Final poll results – table

All five pollsters have released their final week results, with results narrowing.

Election 2014 final poll results

Notes:

  • Polls ask “If an election was held today who would you vote for?”, they don’t try to predict election day voting.
  • It is common for movements in support late in campaigns due to tactical voting and undecideds deciding.
  • If ACT and United Future win electorates they may add more to seats than their share of vote.
  • If the Maori Party hold all three seats they will get more than their vote share. If they hold two seats they will be about proportional to their party vote according to the poll average.
  • If Conservatives don’t make the 5% threshold the other parties will increase their % share of seats.
  • If Hone Harawira loses Te Tai Tokerau Internet-Mana will not get any seats and their party vote will be ‘wasted’.
  • In 2011 National got 47.31% and with ACT and United Future seats were just able to make a majority.

Most of this polling will have occurred before Monday night’s “The Moment of Truth” meeting. NZ Herald recorded before and after results:

With 60 per cent of the poll done by Monday night, when the event happened, National was polling at 47.8 per cent, down on last week, said DigiPoll general manager Nandan Modak. From Tuesday it jumped to 49.1 per cent.

But I asked Andrew from Colmar brunton if he’d seen any change and he responded:

Was looking the whole time, even during.

Impossible to tell if any impact, with any degree of certainly.

I saw no increase for National compared to first two days, but it’s not that simple, as party support differs by day normally.

- @Unimatrix_0

Colmar Brunton explain ‘margin of error”:

The maximum sampling error is approximately ±3.1%-points at the 95% confidence level. This is the sampling error for a result around 50%. Results higher and lower than 50% have a smaller sampling error. For example, results around 10% and 5% have sampling errors of approximately ±1.9% points and ±1.4% points respectively, at the 95% confidence level.

See full final results – Final pre-election poll results

See also Coalition possibilities many and varied

Final pre-election poll results

All five polls have been published in the final week of the election campaign.

Stuff/IPSOS
No polling period given, poll date 17 September

  • National 47.7% (-5.1)
  • Labour 26.1% (+3.7))
  • Greens 12.0% (-1.0)
  • New Zealand First 6.6% (+2.2)
  • Conservative Party 4.5% (+0.9)
  • Maori Party 1.0% (+0.7)
  • Internet-Mana Party 0.9% (-0.5)
  • Act NZ 0.3% (-0.4)
  • United Future 0 (no change)

Poll results and poll report: Tight race ahead for Key and Cunliffe

Our poll provides a maximum sampling error of +/-3.1%-point, at the 95% confidence level. This means we can be 95% confident that the survey results are within 3.1% of the result had we surveyed the entire population of the NZ population, when the analysis is based on all respondents surveyed.

One News/Colmar Brunton
13-17 September

  • National 45.1% (-1)
  • Labour 25.2% no change)
  • Greens 12.5% (-2)
  • New Zealand First 8.1% (+1)
  • Conservative Party 4.4% (+0.4)
  • Maori Party 1.6% (+0.8)
  • Internet-Mana Party 1.8% (+0.4)
  • Act NZ 0.6%
  • United Future 0

Summary and Detailed Report (PDF)

The maximum sampling error is approximately ±3.1%-points at the 95% confidence level. This is the sampling error for a result around 50%. Results higher and lower than 50% have a smaller sampling error. For example, results around 10% and 5% have sampling errors of approximately ±1.9% points and ±1.4% points respectively, at the 95% confidence level.

NZ Herald/Digipoll
11-17 September

  • National 48.2% (-0.4)
  • Labour 25.9% (+1.3)
  • Green 11.1% (-0.4)
  • NZ First 8.4% (+0.3)
  • Conservatives 3.3% (-0.5)
  • Internet-Mana 1.0% (-1.3)
  • Maori 1.1% (+0.4)
  • ACT 0.5% (+0.2)
  • UnitedFuture O.2% (+0.2)

Pre-poll report: DigiPoll: Conservatives fail to make 5 per cent threshold – again

Poll report: Moment of Truth gifts Team Key a late bounce in polls

The poll of 775 eligible voters was conducted between September 11 – 17 The Party Vote is of decided voters only. Undecided voters were 5.6 per cent. The margin of error is plus or minus 3.5 per cent.

“The Moment of truth” split:

The Kim Dotcom-inspired event in Auckland’s Town Hall that was supposed to end John Key’s career gave the National Party an immediate bounce in support this week, according to polling for the last Herald DigiPoll survey.

With 60 per cent of the poll done by Monday night, when the event happened, National was polling at 47.8 per cent, down on last week, said DigiPoll general manager Nandan Modak. From Tuesday it jumped to 49.1 per cent.

3 News/Reid Research
September 9-15

  • National 44.5% (-2.2)
  • Labour 25.6% (-0.5)
  • Greens 14.4% (+1.4)
  • New Zealand First 7.1% (+1.2)
  • Conservative Party 4.9% (+0.2)
  • Maori Party 1.1% (-0.2)
  • Internet-Mana Party 2.0% (0.3)
  • Act NZ 0.1% (-0.2)
  • United Future 0.1%

Report: Poll: Winston holds balance of power
(it’s far to close to call specific outcomes with Conservatives on 4.9% which is teetering either way)

Poll of 1000 voters was taken between September 9 and 15 with a margin of error of +/- 3.1 percent. 

Result table: http://www.reidresearch.co.nz/TV3+POLL+RESULTS.html

Roy Morgan:
September 1-14

  • National 46.5% (+1.5)
  • Labour 24.0% (-2.0)
  • Greens 13.5% (-2.5)
  • New Zealand First 8.0% (+2.0)
  • Conservative Party 3.5% (no change)
  • Maori Party 1.5% (+ 1.0)
  • Internet-Mana Party 1.0% (no change)
  • Act NZ 0.5% (-0.5)
  • United Future 0.5% (+0.5)
  • Independent/ Others 1.0% (no change)

Roy Morgan rounds to the nearest 0.5%

Electors were asked: “If a New Zealand Election were held today which party would receive your party vote?” This latest New Zealand Roy Morgan Poll on voting intention was conducted by telephone – both landline and mobile telephone, with a NZ wide cross-section of 935 electors from September 1-14, 2014. Of all electors surveyed 5% (up 1.5%) didn’t name a party.

Five Poll Summary

Polls are coming out thick and fast leading up to the election showing quite a bit of variability.

Here are the latest results for each party.

National

  • Roy Morgan 45% (-0.3)
  • 3 News/Reid Research 46.4% (+1.4)
  • Stuff/IPSOS 54.2% (+3.4)
  • NZ Herald/Digipoll 50.1% (-0.6)
  • One News/Colmar Brunton 50% (+2)

Labour

  • Roy Morgan 26% (-1.5)
  • 3 News/Reid Research 25.9% (-0.5)
  • Stuff/IPSOS 24.3% (-1.8)
  • NZ Herald/Digipoll 23.8% (-0.3)
  • One News/Colmar Brunton 26% (-2)

Greens

  • Roy Morgan 16% (+4.5)
  • 3 News/Reid Research 12.6% ()-0.9)
  • Stuff/IPSOS 12.9% (+1.1)
  • NZ Herald/Digipoll 11.4% (n/c)
  • One News/Colmar Brunton 11% (-1)

NZ First

  • Roy Morgan 6% (-0.5)
  • 3 News/Reid Research 5.8% (-0.5)
  • Stuff/IPSOS 3.6% (-0.4)
  • NZ Herald/Digipoll 6.0% (+1.0)
  • One News/Colmar Brunton 7% (+1)

Conservative Party

  • Roy Morgan 3.5% (+2.5)
  • 3 News/Reid Research 4.2% (-0.4)
  • Stuff/IPSOS 2.4% (-0.3)
  • NZ Herald/Digipoll 3.8% (+0.5)
  • One News/Colmar Brunton 3% (n/c)

Internet-Mana

  • Roy Morgan 1% (-1.5)
  • 3 News/Reid Research 1.7% (-0.4)
  • Stuff/IPSOS 1.3% (-0.9)
  • NZ Herald/Digipoll 3.5% (+0.1)
  • One News/Colmar Brunton 2% (n/c)

Maori Party

  • Roy Morgan 0.5% (-0.5)
  • 3 News/Reid Research 2.0 (+1.3)
  • Stuff/IPSOS 0.3% (-0.4)
  • NZ Herald/Digipoll 0.4% (-0.6)
  • One News/Colmar Brunton <0.5%

ACT Party

  • Roy Morgan 1% (+0.5)
  • 3 News/Reid Research 0.6% (+0.3)
  • Stuff/IPSOS 0.2% (-0.5)
  • NZ Herald/Digipoll 0.4% (+0.1)
  • One News/Colmar Brunton <0.5

UnitedFuture

  • Roy Morgan 0% (-0.5)
  • 3 News/Reid Research 0.1% (-0.3)
  • Stuff/IPSOS 0.1% (n/c)
  • NZ Herald/Digipoll 0.3% (+0.1)
  • One News/Colmar Brunton <0.5

Polling periods vary but are ordered oldest to latest.

Rounding:
- Roy Morgan rounds to the nearest 0.5
- One News/Colmar Brunton rounds to the nearest whole number so parties not included are <0.5%

Roy Morgan poll two weekly, the others are currently polling weekly.

Updated 5/8/14 with latest One News/Colmar poll.

 

 

 

 

 

 

 

 

Bad poll for National, Labour

The latest Roy Morgan poll has both National and Labour down into worrying territory with Greens and Conservatives being upwards movers:

  • National 45% (down 3%)
  • Greens 16% (up 4.5%)
  • Labour 26% (down 1.5%)
  • Maori Party 0.5% (down 0.5%)
  • Act NZ (1%, up 0.5%)
  • United Future 0% (down 0.5%)
  • NZ First 6% (down 0.5%)
  • Internet-Mana Party 1% (down 1.5%)
  • Conservative Party of NZ 3.5% (up 2.5%)
  • Independent/ Others is 1% (unchanged)

3% isn’t a huge move but 45% is into real worry territory for National, especially if it’s part of a downwards trend.

Labour appear to be either also tainted by Dirty Politics or just seen as crap. Greens are finally benefiting with their best result ever.

Internet-Mana seem to have lost traction while Conservatives are on the rise.

Electors were asked: “If a New Zealand Election were held today which party would receive your party vote?” This latest New Zealand Roy Morgan Poll on voting intention was conducted by telephone – both landline and mobile telephone, with a NZ wide cross-section of 762 electors from August 18-31, 2014. Of all electors surveyed 3.5% (down 3%) didn’t name a party.

Results: http://www.roymorgan.com/findings/5767-roy-morgan-new-zealand-voting-intention-september-3-2014-201409030312

PDF including table of results

Three polls

There have been three polls in three days with some different movements and results showing that polls are indicative of current opinion only and can’t be taken as definitive election predictions – they ask people’s opinion if the election was held today and not how they might vote at the election.

  Stuff
IPSOS
NZ Herald
Digipoll
3 News
Reid Res.
Average
National            50.8            50.7            45.0            48.8
Labour            26.1            24.1            26.4            25.5
Greens            11.8            11.4            13.5            12.2
NZ First              4.0              5.0              6.3              5.1
Conservatives              2.7              3.3              4.6              3.5
Internet-Mana              2.2              3.4              2.1              2.6
Maori Party              0.7              1.0              0.7              0.8
ACT Party              0.7              0.3              0.3              0.4
UnitedFuture              0.1              0.2              0.4              0.2

Margins of error are about +/-3 for National but less for lower results see this table from Stats Chat:

Lower and upper ‘margin of error’ limits for a sample of size 1000 and the observed percentage, under the usual assumptions of independent sampling

Percentage lower upper
1 0.5 1.8
2 1.2 3.1
3 2.0 4.3
4 2.9 5.4
5 3.7 6.5
6 4.6 7.7
7 5.5 8.8
8 6.4 9.9
9 7.3 10.9
10 8.2 12.0
15 12.8 17.4
20 17.6 22.6
30 27.2 32.9
50 46.9 53.1

 

 

Stuff/IPSOS poll

The third poll in three days shows that polls can only be taken as indicators of current support and are nowhere near being definitive indicators of election results.

Stuff/IPSOS results:

  • National 50.8% (-4.3)
  • Labour 26.1% (+3.6)
  • Greens 11.8% (+0.5)
  • NZ First 4.0% (+0.6)
  • Conservative Party 2.7% (-0.6)
  • Internet-Mana 2.2% (+0.1)
  • Maori Party 0.7% (-0.3)
  • ACT Party 0.7% (+0.2)
  • UnitedFuture 0.1% (+0.1)

These show different movements but with similar end results to yesterday’s NZ Herald poll.

The poll has a maximum margin of error of plus or minus 3.1 per cent

Our poll provides a maximum sampling error of +/-3.1%-point, at the 95% confidence level. This means we can be 95% confident that the survey results are within 3.1% of the result had we surveyed the entire population of the NZ population, when the analysis is based on all respondents surveyed.

Note that maximum margin of error of 3.1% means it only applies to National on 50.8% – parties on lower percentages have smaller margins of error.

Stuff commentary: Nats riding high, Key sliding down

Details: IPSOS Polling Station

xx

Disappointing debate, pointless polls

I was disappointed with the leaders debates. Maybe I just wasn’t in the mood for it, but I didn’t see anything that I thought wouold make difference.

Cunliffe held his own so won’t have been harmed by it. He interrupted too much and too many preachy wee speeches. Pros and cons on points made.

Key looked strained but probably won’t have harmed his chances, Some pros and cons as well, nothing remarkable.

Hosking allowed too much talking over each other, sometimes all three were trying to compete. Often seemed messy.

I doubt many minds would be changed.

The online and text polling and online metering were a farce, totally meaningless. There is no way of knowing who was voting or measuring responses so no way of knowing how biased the participants were.

The National minions seem to have been busier on TV1’s text poll and Labour’s on Newstalk ZB’s online poll. Or something. Tells us nothing useful.

They were worse than pointless, they add useless noise to commentary on the debate.

 

Excellent poll coverage from NZ Herald

NZ Herald (Audrey Young) poll coverage starkly contrasts with 3 News, giving clear results with good consideration of wider things like trends and polls from other sources in Digipoll: Minor parties surge as Labour sinks lower.

Summary of results:

The full party vote results compared with last week’s:

National 50.7 (up 0.7)
Labour 24.1 (down 1.1)
Greens 11.4 (down 2.3)
NZ First 5 (up 0.7)
Maori Party 1 (up 0.3)
Internet Mana 3.4 (up 1.3)
Conservatives 3.3 (up 0.7)
Act 0.3 (down 0.3)
United Future 0.2 (down 0.2)
PREFERRED PRIME MINISTER (compared with last week)
John Key 67.8 (up 3)
David Cunliffe 11.6 (down 2.8)
Winston Peters 8.2 (up 3.1)
Russel Norman 3.8 (up 0.3)

The poll of 750 eligible voters was conducted between August 21 and 27. The Party vote is of decided voters only and 7.9 per cent were undecided compared to 12.5 per cent last week.

This contrasts in some aspects with the 3 News/Reid Research poll yesterday – National, Labour down in 3 News poll  – which shows that a single poll can’t be taken as a definitive indication of an election outcome.

National are significantly higher and rising in this poll, and Labour are 2% lower and will be worried.

NZ First and Conservatives have risen but not as much, and Internet-Mana has jumped up potentially significantly.

Poor poll coverage by Gower and 3 News

3 News via Patrick Gower covered their latest poll result poorly. They showed either an awful knowledge of how polling works or they deliberately ignored polling 101 to promote misleading headlines.

Their online headline: Latest political poll big blow for John Key

National were down 2.5 to 45% – this will be a concern for National but it is a fairly moderate movement, well within the margin of error.

It probably reflects a reaction to how John Key has handled the fallout from ‘dirty politics’ (poorly) but it’s too soon to tell whether it is a temporary blip or a lasting change in support.

Labour were also down 2.5, to 26.4% – this should be at least as big a concern for them but the online article doesn’t mention it.

The lead paragraph and most of the article is on one theme:

John Key’s nightmare is suddenly real. Tonight’s 3 News political poll shows the Prime Minister can no longer do without New Zealand First leader Winston Peters.

That’s nonsense. This is just one poll three and a half weeks out from the election. It is far from certain that the Prime Minister “can no longer do without” Winston Peters.

NZ First were up 1.7 to 6.3% and look a good bet to remain in Parliament but there’s no certainty they will maintain this level of support nor any certainty they will hold the balance of power post-election.

Gower led his 6 o’clock news coverage with…

Yes dirty politics has caused political chaos.

One poll over-reported is not political chaos.

Look at these numbers, National takes a hit down 2.5 to 45.5, this hurts, John Key knows he could be gone on this…

Key already knew (and has said so a number of times) he will be ‘gone’ if the election doesn’t go his way. National 44.5% in January, 45.9% in March.

…but Labour also take a hit, down to 26.4%, that’s a drop of 2.6, voters simply do not see them as an alternative.

What? National could be gone but there’s no alternative? It’s Labour’s lowest 3 news result since November 2011 but hey, that’s not significant.

The Greens go up, one of their best results on this poll…

Up 0.5 (a negligible change in polling terms) to 13.5% which is their best result in the Reid Research records back to Sep/Oct 2010. Trending up since May.

…and just look at this, Winston Peters up to 6.3 percent, over the threshold, Peters is back, and just wait to see the mess that this could cause.

Of course a mess is great for headlines but is far from assured despite Paddy’s best efforts to talk one up.

While Peters is obviously dominant – and Gower seems to have an obsession with him – it is actually NZ First, the party, not just Peters. There is no guarantee Peters will be back, despite Gower’s wishes.

Turning now to the minor parties and the Conservatives, up to 4.6% – obviously picking up a protest vote from right wing voters turned National…

They quite likely got some protest vote but Colin Craig has claimed there jump is due to their campaign mail box drop. Opinion shifts can be for many reasons.

…not over the five percent threshold, remember John Key did not do that electorate deal in East Coast Bays, and won’t he be regretting that right now.

Why would Key regret making a sensible decision based on one poll? If Conservatives can sustain this support it’s quite possible they could keep surging past the 5% threshold. Or their support could drop back again, polls can be temporary and fickle.

Gower reported Key “pleading with his supporters for help” (like every other party does in an election campaign) – Key: “if there’s any National supporter that was thinking about the fact that they might be able to stay at election day leave it up to everyone else, um, maybe this poll will give them a wake up call”. 

As for Internet-Mana, they are steady on 2.1%…

I think that’s quite notable. Internet-Mana were 0.2 in May and jumped to 1.8% in June when they combined and started campaigning, but since then have got 2.3 (July) and 2.0 (early August). Apart from one poll (Stuff/IPSOS having them up to 4% indications are Internet-Mana may have plateaued. Roy Morgan poll fortnightly and have always had them in the 1-2.5% range.

…and the other minor parties simply doing nothing and not worth talking about.

Small party support often moves late in an election campaign but Gower is totally dismissive, choosing to exclude them from any coverage. This makes it very difficult for small parties to be heard.

Turning now to seats in the house, this is worth talking about…

Some of the minor parties could be critical in this so are suddenly worth talking about.

…look at this, National 57 plus it’s allies that we say would get electorate seats, that would not be a majority in a 122 seat Parliament…

It’s far from certain it would be a 122 seat Parliament.

…they would only have 61 seats.

If a lot of uncertain results turned out as predicted by 3 News.

Turning now to the left wing block, they would get to 53 seats. It would all come down to New Zealand First’s 8 seats…

Again, far from certain that this exact result would be replicated in next month’s election.

…they would be needed to give John Key the power to govern, but if Winston Peters went to the left it would be a 61 all draw, now this is a hung Parliament, a tie, nobody could govern.

Apart from the many variables and assumptions involved this wouldn’t necessarily result in a hung Parliament. For example NZ First or Greens could agree to guarantee confidence and supply and remain on the cross benches. As much as Gower would love his poll to be definitive there are many possibilities.

It’s a realistic election night result.

It’s a possible result, but a non-tied or non-hung Parliament is far more likely.

Key would have to give Peters what he wants…

No, they would negotiate, and even on these numbers 57 versus 8 is a strong position. Some concessions would be made but Peters couldn’t dictate everything, that would be democratically ludicrous.

…or Peters could force another election.

Possibly, if Greens don’t step in. But there would be a lot of pressure on Peters to enable a stable Government. If he forced another election he and NZ First would be at high risk of being booted out by voters.

So first dirty politics, and now a political mess.

This is dirty poll reporting with a mess of assumptions and poorly considered assertions.

The report then showed sound bite edit of interviews with Key, Cunliffe, Peters, Craig. Typically Russell Norman described it sensibly as “dynamic is how…it’s dynamic at the moment”. Not in Gowers world, just before this Gower repeated his hung Parliament dream. Then 3:45 into the item Gower says:

But the political reality is it comes down to Winston Peters prefers as Prime Minister…

That’s just one amongst a number of possibilities so it’s far from a reality.

…he could choose Key or force a new election.

Dirty politics changing everything.

Not really, two or three months ago when NZ First rose to 5.1% Gower was just as adamant that that changed everything.

So Key has suffered a blow from dirty politics but it is not fatal, he could still govern on this, he’d just need Peters would be unlikely to force another election knowing he would be punished by voters and probably annihilated if he did…

Why mention that now rather than in earlier predictions? 

…but just imagine the negotiations, a true nightmare scenario.

Gower’s nightmare is he would be shut out of the negotiations and would have to wait a week or two for a dramatic story.

And remember Labour on 26%, that’s a shocker, they are actually going backwards here.

So if they are going backwards by the time we get to the election the actual result could be different to Paddy’s poll?

Yet voters clearly think that Key has not handled dirty politics properly…

The poll does nothing to prove that.

So, dirty politics, a crazy campaign, and now chaotic results.

There is no result until September 20. The only thing that’s chaotic here is Gower’s reporting of just one of many pre-election polls.

Remember that polls ask “who would you vote for if an election was held today” – and the polling period is already historical. Polls don’t try to predict how people would vote on election day, they can’t look into the future. Neither should Gower.

There are 24 days to go.

That’s one accurate thing Gower has reported. He mentioned ‘dirty politics’ throughout his report but ignored this only fact in making all his dramatic projections.

Well informed voters rely a lot on media coverage. Reporting like this is grossly misleading and serves voters very poorly. It serves our democracy very poorly.

Gower is playing dirty politics with our election, putting his own ill-informed dramatics ahead of reason, facts and polling basics.

A political editor should be much better than this. 

UPDATE: And Gower is just now talking on Firstline admitting he got previous predictions wrong, particularly regarding Colin Craig and the Conservatives. He has mentioned that Conservative advertising may have affected this poll result.

 

National, Labour down in 3 News poll

Both Labour and National are down over 2% in the latest 3 News/Reid Research poll. NZ First and the Conservative have the biggest rises.

  • National 45% (-2.5)
  • Labour 26.4% (-2.6)
  • Greens 13.5% (+0.5)
  • NZ First 6.3% (+1.7)
  • Conservative Party 4.6% (+2.1)
  • Internet-Mana 2.1% (+0.1)
  • Maori Party 0.7% (-0.1)
  • UnitedFuture 0.4% (+0.2)
  • ACT Party 0.3% (no change)

This may reflect public annoyance at dirty politics – both the large parties tend to be blamed together.

Colin Craig claims a letter box drop of a campaign pamphlet is behind his party’s rise but that will be only part of the reason.

3 News: Latest political poll big blow for John Key

Details as per Curia:

Polling Company: Reid Research

Poll Method: Random Phone

Poll Size: 1,000 respondents (3.1% maximum margin of error)

Undecideds:

Polling Company: Reid Research

Poll Method: Random Phone

Poll Size: 1,000 respondents (3.1% maximum margin of error)

Undecideds:

Dates: 20 to 25 August 2014 approx

Client: 3 News

The new results aren’t on the table here yet but sit should be updated: http://www.reidresearch.co.nz/TV3+POLL+RESULTS.html

Follow

Get every new post delivered to your Inbox.

Join 237 other followers