Psephology-o-rama: Hangover nerdery edition

Written By: - Date published: 1:51 pm, September 21st, 2014 - 23 comments
Categories: election 2014 - Tags: ,

polity_square_for_lynnReposted from Polity.

I was really privileged to be able to work with the TV3 election night team last night, providing some quantitative analysis on the results as they came in. One of the things we put together was a tool that could do demographic splits on the results while they were still uploading. I’m reasonably pleased with how the tool performed, in that it let TV3 do first cuts on how the vote came together before other media, and even while the count was still in progress. I mentioned a couple of these results on The Nation this morning.

Now that the battle for election night ratings is over for another three years, here is a table for your entrail-reading pleasure showing the three biggest parties’ vote shares in a few different geographic / demographic segments.

Labour National Greens
Area characteristic 2011 2014 2011 2014 2011 2014
Renters > 50% 41.6 39.7 32.7 34 11.6 11.7
Renters 35 to 50% 30.6 29.1 43.3 43.9 10.3 10
Renters 20 to 35% 22.7 20.7 52.8 53.5 10.4 9.7
Renters < 20% 18.5 16.7 58.7 59.5 11.2 10.1
Maori > 20% 34 34.2 35.4 35.8 8.3 7.9
Pacific > 20% 63.5 60.6 18.4 20.2 4.9 5.4
Asian > 20% 31.5 28.4 48.2 48.8 8.8 9.1
European > 80% 19.8 18.1 55.8 56.8 11.7 10.7
Degrees > 30% 23.1 21.5 51 51.9 16.3 16.3
No quals > 30% 35.9 34.6 37.6 38.7 7.6 7.1
Current students > 20% 30.7 28.4 44.4 44.8 14.3 14.3
Elderly > 25% 23 21.1 51.2 52.6 9.4 8.7
Kids > 25% 41.6 40.8 34.4 35.4 6.6 6.5
Non-religious > 50% 24.5 23 47 47.6 15.5 14.9
Christian > 60% 39.6 37.6 39.6 41.2 6.5 6.4
HH median income < $50k 45.6 45.6 25.5 26.6 8.3 8.2
HH median income $50k to $75k 31.3 29.1 43 43.8 9.7 9.1
HH median income $75k to $100k 21 19.3 55.3 55.9 11 10.4
HH median income > $100k 18.2 16.8 58.2 59.1 14.2 13.8
Auckland 29 26.7 49.7 50.1 9.1 9.3
South Auckland 52.5 50.1 27.6 28.8 4.6 4.9
Wellington 32.7 29.6 40 41.8 16.2 16.3
Christchurch 25 23.2 51.1 51 13 12.6
South Island 25.9 23.9 49.6 50.2 12.5 11.5
North Island 27.3 25.6 47.6 48.4 10 9.7
South Island town 31.4 30.1 41.6 41.8 14.6 13.1
North Island town 27.4 26.3 45.8 46.2 9.7 8.6

Notes

  1. These cuts are based on matching census data to booths, under the maintained assumption that people vote to a booth close to where they live. I realise that assumption is not always true, but it is true on average, which is all you really need to examine the over-time trends above.
  2. The correct interpretation is: “In areas of New Zealand with [area characteristic], the average booth-level party vote for [party] was X% in 2011 and Y% in 2014.”
  3. These data should NOT be interpreted at the individual-level (ie. Any claim about individual people, such “Asian New Zealanders moved against Labour by 2.9%,” cannot be supported by these data.

 

 


With apologies to Rob, I’m not going to add the phrase “Polity:” to the front of that mess of a title. How much did you have to drink last night?

23 comments on “Psephology-o-rama: Hangover nerdery edition ”

  1. karol 1

    Interesting stats. I never saw them last night as I was watching Maori TV. TV3 and TVNZ’s line ups were a turn off. All those right wing, mostly male, mostly pakeha commentators with a token female and a token leftie or two were a turn off.

    The stats show students went more for National than Labour. So maybe IMP’s targeting of students wasn’t such a clever approach?

    • Lanthanide 1.1

      “The stats show students went more for National than Labour. So maybe IMP’s targeting of students wasn’t such a clever approach?”

      It’s *very* tempting to make these assumptions, but as the post says, it’s actually about areas of voters that have that characteristic, not the voters themselves.

      For example, the Ilam electorate has a lot of students, but it also has a lot of rich white folk that vote National.

      It’s similarly tempting to say that families with children are very pro-Labour, but actually this is more likely confounded with the race – Pacific and Moari more likely to have larger families, and areas with Pacifica and Maori are shown to be very pro-Labour.

      • Yoyo 1.1.1

        Yes.Thus why these stats are next to useless. Not sure why anyone would choose to break them down this way. Far better ways to work out what students think.

        • Lanthanide 1.1.1.1

          Because these are the only stats we have.

          The US has lots of demographic information about votes, because they have exit-polling to capture it.

          NZ doesn’t do exit polling, at least nowhere near the scale that it occurs in the US.

          This data was compiled from matching census data with booth data.

    • according to my son..int/mana were invisible on ak uni campus..

      ..’the greens ‘were everywhere’..

      ..but not much int/mana..

      ..given free education was a cornerstone-policy..

      ..that wd seem to have been a fail…

      ..

  2. karol 2

    I would also like to see a gender break down of voting (and non-voting). I think the current very masculine style of politics is a turn off for a lot of women: ie the focus on economics over people and relationships; the focus on stats over stories about impacts on communities and relationships; some very aggressive attack style politics (part of the Lusk plan), etc.

    • Carol 2.1

      I strongly disagree with your comment. I know a considerable number of women who voted and in some instances even voted with their daughters this election. To say that women are so weak or shallow as to be turned off by a ‘masculine style of politics’ is utter nonsense. The women who fought for us to vote were ‘not put off’, they rose up and met the challenge and gave us a voice. We have not got weaker, we have got stronger. We know our own minds, and we are empowered by our democratic society to have our say on election day to ensure we speak for ourselves and our children. Those that didn’t vote on Saturday suffered from indifference, not because they were female.

  3. Lanthanide 3

    Few interesting points:
    1. Support for the Greens increases based on HH income
    2. Support for the Greens + Labour is slightly greater for $100K+ HH than for $75k-$100k
    3. Greens more support in the south island, Labour in the north
    4. National marginally dropped votes in Christchurch, while Labour and Green support went elsewhere (Internet? NZFirst?)

  4. Maori kept the faith with Labour.

    Labour better damn well keep that in mind when rebuilding.

    • Skinny 4.1

      Yes I agree Maori support was solid the policies hit home with Pacifica too. The Chinese bloc not so well.

  5. Once was Pete 5

    Actually, I thought his contribution to TV3 was very good, and it was surprising the level of agreement between Rob and David Farrar on what the numbers were saying.
    The interesting point about these numbers above is that Labour went back on almost every segment from 2011 to 2014 whilst National improved in all but one.

  6. Reddelusion 6

    There is misconception that all students are left leaning. The student union and left voice is definitely the loudest on campus but I am sure as it was in my day( and before anybody ask i paid fees and funded my own way through) the silent student majority hold very differnt views or just found the so call left leaders a bunch of show ponies

    • The Lone Haranguer 6.1

      Or, (as in my days of free University education) the vast majority of students say nothing and vote based upon their upbringing.

    • greywarbler 6.2

      @ Reddelusion 6.06
      You funded your own way through uni. It was hard to do that if you had to earn money as well as study, or work hard and save up to manage, or get a loan from Mum and Dad, or get a loan from a bank eager to lend to students. But years back you could get jobs fairly easily and the pay and working hours hadn’t been sliced and diced. So don’t be too self-admiring will you. You may be an exceptional person who did it all your own way, but others will have different strengths and abilities that matured at a different age to yours.

      And to be left leaning is usually to be applying idealistic intellectuality to one’s society thinking and questioning. Most students don’t set aside time for this, or maybe aren’t bright enough to consider anything beyond the memes they meet, or to manage studies and question other aspects of the culture.

  7. venezia 7

    Re TV coverage of election results – it was the same for me Karol. Maori TV was the most professional presentation but they covered only the Maori seats. TVNZ & TV3 were a real turn off. Egotist city. I watched TV3 with the mute button on, to see the rolling parade of results.

  8. The Lone Haranguer 8

    I found the TV3 coverage to be very good and was pleasantly surprised by John Campbells efforts.

    TV One was lamentable.

  9. bearded git 9

    Anyone know the percentage turnout this time and last time?

    • greywarbler 9.1

      @ Bearded git 9
      Just over 77 per cent of voters took to the ballot boxes this weekend – a small increase in turnout from the last election.
      More than 2.4 million people voted in this year’s election, representing a 77.04 per turnout.
      http://www.nzherald.co.nz/nz/news/article.cfm?c_id=1&objectid=11328886

      And from the Guardian (did anyone follow their presentation and compare it to our local media?)
      http://www.theguardian.com/world/live/2014/sep/20/new-zealand-election-live
      I’m manning our live coverage with the help of politics buffs Anna Rankin, Thomas Carnegie (@tomcarnegie_) and Craig Robertson (@CraigTRobertson), who between them have 15 years of the finest undergraduate education New Zealand’s tertiary institutions can offer.
      Out in the field is my colleague, Toby Manhire. He’s written for the Guardian extensively on the decision facing Kiwis today and will file this evening as soon as a result is known. Dr Bryce Edwards, a politics specialist from Otago University, will also be providing analysis throughout the next few hours.

      Inclement weather in parts of the country is expected to depress voter turnout, but the NZ electoral commission today revealed that the number of advance votes – those cast before today – number around 717,000, more than double 2011’s poll. That figure represents as astonishing 23% of enrolled voters.

      • bearded git 9.1.1

        thanks grey.

        I think a higher turnout would hve helped the left.

        • greywarbler 9.1.1.1

          Bearded git
          Perhaps a message to those glued to their cellphones, do they receive ads? Coffee shop meetings with half price coffee by the shops on slow nights, as a promotion for themselves and the Party. And some music and discussion and questions and quips to keep it good humoured. Getting to the young, important. Trouble is 30 years after Rogernomics has allowed a generation and a half to evolve a confused idea of what life and politics is about, all mixed up with celebrity people with cynical views, and a desire to take the piss out of anybody if they are blokes, and to be a sex symbol well-spoken clothes horse if they are women. Almost as if the wave of feminism never happened or got to tv.

          This is a bit way out but what a contrast.

  10. Adrian 10

    For Beared Git, if you consider that around 3.3 million were eligible but Judith Collins ( in charge of getting the vote out ) seems to have deliberatly done a bad job, as her minions only managed to sign up 20,000 over 2011 figures, but they seemed really proud of that appalling effort, falling about 300,000 short.
    That meant that turnout was only 69.8 % of eligible voters. Don’t believe anything that The Herald or Fairfax say to the contrary.
    The data is impressive but the student stuff is dodgy I think because most voting students live away from their home electorate temporarily, I’ve got three.
    They are all special votes that are not counted yet.

  11. SeanExile 11

    Its a start. I assume that the secret polls the parties do have alot more data in regards to this.
    I actually think that the student vote isn’t as off as it may seem. A lot of the students don’t engage in campus activities. Few, outside of Otago, care the slightest about political life on campus or even about student committees. Most, id say, just visit campus and then returns home and vote for the same party their parents or peers do. So for me the results don’t look that off what id expect.

    Two things the left has failed spectacularly in
    Young voters
    Chinese voters.

    Youth usually vote left. They are still idealistic, still vote with their hearts. How can we fail in this category?
    Another thing I have noticed from Auckland is that Labour is really missing out on the Chinese vote. That we would loose out among the HongKong:nese are pretty given, but that were loosing the overall Chinese vote so spectacularly is bad for us. This is one segment where Labour should be strong and we really seem to lack the ability and personnel to handle this. And to be honest its not that hard to corner this segment. Clear policies, advertised on mandarin Chinese TV and news for how those with income can bring their parents to NZ. Simplified such policies and more support for education in Chinese in schools etc. (Why do most NZ schools offer Japanese classes but not Mandarin? – it makes zero sense)

    On the other hand from my 100% unscientific knowledge were doing better with the Indian community.

The server will be getting hardware changes this evening starting at 10pm NZDT.
The site will be off line for some hours.