In my previous post I described how I harmonised ABS census data across three editions onto consistent 2021 boundaries. That was the plumbing. This is the payoff.
I used the resulting dataset to look at how Metro Melbourne's demographic makeup has shifted across birthplace, language spoken at home, and religion - and how it compares to other Australian metro cities and to Regional Victoria.
The short version: Melbourne is diversifying fast, and the rate of change is accelerating.
Birthplace: India overtakes England and China
The most striking trend in the birthplace data is the rise of India as Melbourne's top overseas birthplace.
In 2011, England-born residents led at around 134,000, with India at 106,000 and China at 90,000. By 2016, India and China had both surpassed England. By 2021, India had pulled clearly ahead at 242,000 - nearly double the England-born population, which had barely moved across the decade.
The 2026 projection puts India-born residents at over 307,000, with China at 212,000 and England flat at 132,000. England's population is essentially static, while India's growth is steep and sustained.
China's trajectory is worth watching. Growth was rapid between 2011 and 2016, then slowed noticeably between 2016 and 2021. That slowdown coincided with tighter migration settings and COVID-era disruption, but the Census data alone cannot identify the cause. The 2026 projection simply extends the more recent trend, so this is one to revisit once actual 2026 Census data arrives.
Melbourne's birthplace mix vs other metro cities
When comparing Melbourne's overseas-born profile to other Australian metro cities, a few things stand out.
Melbourne has a notably higher share of India-born residents (15.1% vs 9.6% nationally across other metros), and a stronger representation from Vietnam, Sri Lanka, Greece, Italy, and Malaysia. Other metro cities lean more heavily toward England (14.1% vs Melbourne's 8.3%), New Zealand, and the Philippines.
This isn't just a story about which countries are represented - it's about the shape of the distribution. Melbourne's overseas-born population is more evenly spread across South Asian, Southeast Asian, and Southern European origins, while other metro cities tend to concentrate more in Anglosphere and Pacific sources.
Melbourne vs Regional Victoria
The metro-to-regional comparison sharpens the picture further.
Regional Victoria's overseas-born population looks structurally different. England accounts for over a quarter of it (25.5%), compared to 8.3% in Metro Melbourne. New Zealand is also much more prominent regionally (9.7% vs 5.1%). Meanwhile, China and Vietnam have a much smaller regional footprint - China is just 2.9% in Regional Victoria compared to 10.3% in the metro area.
The practical implication: the services, infrastructure, and cultural fabric that support diverse migrant communities are overwhelmingly concentrated in the metro area.
Language: Mandarin leads, but the spread is wide
Looking at language spoken at home tells a complementary story.
Mandarin is the most common non-English language in Metro Melbourne at 13.2%, essentially on par with other metro cities (13.4%). But Melbourne distinguishes itself with higher shares of Vietnamese (7.1% vs 6.0%), Greek (6.4% vs 3.5%), Punjabi (6.0% vs 3.5%), and Italian (5.1% vs 3.5%). Other metro cities lean more toward Arabic (8.4% vs Melbourne's 5.4%) and Cantonese (6.4% vs 4.9%).
The Greek, Italian and Vietnamese language shares are consistent with Melbourne's long-established communities, while Punjabi rises alongside the growth in India-born residents. The Census shows those patterns moving together, but it does not establish why language retention differs between communities or generations.
The regional language picture has a couple of surprises. Punjabi is proportionally more prominent in Regional Victoria than in Metro Melbourne at 7.2% compared with 6.0%, while Italian is also stronger regionally at 8.1% compared with 5.1%. This dataset establishes the difference, but it does not show which industries, migration programs or settlement choices produced it.
Religion: The secular crossover
The religion data captures one of the most significant structural shifts in Australian society.
By 2021, the gap had narrowed sharply, but Christianity remained ahead at 1.98 million compared with 1.84 million people reporting no religion or a secular affiliation. The linear projection puts the crossover between 2021 and 2026.
Islam is the third-largest category and grew from 144,000 in 2011 to 258,000 in 2021, with the linear projection reaching 310,000 by 2026. That growth occurred alongside broader changes in birthplace composition, although these Census aggregates cannot tell us how much one explains the other.
Compared with other metro cities, Melbourne's religion profile does not show dramatic differences: Christianity is 42.9% compared with 46.8%, while secular beliefs are 39.9% compared with 40.2%. Melbourne has slightly higher shares of Islam, Hinduism and Buddhism, which is consistent with the birthplace patterns described above but does not establish a direct relationship.
The metro-regional split is more interesting. Regional Victoria is already majority secular at 49.4%, well ahead of Metro Melbourne's 39.9%. Meanwhile, Islam, Hinduism, and Buddhism barely register outside the city - all under 1%. Christianity is slightly higher regionally (47.2% vs 42.9%), but the dominant story is just how secular Regional Victoria has become.
Where is the diversity?
As a final cut, I used a normalised Shannon diversity index to identify which Victorian suburbs have the most evenly distributed mix of overseas birthplaces, basically answering "how diverse is this suburb versus others"?
The top 20 list is dominated by middle-ring and outer-ring Melbourne suburbs - places like Gowanbrae, Attwood, Brooklyn, and South Kingsville in the west, and Niddrie, Bellfield, and Watsonia in the north. These are neighbourhoods where no single birthplace group dominates, and the population is spread across many origins.
A few regional entries make the list too - Kennington (Greater Bendigo), North Geelong, Hamlyn Heights, and Bell Post Hill (Greater Geelong) - suggesting that diversity in some regional centres is catching up with parts of Melbourne.
The Shannon index filtered for suburbs covering 95% of Victoria's in-scope population, which removes very small suburbs where a handful of residents could distort the score. That keeps the ranking grounded in places where the numbers actually mean something.
What this doesn't tell you
Census data captures who lives where and what they report on the form. It doesn't capture the lived experience of those communities - how connected or isolated people feel, whether services are adequate, or how identity evolves between generations.
The 2026 projections are linear extensions of 2011–2021 trends. They're useful as a direction indicator but shouldn't be treated as forecasts. Migration policy changes, economic shocks, and global events can all bend these curves.
The real value of this dataset is that it provides a consistent, comparable base. When the actual 2026 census data drops, I'll be able to slot it in and see how well the projections held - and where the surprises are.
The methodology piece explained how to make census geography comparable over time. This analysis is what that effort was for.
To move from the city-level story to a specific suburb or LGA, the LGA Language Explorer lets you examine the underlying data directly. I use the same evidence-bounded approach across my Australian public data and demographic analysis work.