLCD, compared with the rest of the population. Further information on the SLCD can be found in Census Data Enhancement Project: An Update, Oct 2010 (cat. no. 2062.0).

Some options for improving both enumeration and estimation in future Censuses have been discussed in the ABS Position Paper - ABS Review of Counting the Homeless Methodology, 2011 (Australian Bureau of Statistics, 2011b).

Another aspect of potential future improvement in homelessness estimation from the Census will be the possible inclusion of new content around, for example, health status, so that the homeless population can be compared with the rest of the population. A longitudinal view through Censuses may therefore be able to be studied in terms of both the pathways into homelessness arising from social factors, and its consequences for other outcomes in later life.

The ABS anticipates opening the public consultation to seek input into the priorities for Census topics and procedures for Census 2016 in early November 2012, after the second data release for the 2011 Census. An information paper with more details about the submission process will be available at that time from <www.abs.gov.au/census>.

For the next Census, ABS will be considering user needs across the full range of topics, including homelessness. The ABS is still setting bounds for the scope, timing and capacity for changes to Census 2016, and has already flagged that strategies for reaching remote or hard-to-get groups will need further improvement. The public consultation process is expected to close in the first part of 2013.

ABS Household Surveys

ABS has so far included a module on past periods of homelessness in two surveys, from which a picture of the incidence of homeless can be derived, as well as trends in the homelessness over time, at least for those who have transitioned out of homelessness at the time of interview. The homelessness module has been run in the 2010 General Social Survey and is being run in the 2012 Survey of Disability, Ageing and Carers. The module will be improved for use in the 2014 General Social Survey and, once adjusted for cultural appropriateness, will be considered for the next National Aboriginal and Torres Strait Islander Social Survey. The ABS has also collected information in the 2012 Personal Safety Survey about where people go when they leave a violent partner, and future developments may expand this information set for later cycles of the survey.

The GSS provided information on people who have been homeless in the past but who are now usual residents of private dwellings. The module also provides information about how long ago the person had been homeless, the number of times they have been without a permanent place to live, whether they sought assistance from services and whether the services were useful.

In March 2012, the ABS released an article entitled Life after homelessness in the publication Australian Social Trends (cat. no. 4102.0). This article drawing on the GSS results. The article examines a range of socio-economic indicators of those who had experienced at least one episode of homelessness in the 10 years prior to the survey, but were no longer homeless.

The development for the 2014 General Social Survey (GSS) will shortly commence. The past experiences of homelessness module will again be included in the 2014 GSS, and again oversampling is expected to occur in lower socio economic areas (using SEIFA). As the GSS 2010 only enumerated usual residents of private dwellings, the past experiences of homelessness module did not include: people currently living in shelters; people sleeping rough; people 'couch surfing' (staying temporarily with other households); nor people staying in boarding houses. It may have included some people staying in Transitional Housing Management (THM) properties, if the adult staying there at the time of the survey considered that it was their usual residence at that time.

The ABS will convene a sub-group from its HSRG to provide advice to the ABS in the development/improvement of these modules.

Key population groups


Guided by its Homelessness Statistics Reference Group, the ABS is continuing to undertake research and development to improve the estimation of homelessness, including youth homelessness. In particular, the ABS is undertaking a quality study to inform the potential development of a nationally representative homeless school students survey.

Until a robust methodology is developed to measure the level of youth homelessness, ABS will focus on producing transparent, consistent and repeatable estimates that can be used to monitor change over time. Because the ABS methods are transparent, users can assess whether there is any evidence to suggest that the components of homelessness that cannot yet be estimated reliably are likely to be moving differently over time to those elements that can be measured.

As part of the development for Census 2016, the ABS will consider how they can improve the correct identification of homeless youth in the Census but recognise the need to use other data sources to gain a more complete picture of youth homelessness.

Aboriginal and Torres Strait Islander Australians

The ABS definition of homelessness has been developed for application to the general population in Australia. While Aboriginal and Torres Strait Islander people are over-represented in the measures of homelessness developed with that definition, there are likely to be additional aspects to homelessness from an Indigenous perspective that the definition does not currently acknowledge or take account of.

The ABS will shortly commence further work and consultation on homelessness concepts and measurement in regard to Aboriginal and Torres Strait Islander peoples. This work may have impacts on how homelessness for Indigenous Australians are estimated.

The ABS is also considering development of a culturally appropriate module on previous experiences of homelessness for Aboriginal and Torres Strait Islander surveys so that this data can be used to understand Indigenous homelessness and allow comparison with the data collected for the general population from the General Social Survey.

As part of the development for Census 2016, the ABS will consider how they can improve the identification of homeless Aboriginal and Torres Strait Islander Australians in the Census but recognise the need to use other data sources to gain a more complete picture of homelessness.

People fleeing domestic and/or family violence

As outlined above (Personal Safety Survey), in late 2013 the ABS will have data from the Personal Safety Survey on people who left a violent current or previous partner and whether they 'couch surfed', sleep rough, stayed in a shelter etc. This will provide will provide an indication of what accommodation was used by people the last time they separated from their violent partner/s and will be an important new source of information about homelessness and domestic violence.

As part of the development for Census 2016, the ABS will consider how they can improve the enumeration of, and the identification as homeless of those who were fleeing domestic and/or family violence in the Census. However the ABS recognises the need to use other data sources to gain a more complete picture of homelessness.

Previous Page