January 23, 2004

A tale of two polls

We finally have proper analysis of our own poll. Newspoll has also obliged by coughing up their latest poll at approximately the same time. At this stage I should put in the standard disclaimer – our poll does not pretend to give quantitative results. If it did I wouldn’t be drawing attention to Newspoll, let alone relying on it. However it is intriguing that Newspoll has Peter Beattie on 57.5% of the vote after preferences, while our poll has him on 59.7%. As Newspoll keeps accurate records over time, they can tell us that this represents a swing of 2.5% against Beattie since the last election. We, on the other hand, don’t even know whether we got our result by accident or not, let alone whether it means any sort of move one way or the other.

But we can do something much more important than they can – we can make predictions about what is likely to happen in the future. In the pollster’s version of the uncertainty principle they can only tell you very precisely what happened between the 16th and the 20th of January (the dates when they surveyed respondents). It is here that the coincidence between their figures and our analysis is significant. While qualitative polling can be very enlightening, to be certain that it is right you need to confirm it with quantitative polling, and the Newspoll does tend to confirm our analysis.

What it shows is that there is very little movement in primary support for Peter Beattie – he retains an extraordinarily high vote. It also shows that One Nation (down from 8.7% to 2%) has virtually vanished, but that the Greens (2.5% up to 4%) have increased. Others (11.4% down to 8%) have also decreased. There is quite a divide between city and country. In Brisbane, Beattie’s primary vote is 58% while outside Brisbane it is only 44%. This 44% is actually only a small 2% improvement on his result last time which is dwarfed by the surge in the Coalition vote from 31.2 to 41.

I cannot tell from the report what the two party preferred vote is dissected by region but I have a suspicion that the swing against Beattie is stronger outside the capital than inside. Certainly the Coalition vote has grown most strongly in the regions, but in a sense this is a transference of a vote they already had with the increase in the Coalition vote almost the same as the decrease in the One Nation vote. Not quite the same - 2% less in fact. As Beattie’s vote also increased by a couple of percent and we know that approximately a third of One Nation votes preferenced Labor last election he would also have benefitted from some of the collapse of One Nation.

Whatever the case, this is very much a status quo result. Reference to the David Fraser pendulum shows that 10 seats are vulnerable with a 2.5% swing. However, one has to be careful because 3 of these are held by Independents or One Nation. For technical reasons to do with consistency, the pendulum shows these on the Labor side of the pendulum, but they are not held by the Labor party. One of them, Nicklin , is not a likely Coalition win, Gympie and Tablelands will also both be difficult, leading to the conclusion that this will lead to the Coalition picking up maybe 7 seats, of which the Liberal party will only win 2. Beattie would still have a huge 29 seat majority.

This is a result which confirms what we are finding through our qualitative research. Beattie has a good favourable/unfavourable rating (48% to 32%), but there is a large degree of cynicism about what he says and promises, and concern about his team. One participant commented that his concerns about Beattie were “Media Tart, great spin doctor, [s]moke and mirrors, not trustworthy…” There is little evidence of widespread community concern about specific issues such as the Families Department or the Ambulance Levy. Most voters nominate Health as the most important issue, followed by the Environment, Economy, Accountability and Education. However, concerns about the style of government collectively rate more highly than any other issue than Health.

This does not mean voters are moving to embrace the Opposition. Even Coalition voters do not think they are ready for office, believe that Springborg is still learning, don’t know what the Opposition stands for, have doubts about relations between the Liberal and National Parties (as well as Liberal Party infighting), are concerned about possible links with One Nation and still harbour bad memories from as long ago as the Bjelke-Petersen era.

As a result the protest vote appears to be going to the Greens, with around half Greens voters being former ALP voters. This protest vote does not appear to be particularly large at this stage and this is reflected in the size of the Greens vote. After their performance in recent by-elections and the NSW election they would have been hoping for a better result. 4% is unlikely to see them win any seats, even the seats of Mt Coot-tha, South Brisbane, Sandgate and Townsville where they think they will do well.

One note of caution. These figures are not truly representative of Queensland. As a result of the Newspoll figures, and also our knowledge of issues specific to particular areas, we will be paying attention to rural and regional voters in the next couple of weeks to try and determine whether there is a protest vote there, and whether it is returning to the National Party from One Nation, where it rested for a while.

It is in Peter Beattie’s interest to do less well outside Brisbane. An election result which delivered say 10 seats to the Nationals, including some taken from Independents, and only 2 or 3 to the Liberals would continue to institutionalize the friction between the Coalition partners.

Posted by Graham at January 23, 2004 03:02 PM
Graham Young
John Black
Mark Bahnisch
Michael Lee