Rumours flew yesterday, here in Crikey and also on Bryan Palmer’s Oz Politics, that the Galaxy poll taken in Queensland on the weekend (published in News Ltd tabloids) was a little dodgy.
The charge was that Galaxy had first asked respondents questions that encouraged negative feelings about the prospect of a Rudd Prime Ministership, and then asked how they would vote. The cited evidence was friends of friends who had been phoned by the pollster.
There are two issues here. One is the ordering of questions. It is an absolute no-no for pollsters to ask for voting intention anywhere other than at the beginning of a political poll. This is so they remain “unpolluted”. Any pollster caught transgressing thus would go straight to market research purgatory.
Get Crikey FREE to your inbox every weekday morning with the Crikey Worm.
The good news is that Galaxy always does ask for voting intentions first, and did on this occasion. The confusion may have come from the fact that the survey asked about both state and federal politics.
The Galaxy questions dealt with, in this order:
- Federal voting
- Federal issues
- State voting
- State issues
And so respondents were asked about federal issues before state voting intentions.
The other matter was the questions themselves, which were rightly excoriated in Crikey yesterday by Irving Saulwick and Denis Muller.
In their excellent recently published book on public opinion, Divided Nation: Indigenous Affairs and the Imagined Public, academics Murray Goot and Tim Rowse write that “[w]hile the press determines the issues that the polls are to cover, it is the polling organisations that craft the questions.”
Sadly, while that was probably once true, it ain’t necessarily so any more.
These days, market researchers often fall over themselves to conduct political opinion polls for newspapers. The PR, cascading around all corners of the media, is just too alluring.
Often the pollsters work at cost price or below.
And some, it seems, allow the newspapers to influence the wording of questions.
This is a tricky area. The papers can argue that they pay the bill and so call the shots. It is up to market researchers to stand firm and maintain the integrity of their product. This may be easier said than done with News Ltd editors – not a species known for conceding the other’s point of view.
But as Saulwick and Muller showed yesterday, others in the industry do notice these things. And they may not like it.