How the pollsters got it right on the Voice

Welcome to Tuesdata, our weekly analysis for Unmade’s paying members.

Below, we examine how the factors that led the Yes campaign to falter in the months leading up to its defeat in the Voice to Parliament referendum.

Further down, the Unmade Index slides for a second day.

A poor campaign strategy made Voice defeat inevitable, but the pollsters got the trend consistently correct

Seja Al Zaidi writes:

On Saturday, a majority of Australians rejected the Albanese government’s proposal for an Indigenous and Torres Strait Islander Voice to Parliament. The debate consumed headlines for the majority of the year, fuelled by a lack of bipartisan support with the Liberal Party campaigning against the proposed constitutional change.

The numbers reveal a story of Australians increasingly sceptical about the Yes camp’s inability to articulate the tangible benefits of systemic reform, while voters were primarily consumed by ‘bread and butter’ economic concerns.

However, the pollsters – often commissioned by news organisations – will be able to point out that they correctly forecast the outcome. Unlike Scott Morrison’s shock Coalition victory in 2019, the polling got the result correct.

The Guardian Australia compiled a data tracker following predictions from 11 different polling companies over the last 12 months. All of the polling trends showed the result trending towards the eventual 59% no vote and 41% yes vote.

The chart shows a clear decline from April – when the Liberal Party announced its opposition – and a further fall in support in June and July. In the first week of June, polling company Resolve Strategic recorded a 51% ‘no’ and 49% ‘yes’ result from a sample of 1606 respondents. It was the first poll since the announcement of a constitutional Voice that saw the ‘no’ vote higher than ‘yes’.

Public opinion and data company YouGov estimated in its final Voice poll, recorded between October 6th and October 10th 2023 with a national politically representative sample of 1,519 voters, that 56% of Australians intended to vote no, 38% intended to vote yes, and 6% were still undecided. Yes was found to narrowly take the lead in inner metropolitan electorates by (48% vs 45%), but no had clear leads in outer metropolitan electorates (56% vs 34%), provincial city electorates (62% vs 35%) and rural electorates (65% vs 33%).

Indeed, in the final polling, the two electorates with the highest proportion of Yes votes were the Division of Melbourne electorate (which comprises inner city suburbs like Fitzroy, Carlton, Brunswick, Cremorne and Docklands) and Sydney’s Grayndler, Anthony Albanese’s seat which comprises of suburbs including Balmain, Lilyfield, Enmore, Newtown, Petersham and Marrickville.

“The people most concerned about cost of living were most likely to vote No,” said Paul Smith​, YouGov’s director of government and social.

“In NSW, only six of 26 Labor held seats returned a yes vote, and Labor holds most of the suburban and provincial town seats. People saw the questions differently. People who voted ‘no’ saw the question wasn’t addressing their issue, which is living standards, jobs, housing. Their view is very much that these issues are being led by Anthony Albanese.”

According to Roy Morgan Research’s Julian McCrann: “If you look at the voting patterns, you see that the inner city electorates were the most likely to vote yes. The closer that you are to the centre of the city, the more likely you are to vote yes, and the Green electorates, the Teal electorates, and some of those inner city electorates voted yes.”

“When we ask people what’s the biggest issue, it’s always cost of living, rising prices, inflation, higher interest rates, high grocery prices, they’re are always at the forefront of people’s minds, so they always sort of focus on the referendum. A lot of these people, out of suburban areas in particular, they’re wondering why the government’s focusing on this referendum issue when, hang on, the cost of living is a real problem. ‘Why aren’t you looking at ways to reduce prices and stop inflation? That’s what’s more important to me’.”

A Roy Morgan survey conducted with an Australia-wide cross-section of 1,419 Australian electors aged 18+ from October 2 to October 12 estimated that the ‘no’ vote was leading in five states while the ‘yes’ vote would lead only in Victoria. It also estimated that Australians in country regions had a 63% likelihood of voting no, whereas Australians in capital cities were likelier to vote yes.

“Younger groups were more likely to say yes, and while age is definitely not the biggest predictor, the biggest predictor without a doubt was university education. So those that have been to university, have the experience in life, and then have gone on to professional occupations, living in the capital cities, are much, much more likely to vote yes. That’s the most significant predictor,” says Freshwater Strategy’s polling director Dr Michael Turner.

Smith also emphasises that the failure of the Yes campaign could be at least partly chalked up to the campaign’s failure to offer a clearly communicated, cut-through solution to an existing problem.

“What problem are you solving? You have to identify a problem, and the best solution which people see as the clincher. The yes vote didn’t do either of those things effectively. The campaign didn’t cut through as it should’ve because it didn’t have those two elements,” Smith says.

“It’s not your supporters the message is for, it’s for the people who will make the difference between winning and losing. You have to have something that cuts through to them. You need a majority and a majority in four states to win so you really need a compelling message, one that’s been researched and properly looked at.”

The failure of the yes camp to communicate their position effectively was one component of their strategic failure – ‘hubris’ and a disconnect from the majority was the other.

“I do think that there was a sense of hubris and overconfidence. I saw this happening in the States when Trump was elected, and there was almost this sense amongst the left that he was unelectable, that no one would be morally corrupt enough to elect Trump as their president,” says Gual Barwell, chief strategy officer at creative agency Innocean.

“There was a lack of empathy on the left. A lot of the arguments they presented were around this moral need for change, without actually addressing any of the objections that people in the No camp had or all the misinformation that the No camp was disseminating.”

“I think you saw that Trump playbook of staying the same is safer than change over and over again. You saw that being mobilised really really powerfully with the ‘no’ camp – ‘if you don’t know, vote no’.”

Barwell noted that the no camp leveraged Australian’s minimal knowledge of what a constitutional change would confer and used that to fuel a campaign led by inertia and fear of change.

“Preying on the fear of change would be the other really strong aspect of the No campaign. I think that they leveraged that really consistently, really powerfully. They actually leveraged minority Indigenous voices so it looked like there was more indigenous support or at least conjecture around what this constitutional change was and the impact it would have,” he added.

“They put indigenous peoples and people of minority at the forefront of their campaigns, which I thought was a powerful, yet manipulative tactic nonetheless, knowing that 80 percent of Indigenous people agreed with the need for a voice in Parliament.”

“If you’re saying to someone well, it’s just right to vote Yes, and you should do so from a kind of moral compass perspective and there is almost no room for objections, it’s a little obnoxious really. I think this was very much like the Trump elections. This was cities speaking to rural. This was urban people, the wealthy, educated, almost talking down to those in the suburbs and the country.”

The choice of representatives and public-facing campaigners on each side also played a significant role in determining the outcome of the referendum. Northern Territory Senator Jacinta Nampijinpa Price and businessman Nyunggai Warren Mundine were at the forefront of the no campaign, rising to national prominence by writing near-daily opinion editorials, making television appearances and travelling across the country with ideologically-aligned think tanks to spread the message of the ‘no’ camp.

“The lack of Indigenous voices, figureheads within the Yes campaign had an impact. If you look at the role Albanese played, the contrast between Albanese’s role in the Yes campaign compared to something like Warren Mundine where Peter Dutton was very much in the background and let other indigenous voices lead the charge when it came to the campaigning around around no,” Barwell added.

“There was a bifurcation of messaging from the yes camp, where there seemed to be high degrees of consistency from the no camp. I think if I was a cynic, I would say that many of the campaigns that were created for Yes satisfied a role of creative ego rather than genuine impact,”

“The no camp really did just take the Brexit and Trump rule book; they were consistent and ultimately pretty manipulative of people’s underlying fears around things like land rights, Indigenous rights in general and really stoking an environment of disunity, but using that ironically as a successful campaign platform.”

“This was 100 percent our Trump moment, and we failed miserably,” adds Barwell.

Bipartisan support the missing key to referendum success

In the run up to the Voice referendum, the final Freshwater Strategy Poll for the Australian Financial Review showed support for a Yes vote at just 33 percent, a fall of 15 percentage points since May, while the No vote was 50 per cent support and 17 per cent were undecided. When the undecideds were excluded, the No vote was just over 60 percent and the Yes vote just below 40 per cent.

“Our latest AFR poll showed that the 15 percent at the time that we picked up of direct switches – so people who once said yes, but then now they’re in favour of no – a clear majority of them said that the reasons why they switched were they felt that there were other priorities. That was the top reason,” Dr Turner says.

“The second reason was that as they found out more details, they just didn’t like the proposition as much. The third reason was that they felt the campaign on the yes side was ineffectual. It was less than persuasive.”

Negative start to the week on the Unmade Index

It was a poor day for most of the ASX-listed media and marketing companies on the Unmade Index yesterday. The overall Index dropped 1.27% to 6.117 points. That was a worse performance than the wider ASX All Ordinaries, which lost 0.4% yesterday.

IVE Group had the biggest drop, falling 2.46%. Domain fell 2.34%, while Seven West Media lost 1.64%. Southern Cross Austereo saw a 1.34% decline, Ooh Media 1.05% and Nine 1%.

The smaller stocks had a slightly better day. Pureprofile lifted 3.70%, while Craig-Hutchison owned Sports Entertainment Group rose 2.56%. Enero Group rose 0.60%.

That’s all for today. Thanks, as ever, supporting us through your membership.

We’ll be back with more tomorrow.

letters@unmade.media