Homelessness Operational Groups (OPGP)

This variable records the operational group of persons estimated to be homeless or marginally housed on Census night. It is applicable to all persons who are estimated to be homeless or marginally housed.

The 'statistical definition of homelessness' underpins these groups. In brief, the Australian Bureau of Statistics (ABS) statistical definition is that:

When a person does not have suitable accommodation alternatives they are considered homeless if their current living arrangement:

• is in a dwelling that is inadequate;
• has no tenure, or if their initial tenure is short and not extendable; or
• does not allow them to have control of, and access to, space for social relations.

The definition has been constructed from a conceptual framework centred around the following elements:

• Adequacy of the dwelling;
• Security of tenure in the dwelling; and
• Control of, and access to, space for social relations.

People must lack one or more of these elements to be defined as homeless. However, people who lack one or more of these elements may not necessarily be classified as homeless if they are living in special circumstances (for example, in hospitals, prisons, student halls or religious orders). While homelessness is not a choice, some people may choose to live in situations that might parallel the living situations of people who are homeless. For example, people may be living in a shed while building a home on their own property, or on holiday travelling and staying with friends. These people have choice because they have the capacity to access other accommodation that is safe, adequate and provides for social relations. Having access to accommodation alternatives is contingent on having the financial, physical, psychological and personal means to access these alternatives (see the Information Paper - A Statistical Definition of Homelessness (cat. no. 4922.0)).

The homeless and marginally housed groups represent an operationalisation of this definition of homelessness within the limits of the data collected in the Census.

How this variable is created

Homelessness itself is not a characteristic that is directly measured in the Census. Instead, estimates of the homeless population are derived from the Census using analytical techniques, based on both the characteristics observed in the Census and assumptions about the way people may respond to Census questions.

Homeless Operational Groups are as follows;

1. Persons living in improvised dwellings, tents or sleeping out;
2. Persons in supported accommodation for the homeless;
3. Persons staying temporarily with other households;
4. Persons living in boarding houses;
5. Persons in other temporary lodgings; and
6. Persons living in 'severely' crowded dwellings.

The following groups are considered to be 'marginally housed', that is, whose living arrangements are close to the statistical boundary of homelessness and who may be at risk of homelessness.

Other Marginal Housing Groups are as follows;

7. Persons living in other crowded dwellings;
8. Persons in other improvised dwellings; and
9. Persons who are marginally housed in caravan parks.

While these categories will overlap in a small number of circumstances, people are only assigned to one category to avoid double counting. This is done by only including them in the group that is the highest on the hierarchy. For example, a person in supported accommodation for the homeless may also be living in 'severely' crowded dwellings, but the person will only be coded to the group '2. persons in supported accommodation for the homeless' as it is higher on the hierarchy than group '6. Persons living in 'severely' crowded dwellings'.

The 'Not applicable' category refers to persons who were not estimated to be homeless or marginally housed on Census night.

Variable history

The ABS has a long history of collecting information relevant to identifying homeless people in the Census. In the 1933 and 1947 Censuses, a question was asked for the ‘number of persons (if any) who slept out throughout the year on verandahs (not enclosed sleep-outs)’. In 1986, a dwelling structure category ‘Improvised dwelling’ was provided on the form for the collector to mark. This category was changed to ‘Improvised home, campers out’ in 1991, but there was no distinction between homeless people and those who were camping (such as on holiday).

From 1976 to 1991, collectors were instructed to seek out all people camping or sleeping out by visiting ‘any places in your Collection District (CD) on Census night where it is likely that persons may be sleeping out, e.g. camping areas, park benches, derelict buildings etc’. They were instructed if they found such a person to issue a Household Form and help them fill it out on the spot. They were assigned to a non-private dwelling (NPD) type ‘campers out’. However, some collectors may not have followed this instruction if they did not believe there were people in their area, or for fear of their own safety. Prior to 1996, some Divisional Managers undertook additional measures to enumerate the homeless, such as providing refreshments.

The 1996 Census was the first Census to target Australia’s homeless population using a special enumeration strategy. This strategy aimed to not only maximise the coverage of the Australian population but also to provide information to policy makers and service providers on the number and characteristics of people experiencing homelessness. In the lead-up to the Census, the ABS liaised with relevant state/territory organisations to gain their assistance in identifying accommodation likely to cater for people experiencing homelessness. In addition, prior to Census night ABS staff from regional offices contacted groups providing services for the homeless to identify possible sites where homeless people were likely to be located. Where possible, members of the homeless community were engaged to enumerate areas where significant numbers of homeless people were likely to spend Census night. For more information see 'Appendix 3: 2016 Census Procedures' in the publication: Census of Population and Housing: Estimating homelessness, 2016 (cat. no. 2049.0).

The current statistical definition of homelessness was first developed in 2012 (see the Information Paper - Methodology for Estimating Homelessness from the Census of Population and Housing (cat. no. 2049.0.55.001)), and with it the methodology to apply it to Census data to obtain national homelessness estimates. The first official estimates of the prevalence of homelessness were published in 2012 using data from the 2001 and 2006 Censuses. Estimates have subsequently been produced for the 2011 and 2016 Censuses.

Data usage notes

To calculate the total homeless population, use only the Homeless Operational Groups 1 to 6.

Further information

More information including detailed estimates can be found in the publication:Census of Population and Housing: Estimating homelessness, 2016 (cat. no. 2049.0).

Detailed methodology used to estimate persons in each operational group can be found in 'Appendix 2: Estimation Methodology' in the publication: Census of Population and Housing: Estimating homelessness, 2016 (cat. no. 2049.0).

More information on the methodology can be found in the publication: Information Paper - Methodology for Estimating Homelessness from the Census of Population and Housing (cat. no. 2049.0.55.001), and information on the statistical definition of homelessness can be found in the Information Paper - A Statistical Definition of Homelessness (cat. no. 4922.0).