14 August 2023

What should we make of the Voice referendum polling?

| Chris Johnson
Join the conversation
Peter Dutton speaking in Parliament

Opposition Leader Peter Dutton would prefer an alternative to an actual referendum on the Voice. Photo: Screenshot.

Opinion polls gauging public sentiment over the Indigenous Voice to Parliament referendum aren’t returning anything positive for the Yes campaign, but no one should be fooled into thinking the result is already in.

While the No campaign has been out and about for some time with its message of uncertainty and fear, the Yes side has one clear advantage.

The Yes side’s biggest cheerleader – Prime Minister Anthony Albanese – knows when the referendum is going to be held. The No campaign doesn’t.

That’s a hugely significant strategic advantage.

It’s the same as an incumbent PM having the luxury and benefit of deciding the exact date (within a legally specified timeframe) of a federal election.

To put it a little more bluntly – the Yes campaign is not calibrated to peak in August.

It is designed to peak much closer to when the vote will be held.

And that date only they know, giving them a tactical advantage.

By they, I mean the PM and his closest allies on the referendum.

READ ALSO The long history of conservatives and treaty fearmongering

They will play that card to give them the optimum opportunity to sell the case to the nation before the vote is held.

The Yes camp is plugging away now, but it won’t peak until the date is set and the referendum called.

The No campaign is peaking now.

And on the polling itself – when was the last time any current pollster had to fathom the public perception over a referendum?

Today’s pollsters haven’t had a referendum to poll about before.

They were all likely interns when the last referendum was held in 1999.

They are right now attempting to traverse uncharted territory.

At best, anyone polling on the referendum is playing a guessing game trying to analyse survey results.

Pollsters have been famously getting it wrong when predicting state and federal elections in this country for some time now.

They have had too many misses to inspire faith in their predictions.

And that’s purely with elections.

They have far less experience with referendums – like, none.

READ ALSO Official Rudd portrait unveiled, complete with unofficial prime ministerial cat

The other thing to be mindful of is who some of these ‘credible’ pollsters predicting the doom of the referendum were actually working or ‘interning’ for before they got into taking the political temperature of the voting public.

Biographies of the founders of some polling companies currently telling us that the No vote has it in the bag reveal a pedigree of working for right-wing consultancy firms, think tanks and MPs.

And the right-wing media are running with their analyses with jubilation.

It’s all too close to really believe it is truly independent polling and reporting.

The most accurate polling is what the political parties do internally for themselves.

The public doesn’t get to see that, but it is what actual political strategies are designed around.

Those figures, which only the hierarchy of the parties get to see, pull no punches.

They are not skewed to present a desired result.

With that in mind, it is worth asking one question.

If Opposition Leader Peter Dutton is so confident the No campaign is going to win, why are so many of his utterances on the Voice aimed at urging the government to find an alternative to an actual vote being held?

He doesn’t want the referendum, he wants some sort of legislative agreement to be reached.

Why? What internal polling has he seen?

Join the conversation

All Comments
  • All Comments
  • Website Comments
Maggot Smith8:26 pm 18 Aug 23

Whilst the mortgage belt burns Albo worries about The Voice instead

That is a false dichotomy. They are independent issues, although I don’t think he is doing well on either.

What an insightful article. The master tacticians of the Yes campaign have got this. Just wait out those awful no voting oiks.
By sharing their wisdom and virtue, our elites will lead them to the light.
Then they will smite the unbelievers with the date of the referendum.

HiddenDragon6:50 pm 15 Aug 23

“The No campaign is peaking now.”

Of course it is, because the mortgage reset awaiting many households, along with increasingly terrifying bills for big items like home insurance and electricity, all to be dealt with by real incomes which continue to fall (as illustrated by today’s ABS release) and with no assistance from the now abolished Low and Middle Income Tax Offset, will fill the hearts of voters with sweetness and light and they will be warmly receptive to the tidal wave of pro-Voice b/s which is coming their way.

As to what the private polling is saying, if the government’s private polling on the Voice is more favourable than the public polls, including the latest from Resolve (obviously another part of the vast right wing conspiracy……) –


I would not want to see how strident and desperate Albanese and other leading lights of the yes case would become if the private polling was less favourable.

This “YES vote yet to peak” line sounds like imagination to me. The YES vote peaked when there was very little known about the referendum question and little public discussion. Its backers, beginning with the PM, were big on emotion and comparisons with the 1967 vote that removed Indigenous Australians from the Constitution.

There has now been significant public discussion, the YES case is still running on emotion, the NO case is picking apart the idea, and Indigenous Australians are divided.

Sound a bit like the Republic?

In 1999 there was near total media support for a YES vote. It went down when the people got a say.

In 1999 there was also a question aimed at reinserting Indigenous Australians into the Constitution. That fared even worse, with no State or Territory in support and a 39% YES vote.

The reasons why this vote will also fail are that whether Albo releases the details or not, the NO team is going to run a scare campaign and that’s a vote-winner. Secondly, Australians have consistently shown that they want everyone treated as equals, not singled out for special treatment on the basis of race or ethnicity.

“In 1999 there was also a question aimed at reinserting Indigenous Australians into the Constitution. That fared even worse, with no State or Territory in support and a 39% YES vote.”

For the record, the question wasn’t “reinserting” Indigenous Australians into the Constitution, it was about including them in a reworded preamble. You tried to obfuscate the reality of why it failed in a previous post. Here’s my response to your previous attempt to massage the truth and imply it was a rejection of recognition of Indigenous Australians:

‘This 2001 paper (https://www.unswlawjournal.unsw.edu.au/wp-content/uploads/2017/09/24-2-17.pdf) published in the Law Journal of NSW, provides an expert (and honest) analysis of the reasons the 1999 “Preamble question” in the referendum failed.

One of the main characteristics of the 1999 ‘Preamble question’, which you ‘conveniently’ failed to mention, is that the preamble (written by John Howard, with assistance from others) contained a number of clauses, covering a number of matters, including a single clause “… to recognise Indigenous Australians in the Constitution”. The above article provides expansive explanation as to the reason the wording of the preamble was unacceptable to so many Australians.’

I’m sorry? A Constitution Amendment Bill – by definition – is about amending the Constitution.


Ummm OK … I’m sure you know what you are talking about because I have no idea.

When did I suggest anything about a Constitution Amendment Bill not being about amending the Constitution?

What I did was challenge your statement about ‘reinserting Indigenous Australians into the Constitution’ – which is clearly a furphy. Secondly I refuted your implication that it was a rejection of Indigenous Australians (fiction you had tried on previously), by giving you a link to a paper which actually analysed the reason the question failed.

You probably didn’t even bother reading it, because it doesn’t align at all with your biased narrative.

Jamie Michael2:47 pm 15 Aug 23

This wouldn’t have been written if the polls had shown Yes leading

With polls always ask what is the sample size and how representative of the general population is the polled sample. If the poll is of few people, who don’t represent the general population, then the poll is unreliable if not useless in determining the views of the general population.
Major polls are no longer conducted on the street. Random sampling of the population by phone is problematic for various reasons – fewer land lines, difficulty accessing representative groups by location, age, gender, ethnicity etc, public suspicion of phone scams….
Now this may surprise you, but many opinion polls are based on answers provided by people who applied to be polled. People join these poll panels to answer questions on various topice for rewards and prizes. Immediately there is a bias in the sample through self-selection. The polling companies weigh answers to reflect general population characteristics, so there is also a manipulation of data.
The media breathlessly report these opinion poll results, despite their dubious accuracy, small sample size and questionable participant selection.
When was the last time you were polled? Me, never.

What do I make of the polls? That there really is such a thing as a Canberra bubble. Canberran’s aren’t representative or the rest of the country and regardless of the vote’s outcome, we need to come to terms with that.

And people talk about the No side promoting conspiracies.

The same people crowing about the polls over many years are now decrying the same ones as unreliable. What a surprise.

Yes, I’m sure the Guardian which has a defined position promoting Yes and their Essential polling is under the control of the evil right wing No conglomerate.

The depressing thing about this article is that it is full of real relish that the “yes” side will manage to win, not by strength of argument, but by spin, manipulation, and raw (bordering on vicious) political partisanship. It’s both depressing that such a cynical attitude to democracy is prevalent (at least among “yes” voters and the Guardianista elite), and to think that this is how Voice advocates might well understand power if the Voice gets up.

The referendum will be held in mid October, it’s no secret, probably on the 14th October, and both campaigns would be aiming for that date.

The most credible recent poll, at least in terms of size, was the Labor Party poll of 14,300 in mid July that found respondents were yes-51 49-no. The latest Essential poll by the Guardian (definitely not a right-wing outfit) found respondents were yes-48 52-no, which had dropped from yes-52 48-no the previous month. The most recent poll is from Resolve (SMH and Age, again not a right-wing outfit) who found respondents were yes-46 54-no.

Support for the referendum is dropping like a stone.

The yes side complains about misinformation, and at the same time writes this loaded with misinformation.

Stephen Saunders10:25 am 15 Aug 23

“What should we make of Voice polling.” Here’s a tip – don’t dismiss it out of hand, as if “silly” voters will have to come round to sublime Albanese “truth”.

15-20% falloff, in a year? Losing now, in all but two states? Even woke SA? Those are bad numbers, Chris. Not quite looking like a polling or a rounding error to me.

Daily Digest

Want the best Canberra news delivered daily? Every day we package the most popular Riotact stories and send them straight to your inbox. Sign-up now for trusted local news that will never be behind a paywall.

By submitting your email address you are agreeing to Region Group's terms and conditions and privacy policy.