Media Release: Action needed to correct bias of Census question on Religion

With an operational cost of $670 million, the 2021 Census needs to be accurate. It won’t be. On August 10 this year, Australia will hold its 18th Census since 1911. For at least one question the data collected by the Australian Bureau of Statistics (ABS) will be knowingly corrupted, with tangible consequences.

Former senator Chris Schacht, an Ambassador for the National Secular Lobby (NSL), says that for decades secular groups in every state have made submissions to amend the flawed question on religion. The ABS has ignored them all.

"Census Question 23, on 'Religious Affiliation', is misleading. It inflates Australia's level of religious belief to be much higher than comparable western countries such as New Zealand, and those in the UK and Scandinavia," he said.

"Consequently, our current 'No Religion' tally is an artificially low 30 percent, compared to the 40 to 80 percent of our European cousins. Several factors contribute to this discrepancy, which has negative social implications."

"This is manifestly wrong, on several levels, allowing governments to gift billions in extra taxpayer dollars each year to fund questionable private religious businesses -- all based on invalid Census data," Mr Schacht said.

Mr Schacht references four specific errors in Question 23, on Religious Affiliation. "Collectively, they steer the public away from stating they have 'No Religion' and influence citizens to nominate their affiliated childhood religion -- which they may no longer believe or practise in any way," he said.

1. "Religious Affiliation" headlines Question 23 of the Census. By seeking 'affiliation', the ABS is in breach of the Census and Statistics Regulation 2016. Regulation 9 specifically requires ABS to ask Census respondents for their "religious denomination". It does not infer the collection of 'affiliation' with some past family or childhood religion.
2. The question on religion, "What is the person's religion?" is a leading question. By any measure, this is a leading question which assumes everybody has a religion and gives the impression that one is required to be recorded.
3. The ABS concedes a 'lapsed family religion' may be included in responses, rather than a firm and current non-belief. This was stated in a letter to NSL. It defeats the whole purpose of the religious question.
4. Listing Atheism as a "religion" goes against the High Court definition of religion (item 3.3), framed in 1983: "...for the purposes of the law, the criteria of religion are twofold: first, belief in a supernatural Being, Thing or Principle; and second, the acceptance of canons of conduct in order to give effect to that belief..."

"ABS has a duty of care to collect accurate and wholly valid statistical data through all the questions it asks in the five-yearly Census -- that includes question 23 on religion," said Mr Schacht.

The significance of these problems is supported by the recent Religiosity in Australia report, which found that:

"Adjusting the original 2016 census religious affiliation data by the proportion of Australians who identify as a member of their religious organisation (whether active or inactive), the 30% No Religion figure becomes almost half, 49%.

"Adjusting further for only those who are active in their religion, the No Religion figure becomes nearly three quarters (73%). Therefore, only a small minority (27%) of Australians are active in and specifically endorse their religious organisation..."

"It can well be argued that, for decades, results from the Religious Affiliation question have contributed to the rise of private religious schools in Australia -- initially driven during the era of Prime Minister Robert Menzies," said Mr Schacht.

"And they flourished under Prime Minister John Howard, who also introduced the program of Christian chaplains in public schools, breaking a one hundred year tradition that public education was “free, compulsory and secular."

"As the Census is a basis for government funding of education, health, and community services -- much of which is run by religious institutions -- it is of great concern that the question of Religious Affiliation continues to be structured in such a way as to directly and financially benefit those same religious organisations."

"The fundamental misrepresentation of the Census question on religion requires examination to determine the extent to which the ABS is in breach of the Census and Statistics Regulation and other relevant legislation," Mr Schacht said.

The NSL is one of a number of Australian pro-secular groups currently involved in the Census21 campaign, aiming to improve the accuracy of religion statistics in the Census.