National demographic indicators seriesAustralia

National Demographic Indicators Series
Data notes

Estimated resident population (ERP)

Estimated Resident Population is the official population of a Local Government Area. It adjusts for the net undercount found in Census data, people overseas on Census night, and is updated annually based on the number of registered births, deaths, and an estimate of overseas, interstate and intra-state migration.

While ERP is the most accurate measure of population at any point in time, it is subject to revision. Minor revisions are made each year to previous years' populations, and a final revision to the previous 5 years' results happens after each Census when the results are 'rebased' to the results of the most recent Census. This rebasing can alter populations significantly, depending on the Census findings, and indeed this is one of the reason we have a Census every 5 years.

Despite this revision, the ERP remains the official population count, and is used in allocation of funding at all levels of government, and the distribution of electorates by the Australian Electoral Commission.

Aged under 5

Derived from the Census question:
'What is the person's data of birth (or age of last birthday)?'

This dataset counts the number of persons aged 0-4 inclusive at the 2021 Census, as a percentage of total population in the area. A high percentage indicates an area of family formation (eg. first home buyers), or large families, or many single parent families. A low percentage may indicate an area with few families – it could be an elderly population, or a population of young adults attending university, for instance.

If an answer to the Age question is not provided, the Australian Bureau of Statistics imputes the age of the respondent, so there is no "Not stated" category for this variable.

For more information on the ABS Census dictionary entry of Age, please refer to the Age ABS Census dictionary entry on the ABS website.

Aged 65+

Derived from the Census question:
'What is the person's data of birth (or age of last birthday)?'

This dataset counts the number of persons aged 65 years and over at the 2021 Census, as a percentage of total population in the area. A high percentage indicates an area with a lot of retirees and elderly population. This may be due to people retiring to the area (common in coastal communities) or ageing in place in family households. A low percentage indicates a younger population, of young adults or families with children.

If an answer to the Age question is not provided, the Australian Bureau of Statistics imputes the age of the respondent, so there is no "Not stated" category for this variable.

For more information on the ABS Census dictionary entry of Age, please refer to the Age ABS Census dictionary entry statement on the ABS website.

Median age

Derived from the Census question:
'What is the person's data of birth (or age of last birthday)?'

This dataset shows the median age in years of the total population. The median is derived by arranging the entire population in an ordered list from youngest to oldest. It is the age of the middle person (or the average of the two in the middle in a list of even size).

Median is a useful summary measure of average age and very convenient for ranking areas – places with a large elderly population will have a higher median age while those with a lot of children and younger adults will have a lower median age. However a lot of detail of the population is lost. It is possible to have two areas with identical medians but very different age distributions.

If an answer to the Age question is not provided, the Australian Bureau of Statistics imputes the age of the respondent, so there is no "Not stated" category for this variable.

For more information on the ABS Census dictionary entry of Age, please refer to the Age ABS Census dictionary entry statement on the ABS website.

Language other than English

Derived from the Census question:
'Does the person use a language other than English at home?'

This indicator shows the number of people who speak a language other than English at home, as a percentage of the total population.

Language spoken at home is designed to measure 'first' or 'native' language, though some migrants who have been in Australia for many years may speak English at home. It can include second and later generation migrants who speak a language other than English.

Sign-language speakers are included as speaking a language other than English. It excludes multi-lingual populations. (e.g. If I speak English and French, but mainly speak English at home, the fact that I speak French is not captured.)

Areas with a high proportion of people speaking a language other than English are likely to have a high number of migrants, including recent arrivals, or a large second generation migrant population maintaining their home language. Areas with a low proportion probably have predominantly Australian-born populations or people born in English speaking countries.

Language spoken at home is coded using the Australian Standard Classification of Languages, 2021 (ABS Cat. No. 1267.0).

First Nations

Derived from the Census question:
'Is the person of Aboriginal or Torres Strait Islander origin?'

People included in this category include all those who answered “Yes, Aboriginal”, “Yes, Torres Strait Islander” or both. It is not derived from the “Australian Aboriginal” response to the Ancestry question, and this population can have any birthplace.

This indicator presents the number of people of Aboriginal or Torres Strait Islander origin, expressed as a percentage of the total population.

Note that due to greater identification over time, comparing indigenous populations between Census years should be done with caution. For more information please see the ABS Census dictionary entry for Indigenous Status.

Changed address in last 5 years

Derived from the Census questions:
'Where does the person usually live?' and 'Where did the person usually live five years ago (at 10 August 2016)'

This indicator measures the mobility of the population over the 5 year period ending on Census night, August 10th, 2021. It presents the number of people who changed address at least once within this 5 year period, as a percentage of the total population aged 5 years or more.

People may have moved locally, interstate or from overseas.

A high percentage indicates a mobile population, with people staying a relatively short time. This is often associated with inner city, rental and student areas with a young population.

A low percentage indicates a stable, sedentary population, with people staying in place over many years. This is often associated with older suburban and rural populations and full home ownership.

For more information please see the ABS Census dictionary entry statement on Usual Address 5 years ago.

Worked at home

Derived from the Census question:
'How did the person get to work on Tuesday, 10th August 2021?'

This dataset relates specifically to those people whose main method of travel to work was “Worked at home”. This includes both people running a home based business and employees working from home.
For the 2021 Census, the number and percentage of population working at home has increased in most areas due to the COVID-19 pandemic. This effect is strongest in Sydney and Melbourne which were under lockdown and employees were required to work from home where possible. Worked at home relates to the day of Census only and is not necessarily indicative of a long-term trend.
The population for calculation of percentages is all employed persons.

For more information please see the ABS Census dictionary entry for Method of Travel to Work on the ABS website.

Attending university

Derived from the Census question:
'What type of educational institution is the person attending?'

This indicator looks at the number of people attending a University during the second semester of 2021 (when Census was conducted) as a proportion of the total population.

There is no age restriction on the denominator, which includes both children and the elderly. For this reason, university attendance may appear surprisingly low.

'Tertiary education' is usually taken to mean University and TAFE education, however this indicator measures university attendance only.

Areas with a high proportion of university attendance are likely to be areas near universities, and affluent areas in capital cities. Low proportions of university attendance are associated with areas a long distance from universities and low socio-economic areas.

For more information about this topic, please refer to the ABS Census dictionary entry for Type of Educational Institution Attending on the ABS website.

University qualification

Derived from the Census question:
'What is the level of the highest qualification the person has completed?'

This indicator looks at the number of people with a university level qualification, as a percentage of the total population aged 15+.

University qualifications are defined as non-school qualifications which are Bachelor Degrees, Graduate Certificate or Diplomas, Masters Degrees and PhDs. Only the highest qualification is counted.

Qualifications attained overseas may be counted, if they are within the scope of the Australian Standard Classification of Education.

Educational qualifications are highly correlated with income and socio-economic advantage. A high proportion of university qualifications indicates a highly educated area, most likely with relatively high incomes. Low proportions of university qualifications may be correlated with lower incomes, or older populations (university attendance has increased in recent decades).

For more information please refer to the ABS Census dictionary entry statement for Highest Level of Educational Attainment on the ABS website.

Median household income

Derived from the Census question:
'What is the total of all wages/salaries, government benefits, pensions, allowances and other income the person usually receives?'

Household income data presents the total combined weekly incomes of all persons over the age of 15 in the household. 'Other non-classifiable households' are excluded from the calculation.

Median Household income is an imputed midpoint obtained by arranging all reported household incomes in an ordered list and taking the middle value. Because incomes are collected in ranges, the range containing the middle value may be broad, and the approximate dollar value of the median is calculated by assuming a uniform distribution of incomes within that range and calculating the point at which the 50th percentile of population distribution would fall within this range based on this assumption, and the cutoffs of the surrounding 2 ranges.

As individual income is collected in ranges, in order to calculate household income, a dollar value has to be imputed by the ABS to each range, then the individual incomes are aggregated, and output into ranges again. There is an inherent uncertainty in this process, as well as the calculation of medians from the derived ranges, so household incomes should only be treated as a guide to the income level in an area, not an exact calculation.

For more information on this topic, please see the ABS Census dictionary entry statement on Total Household Income (HIND).

Average household size

'Average household size’ consists of the number of persons counted in private dwellings divided by the number of occupied private dwellings on Census night.

It is a measure of the mean number of persons per private household and excludes non-private dwellings.

Average household size is likely to be high in family areas with large numbers of children, large houses, or many people sharing dwellings in group households. It is likely to be low in elderly areas with a lot of 1 and 2 person households, and areas with very small dwellings (0-1 bedroom) and a lot of young people living alone.

Households renting

Derived from the Census questions:
'Is this dwelling [owned outright, owned with a mortgage etc.]' and 'If this dwelling is being rented, who is it rented from?'

This indicator presents the number of dwellings being rented, as a percentage of all occupied private dwellings (on Census night).

Renting includes households renting from a State/Territory Government housing authority (generally referred to as public housing), households renting from a housing co-operative, community organisation or church group, private rentals through a real estate agent or private landlord, and renting from an employer (eg. Defence).

It excludes households with a mortgage or full home ownership by the occupants. Areas with a high proportion of renters are also likely to have high population mobility and a young population. Young adults are more likely to rent than people in later life stages.

For more information on this topic please see the ABS Census dictionary entry statement on Tenure Type.

Medium and high density

Derived from the Census:
'Dwelling Structure is derived from the ABS address register supplemented with information from Census Field Officers.'

This is the only Census output which is classified directly by the ABS via a register, and field officers, and not by respondents to the Census.

This indicator measures the number of dwellings which are classified by .id’s dwelling classification system as “Medium density” or “High density”, as a percentage of total dwellings in the area.

'Medium density' includes all semi-detached, row, terrace, townhouses and villa units, plus flats and apartments in blocks of 1 or 2 storeys, and flats attached to houses.

'High density' includes flats and apartments in 3 storey and larger blocks. This is a broad indicator of the amount of housing which is not in the form of detached housing, the dominant form of housing Australia-wide.

As well as separate houses, caravans and cabins, tents and flats attached to shops are excluded from the Medium and High Density category.

Medium and high density housing is likely to be highest in built-up urban areas, particularly inner cities and inner suburbs, and areas with a lot of retirement housing.

For more information on this topic, please see the ABS Census dictionary entry statements on Dwelling Structure and Dwelling Type.

SEIFA index of disadvantage

The SEIFA indexes are derived from Census data by a method called Principal Component Analysis which is a regression technique that derives an index from a set of variables related to the concept of disadvantage, based on the level of correlation between those variables.

There are four indexes in the SEIFA set:

  • Index of Relative Socio-Economic Disadvantage
  • Index of Relative Socio-Economic Advantage/Disadvantage
  • Index of Economic Resources
  • Index of Education and Occupation

Of these, by far the most commonly used is the Index of Relative Socio-Economic Disadvantage (IRSED), and this is the one presented in the National Demographic Indicators Series.

The IRSED compares the level of disadvantage between areas, and is not skewed by a high level of advantage. Technically a high score only measures a lack of disadvantage – NOT evidence of advantage).

Disadvantage is defined by a range of characteristics, including the relative proportions of 17 Census characteristics, for example:

  • Low income
  • Low educational attainment
  • High unemployment
  • Residents working in relatively unskilled occupations
  • High proportion of residents with poor English proficiency
  • High proportion of single parent families
  • High proportion of residents paying low rent

On the Index of Relative Socio-Economic disadvantage scale, low numbers indicate a higher level of disadvantage, and high numbers indicate a low level of disadvantage (more advantaged or affluent areas).

For more details on the construction of the index, plus further information on its use, see (Australian Bureau of Statistics, 2021, Socio-Economic Indexes for Areas (SEIFA 2021) cat. no. 2033.0.55.001) on the ABS website.

A low SEIFA score for an area does not necessarily imply anything about individuals living in the area as the score is for the area overall. While a low score probably indicates many low income people living there, it does not imply that any particular resident is low income.

For more information about the use of SEIFA please refer to the ABS publication above or contact .id.