ABS now testing proposed filter for religion question
The Australian Bureau of Statistics (ABS) has confirmed it is testing the proposed changes to the religion question.
Earlier this month, the ABS reported on the outcomes of Phase 2 of the 2026 Census topic consultation, and confirmed the status of topics identified for change during the first phase of public consultation.
Having proposed the introduction of a filter question to ascertain whether a respondent has a religion before asking about their religion, the ABS’ statement confirmed that work on the possible change to “support more accurate data collection” was “progressing”.
“The ABS is testing the change of question wording to ‘do you have a religion’ and using a free text field for responses,” it said.
Michael Dove, spokesperson for the Census21 – Not Religious? Campaign, welcomed the news.
“This is a positive sign from the ABS that it has taken on board the public’s concerns regarding the biased nature of the religion question and the desire for more acceptable evidence-base to support policy and funding decisions related to religion,” he said.
In its update, the ABS said the religion question was the most common topic addressed in feedback as part of the phase 2 consultation. In total, 101 submissions addressed the religion question, while the next commonly cited topic was ancestry and ethnic identity (with 67 submissions).
In an analysis of submissions to the Phase 2 Consultation, Mr Dove observed: “Of the 74 published submissions that specifically referenced the wording of the question on religion, 63 were fully supportive of the proposed change with a further 8 being partly supportive. This demonstrates the weight of community feeling on the matter and the campaign is very grateful to those who expressed their views.”
In a submission to the ABS in August, the Census21 – Not Religious? said the proposal to introduce a filter question would have a number of benefits, including removing some inherent bias in the religion question and ensuring more accurate and meaningful data.
The campaign also urged the ABS to discontinue with the use of its conceptually confused concept of religious ‘Affiliation’, and to discontinue placing the religion question following questions about cultural background.
The ABS will now continue refining a shortlist of potential content changes during the development and testing stages. It will then make a recommendation in regards to 2026 Census topics to the Albanese government in mid-2024.
The ABS will publish the final list of topics on its website in late 2025.
Media inquiries: [email protected]