News Limited Sunday papers are this morning reporting a 'monster' poll of 4,000 voters by Galaxy. Galaxy has polled 200 voters in four key marginals in each state. The following table summarises the result.
Galaxy Marginal Seat Poll - News Limited Sunday Papers 15 August 2010
State (Seats) |
2007 2PP |
2010 2PP |
Swing |
NSW (Eden-Monaro, Gilmore, Macarthur, Macquarie) |
50.9 |
48.5 |
2.4 to Coalition |
VIC (Corangamikte, Deakin, La Trobe, McEwen) |
50.4 |
52.0 |
1.6 to Labor |
QLD (Bowman, Dawson, Dickson, Flynn) |
51.4 |
46.0 |
5.4 to Coalition |
SA (Boothby, Grey, Kingston, Sturt) |
49.0 |
49.0 |
no swing |
WA (Cowan, Hasluck, Stirling, Swan) |
49.6 |
47.5 |
2.1 to Coalition |
On Saturday evening (14 August), this was being reported as a poll showing the Coalition ahead 51.4% to Labor 48.6%. See the screen grab below from Skynews.
(UPDATE: See comments at end of this post. The Sunday Telegraph itself hasn't given any prominence to these national figures.)
When I entered the swings into the ABC calculator, I got the following result
http://www.abc.net.au/elections/federal/2010/calculator/?swing=state&national=0.3&nsw=-2.4&vic=1.6&qld=-5.4&wa=-2.1&sa=0&tas=0&act=0&nt=0&retiringfactor=1
This produced a national 2PP figure of 51.1% for Labor, completely at odds with the figure of Coalition 51.4% reported above. Then it dawned on me that whoever calculated 51.4% has made a very basic error of political analysis.
The five 2PP figures for 2007 and 2010 reported above apply only to the four electorates in each state. They are not 2-party preferred figures for each state, and the Galaxy tables have been very precise in setting out the 4-seats per state nature of the poll.
What someone has done is take the five entries in the 2010 2PP column and average them to get a national figure. Wrong. Very wrong.
There are two serious errors commited here. First, the figures are for four electorates, NOT the states. Second, while the state samples are the same size, the state populations are not. To get a National 2PP figure, you need a weighted average of the state swings, NOT a simple average of the survey 2PPs by state.
The method I used in the calculator is the correct way to produce a national 2PP %. If you assume the swings in the four states are uniform in each state, the new state Labor 2PP %'s are NSW 51.3%, VIC 55.9%, QLD 45.0%, SA 52.4% and WA 44.6%.
These produce a National Labor 2PP % from this Galaxy poll of 51%, not 48.6%. Someone has made an absolute howler in trying to turn polls in 20 marginal seats into a national figure.
UPDATE 1: I was the victim earlier this morning of a headline that converted my post into a headline suggesting "The Galaxy poll is wrong, very wrong". I've had that corrected as it is not what I am suggesting. What was wrong was the headline national figure reported last night, and I see nothing to suggest Galaxy produced that figure.
As has been shown by state breakdowns of Nielsen and Newspoll, Galaxy suggests that Labor can lose the election in marginal seats in Queensland and NSW. The Galaxy poll is very much in line with what Nielsen and Newspoll have been reporting for three weeks.
However, the way the incorrect national figure was reported last night made this Galaxy poll look like some shock result. If it had been reported last night as a 1.7% swing against Labor costing it the key marginal seats it would have been a repeat message of the other polls. But instead it was reported with an incorrect Coalition 2-party result of 51.4%, which made it look like a shock departure from other polls.
UPDATE 2: Now I know about the print versions of the Sunday Telegraph, I know that the paper itself hasn't highlighted the national figure. It has treated the analysis properly as a seat by seat comparison. What appears to have happened is that the press release to several media outlets included a list of the seats and state swings and a total figure for these seats that was listed with caveats. That number has been taken out of context and turned into a television graphic attributing it as a national figure. As I thought, stuff-up rather than conspiracy.
I'm now going to close comments on this post as all that is coming now is endlessly repeated fanciful News Limited conspiracy theories and personal abuse directed at me. I have better things to do on a Sunday than try and moderate abuse.
Recent Comments
Im with Antony rather than Nathan. Nathan asked if any Westminster parliament without fixed terms has legislated for fixed terms? Indeed: Westminster is about to! Theres no reason to imagine that the GGs power in s 28 to dissolve the House ...
If alternate half Senate elections are triggered by general elections including the full HoRs would a new election for the HoRs spill those Senate seats not put to the people at the most recent election? (This is contrary to what you say above.) COM...
well done antony clear abd unbiaesd again. Care to run for president of aus...
Thanks Antony, great work Just a minor procedural question - how is the election of a speaker governed? Does the prior speaker call for the nominations/ votes, or the prior PM perhaps? Both of these obviously have inherent problems, given these peop...