Skip to content

A Guide to the ABS API Datasets

Source: vignettes/articles/a-guide-to-abs-api-datasets.Rmd
a-guide-to-abs-api-datasets.Rmd


This guide explores the get_abs_data and list_abs_dataflows functions in the scgElectionsAU package, designed to retrieve data directly from the Australian Bureau of Statistics (ABS) Data API. These functions provide access to SEIFA indices, Census data at CED (Commonwealth Electoral Division) level, and other ABS datasets without needing to download large DataPack files. See the ABS Census Datasets Guide for SA1-level Census data retrieval.

Introduction to the ABS Data API

The Australian Bureau of Statistics provides a Data API based on the SDMX (Statistical Data and Metadata eXchange) standard. This RESTful web service allows programmatic access to a wide range of ABS datasets without downloading large files.

The get_abs_data function provides a simple interface to query this API, with data parsed using the readsdmx package. The API is particularly useful for:

  • SEIFA indices at SA1 level (2011, 2016, 2021)
  • Census 2021 data at CED (Commonwealth Electoral Division) and SED (State Electoral Division) levels
  • Census 2011 data at SA1 level (South Australia only)
  • Other ABS statistical datasets

The get_abs_data Function

The get_abs_data function retrieves data from the ABS Data API by specifying a dataflow identifier.

Parameter Description Example Values
dataflow The dataflow identifier to query. “ABS_SEIFA2021_SA1”, “C21_G01_CED”
filter Optional SDMX filter expression to subset data. NULL, “all”
start_period Optional start period for time series data. “2021”
end_period Optional end period for time series data. “2021”

The list_abs_dataflows Function

The list_abs_dataflows function helps discover available dataflows from the ABS API.

Parameter Description Valid Options
filter Filter dataflows by geographic level or type. NULL, “SA1”, “CED”, “census”
pattern Additional regex pattern to search dataflow names. “SEIFA”, “income”, etc.

Example: Finding Available Dataflows 

# List all SA1-level dataflows
sa1_flows <- list_abs_dataflows("SA1")
head(sa1_flows)
#>                            id
#> 162 ABS_CENSUS2011_B01_SA1_SA
#> 165 ABS_CENSUS2011_B02_SA1_SA
#> 168 ABS_CENSUS2011_B03_SA1_SA
#> 171 ABS_CENSUS2011_B04_SA1_SA
#> 174 ABS_CENSUS2011_B05_SA1_SA
#> 177 ABS_CENSUS2011_B06_SA1_SA
#>                                                             name
#> 162                 B01 Selected Person Characteristics (SA1 SA)
#> 165                   B02 Selected Medians and Averages (SA1 SA)
#> 168 B03 Place of Usual Residence on Census Night by Sex (SA1 SA)
#> 171                                      B04 Age by Sex (SA1 SA)
#> 174         B05 Registered Marital Status by Age by Sex (SA1 SA)
#> 177             B06 Social Marital Status by Age by Sex (SA1 SA)

# List all CED-level dataflows
ced_flows <- list_abs_dataflows("CED")
head(ced_flows)
#>                        id
#> 386       ABS_LABOUR_ACCT
#> 387 ABS_LABOUR_ACCT_UNBAL
#> 475           C21_G01_CED
#> 481           C21_G01_SED
#> 484           C21_G02_CED
#> 490           C21_G02_SED
#>                                                                                                                name
#> 386                       Labour Account Australia, Annual Balanced: Subdivision, Division and Total All Industries
#> 387                                               Labour Account Australia, Annual Unbalanced: Total All Industries
#> 475 Census 2021, G01 Selected person characteristics by sex, Commonwealth Electoral Divisions (CED 2021 boundaries)
#> 481        Census 2021, G01 Selected person characteristics by sex, State Electoral Divisions (SED 2021 boundaries)
#> 484          Census 2021, G02 Selected medians and averages, Commonwealth Electoral Divisions (CED 2021 boundaries)
#> 490                 Census 2021, G02 Selected medians and averages, State Electoral Divisions (SED 2021 boundaries)

# Search for SEIFA dataflows
seifa_flows <- list_abs_dataflows(pattern = "SEIFA")
print(seifa_flows)
#>                     id                                       name
#> 410  ABS_SEIFA2016_LGA  SEIFA 2016 by Local Government Area (LGA)
#> 411  ABS_SEIFA2016_POA             SEIFA 2016 by Postal Area Code
#> 412  ABS_SEIFA2016_SA1     SEIFA 2016 by Statistical Area 1 (SA1)
#> 413  ABS_SEIFA2016_SA2           SEIFA 2016 by Statistical Area 2
#> 414  ABS_SEIFA2016_SSC            SEIFA 2016 by State Suburb Code
#> 415  ABS_SEIFA2021_LGA  SEIFA 2021 by Local Government Area (LGA)
#> 416  ABS_SEIFA2021_POA       SEIFA 2021 by Postal Area Code (POA)
#> 417  ABS_SEIFA2021_SA1     SEIFA 2021 by Statistical Area 1 (SA1)
#> 418  ABS_SEIFA2021_SA2     SEIFA 2021 by Statistical Area 2 (SA2)
#> 419  ABS_SEIFA2021_SAL SEIFA 2021 by Suburbs and Localities (SAL)
#> 420      ABS_SEIFA_LGA  SEIFA 2011 by Local Government Area (LGA)
#> 421      ABS_SEIFA_SA2     SEIFA 2011 by Statistical Area 2 (SA2)
#> 422      ABS_SEIFA_SLA SEIFA 2011 by Statistical Local Area (SLA)
#> 1211         SEIFA_POA       SEIFA 2011 by Postal Area Code (POA)
#> 1212         SEIFA_SA1     SEIFA 2011 by Statistical Area 1 (SA1)
#> 1213         SEIFA_SSC      SEIFA 2011 by State Suburb Code (SSC)


API vs DataPacks: When to Use Each

The ABS provides data through two main channels, each with advantages:

Feature API (get_abs_data) DataPacks (get_census_data)
Data Format Direct API query Download ZIP file
Speed Fast for small queries Initial download slow, cached
SA1 Coverage Limited (SEIFA, 2011 SA only) Complete (all tables)
CED Coverage Complete (Census 2021) Not available
Best For CED-level Census, SEIFA Full SA1-level Census

Recommendation: * Use get_abs_data() for CED-level Census data and SEIFA indices * Use get_census_data() for comprehensive SA1-level Census data


SEIFA (Socio-Economic Indexes for Areas)

SEIFA is a suite of four indexes that summarise different aspects of socio-economic conditions. Each index ranks areas from most disadvantaged to most advantaged based on Census data.

Available SEIFA Dataflows

Dataflow ID Description Geographic Level
ABS_SEIFA2021_SA1 SEIFA 2021 indices SA1
ABS_SEIFA2016_SA1 SEIFA 2016 indices SA1
SEIFA_SA1 SEIFA 2011 indices SA1

SEIFA Index Types

The SEIFA data includes four different indexes:

  • IRSD (Index of Relative Socio-economic Disadvantage): Summarises variables related to disadvantage only (e.g., low income, unemployment, no qualifications)
  • IRSAD (Index of Relative Socio-economic Advantage and Disadvantage): Summarises variables related to both advantage and disadvantage
  • IER (Index of Economic Resources): Summarises variables related to economic resources (income, housing costs, assets)
  • IEO (Index of Education and Occupation): Summarises variables related to education and occupation

SEIFA Measures

Each index provides several measures:

  • SCORE: The actual index score (mean of 1000, standard deviation of 100)
  • RWSR: Rank within State/Territory
  • RWAR: Rank within Australia
  • RWSP: Percentile within State/Territory
  • RWAP: Percentile within Australia
  • RWSD: Decile within State/Territory
  • RWAD: Decile within Australia
  • URP: Usual Resident Population

Example: Retrieving SEIFA 2021 Data 

# Get SEIFA 2021 data for all SA1s
# Note: This is a large dataset and may take time
seifa_2021 <- get_abs_data("ABS_SEIFA2021_SA1")
INDEX_TYPE MEASURE OBS_STATUS ObsDimension ObsValue SEIFA_SA1 UNIT_MEASURE
IEO URP NA 2021 152 10402108614 PSNS
IEO RWAD NA 2021 7 40104101425 RANK_DC
IRSAD RWSD NA 2021 10 20804119401 RANK_DC
IRSAD SCORE NA 2021 1162 20902120502 SCORE 
# Filter for IRSD scores only
irsd_scores <- seifa_2021[seifa_2021$INDEX_TYPE == "IRSD" &
                           seifa_2021$MEASURE == "SCORE", ]
INDEX_TYPE MEASURE OBS_STATUS ObsDimension ObsValue SEIFA_SA1 UNIT_MEASURE
IRSD SCORE NA 2021 980 11102121517 SCORE
IRSD SCORE NA 2021 1053 40304108006 SCORE
IRSD SCORE NA 2021 992 30907125329 SCORE
IRSD SCORE NA 2021 NA 60401107706 SCORE


Census 2021 at CED Level

The ABS API provides Census 2021 data at Commonwealth Electoral Division (CED) level through a comprehensive set of tables. This is particularly useful for analysing demographic characteristics of federal electorates.

Available Census 2021 CED Dataflows

The Census 2021 CED dataflows follow the naming pattern C21_G##_CED, where ## is the table number (01-62). Key tables include:

Dataflow ID Table Description
C21_G01_CED G01 Selected person characteristics by sex
C21_G02_CED G02 Selected medians and averages
C21_G04_CED G04 Age by sex
C21_G07_CED G07 Indigenous status by age by sex
C21_G09_CED G09 Country of birth by age by sex
C21_G14_CED G14 Religious affiliation by sex
C21_G17_CED G17 Total personal income (weekly) by age by sex
C21_G19_CED G19 Long-term health conditions by age by sex
C21_G46_CED G46 Labour force status by age by sex

Note: State Electoral Division (SED) data is also available using the pattern C21_G##_SED.

Example: Retrieving Census 2021 CED Data 

# Get G01 (Selected Person Characteristics) for all CEDs
g01_ced <- get_abs_data("C21_G01_CED")

Sample Data:

ObsDimension ObsValue PCHAR REGION REGION_TYPE SEXP STATE
2021 78145 P_1 143 CED 2 1
2021 0 P_1 197 CED 1 1
2021 89519 P_1 310 CED 2 3
2021 75739 P_1 316 CED 1 3


Understanding CED Data Structure

The CED data includes several dimension columns:

  • REGION: The CED code (numeric identifier)
  • REGION_TYPE: Always “CED” for these dataflows
  • STATE: State/Territory code (1=NSW, 2=VIC, 3=QLD, etc.)
  • ObsDimension: The time period (e.g., 2021)
  • ObsValue: The data value
  • Other dimensions vary by table (e.g., SEXP for sex, PCHAR for person characteristics)


Census 2011 at SA1 Level (South Australia)

The ABS API provides Census 2011 Basic Community Profile (BCP) tables at SA1 level, but only for South Australia. These follow the naming pattern ABS_CENSUS2011_B##_SA1_SA.

Available Census 2011 SA1 Dataflows

Dataflow ID Table Description
ABS_CENSUS2011_B01_SA1_SA B01 Selected person characteristics
ABS_CENSUS2011_B02_SA1_SA B02 Selected medians and averages
ABS_CENSUS2011_B04_SA1_SA B04 Age by sex
ABS_CENSUS2011_B07_SA1_SA B07 Indigenous status by age by sex
… … (Tables B01-B46 available)

Note: For complete SA1-level Census data across all states, use get_census_data instead.

Example: Retrieving Census 2011 SA1 Data 

# Get B01 (Selected Person Characteristics) for SA1s in South Australia
b01_sa1 <- get_abs_data("ABS_CENSUS2011_B01_SA1_SA")


Performance Considerations

Large Datasets

Some dataflows contain large amounts of data:

  • SEIFA SA1 datasets: ~60,000 SA1s x 4 indexes x 8 measures = ~1.9 million rows
  • CED datasets: ~151 CEDs x variables = typically 1,000-10,000 rows

For large SA1-level queries, the API may timeout. In these cases:

  1. Use the filter parameter to subset data
  2. Use get_census_data for comprehensive SA1 Census data

Filtering Data

The filter parameter accepts SDMX filter expressions. The exact format depends on the dataflow structure.

# Example: Filter by specific dimension values
# Note: Filter syntax varies by dataflow
data <- get_abs_data("C21_G02_CED", filter = "all")


Quick Reference: Common Use Cases

Get SEIFA Scores for SA1s 

# SEIFA 2021
seifa_2021 <- get_abs_data("ABS_SEIFA2021_SA1")

# SEIFA 2016
seifa_2016 <- get_abs_data("ABS_SEIFA2016_SA1")

Get Census Demographics for Electorates 

# Age and sex distribution
age_sex <- get_abs_data("C21_G04_CED")

# Indigenous status
indigenous <- get_abs_data("C21_G07_CED")

# Country of birth
birthplace <- get_abs_data("C21_G09_CED")

# Religious affiliation
religion <- get_abs_data("C21_G14_CED")

Get Socioeconomic Data for Electorates 

# Medians and averages (income, rent, mortgage)
medians <- get_abs_data("C21_G02_CED")

# Personal income distribution
income <- get_abs_data("C21_G17_CED")

# Labour force status
employment <- get_abs_data("C21_G46_CED")


Additional Resources

For more information about the ABS Data API:

For Census data documentation: