Pat PilcherJust how much credence should we give to recent political polls, asks PAT PILCHER.
It has long been the theory that media was the fourth estate. Aside from informing the public so they can better participate in a democracy, media has also been charged with holding the other three estates (the clergy, the nobility and commoners) to account.
When it works, it is a beautiful thing to behold. The trouble is that the fourth estate is also a very fragile entity. A case in point is the latest Colmar Brunton poll, released in early December, which showed the National party gaining public support, hitting 46% to overtake labour (at 43%) as New Zealand’s preferred political party.
Digging a little deeper, however, reveals a worrying state of affairs. It turns out that the poll had just 1008 respondents.
Putting this into context, a quick check with the electoral commission revealed that 2,605,854 people voted in the 2017 general election. Based on this, the respondents to Colmar Brunton’s poll represent an immeasurably tiny 0.0387% of eligible kiwi voters, yet the survey has been framed by the media as signalling a political sea-change when 99.9% of New Zealand voters did not participate in the poll, so their views were not represented.
Sample sizes are one thing, but when some of New Zealand’s largest media organisations – including TVNZ and Stuff – framed the poll as a political sea-change to their audiences, questions need to be asked.
With the poll presented by media as legitimately showing political change, then it is fair to assume that many viewers/listeners and readers are likely to adopt a noncritical, unquestioning view (after all, why should they question New Zealand’s largest media outlets? They’re legit, right?).
Is the media shaping political reality, or are they simply reporting on it?
Either way, the National party must be heaving a collective sigh of relief. It has been a nightmare few months for them, and things look set to continue to get worse as revelations of bullying within the party’s ranks continue to surface.
That a political party beset with poor polling and controversy is portrayed as surging in the preferred political party stakes should be setting off alarm bells in the minds of media consumers. Sadly, this doesn’t appear to be happening.
Perhaps now might be a good time to introduce media laws governing how polls are reported. It needs to be be mandatory that total poll respondents be disclosed, as well as margins of error and how well a survey represents the views of the overall population. Equally significantly, in an age of rampant fake news and what appears to be an erosion of media ethics and standards, people need to be taught to question media and adopt a critical view towards news reporting.
