Sydney Race Map (2016)

The 2016 Census data was analysed by voomMAPS to graphically illustrate the distribution of racial groups in Sydney as well as Sydney’s population’s geographical distribution and population density. This map visualises the four racial groupings that represent more than 5% of the Sydney metropolitan area’s population, with each dot representing 10 people of a certain race. A larger version of the map can be viewed by clicking the image below:

Click image for full-size view.


Analysis of the map produced of Sydney’s races and ethnicities show that there is indeed segregation and ethnic enclaves in Sydney.  Of course, unlike the past of other countries (e.g. segregated America or apartheid South Africa), the reasons for these enclaves existence is much different.

One of the main reasons these enclaves exist is that, as a foreigner in a new country, the transition is generally much easier when you are surrounded by people of a similar cultural background – language barriers are less of an issue and groceries from one’s home country are also easier to obtain.

Historical reasons can also be responsible for the location of these enclaves. For example, the cluster of Asian people in the Fairfield local government area can be attributed to the fact that Vietnamese refugees from the 1970s were initially placed in immigration centres in the area.

Lastly, facilities and infrastructure can also be a reason for these enclaves to occur. For example, international students attending university would obviously want to live near their educational institution.

Whilst there are indeed ethnic enclaves in Sydney, it would appear that the segregation seen in Sydney is far less than what is seen in American cities. Ethnic groups’ boundaries are far more diffused and soft, whereas in American cities, you can often see roads and freeways acting as hard barriers to a neighbourhood that is black-dominant versus one that is white-dominant.

For comparison, the previous version of this map using the 2011 census data can be found here.

The Methodology

Inspired by Bill Rankin’s Race and Ethnicity map of Chicago, I decided to create a map of race and ethnicity for my home city of Sydney, as I was curious to see whether a major Australian city, such as Sydney, is as segregated as American cities.

The first step was to acquire the data – this was done by extracting ancestry data at the Statistical Area 1 (SA1) level from the Australian Bureau of Statistics’ 2016 census. At the most detailed level, there were more than 300 categorisations for ancestry. As having over 300 different colours would be impractical, this was aggregated into eight main groups. The top five ancestries from each of the groups are listed below:

  • East Asians (e.g. Chinese, Vietnamese, Filipino, Korean, Thai)
  • Indigenous peoples and Pacific Islanders (e.g. Aboriginal Australians, Maori, Samoan, Fijian, Tongan)
  • Latin Americans (e.g. Chilean, Colombian, Brazilian, Peruvian, Salvadoran)
  • Middle Easterns, North Africans, Arabs and Central Asians (e.g. Lebanese, Turkish, Iranian, Afghan, Egyptian)
  • Subcontinental Asians (e.g. Indian, Sri Lankan, Nepalese, Pakistani, Bangladeshi)
  • Sub-Saharan Africans (e.g. Mauritian, Sudanese, Somali, Ethiopian, Zimbabwean)
  • Whites/Europeans (e.g. English, Australian, Irish, Scottish, Italian)
  • Not adequately described or ancestry not stated

Having eight colours was still hard to differentiate, given the overlaps and small size of the dots, so only the groupings representing more than 5% of Sydney’s population were mapped. Inadequately described responses were also removed. This left four groupings to illustrate. The percentages are shown below:

  • Whites/Europeans: 65%
  • East Asians: 18%
  • Middle Easterns, North Africans, Arabs and Central Asians: 7%
  • Subcontinental Asians: 7%
  • Indigenous peoples and Pacific Islanders: 1%
  • Sub-Saharan Africans: 1%
  • Latin Americans: 1%

The methodology of these classifications isn’t perfect, but given the limitations of the data, it would be impossible to have 100 per cent accuracy. The ‘white’ classification was the most difficult, due to ambiguous identifiers such as ‘Australian’ and ‘American’. For the purposes of this analysis classifications such as ‘Australian’ and ‘American’ have been assumed to be ‘white’ owing to more specific classifications also being listed (such as Aboriginal classifications, ‘African American’ and ‘Native American’).

With the data analyses completed, the data was mapped using the most detailed divisions available for ancestry data in the 2016 Census – i.e. SA1s. Within each SA1 boundary, dots were randomly placed geographically, with each dot representing 10 people of a particular ethnicity/race.

To finalise the map, suburb names, main roads and parks were added to give context to the map.

20,039 total views, 8 views today