How 538 got Trump wrong

May 25th, 2016 at 4:00 pm by David Farrar

Nate Silver writes on how he got Donald Trump so wrong (as almost everyone did). He notes:

But it’s not how it worked for those skeptical forecasts about Trump’s chance of becoming the Republican nominee. Despite the lack of a model, we put his chances in percentage terms on a number of occasions. In order of appearance — I may be missing a couple of instances — we put them at 2 percent (in August), 5 percent (in September), 6 percent (in November), around 7 percent (in early December), and 12 percent to 13 percent (in early January).

Silver notes five things:

  1. Our early forecasts of Trump’s nomination chances weren’t based on a statistical model, which may have been most of the problem.
  2. Trump’s nomination is just one event, and that makes it hard to judge the accuracy of a probabilistic forecast.
  3. The historical evidence clearly suggested that Trump was an underdog, but the sample size probably wasn’t large enough to assign him quite so low a probability of winning.
  4. Trump’s nomination is potentially a point in favor of “polls-only” as opposed to “fundamentals” models.
  5. There’s a danger in hindsight bias, and in overcorrecting after an unexpected event such as Trump’s nomination.

The interesting thing is that if you just looked at the polls, then you should have concluded Trump would win. He basically led in every poll for six months. But everyone found reasons to argue why the polls would change. 538 for example places great store on endorsements. And endorsements have been a good predictor in previous elections, but as Silver notes the sample size of previous elections is not great.

So one lesson from this is not to ignore the polls. They’re not always right, but polls vs assumptions, polls tend to win out.

Another source of info I look to is the prediction markets. At the moment they have Clinton at 66% likely to win and Trump 32%.

Latest poll

May 25th, 2016 at 7:00 am by David Farrar

A Newshub Reid Research poll was released yesterday. Details are at Curia.

The Newshub story says Key has plummeted as Preferred PM as he has dropped 1.5% in 6 months. This is of course not even statistically significant let alone a plummet of any kind.

I thought it would be useful to compare the Preferred PM ratings of May 2016, with May 2007 – the same point in Labour’s third term.

In May 2007 the PM was at 30% Preferred PM and in May 2016 the PM is at 37% Preferred PM.

In May 2007 the Opposition Leader was at 32% Preferred PM and in May 2016 the Opposition Leader is at 9% Preferred PM.

So Clark was trailing by 2% in May 2007, while in May 2016 Key leads Little by 28%.

Also for those interested in May 2007, National in Opposition was 12% ahead of Labour in the polls. In May 2016 Labour in Opposition are 16% behind National in the polls.

Yet Newshub trumpet their poll as bad news for National!

Latest poll

May 24th, 2016 at 6:52 am by David Farrar

I’ve blogged at Curia the latest poll (Roy Morgan). It basically reverses the previous month’s poll. Seat projection is CR 57 and CL 51 so NZ First would hold the balance of power.

Another poll shows Trump leading

May 23rd, 2016 at 3:00 pm by David Farrar

There’s been another poll out showing Trump ahead. It is of course early days, but it does show that Trump is consolidating his support.

This poll is the ABC/Washington Post poll – normally one that is highly reported.

Some interesting aspects:

  • Trump net favourability -17%
  • Clinton net favourability -16%
  • 51% say they want a third choice
  • Independents – Trump +13%
  • Moderates – Clinton +10%
  • Whites – Trump +24%
  • Non-Whites – Clinton +48%
  • Men – Trump +23%
  • Women – Clinton +14%
  • Under 40s – Clinton +8%
  • 40 to 64s – Trump +8%
  • Over 65s – Trump +3%
  • North East – Clinton +6%
  • Mid West – Clinton +4%
  • South: Trump +10%
  • West: Tie
  • Urban: Clinton +21%
  • Suburban: Trump +10%
  • Rural: Trump +35%

Beware of geographic breakdown of polls

May 22nd, 2016 at 4:00 pm by David Farrar

Stuff reports:

Hamilton residents and New Zealand First supporters are among four groups which want fewer migrants allowed in the country.

In a recent Colmar-Brunton political poll, people were asked about the number of migrants the Government should let in to the country.

The results showed 42 per cent of those polled in Hamilton, wanted the Government to let fewer migrants in.

Other groups polled which also wanted fewer migrants included 38 per cent of those living in small towns or rural areas, 53 per cent of Maori and 40 per cent of New Zealand First supporters.

Overall figures showed just over half (51 per cent) of New Zealand voters polled believed the number of migrants being let in is about right, close to a quarter, or 27 per cent, said the government should let fewer in and nearly a fifth, or 18 per cent say the Government should let more migrants in. Four per cent said they did not know.

This is a good example of why one should be wary of reading too much into some poll breakdowns.

The poll is of 1,000 people. Around 3.5% of NZ live in Hamilton so I’d assume only 35 people from Hamilton were polled. That has a margin of error of around +/- 17%.

Not a statistical tie

May 17th, 2016 at 3:00 pm by David Farrar

Politico reports:

A new poll of Georgia voters finds Donald Trump and Hillary Clinton nearly tied in a general election matchup.

The Atlanta Journal-Constitution poll shows Trump with a 4-point lead over Clinton, 45 percent to 41 percent, which is within the poll’s 4.26 percentage point margin of error.

The sentiments expressed by independents further contribute to the statistical tie between the two presumptive nominees.

It isn’t a statistical tie. I do wish media wouldn’t use that term. It is good media point out a lead may be within a margin of error, but not good when they suggest that it is effectively a tie.

For a poll of 822 voters, the chance that Trump is actually leading is 89.2% and the chance Clinton is leading is 10.8%. This is less than the normal 95% confidence interval so Trump is not necessarily leading – but 89% probability is not a tie.

50% is a tie and 95% is statistically significant. I’m not sure what the term is for probabilities between 50% and 95% but it is not a tie – ie a 50.1% chance you are leading is not the same as a 94.95 chance you are leading.

April Public Polls

May 16th, 2016 at 3:00 pm by David Farrar

The situation in NZ was:

9 years ago 3 years ago 1 year ago 3 months ago Last month This month
National 47% 45% 49% 47% 46% 46%
Labour 36% 33% 29% 28% 28% 27%
Greens 8% 12% 11% 14% 14% 12%
NZ First 4% 4% 7% 7% 9% 11%
Nat over Labour +11% +12% +19% +19% +18% +19%
Nat over Lab/Gre +3% -1% +8% +5% +4% +7%
Right Direction 55% 62% 60% 61% 58%
Wrong Direction 34% 27% 28% 29% 31%
Net Direction +21% +35% +32% +32% +27%
Preferred PM
National Leader 27% 39% 42% 40% 40% 39%
Labour Leader 37% 15% 11% 8% 9% 7%
NZ First Leader 3% 3% 10% 8% 9% 10%

National’s party vote is lower than a year ago but slightly higher than three years ago.

Labour’s party vote is lower than one and three years ago.

The Green’s party vote is around the same as a year ago.

NZ First party vote is up from a year ago and up from three years ago.

You can subscribe to the full newsletter at http://curia.us10.list-manage.com/subscribe?u=e9168e04adbaaaf75e062779e&id=8507431512 for fuller details of polls in NZ, the US, the UK, Australia and Canada.

Latest poll

April 26th, 2016 at 8:28 pm by David Farrar

Roy Morgan’s monthly poll is out and blogged here.

Quite different to Colmar Brunton. CB has National up while RM has National down.

Both CM and RM have Labour down, and well under 30%.

CM has NZ First down 1% while RM has NZ First up 3.5%.

KDS and polls

April 17th, 2016 at 10:00 am by David Farrar

Pete George notes a tweet from a pollster:

Andrew from Colmar Brunton just tweeted:

Wow this week’s poll! The criticism has been much worse than usual. NEVER happens when Labour support increases.

Labour is down 4 to 28%, the first time they have dipped below 30 since the 2014 election, and National is up 3 to 50% – see One News/Colmar Brunton April 2016.

The reaction from the left, apparent on Twitter and at The Standard, ranged from disbelief to  blame, of everything from bad or corrupt polling methods, misleading or corrupt media and John Key.

Hard core Labour supporters have now had nearly eight years of post-Clark frustration and disappointment and daashed hopes.

On current performances (of the party and of leader Andrew Little) this looks unlikely to change any time soon.

Labour has faded from a major party with a widely respected leader to a struggling party with diminishing status.

They are on to their fourth leader and their latest one seems to be heading towards failure, probably hastened by this week’s lurch into dirty politics.

It’s the problem again of people living in a bubble. They and all their friends hate John Key and adore Labour, so how can the polls possibly be right.

Of course the polls are not always accurate. At the last election they under-estimated support for National and over-estimated support for the Greens!

2014 NZ Election Study on the issues

April 14th, 2016 at 9:00 am by David Farrar

The final post looks at views of New Zealanders on various issues from the NZ Election Study.

The first two, and most important are has the Government done a good job and has the economy been good or bad.

73% of NZers said the Government had done a good job in the last three years and only 21% said a bad job.

That’s a staggeringly low level of people saying the Government had done a bad job.

NZES asked about whether we should be spending more,  less or the same in various areas. The table below summarises this:

nzesspending

Health and education stand out as the two areas NZers most want more spending. This is no surprise. They have a net 65% support.

After that is housing and law enforcement on around net 40%.

Then environment on 34%, superannuation 27% and business & industry 19%.

The three areas where NZers want less spending is defence at -1%, welfare -20% and unemployment benefits -29%.

Then they asked NZers if they agreed with a series of statements, that are below.

nzesissues2

The biggest agreement at net +63% was people should have to work for the dole.

Next biggest agreement was that Government should subsidise or assist companies with research and development followed by income inequality is too large and should be reduced, unions are necessary to protect workers and exporters should get financial assistance.

NZers also think big business has too much power, we should assist international sportspersons and film makers and SOE privatisation has gone too far.

There is modest agreement that trade unions are necessary to protect workers at +9% and raise the super age to 67 at +4%.

Minor disagreement at -1% that many on welfare don’t deserve help, that NZ needs a Capital Gains Tax, unions have too much power (-3%) and lowering benefits helps people stand on their own feet (-4%).

More significant disagreement that the Government should help banks in times of crisis (-15%), we should have more immigration (-35%) and abortion is always wrong (-40%).

Also some voting issues were canvassed.

  • On compulsory voting 44% were in favour and 53% against.
  • Lowering the voting age to 16 had 7% support and a massive 90% opposition.
  • 48% supported keeping the Maori seats and 39% opposed.
  • 35% would vote on the Internet if they had a choice, while 595 would still choose a polling place.
  • 45% were confident that Internet voting would be secure and private and 46% were not.
  • 14% were more likely to vote if they could vote online and 10% said they were less likely to vote.

Huge opposition to reducing the voting age to 16.

2014 Election Study on Leaders

April 13th, 2016 at 9:00 am by David Farrar

Now we look at the favourability ratings for the party leaders in 2014. Again first those who have a favourable view of each leader.

leaderfav

No surprise that Key had the highest favourability at 60%. But considering this was done after two months of Hager and Dotcom allegations against Key, still a remarkable level.

The 2nd most popular leader is Winston. 30% of NZers like him. Will be interesting to see how that breaks down by which party they voted for.

The two Green leaders were next – Turei on 22% and Norman 20%.  In between then was David Cunliffe on 21%. When only one in five NZers like the alternative Prime Minister, then the outcome may not be a huge surprise.

Flavell does quite well for a minor party with 18% favourability, then Dunne on 13%, Craig 12%, Harawira 10%, Harre 7% and poor Jamie Whyte last on 4%.

leaderunfav

This shows how many New Zealanders dislike each leader. The leader with the lowest level of dislike was John Key at 28%. One can have high favourability and unfavourability (think Muldoon), but this shows Key didn’t have a high level of dislike in 2014 – lower than any other leader.

Next lowest was Flavell on 32%, then Turei 37%, Norman 38%, Peters 42%, Whyte 42% and Dunne 46%.

Colin Craig had high unfavourability at 49% but this was less than David Cunliffe at 54%. That is a very high level of unfavourability for the proposed alternative Prime Minister.

The two most unpopular leaders were Laila Harre with 60% dislike and Hone Harawira with 64%,

leadernetfav

This graph shows the net favourability for each leader. Key was the only leader whom more voters liked than disliked – at +32%.

The least unpopular was then Peters at -12%, Flavell -13%, Turei -15%, and Norman -18%.

After that Cunliffe was at -32%, Dunne -33%, Craig -37% and Whyte -39%.

Finally Harre on -53% and Harawira on -54%.

A pity they didn’t survey opinion on the true Internet Party Leader, Kim Dotcom – I suspect would be even lower.

Houston we may have a problem

April 12th, 2016 at 1:00 pm by David Farrar

labopp

This is the Preferred PM rating of the four Labour Opposition Leaders. Rather speaks for itself.

2014 Election Study on Party Position

April 12th, 2016 at 9:00 am by David Farrar

Another question asked in the 2014 Election Study was where voters say each party on a 0 to 10 scale where 0 is hard left and 10 is hard right. They were also asked where they assessed themselves.

Party LR

ACT was assessed as the most right party with a median score of 8. Both the Conservatives and National had a median score of 7.

The median voter score was a 6 – so slightly centre-right. United Future was placed there also.

NZ First is in the middle on 5. The Maori Party slightly centre-left on a 4.

The Mana and Internet parties were judged to be hard left.

I find it interesting that Labour and the Greens both have a median of 3. Once upon a time the Greens would have been seen as more to the left of Labour, but now they are seen as not far apart.

Now consider all the left activists who claim Labour need to go further to the left in order to win. They are already three away from the median vote on six. National is only one away from the median voter. So going further to the left for Labour just makes them look more extreme.

As the median voter is a six, then a three is as far away from them as a nine is.

For Labour to win, they either need to shift the median voter from a six to a five (very difficult to do) or they need to shift the perception of themselves from a three to a four. I doubt they will, and they will be surprised when they fail again.

2014 Election Study on Party Favourability

April 11th, 2016 at 9:00 am by David Farrar

Every election a major study/poll is done by the academics involved in the NZ Election Study. They’ve started to release data from their 2014 study and there is a trove of interesting data, which all parties should be looking at (if they want to increase their vote).

The first set of data I want to look at is party favourability. Respondents were asked to assess how much they like each party on a 0 to 10 scale. So a 0 to 4 is unfavourable and a 6 to 10 is favourable.

So how many people had a favourable opinion of each party? This is an important measurement as it effectively shows the current maximum likely support for each party. If you don’t like a party, there’s no way you’ll vote for it. If they do like the party, they may vote for it.

PartyFav

National is viewed favourably by 58% of NZers. That helps explain why 47% voted for them.

Labour is viewed favourably by just 35% of NZers. This is very significant as unless they can fundamentally change the brand and image of their party, that is their effective lid.  And as they really need to get into the high 30s to be able to form a clear centre-left Government. So they have a favourability rating lower than the minimum they need to win.

The Greens are also viewed favourably by around a third of NZers. This is possibly the same third who view Labour favourably, which suggests Greens could get more votes off them. We won’t know for sure how much they overlap until NZES release the full data set.

I was fascinated that 15% had a favourable view of the Conservative Party. If things hadn’t fallen apart after the election, this indicates they had definite potential to break the 5% barrier. Also of interest is the Maori Party have 19% who view them favourably – suggesting again they could pick up more votes.

ACT and United Future both have less than 10% seeing them favourably. The change of leader of ACT may have changed that, but regardless a real challenge.

Now we turn to the Unfavourability – how many NZers say they don’t like that party.

PartyUnFav

National has the least unfavourable – only 28% of NZers dislike National. This will come as a surprise to hard left activists who live in a bubble where 100% of their friends dislike National.

At the other extreme the Internet and Mana parties were massively disliked and toxic. The thought of them deciding the Government was a turn off for many voters.

ACT have the next highest unfavourability at 55%. Not a big surprise with the Banks prosecution.  Also on 50% are the Conservatives.

All the other parties are in the 40s for Unfavourability.

Now we look at the net favourability – those who like less those who dislike.

PartyNetFav

Only one party has positive net favourability – National at +30%.

Labour and Greens are both slightly negative net favourability at -5% and -7%.  Following them is NZ First on -17%, and Maori Party -22%.

The next five are Conservatives on -35%, United -38%, ACT -37%, Mana -57% and Internet Party -72%.  The Internet Party may be the most unpopular party in the history of NZ politics.

Overall the data shows the huge gulf between National and other parties when it comes to both favourability and unfavourability.

Latest poll

April 10th, 2016 at 6:31 pm by David Farrar

I’ve blogged at Curia the results of tonight’s One News Colmar Brunton poll.

It is almost half way through National’s third term, so it is useful to compare it to the same poll halfway through National’s second term.

In April 2013 National was at 43% and Labour 36% – a 7% gap.

In April 2016 National is at 50% and Labour 28% – a 22% gap.

This is the lowest Labour has been since the election, and also the highest National has been. And taken after the flag referendum which Labour were convinced would damage National, so they opposed change against their own policy. Nice outcome guys.

Key Little is also at 7% Preferred Prime Minister. This is the lowest for a Labour Leader since May 2010. He is now 3% behind Peters.

Labour said in December 2014 their aim was to be polling at 40% by the end of 2015. It’s April 2016 and they’re in the 20s.

If I was a Labour MP, I’d be asking what exactly do they think will change in the next 18 months, so it isn’t a repeat of the last 18 months?

 

Public Polls March 2016

April 6th, 2016 at 11:00 am by David Farrar

pollsmar16

Curia’s monthly newsletter is out. The executive summary is:

There was one political voting poll in March 2016 – a Roy Morgan.

The average of the public polls has National 18% ahead of Labour in March, the same as February. The current seat projection is centre-right 58 seats, centre-left 51 which would see NZ First hold the balance of power.

We show the current New Zealand poll averages for party vote, country direction and preferred PM compared to three months ago, a year ago, three years ago and nine years ago. This allows easy comparisons between terms and Governments.

In the United States Obama now has positive approval ratings for the first time in many years. The prediction markets have Trump at 44% likely to win the GOP nomination and Cruz at 36%. Clinton is at 83% to be the Democratic nominee.

In the UK polls show Remain just 3% ahead of Leave in the EU referendum.

In Australia the Coalition remain ahead of Labor but Turnbull’s approval rating continues to decline.

In Canada the Liberals have dropped 4% this month but remain well ahead in the polls.

We also carry details of polls on the NZ flag and cannabis 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 go to http://curia.us10.list-manage.com/subscribe?u=e9168e04adbaaaf75e062779e&id=8507431512 to subscribe yourself.

Public Polls February 2016

March 11th, 2016 at 12:00 pm by David Farrar

pollsfeb16pub

The newsletter is out:

There were two political voting polls in February – a Roy Morgan and a One News Colmar Brunton.

The average of the public polls has National 18% ahead of Labour in February, down 1% from January. The current seat projection is centre-right 61 seats, centre-left 50 which would see the Maori Party hold the balance of power.

We show the current New Zealand poll averages for party vote, country direction and preferred PM compared to three months ago, a year ago, three years ago and nine years ago. This allows easy comparisons between terms and Governments.

In the United States Trump is a 4/11 favourite to win the Republican nomination against Cruz at 7/2 and Rubio 20/1.

In the UK polls show Remain just 4% ahead of Leave in the EU referendum.

In Australia the Coalition’s lead over Labor has decline, along with Turnbull’s approval ratings.

In Canada Trudeau has declined slightly, but the Liberals remains dominant on 49%.

We also carry details of polls on NZ police pursuits and TPP 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 go to http://curia.us10.list-manage.com/subscribe?u=e9168e04adbaaaf75e062779e&id=8507431512 to subscribe yourself.

1st post TPP poll has bounce for National

February 20th, 2016 at 7:00 am by David Farrar

Roy Morgan released their monthly poll yesterday (on Curia here) and the first public poll post the TPP protests has National up 1.5% to 48.5% and Labour languishing on 27.0%.

My opinion is that the mnature of the protests alienate most New Zealanders and only appeal to those already hostile to the Government. For others, they are a big turn off.

January Public Polls

February 1st, 2016 at 11:00 am by David Farrar

janpolls

The January Curia newsletter is out. The summary is:

Curia’s Polling Newsletter – Issue 93, January 2016

 

There was one political voting poll in January – a Roy Morgan.

 

The average of the public polls has National 19% ahead of Labour in January, down 1% from December. The current seat projection is centre-right 59 seats, centre-left 51 which would see the Maori Party hold the balance of power.

We show the current New Zealand poll averages for party vote, country direction and preferred PM compared to three months ago, a year ago, three years ago and nine years ago. This allows easy comparisons between terms and Governments.

In the United States as voting in primaries is about to start Donald Trump leads by 7% in Iowa, 215 in New Hampshire, 16% in South Carolina and 14% in Nevada.

On the Democratic side Clinton leads by 4% in Iowa, Sanders by 13% in New Hampshire and Clinton by 30% in South Carolina.

In the UK there is only a 6% chance of a Labour-led Government.

In Australia since the accession of Malcolm Turnbull, the Coalition has maintained a strong lead over Labor, with an election due within the year.

In Canada despite gloomy economic news, a plurality of Canadians think Canada is heading in the right direction.

We also carry details of polls on US ship visits 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 go to http://curia.us10.list-manage.com/subscribe?u=e9168e04adbaaaf75e062779e&id=8507431512 to subscribe yourself.

d

Labour in Roy Morgan polls

January 26th, 2016 at 7:00 am by David Farrar

RM Labour

The 1st poll of 2016 saw Labour at 27.5%. This is after we hear what a great year Labour had last year from the media.

The graph above shows Labour in the Roy Morgan poll since they started in 2005.

It is no surprise that they are lower than when they were in Government but they are lower than most of the time they have been in opposition.

In January 2010 they were at 32% and in January 2013 they were at 31.5%. These are at the same stage of the electoral cycle.

So they are polling 4% to 5% worse than they were at the same stage as their 1st and 2nd terms in opposition.

And how were National one year into their third term in opposition?

They were polling 42.5%.

December public polls

January 12th, 2016 at 7:00 am by David Farrar

decpolls

Just done my monthly polling newsletter.

This graph of the last three years is quite telling. National is polling around 5% higher than three years ago and Labour around 5% lower.

A trick question poll

December 26th, 2015 at 10:00 am by David Farrar

The Guardian reports:

More than 530 Republican primary voters were polled this week on their support for Republican candidates and foreign policy issues including banning Muslims from entering the US, Japanese internment camps from the second world war and bombing Agrabah, the kingdom from Disney’s animated classic, Aladdin.

In its poll, Public Policy Polling asked the 532 Republicans: “Would you support or oppose bombing Agrabah?” While 57% of responders said they were not sure, 30% said they supported bombing it. Only 13% opposed it.

Public Policy Polling also polled Democratic primary voters: only 19% of them said they would support bombing Agrabah, while 36% said they would oppose it.

People are getting excited about this, but it doesn’t really say a lot except that people don’t like to admit they don’t know where a place is.

I suspect most respondents just thought Agrabah is a city in Syria. And the question was asked as a direct support/oppose. The majority of Republicans actually said they were not sure, which is what everyone should say.  Also of note is 19% of Democrats said they support bombing it.

Thomas Lumley at Stats Chat damns the poll question:

I’m pretty sure that less than 30% even of Republican voters really support bombing a fictional country. In fact, I’d guess it’s probably less than 5%. But think about how the question was asked.  You’re a stereotypical Republican voter dragged away from quiet dinner with your stereotypical spouse and 2.3 stereotypical kids by this nice, earnest person on the phone who wants your opinion about important national issues.  You know there’s been argument about whether to bomb this place in the Middle East. You can’t remember if the name matches, but obviously if they’re asking a serious question that must be the place they mean. And it seemed like a good idea when it was explained on the news. Even the British are doing it. So you say “Support”.

The 30% (or 19%) doesn’t mean Republicans (or Democrats) want to bomb Aladdin. It doesn’t even mean they want to bomb arbitrary places they’ve never heard of. It means they were asked a question carefully phrased to sound as if it was about a genuine geopolitical controversy and they answered it that way.

When Ali G does this sort of thing to political figures, it’s comedy. When Borat does it to unsuspecting Americans it’s a bit dubious. When it’s mixed in with serious opinion polling, it risks further damaging what’s already a very limited channel for gauging popular opinion.

I agree.

YouGov on why their UK polls were wrong

December 21st, 2015 at 10:00 am by David Farrar

YouGov have published why they think their polls were out in the UK election. Their conclusions:

The younger age range within the samples over-represent those who are more engaged in politics and are therefore more likely to vote. As younger people, they disproportionately supported Labour, so having too many young voters in our likely voter sample skewed the overall result towards Labour. We believe we had the party voting proportions for this age group correct but that fewer of them actually voted than our sample suggested. This can be corrected in the future in two ways: a) interviewing the correct proportion of people who are less interested in politics, and b) weighting the sample to the expected turnout for different demographic groups. The problem with both of these is that, unlike in the US where detailed exit poll data is publicly available, in the UK no detailed information is available by which we can know the correct target proportions for each age group. However, we can make better estimates of them.

Youth turnout is low almost everywhere. And if you only get to poll the politically motivated youth, then you will over-estimate their likely turnout.

In NZ the Electoral Commission has released turnout by age, so pollsters should be able to take this into account when weighting.

The oldest demographic group, the over-seventies, were under-represented in our samples. They voted disproportionately for the Conservatives, and having too few of them in our samples skewed it slightly against the Conservatives. This can be corrected in the future in two ways: a) interviewing the correct number of over-seventies, and b) weighting the over-seventies in our samples to the correct target weights.

Elderly people vote far more than younger voters.

In NZ only 62% of under 30s enrolled, voted. For over 65s it is over 85%.

One cannot discount misreporting (“shy Tories”), but we can find no direct evidence for it. In this election, polling showed dissonance between the outcome which people (in aggregate) said they wanted, and their underlying party preference. There was a strong overall preference for a Cameron-led government over an SNP-influenced government led by Labour leader Ed Miliband, although stated voting preferences would not have delivered that. It is possible that this led to some respondent misreporting, if people wanted to express their party preference and not their actual tactical vote, but it is impossible to establish this objectively as we can never know how individual respondents really voted.

This is that basically people changed their mind at the last minute as some Labour voters didn’t want Labour propped up by the SNP so chose Conservative as the lesser evil. Same in NZ where some left voters hated the idea of a Labour Government propped up by Kim Dotcom, so voted National.

Final poll for year has National at 51%

December 15th, 2015 at 6:18 am by David Farrar

I’ve blogged at Curia the results of the Herald DigiPoll which should be the final poll of 2015.

Their seat projection has National with 62 seats, enough to govern alone, and Labour/Greens on 47 seats.

The interesting comparison I like to make is with three years ago, as you are comparing similar points in the electoral cycle.

In the average of all polls, the change from December 2012 to December 2015 is:

  • National up 4.3%
  • Labour down 4.2%
  • Greens down 0.1%
  • NZ First up 1.5%

Final public poll for the year

December 11th, 2015 at 1:00 pm by David Farrar

Roy Morgan have released what is probably the final public poll for the year.

The seat projections on their poll are:

Centre-Left

  • Labour 34
  • Greens 16
  • Mana 0
  • CL 50

Centre

  • NZ First 7
  • Maori 2
  • C 9

Centre-Right

  • National 59
  • ACT 1
  • United Future 1
  • CR 61

Not bad for the start of the eight year of office.