I find these books so fascinating, I was even reading it during the speeches!

Bryce Edwards has reviewed (h/t iPredict Blog) a chapter by Shaun McGirr and Rob Salmond on what sources of information best predicted the election outcome. Was it an individual poll, the iPredict markets or the polls of polls.

The amount invested in iPredict was considerable:

• $64,500 was traded over the likely nature of ‘the Maori Party’s post-election relationship with National’
• $25,800 was traded over the Wellington Central battle between Grant Robertson and Stephen Franks
• $132,100 was traded over whether ‘there will be a National prime minister after the 2008 election’
• $413,000 in total was invested in election-related predictions

And how did iPredict do”

So, how accurate was iPredict in 2008? McGirr and Salmond conclude that although iPredict overestimated the eventual support for both Labour and National, it was more accurate any individual polling company.

And the individual polls:

In reality in 2008, McGirr and Salmond found this to be the case – with Colmar Brunton and DigiPoll exaggerating public support for National, and Roy Morgan exaggerating support for Labour (p.264).

So which polling companies were most accurate and inaccurate? McGirr and Salmond say that TV3’s TNS poll was the best (as it was in 2005 as well), and Fairfax’s Neilson pool was the poorest.

The TV3 poll is the one that currently shows a 27% gap! Mind you they are now with Reid Research, so there may be a different methodology used now.

Then they look at the polls of polls published by three outlets – NZPA, Rob (at 08 Wire) and myself (at curiablog).

In addition to the five opinion polls, some observers attempted to average out the idiosyncratic errors of the individual polls by aggregating them into a “poll-of-polls” using different methods. The New Zealand Press Association simply took the average of the estimates of the six most recent polls, while The New Zealand Herald took the average of the last four polls. Two blog-based polls-of-polls – one run by David Farrar of New Zealand’s premier political blog Kiwiblog, and one hosted at a smaller blog [run by author Rob Salmond] called 08wire – weighted more recent polls with larger sample sizes more heavily (p.257).

And how did the poll of polls do?

McGirr and Salmond say that ‘Poll-of-polls consistently performed well during the 2008 campaign, outperforming most of the opinion polls and the prediction markets’ (p.270). They therefore advocate that both the media and public should pay much more attention to this highly accurate source of political information.

Tis has prompted me to update the poll of polls widget, which is below.

Salmond ranks the different outlets for their accuracy to the final result. In order they were:

1. NZ Herald poll of polls 6.1 (error from result)
2. NZPA poll of polls 6.8
3. Curiablog poll of polls 8.1
4. TV3/TNS poll 9.6
5. 08 wire poll of polls 13.6
6. iPredict 15.7
7. TVNZ/Colmar Brunton poll 16.8
8. NZ Herald/Digipoll poll 19.8
9. Roy Morgan poll 20.8
10. Fairfax/Neilsen poll 29.6
11. NZ Political Stockmarket 109.5

The NZ Political Stockmarket used virtual money, so it shows what a difference real money can make.

The authors conclude that media outlets should not just report the individual poll results when they commission a poll, but also publish regular info on a poll of polls and on iPredict.

Incidentally I will probably review and tweak the curiablog methodology a bit when I have some spare time.

So far we have:

• Green’s 2008 campaign
• Labour’s 2008 campaign
• National‘s 2008 campaign

How well were electronic forms of politics utilised in last year’s general election? How effectively did the political parties and electorate candidates use websites, email, social networking in their campaigning? What about bloggers and the mainstream media? These questions are addressed in a chapter by Peter John Chen about ‘the role, use and impact of online media in New Zealand’s 2008 election’, published in Informing Voters? Politics, Media and the New Zealand Election 2008 (edited by Chris Rudd, Janine Hayward and Geoff Craig of the University of Otago Politics department). This blog post is the fourth of a series of explorations of the chapters from the new book (which I also have a chapter in).

Most of the focus is on how parties and candidates used online media, rather than the role of blogs by non candidates. Still very interesting. Labour gets caned for their 2008 e-campaign. No surprise as they had three different websites running.

They also reveal Labour spent around 10% of its advertising budget online and National spent zero.

Dr Catt mentioned that a paper was due out today on how a text reminder on election day significantly increased vote turnout. The paper is here.

There were 15,662 who had texted the EEC requesting an enrolment pack.Some of them were sent a text message on election day reminding them to vote, and some were not.

The turnout rate of those who got a text message was 75.5% – a significant 4.7% higher than the 70.8% who did not get a text message.

The biggest increase in voting was in low population density electorates (rural), electorates with high Maori populations and amongst people enrolled in the last month – they had a greater than 10% difference

I was amused that a few people replied to the text message, and this was one response:

HEY ORANGE GUY WE HAVIN FLAT WARMING 2NYTE BRING THE ORANGE JUICE AND ALL YR FRIENDS.

Anyway the results are strong enough that I think the Government should seriously look at doing a mass text on election day to enrolled voters. This would mean the EEC would have to collect cellphone numbers, but that can be done with a small law change.

1. Labour got $431,917 in big donations. Their biggest individual donor was the Vela Family who gave them $100k. The unions gave $117,500 and individuals (mainly from arts/culture sector) gave $134,830. Also a total of $60,000 from corporates and lobby groups, $10,587 from Helen Clark and $9,000 anonymously through the Electoral Commission.
2. ACT got $315,906 in big donations. John Boscawen gave $101,000, Alan Gibbs $200,000 and $14,906 from Virtual Bucket Ltd!
3. National got less than half as much as Labour in big donations. They got $207,001. John Key put in $30,000, some other individuals $30,001, $70,000 from corporates and a lobby group, and $77,000 anonymously through the Electoral Commission.
4. Greens were 4th largest for total big donations on $184,693. These were all individual donations with MPs putting in 70,725 and others $114,238.
5. NZ First got (assuming this return is accurate – the last three years were not) $111,999. The Velas gave $100,000 to their favourite boy and Sir Patrick Hogan also gave his thanks with $11,999.
6. The Family Party had $88,044. Destiny Church handed over $10,926, Paul Adams $41,037 and an Elias Kanaris $36,081

National’s level of “big” donations is very small. Part of that will be that some gave in 2007 (2011 will be interesting) but part is that the vast majority of National’s revenue comes from smaller donations. The members alone contribute between $1.5 and $2.0 million a year in small donations I would estimate.

Also somewhat amusingly, NZ First broke the law – again! They received a $2,190 donation from a foreigner on 22 October 2008. As it exceeds $1,000 it has to be paid to the Electoral Commission within 20 working days. They only paid it over on 12 December 2008.

Labour also received an overseas donations in excess of $1,000. They got $10,000 which they had to give up.

This spreadsheet above looks at what correlation there is between advertising spend on behalf of a party, and how many votes they got. The first column is the party’s declared advertising (it does not include candidate advertising) and the second column is how the costs of any broadcasting paid for by the Electoral Commission. So the third column is the total advertising spend on behalf of the party, and then divided by the number of votes they cost, what he spend per vote is.

The variation is huge – from 29c a vote to $55 a vote. Take the two extreme minor parties. The Bill and Ben Party got 13,016 votes with under $4,000 of spending. While Social Credit got one tenth as many votes despite spending 20 times as much money.

Most focus will be on the parties that do make it into Parliament, or have a serious chance of doing so. Of those National spent the least per vote – $3.04. So hardly buying your way to power.  ACT spent the most at $14.57 a vote, followed by Progressive at $12.80 a vote.

The Greens and NZ First both spent over $10 a vote, two to three times that of Labour and National.

Money is useful in politics. But it is far from a dominant factor. Policies, leadership, media reporting, volunteer effort, membership levels, smart use of IT, MPs behaviour all have (in my opinion) a greater influence on electoral outcomes than merely money spent.

• Total candidate expenditure is $2.26m
• Total disclosed donations to candidates is $1.26m
• Average spent for a winning candidate is $12,836
• In only 38 out of 70 (54%) electorates, did the candidate spending the most money win!!
• The top five spending candidates all lost – Russell Fairbrother, Paul Adams, Nicky Wagner, Ron Mark and Stephen Franks
• Only 2 of the top ten winning candidates won their seats, and only seven of the top 20.

Bryce also has calculated the average spending per candidate for each party.

The amount spent by candidates on Internet advertising was interesting for me. The top spenders:

1. Charles Chauvel $5,551
2. Jills Angus Burney $2,658
3. Brendon Burns $2,250
4. Pita Sharples $,2000
5. Aaron Gilmore $1,318

What did Charles spend $5,551 on?

The facts booklet has 22 pages of info, and the stats has lots of useful stuff, including:

• Enrolments stats
• Turnout Stats
• Voting stats (did you know National won the 120th seat in the last three elections?)
• That United Future was just 504 votes off an extra seat in 2002, Greens 1,250 off another in 2005 and Labour just 39 votes off an extra in 2008
• Split Voting stats
• Candidate stats (160 list only cands)
• MPs stats (in last three elections a total of 59 MPs have lost their seats)
• Electorate stats (they have ranked them into quartiles for various parties)

You can spend hours in this stuff!

It is only six pages, but a very good summary.

The study is based on an analysis of the leading New Zealand newspapers and television news programmes (TV1 and TV3 evening news) over an eight-week time period. The final results are uploaded here as a pdf – new-zealands-media-coverage-of-the-2008-election-study-final-results

The results:

Overall Media coverage:

Labour 38%, National 34%, Maori 8%, Greens 7%, NZF 6%, ACT 4%, UFNZ 2% and Progressive 1%

Negative Media Coverage

1. National 38%
2. NZ First 37%
3. Labour 36%
4. Progressive 31%
5. ACT 26%
6. United Future 21%
7. Greens 18%
8. Maori 17%

So National had a higher proportion of negative stories than NZ First!

Net Positive less Negative Coverage

1. Greens +14%
2. Maori +13%
3. United Future +1%
4. ACT -2%
5. Labour – 10%
6. Progressive – 12%
7. National – 16%
8. NZ First – 16%

This suggests to me the Greens and Maori Party were not held to the same level of scrutiny as other parties.

What I found interesting was the further breakdown that found news stories were more negative on National, but analysis stories more negative on Labour.

They also compared total media coverage to the result a party got at the election. The difference between the two was:

1. Maori +6%
2. Labour +4%
3. NZ First +3%
4. United Future +2%
5. Greens +1%
6. Progressive 0%
7. ACT -0%
8. National – 11%

Then we have the tone of media coverage of the leaders. And the net positive over negative was:

1. Anderton +14%
2. Turia/Sharples +9%
3. Dunne +5%
4. Clark +3%
5. Fitzsimons/Norman -4%
6. Key – 7%
7. Hide – 15%
8. Peters -19%

In the preliminary report, Key and Hide had more net negative coverage than Winston. The re-emergence of the Owen Glenn affair obviously resulted in a change for the last two weeks.

A breakdown by media, finds that Key got much more negative coverage on TV than Helen Clark.

In terms of issues, the major ones were:

• Economy 28%
• Law & Order 10%
• Tax 9%
• Maori Issues 7%
• Education 7%
• Immigration 6%
• Health 6%
• KiwiSaver/Super 6%

A very interesting, easy to read report.

2008-final-election-results

The file is around 3 MB and in MS Word. Almost every piece of data you can want is in there. I have not yet added on the split voting statistics though.

Apart from the detailed results per polling place, they have the vote splitting statistics. The summary tells us 70.4% did not split their votes and 29.6% did split. Of the larger parties, these were the percentage who split their votes (voted for a different electorate candidate to their party vote):

1. National 14.5%
2. Labour 22.5%
3. Maori 39.3%
4. Green 66.7%
5. United Future 75.8%
6. Progressive 77.6%
7. NZ First 81.9%
8. ACT 83.6%

National’s low level of vote splitting will help explain why they won so many more electorate seats than Labour. Now let us look at how those who split their votes, did so:

1. ACT – 72.6% voted for the National candidate, 4.9% Labour cand, 16.4% ACT cand
2. Green – 10.5% National cand, 47.2% Labour, 33.3% Green
3. Progressive – 16.8% National, 43.3% Labour, 22.4% Progressive
4. Labour – 5.0% National, 77.5% Labour, 6.1% Green
5. Maori – 6.5% National, 18.7% Labour, 60.7% Maori
6. National – 85.5% National, 3.5% Labour, 4.2% ACT
7. NZ First – 20.1% National, 43.2% Labour, 18.1% NZ First
8. United Future – 50.1% National, 12.3% Labour, 24.2% United Future

With the two largest minor parties (in Parliament) we see that 73% of ACT voters voted for the National candidate while only 47% of Green voters voted for the Labour candidate.

We also saw twice as many NZ First voters voted for a Labour candidate than a National candidate.

Also most importantly 86% of National voters supported the National candidate, while only 78% of Labour voters supported the Labour candidate. Even if you exclude the Maori seats this only increases to 79.6%.

One can also look at the split voting in each electorate. A few interesting ones:

• In Epsom, Rodney Hide got 89% of the ACT voters, 16% of the Greens, 21% of Progressives, 16% of Labour, 33% of Maori Party and 70% of National – over 15,000 National Party voters voted for Rodney.
• In Mangere 73% of Labour voters voted Sua, and 15% Field.
• In Nelson Nick Smith got 95% of the National voters, 22% of Labour voters and 19% of Green voters
• In New Plymouth Harry Duynhoven narrowly lost despite getting 94% of Labour voters, 13% of National voters, 19% of ACT voters and 80% of Green voters.
• In Ohariu Peter Dunne got 91% of United voters, 44% of National voters and 16% of Labour voters. In 2005 he got 34% of Labour voters.
• Otaki was a hard fought battle. Nathan Guy got 92% of the National vote and 4% of the Labour vote. Darren Hughes got 91% of the Labour vote and 6% of the National vote.
• In Tauranga Peters got 90% fo the NZ First voters but only 6% of the National voters and 46% of Labour voters. Sime Bridges got 88% of National voters and 11% of Labour voters plus even 5% of NZ First voters. He even got 27% of Green voters – double what the Labour cand got. Simon also got 30% of the Maori Party voters.
• In Wellington Central Stephen Franks got 84% of National voters, 77% of ACT, 4% of Labour and 6% of Greens. Robertson triumphed because on top of his 76% of Labour voters he got 57% of Green voters.  A quite large 16% of Labour voters chose Kedgley over Robertson.
• In WIgram Jim Anderton get 86% of Progressive voters, 58% of Labour voters and 21% of National voters.

It’s great fun analysing these statistics. Now if only we could get split voting stats per polling place!

The MPs will be sworn in on Monday 8 December.

And those who want a smaller Parliament will be delighted with the results – a Parliament of only 107 MPs!

How is this possible? It is an obscure feature called underhang – the opposite of overhang. Let us look at the results:

1. National 28.4%, 36 seats, 29 elects, 7 list
2. Labour 25.0%, 32 seats, 16 elects, 16 list
3. Bill and Ben Party 11.7%, 15 seats, 0 elects, 15 list
4. Greens 11.2%, 14 seats, 0 elects, 14 list
5. Legalise Cannabis 10.5%, 13 seats, 0 elects, 13 list
6. Maori Party 4.9%, 6 seats, 6 elects, 0 list
7. ACT 1.9%, 2 seats, 1 elect, 1 list
8. NZF 1.0%, 1 seat, 1 elect, 0 list
9. United Future 0.7%, 1 seat, 1 elect, 0 list

Now the Bill and Ben Party only had two candidates (Bill and Ben!) on their party list, so there would be an underhang of 13 MPs for a Parliament of 107. This means to form a Government you need 54 votes. So what would the Government be?

You could do National 36 + ALCP 13 + Maori 6 = 55

More likely is Labour 32 + Greens 14 + ALCP 13 = 59

The electorate seat won by NZ First was actually in Tamaki, so their sole MP would be a Doug Nabbs.

They have given results for each electorate also. Below are the seats where a different party would have won the electorate vote if the kids were the real electors:

1. Auckland Central – Labour
2. Bay of Plenty – Labour
3. Dunedin North – National
4. Invercargill – Labour
5. New Lynn – National
6. Rimutaka – National
7. Tamaki – NZ First
8. West Coast-Tasman – Labour
9. Whanganui – Labour
10. Wigram – National
11. Ikaroa-Rawhiti – Maori

It is great to see the kids encouraged to take part in mock elections and vote. The results do show though why the Greens are bonkers with their campaign to lower the voting age to include 16 and 17 year old schoold students.

SO SURE OF AN OUTRIGHT WIN WERE NATIONAL STALWARTS LIKE FORMER MP KATHERINE RICH THAT SHE REFUSED TO COMMENT LAST WEEK ON THE POSSIBILITY OF KEY NOT BECOMING PRIME MINISTER “BECAUSE IT’S JUST NOT GOING TO HAPPEN”
865 words
Nov 9th 2008 12:30pm Newspapers/Nz Herald

On page: 26

IT WASN’T meant to happen this way. So sure was John Key that he had this one in the box that he didn’t have an option two.

So sure of an outright win were National stalwarts like former MP Katherine Rich that she refused to comment last week on the possibility of Key not becoming prime minister “because it’s just not going to happen”.

Now Key has fluffed his big chance, not by losing outright but by allowing Helen Clark enough room to stick a Labour toe in a door that he was supposed to have slammed shut.

Now the voters can do nothing but wait while the negotiators do the deals that will ultimately decide who governs. While Key is a master dealer when it comes to investment banking, it is Clark who has an extraordinary ability to negotiate coalitions, partnerships and support agreements that most people wouldn’t have thought possible. If the Maori Party emerges as the kingmaker, punters are predicting Clark will have the upper hand.

Now the Dream Team, Key and his sidekick Bill English, face another three years in Opposition, another three years before the ambitious 47-year-old leader of the National Party gets another crack at what has been a lifelong dream.

And have another crack, he will.

Broadcaster and political commentator Bill Ralston predicts Key will be “pissed off to hell” with the loss but will treat it as a setback rather than a failure. He would use the next three years to hone his skills, Ralston said. “This is a guy who has never failed at anything.”

Party insiders doubt there is a chance Key will spit his dummy, quit politics and go back to the lucrative investment banking career that made him a very rich man.

That there will be no leadership challenge is in little doubt. “There is no question he will remain a strong uncontested leader of the Opposition,” MP Murray McCully said. “He will be unrivalled, unchallenged.”

He, like others, doubts Key had thought about a Plan B.

“He’s always been the sort of guy who from the time he thought he might take the job on has only ever had one scenario in mind.”

That innate confidence grew more evident as the campaign progressed, as Key’s media training paid off and as he took more risks _ like discounting a union with New Zealand First leader Winston Peters outright.

Ralston said the only way the National Party would lose Key now was if Caucus voted him out. “And I don’t think they have another alternative leader there. Bill English might think he is but the Caucus are aware that Key is the man.”

One National MP said he had never seen a more disciplined and unified caucus.

Whether English would even want the job is up for discussion. A source close to the inner circle doubted English would make a run for the leadership.

“He’s got the best of both worlds. He’s a family man, a committed Catholic with a busy working wife. He’s got plenty of power and influence without carrying the top dog title.”

And despite this loss, National’s polling leaves the party as a strong Opposition, a position Key is credited with achieving. In the next three years Key would have “more mana and authority than any leader of the National Party going back a long way,” one insider said.

Yesterday’s achievement in terms of the party vote would be recognised as being “substantially the result of his [Key's] leadership”, he said.

“His stocks are very high. He’ll have a good deal of authority.”

He predicted he would be a “powerful and untouchable leader of the Opposition.”

Observers say Key will waste no time in reshaping the National Party, purging the old guard in the process. Ralston expects to see a new lineup which includes younger faces and a more diverse mix in terms of gender and ethnicity before long. Many of the old guard will “take the hint and the nudge”.

“I think he will very quickly take a knife to the party… that’s going to be a risky process because at the same time you destabilise your Caucus support.”

That he’s capable of the hard decisions is unquestioned. Back in the 90s, Key earned the nickname “smiling assassin” after implementing mass sackings for Merrill Lynch.

Ralston expects heavy scrutiny by Key of what went wrong in the campaign because there was “no shortage of money this time round”.

That would include looking at what tactics didn’t work, the party marketing and organisation, and the billboards and television ads.

The “Mr Nice Guy” strategy would be called into question, he said, and whether the party should have used “attack” advertising and Key should have come across as more assertive.

Blogger Russell Brown of Public Address described the Labour victory as a “bugger the pollsters” moment. “In the end people have responded to the trust message. Maybe in the end they didn’t trust John Key enough.”

Ralston predicted that Key would spend the next three years making sure he earned that trust.
Source: Herald On Sunday
Credit: Herald On Sunday

Katherine was very smart in refusing to comment in advance on the possibility of National not winning. I presume all the other people quoted in the article did comment hypothetically.

Incidentally congrats to Katherine for her appointment as CEO of the NZ Food and Grocery Council. The FGC represents $15 billion of sales and $3.5 billion of exports, and Katherine’s background in agri-business make it a great role for her.