A new survey suggests that Australia would no longer be a majority religious country if the format of the census question on religion were changed. The Essential Media poll, commissioned by the Census – Not Religious? Mark No Religion campaign, tested two versions of the question.
In the current census format, respondents choose from a list that includes common religions, “no religion”, and “other”. At the 2021 census, about 39% selected “no religion”. In the new survey, when asked in the same format, 43% chose “no religion”. However, when given a simple “yes/no” option first (“do you have a religion?”) followed by a text box for those answering “yes”, 54% of respondents picked “no”. This translates to roughly 2 million adults.
Campaign calls for change
The campaign argues that the Australian Bureau of Statistics (ABS) format overstates religiousness and undercounts those without religion. Spokesperson Michael Dove, a “demography nerd”, said the census is the “gold standard” for data that informs “debate, policy and ultimately funding decisions”. He added, “We trust the ABS to deliver us high-quality data that we can rely on and be confident that the right decisions are being made on the basis of the right data.”
The campaign website lists various surveys finding the “no religion” cohort higher than the census’s 39%. The religion question has long been contentious, with over 70,000 Australians declaring themselves Jedi Knights in 2001 and Pastafarians using colanders as religious headwear to avoid being counted as Christians.
Decline in Christianity
According to the ABS, Christianity has steadily declined over 50 years but remained the most common religion at the last census, with 43.9% listing it, while 38.9% ticked “no religion”. The number of people from non-Christian faiths and no religion has consistently risen.
The poll surveyed over 2,000 people with the existing census question and another 2,000 with the proposed alternative. In the first group, 43% selected “no religion”, 55% a religion, and 2% didn’t answer. With the second option, 43% said they had a religion, 54% said they did not, and 4% didn’t answer.
Demographic differences
Religious affiliation was lower among those aged 18-34 (34%) and higher among university-educated respondents (50%) in the second group. The poll surveyed adults, while the census includes children, with an adult filling out for younger children.
After the 2021 census, the ABS considered changing the question, noting feedback that the wording “assumes you have a religion”, but said changes might affect comparability with previous censuses. Dove said campaign members participated in two rounds of consultation with the ABS, which was responsive, but ultimately rejected the change. “I think they have been compromised by the lobbying from the religious groups [who] have the strongest vested interests,” he said. “It needs to be fixed.”
ABS response
The ABS stated it consulted religious and secular organisations and found it “not possible to design a question(s) that will meet the range of needs identified” while supporting data comparability. It has provided extra instructions, updated examples, reordered categories to reflect the most common religious groups, and changed data processing to record the most granular detail. Dove said the “ship’s sailed” for the 2026 census, adding, “We’ve already got our eyes on 2031.”
