Toby Manhire at the Listener blogs:

The New Zealand prime minister ranks ninth in theGallup survey (see below) of leader approval ratings in 21 nations in the Asian region, with 72% saying they approved of Key’s job performance, 24% disapproving, and 4% confused or profoundly shy or something.

The survey by the Washington-based Gallup, which polled 1,000 people in each country, has attracted widespread coverage across the region, but as far as I can tell has been overlooked in New Zealand.

I only saw it myself thanks to the daily summary by Bryce Edwards. I was surprised that Key ranked only 9th, as I know his ratings are much higher than national leaders in the US, UK, Canada, France, Germany and Australia. But then I looked at the table:

Laos is a one party state. Cambodia has been ruled by the one party since independence as has Singapore. Sri Lanka is one of the most repressive regimes in the world for press freedom. Haven’t looked into details for Philippines, Malaysia, Bangladesh and Indonesia but I suspect in a fair few of those, criticising the head of state can be a risky thing to do

Would be interesting to see such a poll in say the OECD countries.

Great to see so many people attend the inaugural Political Polling Forum on Wednesday, and especially the questions from audience members. There were a number of excellent contributions.

For those who couldn’t make it, but are interested in political polls, a number of documents have been put up on the MRSNZ website. Three which may be of special interest are:

• Rob Salmond’s presentation
• Murray Campbell’s presentation
• The ESOMAR code/guide on opinion polls

The ESOMAR code is 39 pages long (and an excellent resource). One of the challenges going forward is if we can come up with a NZ version which is concise enough to be of practical value to media as well as research companies.

A reader e-mails:

I notice a number of US pollsters appear to adjust their samples to achieve a predetermined proportion of Republican/Democrat voters.

Would you be kind enough to do a blog post on this subject, citing the nature of the practice, the reasons for it, it’s legitimacy and whether or not it skews results.  It seems to me it does.  Also, to your knowledge is such a practice carried out in NZ?

Weighting is very common to help ensure the results are representative of the overall population. If (for example) 20% of adults are aged under 30 and they make up 15% of your responses (as say young people are out more and harder to get hold of), then you weight their responses by 20/15 or 1.33.

Normally weighting is done on demographics such as gender, age, area, income, ethnicity etc.

In the US some pollsters do weight by party identification. This is not really possible in NZ, as most NZers do not publicly declare their identification. The purpose of such weighting is to make sure a sample is not under-represented by Republicans or Democrats.

However there are dangers in doing so. What do you weight to? You see party identification levels change over time. As party gets more or less popular, more or fewer people identify with them, so if you weight by party ID you may miss that there is a trend of decreasing or increasing support.

In NZ there is no polling by party ID, but sometimes you may see a poll weighted by vote at the last election. This is an attempt to have the results reflect the voting population. The problem with this, is people do not always correctly recall who they voted for at the last election. There is a tendency for some to say they backed the winning party, even if they did not.

In my opinion the best thing is to have your sampling done in a way which means you do not have to weight by party identification. If your sample is picking up (for example) 65% Democrats and 35% Republicans there is probably an issue with your sampling, and while weighting mitigates that, it can mean you ignore the underlying issue.

There is another issue with weighting. Sometimes it means you can exaggerate the error in a small sample. Let’s say you have a poll of 500 and 10% of the population is from (say) Timaru. Now if for some reasons only 20 responses were from Timaru, then they would get a weighting of 2.5. Now that small sample of 20 has a huge margin of error, and you have just compounded it by weighting.

Overall weighting is (in my opinion) a good and necessary thing to do. But I prefer to do it only on known factual demographics.

Just published the latest monthly newsletter summarising the polls. You can subscribe to receive it by e-mail here.

Curia’s Polling Newsletter – Issue 56, April 2012

 April saw four political polls published – two Roy Morgan polls a 3 News Reid Research poll and a Herald Digipoll.

The average of the public polls has National 19% ahead of Labour, up from 18% in March. The centre-right has 63 out of 121 seats on the April average.

Australia has the Government 17% to 18% behind the Opposition on a two party preferred basis. Gillard’s approval ratings are at record lows with 63% disapproval.

In the United States Barack Obama is 1..9% ahead of Mitt Romney in the polls, but projected to be well ahead in likely electoral college votes. His re-election chances are now rated at 60%, down from 61% last month.

The Republicans remained favoured to keep control in the House (72%) and gain a majority in the Senate (63%).

In the UK the Conservatives are now 9% behind Labour, and were trounced in the local elections.

In Canada the Conservatives are still impacted by a fake robo-call scandal and the NDP is now level with them in the polls.

The normal two tables are provided comparing the country direction sentiment and head of government approval sentiment for the five countries. New Zealand continues to stand out in both.

We also carry details of polls in New Zealand on Epsom, the 2011 Election, paid parental leave, the Sky City plus the normal business and consumer confidence polls.

This newsletter is normally only available by e-mail.  If you would like to receive future issues, please e-mail newsletter@curia.co.nz with your name, organization (if applicable) and e-mail address or go to http://listserver.actrix.co.nz/cgi-bin/mailman/listinfo/polling-newsletter to subscribe yourself.

I’ve blogged the latest Roy Morgan poll at Curiablog. On that poll the Maori Party would hold the balance of power.

The Herald on Sunday reports:

Banks, copping heavy criticism over donations from Kim Dotcom and SkyCity listed as being anonymous from his 2010 mayoralty bid, has lost a great deal of support, according to the Herald on Sunday-Key Research survey.

Only 10 per cent of electors would vote for him now – down from 44.1 per cent at the election last November – and a National candidate would romp home. 

The poll results are no surprise, and reflect my gut feeling that Epsom voters will vote for a National candidate, no matter what.

I would point out though one historical fact. No poll ever done in Epsom has shown an ACT candidate winning the seat – and they did win in 2005, 2008 and 2011. Epsom voters understand the tactical advantage of doing so.

However that tactical advantage disappears when there is little possibility of getting 1.2% of the party vote and one or more List MPs.

I’ve blogged at Curiablog the NZ Herald Digipoll released this morning. A big boost for Labour, but National still above what they got at the election.

The poll gives the CR a majority as NZ First just under 5% threshold. If NZ First did make 5%, then the Maori Party would hold the balance of power.

The headline in the Sunday Star-Times, at Stuff is “Mr Unpopular?” above a photo of John Key, and a story by Anthony Hubbard proclaiming “People are getting angry, John”.

Rather ironic timing then that tonight’s 3 News Reid Research poll (blogged at Curiablog) has National increasing its support since February by 2.3%, to reach 49.8%.

Personally I think this shows what a small bubble some (not all) of the media live in. Just because their Grey Lynn friends are all up in arms over something, doesn’t mean most of New Zealand is.

As for the Labour wet dream headline of “Mr Unpopular” I should point out that the big movement in the poll was a 9% increase in those saying David Shearer is performing poorly. Those saying Key was performing poorly increased by just 2%.

I really do wonder sometimes whether or not the Sunday Star-Times should just be renamed “The Standard”.

I’ve blogged the latest Roy Morgan poll at Curiablog. Quite a change from the previous one.

I’ve blogged at Curiablog the details of the latest Roy Morgan poll.

It is the first poll since (off memory) 2007 which has shown a centre-left government able to be formed, categorising Labour, Greens, NZ First and Mana as centre-left.

Martin Johnston at NZ Herald reports:

A survey indicates there has been an increase in the number of parents who choose not to smack their children, in line with the controversial “anti-smacking law” implemented in 2007.

The survey was commissioned by conservative lobby group Family First from Curia Market Research, a firm headed by centre-right blogger David Farrar.

It is based on responses from 500 parents of children aged less than 12. It found that 44 per cent reported never smacking their children since the 2007 legislative change to remove the Crimes Act defence of “reasonable … force” for parents who hit their children to correct them.

Twenty-nine per cent told Curia they had smacked rarely since the change, 21 per cent said occasionally, 1 per cent said frequently and 5 per cent were unsure or refused to answer.

The full results are on the Family First website.

The never-smacked figure was higher than found in a 2009 Herald-DigiPoll survey of parents of 4-year-olds.

That poll found that 39 per cent of mothers and 33 per cent of fathers never smacked – and that was more than three-fold higher than the rate during the four decades to 1997.

I’d be a bit careful comparing a poll of parents of four year olds and a poll of parents of children aged from zero to 11 years old. I would suspect that the older the children the less likely parents are to smack.

This is not to say that the “never smacked” figure may not be increasing. It may or may not be. But one would want to compare data of parents of same aged children to be able to say if there is a trend.

Survey findings

66 per centof parents would “smack to correct in future”

44 per centhad not smacked their children since the 2007 law change

49 per cent think law change “caused decline in discipline”

81 per cent would not report someone for smacking

63 per cent think law should be changed to allow hand smacks

75 per cent say 2007 law change has not changed New Zealand’s level of child abuse

One interesting aspect was the views of parents on whether the law change has caused a decline in disciple. Only 42% of parents living in areas in the top three (least deprived) deciles said it had, but 59% of those in the bottom three deciles (most deprived)  said it had caused a decline discipline.

One News have just released their first post-election poll, taken over the last week. National is up 1%.

Full results are at Curiablog.

Have blogged at curiablog the latest Roy Morgan poll out today.

On this poll, there would be a hung Parliament. The CR would have 61 seats and the CL 58 seats. The Maori Party on 3 seats could enable National to govern or force a new election.

I’ve just published Curia’s monthly public polls newsletter. You can subscribe to it at this address. The summary is:

February saw three political polls published – two Roy Morgan polls and a 3 News Reid Research poll.

The average of the public polls has National 16% ahead of Labour. For the first time, the public polls average does not show a clear centre-right Government, but instead the Maori Party holding the balance of power

Australia has the Government only 4% to 6% behind the Opposition on a two party preferred basis. However in the upcoming Queensland state election Labor are 16% behind. Gillard has only 26% approval and 64% disapproval of the job she is doing as PM.

In the United States Barack Obama has had a great three months. His approval ratings have improved by a net 9%, and for economic management have gone up 20%. The country direction sentiment has improved by a net 32%. His re-election chances are now rated at 605, up from 49%. The Republican nomination is Mitt Romney’s to lose.  He only has 166 out of 1,144 delegates and still the preferred choice of only 37% of Republicans. However Intrade has him at 85% likely to gain the nomination.

The polls show Obama ahead of Romney by 5% in a head to head contest. However the Republicans remained favoured to keep control in the House (63%) and gain a majority in the Senate (61%)

In the UK Labour lead by just 2%.

In Canada the Conservatives have dropped 5% to 34%.

We also carry details of polls in New Zealand on the most important issues, top news stories, personal finances, the Occupy movement, low voter turnout reasons, breastfeeding, Maori views and the normal business and consumer confidence polls.

The graph above shows the combined effects of NZ First getting over the 5% threshold, and also National dropping from low 50s to high 40s.

I blog a bit of context around the Preferred PM ratings at Stuff.

I also note that on the 3 News poll the CR would have 60 seats, the CL 60 seats and the Maori Party 3 seats. So they would get to choose the Government, and either way it would have a bare two seat majority.

In my blog at Stuff, I look at the latest Roy Morgan poll, and ask if National is now the underdog?

Roy Morgan have their first poll since the election out.

They say it shows little change for the major parties:

• National 47%
• Labour 27.5%
• Greens 14.5%
• NZ First 5.0%
• Maori Party 1.5%
• Mana 1.0%
• ACT 0.5%
• United 0.5%

To me it shows how things remain close. Labour, Greens and NZ First combined are 47% – same as National.

However the right direction indicator has gone up from 49.5% to 60.5% and wrong direction down from 31% to 28%.

Family First have released the results of some polling done amongst 16 to 21 year old New Zealanders by Curia. The summary is:

When asked “Do you think sex education in schools should teach values, abstinence and consequences such as pregnancy, or just teach safe sex?” only 19% supported just the ‘safe sex’ message currently being taught in schools, with one in three (34%) wanting ‘values, abstinence, and consequences such as pregnancy’ taught instead, and a further 42% asking for a combination of both – especially amongst older teens. The support for just the ‘safe sex’ message dropped even lower for the older teens. …

When asked “Provided it won’t put the girl in physical danger, should parents be told if their school-age daughter is pregnant and considering getting an abortion?” 59% of young respondents thought the parents should be told. 34% disagreed. More young men than women agreed, but both had majority agreement.

When asked “Do you believe an unborn child or foetus has a right to be born?” 56% of youth respondents said they believed an unborn child or foetus has a right to be born. Slightly more young women than young men agreed – 58% to 55%. Those aged 15 to 17 were strongest in support – 66%.

It was intriguing that young women were slightly more supportive than young men of there being a right to be born for an unborn child or foetus.

The parental notification results may surprise some, but thinking about it it is normal that most young people would expect to talk to parents about any unplanned pregnancy. The question did not specifically say “Should the law require parents to be told …” but at a minimum it makes clear that a majority thought parental notification should occur.

Note that of course the results do not necessarily represent my personal views.

I blogged on Friday the final polls by the five public telephone pollsters and the final poll by Horizon. A fuller analysis will be done once we get final results, but for now I’ll do a quick analysis of how each pollster did for each party.

Note that this is not comparing apples and apples entirely. Those pollsters whose final poll was earlier in the election period naturally do not pick up what happens in the final few days. And there are other factors at work such as sample sizes. So this is not about saying who is “best” and “worst” but just a quick look at were they broadly in the right ballpark for the various parties.

This shows the actual result, and the (absolute) difference between the final poll for that pollster and the final result. Where the difference was greater than 1.5%, I have highlighted them in red.

This is just one of several ways to analyse it. One can also total up the differences for each pollster. Also you can count how many had a result within the margin of error for that poll. I’ll comment on each poll result.

Roy Morgan

They were the pollster that got NZ First closest. They had National and Greens too high and Labour too low. They did not record results for the Conservative Party at all, but otherwise were pretty good.

Fairfax Research International

National significantly too high, but Labour pretty accurate. Undershot NZ First and did not report on Conservatives. Other Minors within range.

3 News Reid Research

Like everyone had National too high (but within margin of error) and like most had the Greens too high. All other minor parties within 1.5% except NZ First whom they had at half what they got.

One News Colmar Brunton

Overall seemed to get things closest. National 2% too high and NZ First 2.6% too low, all others less than a 1% variance.

NZ Herald Digipoll

Also did well. National too high and NZ First too low, but did have them over 5%. Slightly more variance with the minor parties but none greater than 1.5%.

Horizon Poll

Of the nine significant parties, Horizon only got two of them within 1.5% – the Maori Party and United Future parties. They were the least accurate with National (14.2% out), NZ First (4.1% out), ACT (1.7% out) and Mana (1.8% out). They also had Conservatives at close to double what they actually got.

Very amusingly, Horizon are boasting how they consider their poll to have been highly accurate. It staggers me how anyone can put out a poll which had National only 5% ahead of Labour and then could claim it was “close to forecast” when the actual result was a gap of 21%.

My Stuff column is on how the election outcome is not settled. I conclude:

So I don’t regard the outcome of the election tomorrow as settled. Certainly National is in a much better position than Labour. But under MMP, even a 23-point lead in the polls does not guarantee you government.

My message to all Stuff readers is to make sure you vote. Do not think the outcome of this election and the identity of the next government is settled. No matter who you support, make sure you have you say and cast a vote today or tomorrow.

Talking of voting, what is the weather forecast for tomorrow?

Roy Morgan released a poll overnight, which some may have missed. I’ve blogged details at Curiablog.

One aspect of the poll makes me a bit doubtful of its salience – those who say the country is heading in the right direction dropped 11% from last week. That is a huge drop for one week, and suggests to me the sample may be over-represeted with “grumpies”. But we will see tomorrow.

What is interesting is the seat projections on their poll:

If all parties hold their current seats, it is:

• National 60
• Labour 29
• Green 18
• ACT 2
• Maori 4
• United Future 1
• Mana 1
• NZ First 8
• Total 123

If Banks does not win Epsom and Dunne does not win Ohairu, it is:

• National 62
• Labour 30
• Green 18
• Maori 4
• Mana 1
• NZ First 8
• Total 123

In this scenario National just has to drop one further seat and it can not form Government, as the Maori Party holds the balance of power.

Phil Goff could almost form Government on 23.5% of the vote if he does deals with Greens, NZ First, Maori and Mana parties.

It would be a good result for Phil Goff as he keeps his job. Not so good for his colleagues though as on this poll, Labour would lose 13 MPs, but get to form Government.

This is a summary of the polls released this week. You can click on it for a larger version.

• The first five columns are the five main public polls.
• The simple average is just that – the mean of their results
• The weighted average is weighted by size and date. The full methodology is at Curiablog.Those polls which cover an earlier period are weighted slightly less. The fact this varies from the simple average shows there is some movement in the final days – National and Greens down a bit and Labour and NZ First up a bit
• The weighted seats is how many seats each party would have if that was an election result, assuming all electorate seats stay with the party that currently holds them (exception is Wigram going from Progressive to Labour)
• For comparison I have the final Horizon poll. The Horizon poll features regularly on Radio Live and on the front page of the Sunday Star-Times. Their published figures include 1.3% undecided which I have adjusted their results for, so like the other polls it is a percentage of decided voters. This allows a comparison to election results.
• I also have the seat projections based on the final Horizon poll.
• Under the weighted average there would be a centre-right Government of the only three parties pledged to support National.
• Under Horizon’s poll there would easily be a centre-left Government with the four centre-left parties having 67 seats.
• The Maori and Conservative parties have said they will go with either Labour or National so are shown in their own bloc

People should be aware of the commentary with the final NZ Herald poll:

Today’s poll also throws up a bizarre possible outcome – National winning more than 50 per cent of the party vote but still needing Act, the Maori Party or United Future to give it a majority in Parliament.

This could happen if today’s poll results were translated to votes.

The revival of New Zealand First – which National won’t deal with – could make the survival of Act crucial.

The Maori Party could be in the box seat to negotiate a confidence and supply agreement to give National a cushion of comfort if Act and United Future don’t make it.

The reason National could get a majority of party votes tomorrow but not a majority of seats in Parliament is the overhang factor.

If today’s poll figures were translated to votes, United Future, the Maori Party and Mana would get more electorate seats than their party vote entitlement.

When that happens, the size of Parliament expands beyond 120 seats, and the parties are allowed to keep the extra seats.

In this case, the “overhang” seats would take Parliament to 126 seats.

In that scenario, a Government would need 64 seats for a majority and in today’s poll, National would have 63 seats – based on the assumption that Act, Mana, United Future and the Maori Party will keep their electorate seats.

Only Act and United Future have pledged to support National.

MMP does not always deliver proportional results. As the Herald says, one could have National get over 50% of the votes, but lose power.

It is quite simple. If you want a National-led Government, party vote National. If you live in Ohariu you should vote for Peter Dunne or Winston may get to pick the Government. If you live in Epsom and want John Key to remain Prime Minister, you should vote for John Banks.

The Conservative Party have not said they will back National. Do not think a vote for them, or a vote for Craig, is a vote for John Key to remain Prime Minister.

It should not be assumed the Maori Party would back National if they hold the balance of power. They have more policy in common with Labour, and would be more likely to retain their seats in 2014 if they go with a centre-left Government.

Despite the fact the average of the polls show Phil Goff is leading Labour to their lowest ever vote share, the way MMP works means he can still emerge Prime Minister. National would never say yes to the huge spending demands that the Greens, and Winston would demand for supply and confidence.

People may also wish to think about how they will vote in the referendum, based on the scenario outlined by the NZ Herald that MMP could deliver government to Labour on 28%, even if National gets 51%. Do you want to lock MMP in as our electoral system for the next 50 years, or do you want to vote to have a referendum in 2014 pitting MMP against the most popular alternative?

Have a look at what John Key and Phil Goff are saying on the referendum, and decide for yourself.

UPDATE: The original version did not include the Roy Morgan poll released overnight. This is now included.

The final Fairfax Research International poll is at Stuff. I’ve done full results at curiablog.

National is up 1.5% from their last poll, Labour effectively unchanged, Greens down a bit and NZ First at 4.0%. Obviously a very positive result. A caution thought. The level of undecided voters is relatively high at 15.6%, so where they end up can change things significantly. The poll covers the period from Thursday to Monday. I would not take this poll as any reason for complacency over the outcome. Turnout is crucial.

What would a Parliament look like on this poll result, assuming no electorates change hands except Wigram going to Labour.

National

69 seats would have National with 41 electorate MPs and 28 list MPs. All 63 electorate candidates would get elected via the electorate or the list. On top of the 63 electorate candidates, list only candidates Lockwood Smith, Steven Joyce, Jian Yang, Alfred Ngaro, Heather Tanner and Denise Krum would also make it.

Labour

Labour would get reduced to 33 seats from 43. With 22 electorates, this gives them 11 list MPs which would be David Parker, Maryan Street, Sue Moroney, Charles Chauvel, Jacinda Ardern, Andrew Little, Shane Jones, Darien Fenton, Moana Mackey, Rajen Prasad, and Raymond Huo.

Andrew Little would be the only new List MP. Current MPs who would be gone are Carol Beaumont, Kelvin Davis, Carmel Sepuloni, Rick Barker, Stuart Nash and Steve Chadwick

If Labour pick up electorates such as Te Tai Tonga and WCT, then Prasad and Huo would miss out on the list. Likeiwse if Labour lose Rimutaka and Palmie, then Hipkins and Lees-Galloway are gone and Beaumont and Davis would make it in.

Greens

Their caucus would go from nine to 15, with eight new MPs.

They would get in all current MPs plus Eugenie Sage, Jan Logie, Steffan Browning, Denise Roche, Holly Walker, Julie-Anne Genther, Moho Mathers and James Shaw.

Maori Party

On this poll their party vote would only entitle them to one seat, so if they retain all four electorates, an overhang of three seats.

ACT

On this poll (and it is only one poll), John Banks would be the only MP if he wins Epsom.

Mana

On this poll Hone Harawira would be the sole MP.

United Future

Their party vote is too low to be “entitled” to a List MP, so if Peter Dunne wins he would be an overhang MP (which actually helps National)

Parliament would have 124 seats, so 63 would be needed for a Government to have confidence and supply.

Te Karere have just released their Te Tai Tonga poll, which I have blogged in full at Curiablog.

Many had been picking it as a seat that Labour may win with Rino Tirikatene standing for then. However this poll of 400 voters has him 11% behind Rahui Katene.

The Maori seats are difficult to poll, so this poll is not gospel, but it will be a big blow for Labour if they fail to win this seat. On the plus side it means one of their List MPs will survive if they do lose it.

Rob Salmond at Pundit looks at the polls, and linear projection of where the parties may end up on election night, being:

• National 50.8%
• Labour 24.8%
• Greens 14.6%
• NZ First 3.8%

He comments further:

Using the analysis above, we project that New Zealand First will get 3.8% of the vote. We can also project with 95% confidence that the party will get between 2.3% and 5.2% of the vote. Importantly, we project that there is only a 4% probability New Zealand First will get more than five percent of the vote.

We project that National will get an absolute majority of the vote, but a pretty small one (50.8%). Based on this analysis, we think there is a 74% chance that National will get over 50% of the vote.

Of course this is all based on the trend of the last few weeks continuing in the last week, and it may not continue if events get in the way.