Socio-Economic Indexes for Areas (SEIFA)
SEIFA is a set of four ABS indexes that rank areas of Australia by relative socio-economic advantage and disadvantage, using Census data. Each area gets a score, a rank and a decile, where decile 1 is the lowest-scoring 10% of areas and decile 10 the highest.
What SEIFA measures
SEIFA takes a wide set of Census answers about income, education, jobs, housing and household circumstances and boils them down to a single score for each area. The ABS builds the indexes from Statistical Area Level 1 (SA1) blocks of a few hundred people each, then publishes them for larger areas too: SA2s, councils, postcodes and suburbs.
The current edition is SEIFA 2021, built from the 2021 Census and released in April 2023. The next edition will come from the 2026 Census.
The four indexes
SEIFA isn't one index but four, and each answers a slightly different question, so the choice of index changes which areas rank high and low.
- IRSAD (Index of Relative Socio-economic Advantage and Disadvantage) counts signs of both advantage and disadvantage: high incomes and skilled jobs as well as unemployment and low incomes. It's the best general-purpose choice for comparing places.
- IRSD (Index of Relative Socio-economic Disadvantage) only counts signs of disadvantage. A high score means an area has little disadvantage, not that it's wealthy, so it can't separate a comfortable suburb from a very affluent one. Governments use it widely for funding and health planning.
- IER (Index of Economic Resources) focuses on money: income, housing costs and home ownership. It leaves out education and occupation.
- IEO (Index of Education and Occupation) focuses on qualifications and the kind of work people do. It includes no income figures.
IRSD and IRSAD usually agree (the ABS puts their correlation at 0.94), but they can split where an area mixes pockets of hardship with pockets of wealth. IRSD picks up the hardship, while IRSAD lets the wealth offset it.
Reading a score, rank and decile
Each area gets a score, a rank and a decile, both nationally and within its state. Scores sit on an arbitrary scale centred around 1,000, so an area scoring 1,100 isn't 10% better off than one scoring 1,000. The ABS recommends working with ranks and deciles instead.
Deciles split areas, not people, into ten equal groups. Decile 1 is the lowest-scoring 10% of areas and decile 10 the highest, and each kind of area is ranked against its own kind: an SA2's decile compares it with other SA2s, and a postcode's with other postcodes.
Using SEIFA for site and location decisions
SEIFA works well as a first screen. If a premium format only trades in well-off areas, filtering a shortlist to IRSAD or IER deciles 8 to 10 takes seconds. A value format might look at the other end of IRSD, and a tutoring or childcare business might care most about IEO.
It works less well as the final answer, for a few reasons.
- It describes areas, not people. A mid-decile SA2 can contain blocks in both the top and bottom deciles. Treating everyone in an area as typical of it is known as the ecological fallacy, and it's easy to fall into with large areas. Looking at the SA1s inside a larger area shows how mixed it is.
- Small places are less reliable. The ABS advises caution with scores for suburbs and postcodes with small populations, because their boundaries don't line up neatly with the SA1s the scores are built from.
- Scores are rankings, not quantities. Averaging or adding them across areas gives a rough number at best, so treat any combined score with care.
- Don't double count. Comparing an index with a figure already built into it, such as IEO with unemployment, tells you very little.
- It ages. SEIFA 2021 reflects the Census taken on 10 August 2021. Areas that have grown or changed quickly since then may rank differently today.
SEIFA and Censible
SEIFA tells you where an area ranks, not why. Two suburbs in the same decile can get there in very different ways: one through high incomes and home ownership, another through a highly qualified workforce on modest pay. For a site decision, that difference matters.
Censible shows the figures underneath: household income, rent and mortgage repayments, qualifications, the mix of occupations and unemployment. You can see them for suburbs, councils, postcodes, statistical areas and catchments you draw around a site, each compared with Australia or another area. Catchments are where this helps most, because the ABS doesn't publish SEIFA scores for areas you define yourself.
Source: ABS, Socio-Economic Indexes for Areas (SEIFA), Australia, 2021.
Statistical Area Level 2 (SA2)
A medium-sized area defined by the ABS to represent a community that interacts socially and economically. Most have between 3,000 and 25,000 residents.
Suburbs and Localities (SAL)
The ABS's approximation of officially recognised suburbs and localities, built from mesh blocks. It replaced State Suburbs (SSC) from the 2021 Census.
Catchment (trade area)
The area a store, clinic, school or other site draws its customers, patients or users from. A catchment can be drawn as a ring around the site, a drive or walk time, a hand-drawn boundary, or from where existing customers live.
Questions people ask
There isn't a good or bad score, only a relative one. Higher scores mean an area is relatively more advantaged (or, for IRSD, less disadvantaged) than other areas. Deciles are easier to read than scores: decile 10 is the top 10% of areas and decile 1 the bottom 10%.
The area is among the 10% of areas of its kind with the lowest scores on that index, either nationally or within its state, depending on which decile you're reading. It describes the area as a whole, and many people living there won't be disadvantaged.
IRSD only counts signs of disadvantage, so a high score means little disadvantage. IRSAD counts both advantage and disadvantage, so it can also tell a comfortable area from a very affluent one. For comparing places for a business, IRSAD is usually the better fit.
Not officially. The ABS only publishes scores for its own areas, from SA1 up. A common workaround is a population-weighted average of the SA1 scores inside the catchment, but because scores are rankings the result is a rough guide, not an ABS figure. Looking at the catchment's income, education and occupation figures directly is usually more useful.
The next edition will be built from the 2026 Census. The ABS hasn't set a date, but SEIFA 2021 was released in April 2023, about 20 months after Census night.