AMSRO Polling Inquiry 2019
Hand raised with Australian flag

Inquiry into the Performance of the Opinion Polls at the 2019 Australian Federal Election final report released

Industry Code of Conduct for election polling among 10 key recommendations

Final Report Executive summary

The Association of Market and Social Research Organisations (AMSRO) today announced that the Inquiry into the Performance of the Opinion Polls at the 2019 Australian Federal Election (AMSRO Polling Inquiry) has released its final report, including recommendations for how to improve political polling in Australia.

The report follows AMSRO’s decision, along with the Statistical Society of Australia, to instigate an independent and impartial review of election polling in Australia to determine why all the published polls incorrectly called the outcome at 2019’s May Federal Election and how polling might be improved in the future.

As evidenced by the Inquiry, from 2007 until 2016, Australian pollsters had, in aggregate, a 96% success rate (that is, 25 out of 26 final week polls called the right result) across four Federal elections. The 2019 May election, however, was a notable anomaly and AMSRO and the Statistical Society of Australia, on behalf of the wider industry, wanted to know what went wrong.  AMSRO invited pollsters, media organisations and others who commission election and political polling to contribute to the inquiry.

The independent inquiry report concluded that the collective performance of the published national election polls in 2019 constituted a ‘polling failure’ rather than simply a ‘polling miss’, as the polls erred in their estimate of the vote in a manner that was statistically significant and erred in the same direction and at a similar level; and found that the source of errors lay in the polls themselves and was not a result of a last-minute shift in preferences among voters.

The AMSRO Polling Inquiry Chair, Darren Pennay (Founder and former CEO of The Social Research Centre) said: “The first preference voteswere either underestimated (LNP) or overestimated (ALP) because of inadequately adjusted, unrepresentative samples. It is very likely the polls were skewed towards over-representing more politically engaged and better educated voters and this bias was not corrected. As a result, the polls over-represented Labor voters. 

“As a consequence, pollsters should seek to better understand the biases in their samples and to develop more effective sample balancing and/or weighting strategies to improve representativeness, by looking at education or other variables.”

The Inquiry Panel also found some evidence that the reporting of the polls failed to consistently meet the basic disclosure guidelines for editors and journalists set out by the Australian Press Council.

The Inquiry report has made 10 key recommendations aimed at improving polling methods and their disclosure. Recommendations include calling for an enforceable Code of Conduct for Election Polling; consistent, enforceable disclosure standards; and recommendations to improve polling methods and better educate the media and public on polling.

Continue reading here

Final Report Executive summary Recommendations