top of page

Understanding voter behaviour at the Voice referendum: A first look

An Octopus Group | Accent Research survey project

In less than eight months, support for the Voice nearly halved. In late February, Newspoll reported that 60 per cent of Australians would vote Yes if the referendum was held then. A number of polls in late 2022 had support even higher, up around 65 per cent. By the night of October 14, it appears as though the case for a constitutionally enshrined Voice to Parliament had lost around a third of its support, declining by approximately 20 percentage points, with the count on the night indicating around 40 per cent voted Yes.

 

How did this happen?

 

This report provides a first look of what Australians were thinking as they voted. This research is the product of a survey run by Octopus Group and Accent Research of 1,204 Australian citizens aged 18 and older who were enrolled to vote in the final stretch of the campaign (between 06 and 11 October, 2023). As we ran the survey, approximately a third of respondents had already voted, so we catch public opinion about as close to the act of casting the ballot as is possible.

 

This is not intended to be the final word on why Australians voted the way they did, but a first look post-referendum at why they voted for and against the Voice. We find at least five main reasons for the outcome. These include:

  • It was the claim by opponents that the Voice would be divisive that appears to have been a major driver of opposition to the Voice, selected by 41 per cent of those voting No as the most important reason for their opposition.

  • Nearly half of voters (47 per cent) do not think that Aboriginal and Torres Strait Islanders face more discrimination than White Australians. Of these, only 20 per cent voted Yes for the Voice.

  • Many ‘No’ voters said they did not understand the proposal. While only a quarter of those who opposed the Voice said a lack of understanding was one of their main reasons for voting No, only a third of the voters said they understood the Voice, while 16 per cent said they did not understand it at all.

  • The No campaign attempted to tie the Voice to political and media elites. Whether or not that was successful, very few voters believe most politicians can be trusted (just 17 per cent), while journalists are more trusted, the absolute rate level is still low (26 per cent), and those with lower levels of trust were significantly less likely to support the Voice. 

  • Voters who relied on Sky News, the daily tabloid newspapers, the Australian, free to air television news and AM radio were less likely to say Aboriginal and Torres Strait Islanders faced discrimination, and (along with those who relied on Facebook and FM radio for news), were more likely to vote No. More than a quarter of voters report that each of free to air television news, Facebook, and the daily tabloid newspapers were important sources of news and information (58 per cent for free to air television).

A full copy of the report can be found below. 

For additional details on the methodology, full question wording and breakouts of all variables in the survey, see the Detailed Methodology Report.

voice-coverpage.png
bottom of page