What the election polls are – and aren’t – telling us
Political junkies and the media are obsessed with opinion polls on the relative standings of the political parties.
Movements within standard statistical margins of error are treated with great respect.
In between elections, it’s probably all irrelevant, but a significant turning point may have been reached in the forthcoming Australian election, which until very recently looked like the election Peter Dutton couldn’t lose – or at least come very close to a majority within a couple of seats.
Back in February, the first YouGov MRP 2025 election model showed the Coalition winning 73 seats, which would make Albanese’s Labor government the first one-term federal government in 94 years.
But now there has been a subtle – if not yet seismic – shift.
A Freshwater poll for the AFR conducted from March 13-15 with a 1051 sample size gave the Coalition a 51-49 lead – a one point gain since February – Labor steady on 31 per cent and the Greens and others up one point.
Insignificant, obviously in many ways, but the Coalition was down.
Albanese’s net approval was up one point (again not really statistically significant) but Dutton slipped down four points. This was Albanese’s best result since September 2024.
Over the past two months, Albanese is up eight points and Dutton down eight. Momentum, as some West Wing character would trumpet.
You might like
Analysis in The New Daily suggests there has been improvement for Labor across a range of polls – and that Labor has gained seven points on economic management and three points on cost of living since February.
All those Albanese cost-of-living announceables – normally ignored by most people other than political commentators – have obviously been heard this time.
The New Daily analysis also highlighted that Labor leads in three of the past national polls (two YouGovs and one Morgan) with the Coalition ahead in Newspoll and Freshwater.
Perhaps most significantly, Albanese has gained six points on net approval while Dutton has dropped. Albanese also led Dutton as a better PM by 45 to 39.
How preferences are allocated is, of course, a difficult thing to estimate.
In the last federal election, the chances of teal candidates being elected were dramatically underestimated because polls didn’t take into account tactical voting where Labor supporters voted for independents in seats they knew they couldn’t win.
There is also some doubt about other preferences. One estimate suggests there will be a pro-Coalition shift in One Nation preferences that would result in a 50-50 tie.
More significant would be the Green vote. But to put it mildly, the Greens are on the nose and people are sick of the party big-noting and undermining the possible by insisting on the impossible.
Indeed, it is possible that the only Green left in the Lower House after the election will be its leader.
Of course, we shouldn’t dismiss the polls even if they have some inherent weaknesses.
Answers to survey questions are always influenced by extraneous factors, from what side of the bed you got out of that morning to significant ones such as wondering whether you are about to lose your job or not.
This would also be a lot clearer if we had better access to the more in-depth focus group research parties undertake.
Stay informed, daily
Focus groups are invaluable, but they too have their problems. Focus groups are a product of the distinctions between qualitative and quantitative research that date back to social sciences developments in the US and in Australia.
In Australia, researchers such as University of Melbourne Professor AF ‘Foo” Davies, and Graham Little – a pioneering political psychologist who was also at UoM – interviewed people at depth for the Images of Australia project.
Hugh Mackay was also an Australian pioneer of such interviews, which he used to develop quite detailed pictures of Australian society.
But over decades, focus-group research has evolved from that approach and Judith Brett’s book, Doing Politics, has an important chapter (“Writing ordinary people’s politics”) that looks at the weaknesses of much political quantitative and qualitative research and outlines alternative approaches.
Graham Chant, former principal at ChantLink consultants and before that an academic psychologist, also published a paper (Qualitative Research: Issues & Techniques) in 2012, which described qualitative research methodological problems and recommended measures to address them.
But drawbacks or not, they do have their advantages.
Meanwhile, there is another important measure: trust and distrust.
The most reliable measure of this is the Roy Morgan Research into who – among politicians, companies and brands – is trusted and distrusted.
The one overwhelmingly striking finding in this research over the years is that Dutton always tops the most distrusted category. So, perhaps the slight signs of an Albanese revival are less about what he is doing and more about voters deciding to take a closer look at Dutton, which reminds them of why they don’t like or trust him?
The research also indicates another profound shift in Australian politics.
During the Menzies and later eras, women voted overwhelmingly for the Coalition parties. That has been steadily declining and has now reversed. It is unlikely Dutton will address that trend.
Yet another important indicator is the odds that bookmakers offer.
Obviously, many keen punters bet on anything. But among them are also a lot of people who have detailed knowledge of what polling and focus groups are telling them.
As of March 25, the Labor is the slight favourite with odds of $1.80 compared with $2 for the Coalition. A tight situation if it was a straightforward two-horse race.
So, trust the momentum? Wonder what happens next when people further consider what they dislike about Dutton? Will Albanese screw up the campaign? Will Dutton release some actual policies?
The odds on those outcomes are probably much tighter than the current odds on the election result.
Noel Turnbull has had a 50-year-plus career in public relations, politics, journalism and academia
This article first appeared in Pearls and Irritations. Read the original here
