WARNING: POLLING ERRORS AHEAD: Make up your own mind!

You can’t believe any media biased, distorted and partisan distortions of public opinion as voters head towards critical local, state and national elections this year.

Although there has been some tinkering around the edges of sample procedures, review of mobile phone ownership, consideration of rear view vision weighting schemes and increase in the frequency of polls, nothing has changed that would justify any belief in the mea culpas.

Nate Silver, Editor in Chief of 538, famous for its electoral mathmatics provides a rule “Almost all polling errors occur in the OPPOSITE DIRECTION of what the conventional wisdom expects”, reflecting the efforts of the media push polling to attempt to shift the vote in the direction favoured by the media owners.

Post election reviews conducted to establish “what we got wrong” around the world by the media pollsters, continue to suggest that they got it right but the electors, for any number of challenged reasons, changed their mind or lied to them.’

AMSRO (The Association of Market and Social Research Organisations) decided, along with the Statistical Society of Australia, to conduct a review of election polling in Australia to determine why all the published polls incorrectly called the outcome at last May’s Federal election and how methods can be improved in the future.

The AMSRO Polling Inquiry Panel, Chaired by Darren Pennay, the founder and immediate past-CEO of one of Australia’s leading social research organisations, the Social Research Centre (SRC), invited pollsters, media organisations and others who commission election and political polling to contribute to the inquiry.

Pennay reports that they still have not completed their search for the reasons for misleading the nation, saying -“There has been no consensus among the polling companies or anyone else regarding ‘what went wrong’ at the last Federal election and the reputation of opinion polling with the Australian public appears to be at a low ebb.”

AMSRO so far appears to have been unable to get the cooperation of many of the media pollsters, unable to explain the consistency of the failure to identify sampling and survey methods and reserved judgement on introducing any standards that might restore confidence to the electoral polls.

It has released a statement of the bleeding obvious-

 “Election and political polling have an important place in Australian society, and it is imperative that the broader polling industry takes this opportunity to improve the accuracy and reporting of the polls because it underpins a modern, well-functioning democracy. 

 It’s also important for the credibility of the polling companies, as well as the wider market and social research industry, that the public has confidence in the results of the major polls.”

In Australia, we have the coming July 4 Eden Monaro byelection that various polls already report that both the Liberal and Labor parties have in the bag. The critical problem with that is the huge gap in both directions between their claimed two party outcomes..

Recent results from private polling companies associated with the union movement heavies and environmental lobbies tell us that the ALP has a clear two party lead in Eden Monaro byelection, (significantly neglecting to provide survey details recommended by AMSRO about which there is more comment below.)

Unpublished polls that have been pushed out by Fox and other conservative sources (equally under reported up to AMSRO standards) show that despite a surfeit of candidates, the Liberal candidate has a significant lead over her Labor alternative.

In the United Kingdom, before decisive victory of Boris Johnson, three elections in a row got the result significantly wrong with huge margins of error on both sides of the House. The polling industry breathed a sigh of relief in 2019 with all 3 parties predicted to within 2 percentage points. 

On average in the U.K., the final polling average missed the actual Conservative-Labour margin by about 4 percentage points-twice the average error in U.S. presidential elections.

In America, we have even the stalwart FOX poll reporting that Donald J Trump is way behind his opponent “Sloppy Joe” Biden at the same time as media analysts are beating out a “ too close to call” representation.

All — that’s right, all — of the polls cited in the venerable USA Real Clear Politics Poll (RCP) average have a bias of at least six points toward Democrats with Independents being underrepresented by at least two percentage points and always leaning more Democratic.

The problem with this projection is the number of polls in the Real Clear Politics average between 10/26/16 through 11/7/16 — 65 percent to be exact (194 for Clinton out of 297 total polls) had Clinton reinforcing the “group-think” that Hillary Clinton would certainly become the 45th President of the United States.

Dr. John Tantillo, a marketing and branding expert, known as The Marketing Doctor, applies his doctoral skills in applied research psychology to analyse the issues and personalities of the day utilizing his marketing and branding lens

Tantillo points out that polls are accurate in the long run, but are not perfect regarding outcomes no matter how much we want them to be true. Within the industry, it’s referred to as predictive validity. And why polls are probably not usable in forecasting results especially when they are weeks away from Election Day. His analysis of the systemic failures of the media polls to provide anything more than their opinion of the mob that they would like to win proposes three groups of commenators – the Apologists, the Metricists, and the Skepticists.

 “Apologists,” those who interpret polls with either an intentional or unintentional bias, are another important group within the industry. These folks include campaign managers, candidate/party spokespersons, or opinion columnists. They usually have an agenda and must be viewed with a high degree of skepticism.

“Metricists” are operationally defined as professionals who are the methodologists within the polling industry (and not a person who writes in or analyses poetic metre). We see them hard at work making sure that every poll they construct is reliable. They understand that polls can do little to predict a future outcome.

“Skepticists” are the people who consider the poll’s methodology, the organization conducting the poll, and the different parties citing the poll before coming to an informed conclusion notionally based on objective thinking rather than subjective feelings in their poll analytics.

Which brings us back to the studied failure of The Association of Market and Social Research Organisations (AMSRO) interim report attempting to correct the obvious failure of NewsPoll and other media owned polls to indicate that PM Morrison was going to pull off “a miracle” reelection against all of the dozens of surveys leading Shorten to believe he would fulfil his life’s dreams.

 (AMSRO) recently released an Interim Discussion Paper recommending the implementation of a comprehensive regime of disclosure standards for election polling Inquiry into the Performance of the Opinion Polls at the 2019 Australian Federal Election.

AMSRO Board member, Craig Young, says: “AMSRO welcomes this Discussion Paper from the Inquiry Panel and looks forward to the full report being released in October. We believe greater transparency and disclosure regarding how the published polls are conducted is critical to re-establishing public confidence in polling. 

“As the research industry association, AMSRO is also supportive of extending these disclosure standards to all published market research, not just polling. We look forward to working with the polling companies and other important stakeholders to discuss the creation and oversight of appropriate standards to bring about these much needed changes.”   

The Discussion Paper and further information on the AMSRO Polling Inquiry can be found at: https://www.amsro.com.au/amsro-polling-inquiry/.

• The existing codes of conduct/guidelines that apply to election and other political polling in Australia;
• International guidelines for election and other political polling, as well as examples of disclosure standards in the USA, UK, Canada and New Zealand;
• Implications for the development of more comprehensive standards for election and other political polling in Australia. AMSRO Interim Report states that

“An important first step in establishing the remit of this Inquiry was to settle on a working definition of the types of polling activity covered by the term ‘published opinion polls’ as set out in the Inquiry’s Terms of Reference. This comprised the following:·

Voting intention poll: This refers to pre-election polls or surveys which aim to gauge how people intend to vote at any one time or in a particular election.·Policy issues poll: This refers to polling or surveying undertaken to assess people’s views on issues that might relate to social policy or politics, such as views on same-sex marriage or the government’s response to the bushfires crisis or the Corona viruspandemic, but which do not involve estimating voting intention.

 ·Private poll: This refers to the polling or surveying undertaken by political parties, individuals, or private and public companies, where the results are only selectively released to the public.·Exit poll: This is a poll conducted of voters as they leave the polling booth.·Informal poll (sometimes called a ‘snap poll’ or ‘straw poll’): This refers to a poll which has been conducted without using robust sampling techniques and where the representativeness of the sample is questionable. An example of this would be a television station or newspaper running a limited poll of their own readers on an issue. There is nothing inherently wrong with this approach unless the poll findings are presented as being representative of the wider population.

·Social survey: This term refers to more comprehensive, longer-running exercises conducted by governments, independent research agencies, academicsand think tanks to measure social and policy issues (for example, the Household Income and Labour Dynamics in Australia [HILDA]).

We will have to wait until October for the release of the findings of this Inquiry into voting intention polls and policy issues polls. A strong argument exists that privately commissioned polls that end up in the public domain should be required to meet the same disclosure standards. Meanwhile, readers should make up their own minds and join the Skepticists rather than wait for the Apologists to explain why there has been no independent reconstruction of the Metricists polling processes.

?