Two of Australia’s major newspapers have ceased running opinion polls in the wake of the weekend’s shock election win by Prime Minister Scott Morrison’s government, which defied every major public survey.
The Sydney Morning Herald and Melbourne Age newspapers decided against renewing a contract with Ipsos, whose final poll published the day before Saturday’s election showed the opposition Labor party leading the center-right coalition 51 percent-49 percent. Newspoll, used by Rupert Murdoch’s News Corp., similarly consistently called a Labor win.
In an article in the Herald Friday, national editor Tory Maguire noted that polling companies “are the main reason” the election result “took voters, the media and many political operatives by surprise.” She further said the paper had a responsibility to put resources into journalism, noting its articles that best reflected the electorate’s mood were done by journalists out on the road.
Australia’s election shock followed similar surprises abroad, including the UK’s vote to leave the European Union and the US election victory of President Donald Trump. All three outcomes contradicted polling and prevailing wisdom and have given credence to Trump’s characterization of journalism he doesn’t like as “fake news.”
Yet while the surprise votes gave the victors ammunition to attack established wisdom -- particularly useful for populists -- there are potentially significant implications for the democratic process. Writing Thursday in the Australian Financial Review, John Utting, a veteran pollster for the Labor party, said it was worth considering whether there had been a 2-3 percent pro-Labor bias in polls over the past few years.
“The last 50 or 60 Newspolls showed Labor comfortably ahead,” he said. “What if, in reality, Labor was at best level pegging or even behind? How has this affected our political commentary and political behavior? Did polling create a parallel universe where all the activity of the past few years, especially the leadership coups and prime ministerial changes, were based on illusions, phantoms of public opinion that did not exist?”
The rise of social media, fewer landline telephones, cost cutting and the difficulty of constructing mobile phone samples had all made it more difficult to get a representative sample of the population, he said.
