Which population figure should I use in Community Profile?

NOTE: This article is for Community Profile customers. Using Placemaker? See: What is Estimated Resident Population (ERP) and how is it relevant to Placemaker?

Community Profile carries more than one population figure for the same area and same year, and they won't match. This article covers what each one counts, which one belongs in a Council Plan, annual report or grant application and why your small areas don't add up to the council total. If your question is why an .id figure differs from the ABS website, that's a different question and it's answered in 'Why Census figures in .id tools can differ from the ABS website'.

The four figures, and what each one counts

Census counts. How many people were counted in the area on Census night. It's a count, not an estimate, and it's a snapshot of one night every five years and it actually comes in two formats, both are reported in Community Profile:
1. Usual residence population (counting people based on their usual address). All the Census characteristics based on counting people in the community profile comes from the Census - age, ancestry, income, travel to work and anything to do with individuals. Usual residence is the main one used for all these person characteristics, unless you are comparing with years prior to 2001, when usual residence counts were not available.
2. Enumerated population (counting people where they are on the night of Census). For counts of dwellings and households (for example household income and tenure type), these are all on a place of enumeration basis.
Estimated Resident Population (ERP). The ABS's estimate of how many people usually live in the area, adjusting for people the Census missed and for residents temporarily away, and updated annually based on births, deaths and known population movement. ERP is usually higher than the Census usual residence count, but not necessarily higher than the enumerated count in areas with lots of visitors. It's the official population of a place. It's updated annually, so between Censuses it's the only figure that moves, and then is subject to revision for the previous five years after the next Census results. See what-is-estimated-resident-population for more information.
Forecast population. The modelled population in future years, produced in Population Forecast rather than Community Profile. It conceptually matches the ERP in the most recent year (but due to timings of forecasts may not match exactly) and is forecast for 25 years ahead from the previous Census.

Note: ERP is an estimate, and the ABS may revise it after the next Census. There are also minor revisions to previous years with every annual release. Say which release your figure comes from whenever you publish it. A figure quoted without its vintage is a common reason two council documents disagree about the same population.

Which population figure to use

What you're doing	Use	Why
Stating the population of your LGA in a Council Plan, annual report or advocacy document	The most recent ERP	It's the official ABS population and it's more current than the Census
Describing the people who live here - age, ancestry, language, income, occupation	Census	Every characteristic in Community Profile comes from the Census
Showing how the community has changed	Census years	The Census is the source with a consistent series of characteristics over time
Anything about future demand, dwellings or service planning	Forecast	ERP and Census describe now and the recent past, not what's coming

If a grant application or a state government template names the figure it wants, use that one, even where you'd otherwise choose differently. Say in the document which figure you've used.

Why your small areas don't add up to the LGA total

Small area ERP is built from experimental SA1 level estimates which the ABS releases each year with the standard geography ERP. Due to rounding adjustments at this lower level, the sum of ERPs at small area level may not match exactly to the LGA total even in the Census year, but they will be very close. In subsequent years, the small area ERPs may drift away from the higher level total due to development occurring in the SA1 - where an SA1 crosses two or more small areas, we don’t necessarily know where the development has occurred and so the totals don’t match exactly. This is reconciled after each Census.

With Census data, the ABS applies random adjustments to protect the confidentiality of individuals. Different tables will be randomised slightly differently, and because .id uses addition of components to provide a table total (rather than using the table total provided by the ABS) these numbers won’t always match between tables. We do this so that the percentages always add to 100%.

Neither is an error, and neither is something to reconcile. Quote the LGA figure for the LGA and the small-area figure for the small area. All these differences are statistically insignificant in any population of more than 1,000 people.

Where to find each figure in Community Profile

The population summary carries the enumerated population, the usual residence population and the annual ERP by small area. ERP is also shown in Population Highlights and on the home page. Every Census-based topic reports its own counts against the Census population. The box at the top right of the table in Community Profile shows the reference population, either usual residence or enumerated for dwellings, and in some cases a subset of the usual residence population, for example people aged 15 and over, or people born overseas.

Forecast figures can be found in the Population Forecast, on a separate site to your Community Profile.

Frequently asked questions

When is ERP released and when will it be on my site?

The ABS releases ERP in late March each year, for the previous 30 June. .id receives it within a few days and aims to have it live on your site within about a week of the ABS release.

Why can't I get ERP by age?

You can, in Housing Monitor, at LGA level. It isn't available from the ABS at smaller geographies. The ABS produces ERP by age and sex once a year, but it's released around August, roughly 14 months after the 30 June it refers to.

Why does my small area's ERP only go back to 2017?

Small-area ERP is normally available back to the year after previous Census only, because that's the earliest point the current SA1 geography can be used. Going back further would mean using 2016 Census SA1s, which don’t always match current boundaries.

Why doesn't the forecast's starting population match the 2021 ERP?

The forecast base reflects the number that was available at the time that forecast was produced, which may differ from the current ERP. The ABS revise past ERP figures each release.

The ABS website gives me a different number. Which is right?

Usually both, measuring different things or built on different geographies. Some of .id’s figures use modified versions of the Census classifications. The reasons are set out in 'Why Census figures in .id tools can differ from the ABS website'. Start there before assuming either figure is wrong.