Public transport in Australian cities

Let’s have a play with some of the working population transport data provided in the 2011 Australian census. I’ll start with a couple of reasonable assumptions, and we can see how they pan out in the data:

1. Assumption #1: The CBD of each respective city acts as a hub for public transport for the whole city
2. Assumption #2: As distance from the CBD increases, the tendency to use public transport decreases; in other words peripheral suburbs should have the lowest level of public transport usage, and inner city areas should have the highest level of usage.

In order to test these assumptions, I have taken each suburb in the four largest Australian cities and determined the public transport usage for workers located in those areas from the census data. I have then calculated the time it takes to drive from those suburbs to the CBD – driving time being used in this case as a real world approximation of ‘distance’ from a CBD.

I have then plotted a graph of PT usage vs driving time for each of the four cities. This graph allows us to:

·         See whether assumption #2 above is true

·         Determine how good a spread of PT usage there is in the suburbs of each city i.e. in effect determine how good the network is for each city

·         Use a trendline to identify any outlier suburbs i.e. any suburbs that have particularly good or bad public transport connections relative to the rest of the city

But before we get into the results of the above analysis, to set the scene, here is a graph of public transport usage for each capital city region in Australia:

Exclusive public transport usage to workplace in Australian capital cities

As you can see above, Sydney has by far the largest amount of public transport usage among its workers. Melbourne and Brisbane have much lower levels of PT usage, and their levels are [perhaps surprisingly] similar. At the other end of the scale, Darwin, Hobart and Canberra have very low levels of public transport usage for their workers.

There does seem to be a strong correlation between population size and PT usage. Indeed they are ranked exactly in the graph above as per their population. This can be explained by the larger population bases providing higher tax revenues which permit more investment in public transport – although per capita, of course, there should be no difference.

·        

OK, as promised, we’ll now have a look at the relationship between the distance (well, driving time) from the CBD and PT usage in each city and see if we can spot anything interesting in the data.

Public transport in Sydney

Working population public exclusive transport usage vs driving time to Sydney CBD, by suburb

The Sydney CBD – Haymarket – The Rocks area has the highest level of public transport usage in Australia. 65% of workers based there exclusively use public transport to get to work. Likewise, the North Sydney/Lavender Bay and Surry Hills areas come in impressively with just under 50% of workers only using public transport. These areas are very central, very high density and have excellent transport links so this is perhaps to be expected.

The graph shows that the Sydney area in general has a relatively high degree of public transport usage. The trend line has a relatively gentle slope compared with other cities – there is still significant PT usage far out from the CBD. For example, a number of suburbs located 30 minutes drive from the CBD have around 10% of their workers only using public transport.

Sydney also hosts perhaps the most impressive suburban PT usage figure in Australia: that of Parramatta. 28% of workers based in the Parramatta / Rosehill area use only public transport to get to work, despite this area being almost a 30 minute drive from the CBD.

The town of Katoomba, up in the Blue Mountains area west of Sydney also deserves special mention. It is a real outlier with 6% of its workers using public transport to get around yet it is almost 1.5 hours from Sydney CBD. Clearly there is some good public transport up there.

Public transport in Melbourne

Ah, Melbourne. With its wonderful trams running up and down its beautiful Victorian streets and wide boulevards, what’s not to like about public transport in this southern city? Quite a lot, at first glance of the graph below.

Working population exclusive public transport usage vs driving time to Melbourne CBD, by suburb

Melbourne's graph has a much steeper curve than Sydney's graph. This suggests that as distance from the CBD increases, PT usage drops rapidly. Suburbs that are 20 minutes drive from the CBD have approximately 5% PT usage – Sydney has almost 10% usage at this distance.

There are, of course, a number of outliers that buck that particular trend, but the trendline describes the general pattern in the city. The best places for PT usage seem to be South Yarra and St. Kilda. Further out, Mount Evelyn manages a fairly impressive 5%, despite being almost 50 minutes away from the city.

The outer reaches of the Melbourne tram network are approximately 20 minutes drive from the CBD. Perhaps the explanation for Melbourne’s weaker spread of PT usage is that the tram network provides excellent covered within a relatively small area, then the quality of PT drops substantially and the incentive to drive increases.

Public transport in Brisbane

Brisbane is the subtropical focal point of the sprawling, 250km-long, 3 million population, South-East Queensland conurbation and home to the eight lane - and often clogged - Pacific Motorway. So how does public transport usage measure up here? The graph above shows that it follows an even steeper curve than Melbourne; public transport usage drops of at a faster rate the further out from the CBD you go. The Brisbane CBD manages a reasonable 51.6% PT usage, with South Brisbane also bang on trend with around half that level of usage.

Working population exclusive public transport usage vs driving time to Brisbane CBD, by suburb

Suburbs punching above their weight include Fortitude Valley, Indooropilly and Carindale. All three are home to large shopping/commercial districts and good public transport links. However the best performer is St. Lucia, home of the University of Queensland. It manages around 22% PT usage among workers based there; this is roughly double the rate of Carindale (also a good performer). St. Lucia is a 13 minute drive from the CBD and for the sake of comparison, Mount Gravatt - which is also a 13 minute drive - has roughly 6% PT usage among its workers.

Another notable performer is Stradbroke Island (‘Redland Islands’ in the graph above) – this is at least 80 minutes (by ferry and road) from Brisbane CBD, yet 5% of its workers use public transport. The obvious explanation for this is that it is an island so is at the very periphery of the socio-economic influence of Brisbane. Its workers probably use its public transport merely to move around the island, rather than commute onto the mainland and into central Brisbane.

Gold Coast and Sunshine Coast suburbs are excluded from the chart above. However the low levels of public transport usage in these areas should be noted. For example, the suburb with the highest rate of PT usage in the Gold Coast is Broadbeach Waters at a mere 7.6%. Commercial areas such as Southport and Robina also have paltry levels of 3.1% and 3.5% respectively. The Shine Coast fares even worse: its best performing suburb is Buddina / Minyama with only 1.6% of workers using public transport exclusively. Both Gold Coast and Sunshine Coast are cities with extremely decentralised structures. These decentralised cities are examples of urban areas where there is no single CBD that acts as a hub for the whole area; this is counter to the first assumption at the start of this article.

Public transport in Perth

Working population exclusive public transport usage vs driving time to Perth CBD, by suburb

The last city in our analysis here is Perth – like South-East Queensland, it is also notorious for its sprawl along its coastline. The gradient of the trend curve for Perth is steeper still than any of the four cities. This means that public transport usage, despite peaking at a paltry 34% in the CBD, drops off precipitously as distance from the CBD increases. For example, in suburbs located only a ten minute drive from the CBD, average public transport usage among workers is only around 5%. In Sydney the equivalent figure is 26%, in Melbourne it is 14% and in Brisbane it is 7.5%.

Freemantle is the best performing suburbs in Perth – despite being 26 minutes away from the city, its public transport usage is over 9%. This is a level significantly higher than, for example, North Perth, which is 7 minutes away from the city and has usage levels under 7%.

In summary

The best and worst Australian suburbs for public transport usage, categorised by into CBD driving time brackets.

Australia's best suburbs for public transport uptake by workers:

Drive Time	Sydney	Melbourne	Brisbane	Perth
CBD	66%	53%	52%	34%
0-10 mins	Pyrmont / Ultimo, 40%	Docklands, 44%	South Brisbane, 26%	Mount Hawthorn – Leederville, 11%
10-20 mins	Bondi Junction / Waverly 38%	Albert Park, 24%	St. Lucia, 23%	Claremont, 10%
20-30 mins	Parramatta / Rosehill, 28%	Box Hill, 11%	Shailer Park, 5%	Freemantle, 9%
30-40 mins	Hornsby / Waitara, 12%	Springvale South, 6%	North Ipswich / Tivoli, 4%	Armadale - Wungong – Brookdale, 4%
40-50 mins	Mount Druitt / Whalan, 8%	Chelsea / Bonbeach, 5%	Morayfield, 4%	Rockingham, 4%

Australia's worst suburbs for public transport uptake by workers:

Drive Time	Sydney	Melbourne	Brisbane	Perth
0-10 mins	Leichardt / Annandale, 13%	Flemington, 10%	Tarragindi, 5%	Como, 5%
10-20 mins	Forestville / Killarney Heights 4%	Altona Hill, 2%	Rochedale / Burbank, 1%	Winthrop, 1%
20-30 mins	Cromer, 2%	Hillside, 0.4%	Cashmere, 0.3%	Forrestdate / Harrisdale / Piara Waters, 0.5%
30-40 mins	Glenhaven, 1%	Sunbury, 0.5%	Greenbank, 0.3%	Byford, 0.5%
40-50 mins	Mulgoa - Luddenham - Orchard Hills, 0.2%	Wallan, 0.5%	Boonah, 0.3%	Chidlow, 0%

Two overall patterns can be derived from this analysis:

1. As cities get larger, public transport usage also increases per capita
2. As cities get larger, the balance of usage also spreads out through the suburbs more i.e. public transport usage is less centralised and more dispersed in larger Australian cities.

In terms of specific cities, if you want to drive a car, Perth looks like your best bet. If you would prefer to commute by public transport to work, Sydney looks to be the best choice. In terms of suburbs, Albert Park in Melbourne, Parramatta in Sydney or St. Lucia in Brisbane seem to be choices for public transport.

Hopefully this information will be of some value to migrants, whether they be inter-city, interstate or international when it comes to selecting a city to settle in, or a suburb to select. These data provide a quantifiable, objective method of assessing the quality of the public transport services available in Australian suburbs. Hopefully your own anecdotal knowledge matches the conclusions here. If not, please let us know! If you found this article - or any of the other articles on this blog – interesting or useful please let us know in the comments section below.

New look and feel for Australian Census Stats

Yep, you're not imagining it. We have redesigned the look-and-feel of our blog. We have also rationalised the tags associated with each article, so hopefully you'll find it easier to navigate. The tags are displayed at the top of the site for easy access.

You may also have noticed that we're working off of a new domain name. Previously we used censusstats.blogspot.com.au, but we have now acquired www.ozcensusstats.com to use for the blog.

If you like this blog, we'd love it if you shared our links with your friends and colleagues and posted them on your favourite social networking sites and forums.

As for the content you're really interested in - i.e. new stats and articles - we're working on one or two things at the moment. Hopefully we'll have a fresh article to you within the next week.

Let us know what you think of the recent changes to the site!

Australian atheism heatmap

"Calling Atheism a religion is like calling bald a hair color." --Don Hirschberg

My previous post, Religion in Australia mapped, treated the 'no religion' census response the same as religion-specific responses (such as Catholic, Islam etc.). This heatmap intends on redressing the balance somewhat by focusing solely on the amount of non-religious people around Australia.

You can peruse the map directly below, or click this link to open a new browser/tab. The colour scheme is quite simple: the darker the shade of green, the more heathens are located in the area.



*Please note: for convenience, I use the word atheist interchangeably with 'non-religious'. However the census response covers alternative views, such as agnosticism.

Religion in Australia mapped

Now for an interactive map that lets you browse the religious landscape of Australia. Each suburb in Australia has been classified based on its religious make-up. The classification is designed to highlight where there is particular concentrations of certain religions. It has looser requirements for rarer Australian faiths such as Islam, Buddhism, Hinduism and Judaism. Areas classified with common groups such as Catholicism, Anglicanism and non-religious demographics have stricter criteria and require significant proportions of residents.

The colour classifications

The map uses color classifications to identify particularly notable religions in a geographical area. The areas without a colour classification are simply not worth classifying with a single religion - because no religion is dominant enough in these areas to highlight.

The map

The interactive map is below. Clicking on a suburb reveals more details. All of Australia is covered therefore the map can be navigated as necessary. Across the whole of Australia, the five most common responses were:

• Catholic 25.3%
• No Religion 22.3%
• Anglican 17.1%
• Uniting Church 5.0% 
• Presbyterian and Reformed 2.8% (categorised as Christian (Other) here).

Some observations from the map:

• Sydney (click the link to open a new window/tab) has a fairly large amount of religious clusters:
• Areas with 30-40% Islamic residents in the west, including Auburn and Greenacre
• Areas with over 40% Buddhists around Canley Vale in the city's west
• Almost half of the residents are Jewish around Dover Heights
• Hindu
• There is a substantial Eastern Orthodox population to the south of the CBD around Rockdale/Banksia
• Inner city residents have more of a tendency to be non-religious
• The dominant faith in the remaining suburbs is Catholicism

• Melbourne also has a number of religious clusters:
• Substantial Islamic clusters exist around Meadow Heights in the north-west and Dandenong in the south east
• There are two Buddhist clusters; one around Sunshine in the west of the city and another around Springvale in the south-east.
• A few Eastern Orthodox clusters exist in the east of the city, for example, the Oakleigh/Huntingdale area
• The two most dominant groups overall are the Catholics and the non-religious residents. There appears to be more Catholics in the west and more non-religious residents in the east of the city.

• Brisbane has a more even spread than Sydney and Melbourne.
• The non-religious dominate the inner city
• The largest groups in the suburbs are generally Catholics
• The only substantial minority cluster is in Kuraby, where 21% of residents are Islamic

• Perth follows a similar pattern to Brisbane in that the inner city is dominated by younger, non-religious residents and the suburbs tend to be either dominated by Catholics (or fairly evenly spread)

• The inner-west and inner-west of Adelaide are dominated by large proportions of Catholics.
• There is a cluster of Eastern Orthodox in the inner-west city centre area of Richmond
• The remainder of the CBD and suburbs are dominated by the non-religious demographic

• Rural and regional Australia contains a broad mix:
• Religions such as Islam, Buddhism and Hinduism are very rare compared to the large cities
• Anglicanism is extremely popular across broad swathes of the country, particularly in rural New South Wales and Queensland
• In some areas, such as APY Lands, the dominant faith is traditional Aboriginal religion
• As with the cities, Catholicism and non-religious demographics remain very common in rural Australia

Data is provided from the 2011 census undertaken by the Australian Bureau of Statistics. If you would like to read more about Religion and the 2011 Australian Census, click here to read a previous post on the subject.

Australian population density heatmap

Here are some interactive heatmaps that let you compare the population density of Australian suburbs.

If you click on a suburb you will see more details of the suburb, including population density figures, plus other demographic stats. The whole of Australia is covered, so you can zoom out and move around the maps as you need.

The densest suburb is Pyrmont/Ultimo in Sydney with 12,563 people per square kilometre (see below). At the other end of the scale, the emptiest area in Australia is the Western region of South Australia, with 0.001 persons per square kilometre, or one person per 1000 square kilometres.

Sydney

Melbourne

Brisbane

Perth

The source of this data is the 2011 Australian Census, undertaken by the Australian Bureau of Statistics.

If you want to read more about population density in Australia, have a look at this article.

Australia: the world's fastest growing large rich country

In the decade between 2000 and 2010, Australia's population grew by 15.1%, more than any other large rich country. Spain was second with 14.4% growth and Canada was third with a 14.4% increase.

Population growth in IMF advanced economies with a population greater than 10 million, 2000-2010

Despite the country facing many of the same demographic challenges as other rich nations - namely an ageing population and a declining birth rate - it has managed to sustain strong population growth. This growth has been largely fueled by immigration; for example, in 2009, Australia had the highest rate of per-capita immigration in the OECD (the world's rich country club) in 2009.

Relative population growth 1980-2015,  top ten advanced economies

If smaller advanced economies are included in the analysis, only Singapore, Israel and Cyprus have outpaced Australia since 1980. When taking the IMF's 2015 population estimates, Singapore's growth rate is 123% during the period 1980-2015. Israel's rate is 119%, the growth rate for Cyprus rate is 78% and Australia comes in at 59%.

As the twelfth-emptiest country in the world, Australia certainly has the physical capacity to keep up its high rate of population growth. This is a key strength of the Australian economy as it can supplement its resource-led productivity gains with population-derived gains.

Population growth of Australian cities 2001-2011

Of Australia's larger cities, the south-east Queensland cities of Gold Coast and Sunshine Coast have been the fastest growing in the country from 2001 to 2011. The Gold Coast grew by 2.5% per year during the decade and the Sunshine Coast grew by 2.4% over the period.

Perth was the fastest growing capital during the decade to the 2011 census, with an annual rate of 2.14%. Queensland capital Brisbane was close behind with a growth rate of 2.11% over the period. The closeness of the figures between these two cities masks the accelerating growth in Perth since 2006. In the first half of the decade, Brisbane significantly outpaced Perth; however since 2006 the roles have reversed and Perth achieved a rate of 2.69% per year between 2006-2011.

The following chart uses the ABS's Greater Capital City Statistical Areas for the capitals, and Statistical Area Level 4 boundaries for the Sunshine Coast, Gold Coast, Newcastle and Wollongong. This ensures consistent boundaries and no crossover. The data source is the 2011 Australian census.

Annual population growth rates of major Australian cities 2001-2011

Australia's largest city, Sydney, only managed a growth rate of around 1% during the decade to 2011. This is around half the growth rate seen in Perth and south-east Queensland. The Brisbane-Gold Coast-Sunshine Coast conurbation grew from a population of 2,357,417 to 2,933,714; equivalent to a growth rate of 2.2% per year during 2001-2011.