A new national poll finds:
- “A clear majority of Australian electors oppose the Gillard Government’s plan to introduce a carbon tax, 37% support the proposed carbon tax and 10% can’t say.”
- “A majority (64%) believes that Australia’s proposed carbon tax will make no difference to the world’s climate.”
- Citizens are ill-informed; indeed they are “rationally ignorant” because, being just one of millions, any individual’s opinion is likely to have so little effect that it makes no sense to put in the effort of becoming well-informed.
- “Opinions” reported in polls are frequently not genuine opinions at all; when people are forced to answer a question on a topic they know little or nothing about, they “choose an option, virtually at random.”
- When people do try to form an opinion on a topic, they tend to talk mostly with people just like themselves, thereby, frequently, just reinforcing their ill-informed and prejudiced views.
- Mass public opinion is vulnerable to manipulation.
So when you get stupid answers like the ones delivered in the poll, its because you’ve asked stupid questions. Or rather, you’ve asked questions stupidly. There’s nothing intrinsically stupid about a question like “Do you support the Gillard government’s plan to introduce a carbon tax?” Rather, what’s stupid is the asking. It is the whole practice of opinion polling as a mechanism for identifying the public’s viewpoint on important matters.
There must be better ways.
Fishkin has devoted much of his career to developing and promoting an alternative: deliberative polling.
We’re working on another.
Update, 5 July:
For an illuminating discussion see Australians and climate change – beliefs about public belief may be quite wrong and Polls, framings and public understandings: climate change and opinion polls by Joseph Reser.
Tim van Gelder 
I remember reading somewhere that thinking pools like this would yield meaningful results was like thinking that adding a sufficiently large number of zeros together would yield a finite sum.
I meant *polls* not *pools*. What can I say, it’s late on Friday arvo…
What do you expect from polls?
I expect to merely find out what people think about something. Do you expect to figure out whether something is right or wrong?
What people think about these issues is obviously very important in determining whether it gets done because in a representative government people like the government doing what they want it to do and will punish it otherwise.
So it is obviously not a stupid question because it tells us something about the future composition of the government and/or the likelihood of a law being made.
It would be a dumb idea to poll people to find out the truth on complicated issues. It would also be a dumb idea to try to figure out the truth on issues by seeing what a bunch of people given minimal education on an issue would think.
People are going to vote the way they are going to vote no matter how they would vote had they been more educated. Politicians only care about how people are actually going to vote.
I don’t see the relevance of any other system of polling people minimally educated on an issue. Better just to poll experts if you are trying to get to the truth.
And what if deliberative polling gave you the same ‘stupid’ answers as garden-variety polling?
Deliberative polling tells you what people would think of an issue if they had a particular experience with regard to the issue.
Garden-variety polling tells you what people think now. Given that votes are likely to be cast on the basis of what people think now, rather than what they would think if they were all as fully informed as Fishkin would like them to be, it is much more useful to our political representatives in that it allows them to gauge where people are with respect to an issue.
Just quickly:
1. There is no way that the answer to the first question can be stupid. People either support the proposal or they don’t, and they might have good reasons for doing so in either case, depending on their personal circumstances. Without knowing those circumstances, you can’t say whether their decision is stupid or not.
2. In its proposed form, Australia’s carbon tax will make no difference to the world’s climate – not directly anyhow. So the majority have actually given the correct answer. Che miracolo! Maybe people aren’t the gullible meat puppets you and Fishkin think they are. Your not liking the answer that most respondents have given makes neither them nor the answer ‘stupid’.
Of course, you may disagree with me on the answer, in which case I invite you to venture where the angels Gillard and Combet fear to tread, and tell us how much of a difference the tax would make to the world’s climate, over the next 5, 10 and 100 years.
And I’d like to meet a few of the 36%, such as yourself, who think the tax would make a difference to the world’s climate – I have a lovely big bridge here in Sydney that I’d like to sell …
In both cases, garden-variety polling has worked well, doing exactly what was expected of it.
Hi Jeremy,
stupid is not the same as wrong/false. Different concepts.
cheers
Tim
Yes, I know. That’s (part of) what I was telling you.
I think I’m with Evan.
1. Every one of the four “problems” with polling are real problems with democracy.
2. If polling is *for* making good decisions, it’s pretty stupid. But if deliberative polling was *for* predicting what a democracy would vote for at an election, *it* would be pretty stupid. What if neither of them was stupid? When you look out at the world and see millions of dollars and thousands of hours being spent on something, and the something seems stupid… the something might not be *for* what you think it’s for.
3. Politicians and pollsters might be more interested in getting elected than in making “good” decisions… and normal polling is almost certainly better at predicting election results than deliberative polling.
Given that you and Fishkin have identified some pretty major problems with democracy… are you going to change your mind about how we should organise government?