Search Toggle

Irresponsible reporting of public opinion polls

By RUSSELL AYRES

OPINION polls are often reported with attention seeking headlines using terms like ‘dramatic slides’, but do they present an accurate picture?

The Australian often uses Newspoll to generate news cycles stating that popular opinion has recently shifted dramatically in regards to the large political players.

In fact the Australian has a section on its website entirely devoted to news stories arising from opinion polls.

On March 13 it reported that Julia Gillard overtook Tony Abbott as preferred Prime Minister by a margin of 39 to 37 per cent.

The Australian is certainly not alone in this type of reporting, and in fact is generally known as a good reporter of opinion polls, but are examples such as the one above, fair and accurate reporting?

The Newspoll section on the Australian states that the error rate for its polls is around 3 per cent.

This means that reporting margins such as 39 to 37 per cent to the Prime Minister may be very inaccurate as the real figure may be 42 to 34 per cent in favour of the Prime Minister or in fact 40 to 36 per cent in favour of opposition leader Tony Abbott.

The 3 per cent error rate is an incredibly important figure that is often glossed over in a bid to write headline grabbing articles.

Reporting swings in popular opinion of 2 or 3 per cent can be very misleading as in reality, such a swing may be due to the error rate in the poll itself.

Such small swings are constantly reported in the media.

Crikey reported on 13 March that Tony Abbott’s approval had gone up a point and his disapproval rating had gone down a point.

The Herald Sun reported on the same day that the coalition’s primary vote had fallen by two points and that Tony Abbott’s rating on the question of who would make a better PM had actually gone down by a point.

Experts on opinion polling agree that reporting such smalls ‘swings’ in opinion is often inaccurate and misleading.

Professor Ian McAllister of the Australian National University said, “To report changes of around one or two per cent . . . is normally well within the sampling error and therefore not particularly precise.”.

Within the journalism industry, McAllister said there are a few journalists expert enough to report opinion polls properly; however there are too many ‘run of the mill’ journalists that are reporting opinion polls incorrectly.

Professor Warwick Blood of the University of Canberra said that Australia often reports opinion polls incorrectly and potentially misleads the public.

“Most of the polling organisations do a great job in conducting their surveys . . . ,” he said. “The problem is that news values often get in the way of balanced news coverage.”

Blood said that most poll results often show ambivalent results, and in his opinion, “ambivalence is not the stuff of headlines”, which can lead to unbalanced news coverage.

The question must then be explored in more depth. Why would news organisations so often report opinion polls so inaccurately?

According to McAllister there is potentially three effects on the public once an opinion poll is reported.

He said, “Basically there are three theories. One is the underdog effect where the public feels they need to give the reported underdog support, the second is the bandwagon effect, where the public feels a party that is doing well in the polls, so the public supports that party even further, and the third is the projection effect, where people project the image of the party they support”.

McAllister said that research points towards the bandwagon effect in Australia.

“People are more likely to support a party that is leading in the polls, so generally opinion polls are likely to shape public opinion,” he said.

This may potentially lead to serious problems in regards to unbalanced media.

Reporting small changes in public opinion is irresponsible and often negligent and it’s an issue that the Australian public should take very seriously.

Recent Comments

1

Thought provoking article. I think we've seen sensationalist journalism on the rise for some time but criticism has been very generic, drawing out such a specific abuse by lazy journalists shines light on this type of behaviour. Thanks for the insights Russell and I hope a more educated reader base will punish this behaviour through sales.

Post Comment

Your email address will not be published. Required fields are marked *