Usual residence standard

Usual residence - country and Usual residence - location in Australia nominally refer to the same concept at different levels of geography. Operationally, however, the two variables are collected and used in distinct ways.

• Usual residence - country is frequently the basis for deciding whether respondents are within the scope of a data collection, including the Australian Census of Population and Housing (Census), Estimated Resident Population (ERP) and ABS surveys.
• Usual residence - location in Australia data is used widely in both government and private sector decision making and planning. Federal, state and local government services and funding, are often allocated according to where people usually live. For example, the drawing of electoral boundaries and calculations for distributing the Goods and Services Tax (GST) to Australian state and territory governments rely on population estimates based on usual residence.

Usual residence is a key collection and output variable for determining the geographic distribution of the population. Other measures of the temporal and spatial status of a population exist for other purposes, including service populations, dual residences, and 'de facto' populations defined on a Place of Enumeration basis. 

Australian Census counts were based on Place of Enumeration until 1961. Australia’s official ERP began using a Usual residence concept from 1981, and backcast estimates on this basis to 1971. ABS Household Surveys are weighted to population benchmarks based on ERP, and also collect and output data on a usual residence basis. 

Recognising the importance of international comparability for these variables, the ABS aims to reflect the United Nations Statistical Division's recommendations for the definition and collection of usual residence. The following two United Nations principles and recommendations (UN, 2017) are especially relevant to defining Usual residence - location in Australia: 

• In general, usual residence is defined for Census purposes as the place at which the person lives at the time of the Census, and has been there for some time or intends to stay there for some time.  
• It is recommended that countries apply a threshold of 12 months when considering place of usual residence. 

The 12-month threshold is determined if one of the following two criteria are meet: 

• The place at which the person has lived continuously for most of the last 12 months (that is, for at least six months and one day), not including temporary absences for holidays or work assignments, or intends to live for at least six months; 
• The place at which the person has lived continuously for at least the last 12 months, not including temporary absences for holidays or work assignments, or intends to live for at least 12 months 

In relation to Usual residence - country, the United Nations Statistical Division recommends a threshold of 'at least a year (12 months)' (UN, 1998). As such, a more stringent definition applies to Usual residence - country, compared to Usual residence - location in Australia.

Usual residence. It includes the variables:

• Usual residence - country
• Usual residence - location in Australia

Usual residence - country

Usual residence - country is the country in which a person resides for 12 months or more. With specific reference to Australia, a usual resident is defined as a person who has been or will be living in Australia for one year or more, excluding in a foreign military, diplomatic or consular capacity (including employees and their immediate family).

Usual residence - location in Australia

Within Australia, Usual residence is the address of the dwelling at which a person considers themselves to currently live, either having lived there for some time or intending to live there for some time. 

This may refer to:

1. The location of the dwelling where the person spends the daily period of rest, the majority of the time (usual residence in a dwelling)
2. The location of the household to which the person belongs (usual residence in a household or the family home) 

Some collections will be specifically targeting one or the other concept, whereas other collections may not need to draw this distinction. The concept of Usual residence assumes each person to have a basic attachment to a particular dwelling. This is, of course, a simplification; see section 'discussion of issues'.

Usual residence - country

The 'one year or more' criteria is operationalised differently in different collections. The examples below relate to the collection of Usual residence - country as 'Australia' or 'other country' only.

Usual residence - location in Australia

The criteria 'for some time' (taken from the United Nations definition (UN, 2017)) may be operationalised as 'the majority of one year', however this time frame is intended to be indicative rather than prescriptive. In cases where a person does not have one and only one possible usual residence response, the definition consciously relies on self-perception of usual residence.   

For example:

The Usual Residence concept can also be applied at a specified time other than the present, such as in the Census which asks where the person usually lived one and five years ago.

Statistical units

The variables 'Usual residence - country' and 'Usual residence - location in Australia' apply to all persons. 

Usual residence links the person to a dwelling address, and/or to a household.

Detailed question module

The detailed question modules for the variables in the standard for Usual residence are shown below. These are the questions used by the ABS and are suitable for different modes of collection. The methods employed by the ABS to process administrative data are also outlined, along with recommendations to ensure the collection of administrative data aligns closely with the definition of each variable.

Wherever possible, full address information should be collected, including automatic verification, to facilitate coding to various geographical standards.

Usual residence - country

The detailed question module supports the collection of Australian and 'other country' usual residence data. Detailed country data are not collected by this module.

Self enumerated collections: e.g. Census

The Census collects 'Usual residence - country' and, where relevant, Usual residence - location in Australia from the same Census question. 

Instructional text:

• For people who usually live in another country and who are visiting Australia for less than one year, mark 'Other country' 
• Additional directions pertain to Usual Residence - location in Australia.
• All other responses are allocated to Australia according to additional directions and response options for collecting Usual residence - location in Australia 

Interviewer-based collections: e.g. ABS Monthly Population Survey

In ABS household surveys, the sample frame is comprised of dwellings rather than persons. Usual residence is collected with reference to the selected dwelling. 

The following series of questions is an example taken from the ABS Monthly Population Survey. The ABS Monthly Population Survey data is collected via self-completion online, telephone interview or in-person interview. For assisted interviews, an interviewer enters the responses into a form on their laptop. Other ABS Household surveys may differ in their specific question wording.

Instructional text:

• if Australia, usual residence country = Australia
• if other, ask next question

Instructional text:

• more than one year ago, usual residence country = Australia
  if current year, ask next question

Instructional text:

• if yes, usual residence country = Australia
• if no, usual residence country = other

Administrative data collections: e.g. Net Overseas Migration

Country of usual residence (Australia or other) is determined in ABS Migration data based on border crossings of individuals, obtained from the Department of Home Affairs' Travel and Immigration Processing System (TRIPS) Database, which contains passport, visa and passenger card information.

Records of each traveller’s overseas arrivals and departures are used to measure their cumulative duration in (or out of) Australia over the subsequent 16 months following an initial arrival or departure. Travellers attaining a combined duration in Australia of 12 months or more, and who are not currently counted within the population, are added to the population. Conversely, residents attaining a combined duration out of Australia of 12 months or more, and who are currently counted within the population, are subtracted from the population.

Preliminary Net Overseas Migration (NOM) for Estimated Resident Population (ERP) purposes is modelled based on a propensity model. This takes a sample of incoming passenger cards and is based on the behaviours (actual length of stay compared to intended length of stay) of past travellers with similar demographic characteristics. 

Usual residence - location in Australia

Self enumerated collections: e.g. Census

In the Census, this question is asked of all persons physically present at the dwelling in question. The same Census question collects information on Usual residence - country (see above). Dwelling address is collected in an earlier question. Suburb/Locality, State/Territory, and Postcode fields are mandatory in the online form.

Instructional text:

• For people who usually live in another country and who are visiting Australia for less than one year, mark ‘Other country’.
• For other people, ‘usually live’ means the address at which the person has lived, or intends to live, for a total of six months or more in [year]. 
• For people who have no fixed or return address (for example, due to family conflict or eviction), write ‘NONE’ in the ‘Suburb/Locality’ box. 
• For boarders at boarding school, write the address of the boarding school or college.

Variations in Census

A tailored Census form (known as the Interviewer Household Form) is used in discrete Aboriginal and Torres Strait Islander communities. This form asks:

Instructional text:

• If another community, write that community name in the ‘Elsewhere, please specify’ boxes. 
• If another town or city, write street number and street name (if known), suburb, rural locality or town in the ‘Elsewhere, please specify’ boxes. 
• If another country, write the name of that country in the ‘Elsewhere, please specify’ boxes.

A tailored Census form (known as the Special Short Form) is used for people experiencing homelessness. This form asks 'Where did you sleep on Census night [date of Census]?'

Interviewer-based collections e.g. ABS Monthly Population Survey

In ABS household surveys, dwelling address has already been established prior to questions determining usual residence. The sample frame is comprised of dwellings rather than persons. Household surveys begin with a pre-determined place of usual residence (the canvassed address) and ask questions to identify to which persons the usual residence value can be attributed. The following questions are examples taken from the face-to-face and telephone interview format of the ABS Monthly Population Survey. Specific question wording and interviewer guidelines may vary across other household surveys according to their specific purposes.

Instructional text:

• The interviewer guidelines for the ABS Monthly Population Survey (MPS) include the following (other collections may have different rules for treating sub-populations):
• If other household members are named as usual residents who otherwise live in non-private dwellings, they should be included - but interviewers are instructed not to probe for this information. 
• If a child is under a 50/50 shared care arrangement, interviewers are instructed to 'Ask the parent to nominate one dwelling as the primary residence' 

Noting that children under 15 years are out of scope of the MPS, the child's inclusion or exclusion at either dwelling is relevant only to household and family level data, not to person level data. 
No further definition (e.g. timeframe) is given for 'usually' live. Coverage rules of the MPS exclude usual residents who are absent from the household for six weeks or more.
Response format: Usual residence - location in Australia is thus derived from the selected dwelling address according to the sampling frame (that is, as listed on the ABS Address Register).

Instructional text:

• if unknown, the interviewer can look up postcode based on suburb and state/territory.

Note: Some household surveys allow one respondent to provide information on all members of the household - in this case the responses reflect the third party's perception of the person's usual residence. 

Response format: may be free form or with sections to provide apartment/unit number, street number, street name, suburb/locality, state/territory, postcode. Where possible (such as online), address validation should be used in order to maximise successful geocoding (see section 'processing the data').

Instructional text:

• Many administrative data sources include no directions, assuming the concept of 'address' to be self-evident. The collection may specify or imply that address is requested for a particular purpose, such as for determining eligibility for a service, or as a communication method. Where usual residence information is of interest, guidance should be provided to this effect.

Short question module

The variables have no short question module.

Minimum question module

If less than a full address is provided in response, the required level of geography will be derived from the available data, or imputed (see below).

In cases such as where the variable 'place of enumeration' is (also) collected and is likely to have a high correspondence to usual residence, the usual residence question may allow a reference to the place of enumeration as one valid response option (see Census and MPS question modules in section 'Detailed question module'). In this instance, for relevant responses, usual residence is derived from place of enumeration rather than separately collected and coded.

Aggregate, modelled household level data is output on a usual residence basis, derived from person level usual residence. In this case household usual residence is a characteristic of modelled aggregate data only, not of individual household units.

Place of enumeration may be used in determining place of usual residence. Census of Population and Housing: Census dictionary

Usual residence is used as a key input variable in determining: 

• household and family composition variables (see Family, Household and Income Unit Variables)
• migration and mobility variables such as the Census variables Usual address indicator 1 year ago (UAI1P) and Usual address indicator 5 years ago (UAI5P) (see Census of Population and Housing: Census dictionary)

Usual residence – country

The standard input categories for collecting 'Usual residence - country' using the self enumerated and interviewer-based detailed question modules are:

• Australia, and
• Other country.

The standard input categories for collecting 'Usual residence - country' using questions that collect detailed country data, including from administrative data sources, are the 4-digit categories in ABS catalogue Standard Australian Classification of Countries (SACC).  

Usual residence – location in Australia

The  input categories for 'Usual residence - location in Australia'  are ASGS Mesh Blocks where complete physical addresses are captured during data collection, or higher level ASGS categories (such as SA2) where basic locality information is known (for example, state/territory, suburb/locality and/or postcode).  

Variables related to 'Usual residence - location in Australia' utilise additional input categories. For example, the Census processing variable UAIORIGP (Usual Address Indicator, Original, Persons), which forms part of the coding process for Usual residence, indicates how a person responded to the usual residence question on Census night. It has the following categories:

Usual residence - location in Australia

The ABS uses both automatic and manual address coding indexes based on the ABS Address Register. The automatic coding index contains individual addresses and their associated ASGS codes.

Administrative data may be coded to ASGS Mesh Blocks by the collecting agency, or by the ABS using ABS Address Coder software.

Usual address - country 

Standard Australian Classification of Countries (SACC) should be used in coding responses where detailed country data have been collected. Countries other than Australia are typically not specified and instead are recorded as 'Overseas'.

Once automatic address coding is completed, edits are performed to ensure the records have been successfully and validly coded. Any records that fail edit validation are raised as edit errors and are required to be coded manually.

Exact processes will vary by collection. By way of example, the complete process used in Census is detailed below.

Step 1: Initial Coding

The Census output variable Usual Address Indicator (UAICP) indicates whether a person was at home or elsewhere on Census night. 

• It is largely captured from mark box responses 

The responses ‘Same as Question 1’ and ‘Same as Person 1’ do not require any further address coding. For these, place of usual residence is derived from the dwelling address or the usual residence of Person 1.

• Where the indicator mark boxes have been left blank by a respondent but they have reported a usual address, UAICP is derived based on 

the address reported. 
the relationship information reported on the form indicating that they are a visitor. 

• For the response ‘Other country’, usual residence is set to ‘Overseas’ (see Usual residence - country standard)
• Where the user has responded ‘elsewhere’, they are required to then state their address where they usually reside. 

This address is then passed through the automatic coding index. 

• Where the user has entered a valid and complete street address, this process will more than likely successfully return a Mesh Block code and require no further coding.

Step 2: Editing Missing or Incomplete data

Incomplete addresses or those on paper forms which may have been mis-scanned may not be able to be automatically coded to Mesh Block level, but may be coded accurately to higher levels of the ASGS (SA1 or above).
Addresses that could not be assigned a Mesh Block code during automatic coding are then subject to manual coding. This includes use of the Interactive Spatial Coder (ISC) to attempt to locate an address on a map.

Step 3: Imputation

Addresses that have not been assigned a Mesh Block code in step 2 are imputed. This usually occurs due to non-response (including item non-response); less frequently a partial or unclear response is provided which can only be partially coded (to SA1 level or above).  

Where imputation is required, a hotdecking method will be used to assign a Mesh Block code for their usual address.  The amount of imputation varies depending on the detail of address information provided by the respondent. For example, if a respondent gives only their state or territory of usual residence then they are imputed to a Mesh Block within that state/territory. 

The variable Imputation Flag for Place of Usual Residence (IFPURP) indicates if a person's place of usual residence was imputed in the Census, and to what level. 

In Census 2021, the following applied:

Once address coding is completed, records are passed through auto-validation processes to ensure they are valid. Any records that fail edit validation are raised as edit errors and are required to be resolved using manual coding processes using inputs from the Interactive Spatial Coder (ISC), and/or the Address Register. 

Examples of variation in coding/editing: 

The requirements of individual collections may vary from the example given above. For example, 

• The ASGS supplementary classifications 'no usual address', migratory, off-shore or shipping are not considered valid usual residence locations for estimated resident population (ERP). Data pertaining to these ASGS classifications are reassigned in estimated resident population (ERP) on a pro-rata basis across the relevant state/territory.
• Some administrative data used by the ABS is pre-coded to Mesh Block outside of the ABS. Where this Mesh Block data is missing (not able to be accurately geocoded by the external party), this data is removed from the file by the ABS in situations where (a) the deleted data represent small, unbiased proportions of the data and (b) the data to be analysed is aggregate and/or modelled rather than unit record level data.

Each collection should carefully consider their own requirements with regard to editing procedures.

The standard output categories for 'Usual residence - country' where detailed country data is not collected are:

• Australia, and
• Overseas.

The standard output categories for collecting 'Usual residence - country' where detailed country data is collected and processed are the categories in Standard Australian Classification of Countries (SACC). The hierarchical structure of the SACC allows users the flexibility to produce statistics at the level of the classification which best suits information needs. Requirements for data quality or respondent confidentiality may preclude output of data at the more detailed level of the classification. Under these circumstances, data can be aggregated and disseminated at the higher levels of SACC.

The Australian Statistical Geography Standard (ASGS) is used to output ABS data about Usual residence - location in Australia.

The ASGS ABS Structures describe the geographies by which 'Usual residence - location in Australia' data are output. While data is ideally coded to SA1 or Mesh Block level to enable other geographies to be defined, these geographies are not commonly used as output categories. 

Not all geographic units defined in ASGS will contain population on a usual residence basis.  For example the shipping, offshore and migratory categories do not contain usually resident population and so population estimates are not output for these categories.

In addition to outputting usual residence on an ASGS basis, Usual Address Indicator (UAICP) indicates whether a person was at home or elsewhere on Census night.