Posts filed under Polls (132)

The Critic, the student magazine of Otago, has an interesting feature on the council elections, rating the candidates on six issues determined by polling to be important to students. One of the ways they present the candidate ratings is with radar plots

These show the rating for each of the issues on six radial axes, connected to form the white polygon.  Candidates who are more ‘student-friendly’ on these issues will end up with larger polygons, and the different shapes show that there are significant tradeoffs between, say, a candidate who is in favour of drinking and (quality) loud music, and one who is sound on environment and transport.

This is a pretty good use of radar plots. Their main limitations are that the ordering of the axes can have a big effect on the visual impression, and that evaluating tradeoffs quantitatively is hard. Neither is a really serious limitation here, and both are problems common to many ways of displaying multivariate data.

Here’s another radar chart, originally from INFOGRAPHIKA magazine, rescued from its unfair banishment to wtfviz.net.

This one shows how people rated the importance of eight factors in success, split up by their income.  It’s interesting to see how much higher  connections, initial capital, and cheating were rated as important by the poor, and how the rich thought hard work was the key factor, not being very impressed even by education.   It’s clear that each group likes the story that makes them look good; less clear who is more correct.  What’s a bit depressing is how small a role anyone thinks is played by luck.

Today is the 120th anniversary of women’s suffrage in New Zealand, with commemorations in a range of places, including the Centenary fountain in Khartoum Place, Auckland

The petition for women’s suffrage, signed by about 24000 women, was submitted to Parliament in July 1893

and the names of the petitioners have been digitised and made available at New Zealand History Online.

I haven’t been able to work out exactly what the adult female population of NZ was at the time, but the digital yearbook says that there were 305287 non-Maori females, that 30.94% were married and 4.11% widowed, and that there were 67000 never-married females 15 and older.  Depending on how many of the never-married 15+ were 21 or older, this gives perhaps 150000, so about 16% of the non-Maori adult female population signed the petition. That compares to modern petitions with about 2.4% of voting-age people opposing marriage equality and about 10% for the anti-asset-sales petition (though these are targeting the entire NZ voting population, not just women).

Presumably, rather more than 16% of women were in favour of getting the right to vote, but it’s always difficult to track people down and get them all to sign.  In 1893 there wasn’t an alternative: sampling hadn’t been invented, and would likely have been impractical  — certainly, calling random phone numbers wouldn’t have got you very far.

Today, we have much more accurate ways of estimating the proportion of people who support some government action. Petitions, like demonstrations, are mostly useful for signalling to the government that some issue they weren’t  aware of is actually important.  For example, the petition against animal testing for legal highs would have been effective to the extent that the government wasn’t aware people cared about the issue. For anyone who was aware this was a political issue, a well-conducted opinion poll would be more informative and should be both less expensive and more effective than a petition.

Referendum petitions, as in New Zealand and some parts of the US, are an example of this principle: if an issue can get the support of 10% of the NZ voting population, it’s probably important enough to be worth serious consideration and debate.  The threshold is weaker in many places. For example, in California a petition need only get 5% of the number of people who voted in the last election for state governor, which currently comes to under 2% of the adult population.

The Dominion Post has a story (via @LewSOS) on the Wellington mayoral elections.

John Morrison is leading incumbent Celia Wade-Brown in the race for the Wellington mayoralty, according to a poll of Dominion Post readers.

Mr Morrison, who has been a city councillor for the past 15 years, had support from 27 per cent of the 635 readers surveyed last week – while Ms Wade-Brown trails on 17 per cent.

They don’t say how the survey was done — it’s not clear how you would get a representative sample of Dominion Post readers.  For all we can tell, it might just be a bogus poll.  It’s also not clear, on this topic, why Dominion Post readers are even the population you would want, since the story continues

Of those surveyed, 275 were eligible to vote in the Wellington City Council elections.

You’d at least expect that the voting preferences would be broken out by eligibility.

Sadly, no.

A couple of weeks ago, I wrote about a Sydney Morning Herald story on new electorate-specific robopolls and their surprising predictions. Tim Colebatch at the SMH wrote

Uh-huh. Lonergan’s own national poll reports only a 2 per cent swing against Labor. Yet in the three seats it polled individually, it found an average swing of 10 per cent. That’s huge, far bigger than we have seen in any Federal election since 1943.

and I was similarly dubious.

We now have the facts: the national (two-party preferred) swing against Labour is just over 3%, and the swing in Kevin Rudd’s seat of Griffith, one of the three specifically polled and predicted to have a swing of 10.5%, was 5.42%. The other two seats mentioned in the story as polled by Lonergan were Forde (robopoll swing 8.5%, actual swing 2.5%) and Lindsay (robopoll swing 11%, voters 3.9%)

It looks as though the robopoll skeptics were right. Even though the national swing was larger than the poll predicted, the swings in the target electorates were much smaller.

Australia, as you may have noticed, is having an election. Some statistical analysis links

• Pottinger.com
• Simon Jackman (Oz political scientist, at Stanford)
• Adrian Beaumont, at The Conversation
• Daily Kos

Two episodes to be noted

First, the GCSB bill.

We’ve had a nomination for Stat of the Week for the Campbell Live bogus poll finding 89% opposition to the bill: you just can’t draw that sort of conclusion from self-selected phone-in polls.  On the other hand, they did get over 50000 identified individuals voting, so as a petition it isn’t completely negligible — that’s a bit more than 1.5% of voters.

The Fairfax/Ipsos real poll found a bare majority who trusted the government to protect privacy and only about 30% who were seriously opposed to the bill.  The pollster or the papers fell down badly by not giving us a party breakdown of these figures.  If half the 30% were National voters, the government should have been concerned, but if, like me, they were mostly Labour/Greens voters already, there isn’t any political problem in ignoring them. It’s also a pity there wasn’t any polling relevant to the most obvious pressure point in the coalition – “Would you vote for ACT if they voted against the bill?” would have been an interesting and important thing to know.

Second, the West Island.

As you may have heard, they are having an election soon. In addition to the traditional election polls there are new automated ‘robopolls’ that are sufficiently cheaper that it’s possible to get a useful sample size in single electorates. Or perhaps not. The Sydney Morning Herald has an interesting report

Lonergan’s own national poll reports only a 2 per cent swing against Labor. Yet in the three seats it polled individually, it found an average swing of 10 per cent. That’s huge, far bigger than we have seen in any Federal election since 1943.

My attention has been drawn (by Joseph Simpson, on Twitter) to the Campbell Live online poll on the GCBSB vote.  These polls are fairly bogus anyway, being collected from self-selected viewers or readers, but in this case there’s an added twist.

In red at the top “Please fill in all fields or your vote will not be counted.”  The fields include name, email, and address.  Wouldn’t you expect that people unhappy with the prospect of increased (legal) surveillance of New Zealanders might be less willing to give all their personal details with their vote?

To be fair, the form says that the personal information is private and won’t be linked to you personally.  If we assume don’t mean this, but actually mean that the personal information won’t be linked to the votes, and that TV3 is able to ensure this, there’s no problem.

The Royal Statistical Society has just commissioned an opinion poll to look at beliefs about policy-relevant issues and how they relate to reality.  The first few results:

1. Teenage pregnancy: on average, we think teenage pregnancy is 25 times higher than official estimates: we think that 15% of girls under16 get pregnant each year, when official figures suggest it is around 0.6%.
2. Crime: 58% do not believe that crime is falling, when the Crime Survey for England and Wales shows that incidents of crime were 19% lower in 2012 than in 2006/07 and 53% lower than in 1995. 51% think violent crime is rising, when it has fallen from almost 2.5 million incidents in 2006/07 to under 2 million in 2012.
3. Job-seekers allowance: 29% of people think we spend more on JSA than pensions, when in fact we spend 15 times more on pensions (£4.9bn vs £74.2bn).
4. Benefit fraud: people estimate that 34 times more benefit money is claimed fraudulently than official estimates: the public think that £24 out of every £100 spent on benefits is claimed fraudulently, compared with official estimates of £0.70 per £100.

You might look up what the actual figures are in New Zealand. To get you started, Paula Bennett claimed about $200 million in benefit fraud in 2010/11 (and about 10% of that was prosecuted) from about $9 billion, or about 22c per $100.

More on the ongoing fluoridation story:

Firstly, there is a very good statement from the PM’s chief science adviser, Peter Gluckman.  As he says, the scientific issues are entirely settled: at the concentrations used in treating water, fluoride reduces tooth decay and does not cause any harm.  At one time there was scientific uncertainty about adverse health effects; this is just not the case any more.  There is still a question of whether you want to treat the whole population in this way.  We add iodine to salt and folate to bread, but these prevent more serious illnesses than fluoridation does, and there are fewer people with an irrational fear of iodine or folate.

Second, the Herald has a Digipoll on fluoridation

The poll showed 48 per cent of New Zealanders supported the addition of fluoride – double the 25 per cent of those who opposed its use. A further 24 per cent believed the issue should be left to local councillors to decide.

Unfortunately, the poll tried to use a single question to address two unrelated issues: do you want fluoride in your water?, and should the decision be made nationally or locally?  As a consequence, it’s hard to interpret the results.  The ratio of for:against is about the same as in the Hamilton referendum that started fluoridation there in 2006, but if you assume all the people who want the issue decided locally  are really against fluoridation, the opinion would be nearly 50:50.  It obviously isn’t reasonable to assume everyone in favour of local decision-making is against fluoridation — I’m on record as a counterexample — but there’s no way to know how these folks would split.

The Herald story goes on to quote an antifluoride lobbyist

Ms Byrne said the group had science to back its claims that fluoride was toxic and harmful when added to water and without applying it directly to teeth offered none of the benefits health authorities claimed.

However, that is hotly disputed within the science community.

It’s not disputed within the scientific community, it’s disputed by the scientific community. The science, as Sir Peter observes, is settled.

In a Stat-of-the-Week nomination, Nick Iversen points out a Herald story about the convention centre deal, under the headline “Support disappears for convention deal”

Public opinion has turned against the Government’s SkyCity international convention centre deal just days before it is due to be signed off, allowing for 230 extra poker machines at the downtown Auckland casino.

The latest Herald-DigiPoll survey shows 61.5 per cent of those polled disapprove of the deal while 33.8 per cent approve.

That’s a sharp turnaround from a year ago when a similar poll found 40.3 per cent disapproved and 57.3 supported it.

As Nick and the SkyCity spokesperson point out, the poll last year had a ‘conditionally approve’ option, so they aren’t really comparable. In fact it’s worse than that. the Herald’s headline last year was “Public opposed to SkyCity deal — or want conditions“, so it’s not just a matter of interpreting last year’s data; the story isn’t even consistent with last year’s headline.

Last year, the poll found 40.3% disapproved, 37.7% approved if the total number of gaming machines across Auckland decreased, and 19.6% approved.   This year 61.5% disapprove and 33.8% approve. It’s hard to decide how the 37.7% of conditional approvals should be apportioned, since the number of machines is going down, but it’s going down for reasons basically unrelated to the SkyCity deal.

Some possibilities

• All the conditional approvals should be treated as approvals. That’s the Herald’s current interpretation and gives 57.3% approval last year, and a roughly 25% decrease in approval
• All the conditional approvals should be treated as disapprovals. That was the Herald’s interpretation last year, and gives 19.6% approval last year and and a roughly 13% increase in approval
• The conditional approvals should be ignored, and we should look at the proportion of approval among those approving or disapproving.  That gives 32.7% approval last year, almost identical to this year.

It’s clear that you really can’t say anything very useful about the change using the two different sets of questions.  On the other hand, it’s also clear that the majority in Auckland is currently opposed to the deal, and if public opinion was relevant to the decision-making process, that, rather than the change since last year, would be the important fact.