The pollsters had a pretty bad election campaign as you can see from the table at the bottom of this subscriber contribution.
Being an independent media player, how about applying the blowtorch of hindsight to the election opinion polls?
1. The pollsters try to relate these figures to events during the election campaign. They seem to claim that they are measuring the ebbs and flows of public opinion. However, when you compare the results from these 3 pollsters, there are many times when the figures are very different and moving in different directions. How can this be if they are measuring real movements in opinion?
2. In the Newspoll and Morgan polls, the average figures over the campaign period are up to 3 or 4 percentage points away from the actual, final results. In fact, the actual results are outside the range of the week by week figures (in other words, the poll results are consistently too high or too low over the campaign period). Please explain.
3. The AC Nielsen average results over the campaign are mostly within a percentage point of the final results, and they show a fairly steady drift to the Coalition over the campaign, which has the ring of truth to it. However, the final poll on election day was wrong by about 1.5 to 2 percentage points. If that is the acceptable margin of error, the pollsters’ certainly don’t trumpet it loudly.
Get Crikey FREE to your inbox every weekday morning with the Crikey Worm.
4. Finally, the issue of the large number of undecided voters which the pollsters talk much about. I have always thought this only represents the punters who are annoyed about being rung up during dinner but are too polite to just hang up. We are told that there was an unusually large number of undecided voters this time, and yet the final result was very decisive, almost a landslide. That doesn’t make much sense to me.