The unusual feature of Taxpayers Union’s latest Curia poll

Written By: - Date published: 10:03 am, December 11th, 2024 - 14 comments
Categories: act, national, polls, spin, taxpayers union, treaty settlements - Tags:

A new Taxpayers Union Curia poll hit the media this week.

The headline feature was stark. During recent times that have seen a mass Hikoi on Parliament and an earth shattering haka in Parliament protesting against Act’s dog whistle rewriting the Treaty of Waitangi Bill Act had seen a surge of support.

Here are the results.

For a minor party a 4.5% point bump is not to be sneezed at.

But another poll was also released this week. The latest One News Verian poll showing no change for Act’s support.

Two polls taken at similar times but one showing a dramatically different result, although both polls showed a surge of support for the Maori Party.

And a slightly earlier poll, taken for Labour by Talbot Mills, suggested that Labour was slightly ahead of National. Again no surge to Act was detected.

The highlight events during the previous couple of weeks were the Hikoi against Act’s Treaty Principles Bill which reached Parliament on November 19 and the Maori Party’s parliamentary haka on November 15. The Treaty Principles Bill was at the forefront of political activity during the relevant period.

But why did Curia suggest a surge of support to Act and has this happened before?

A Curia poll taken during the period including Waitangi Day earlier this year also showed a surge to Act while other polls did not.

Polling is important. Polls can create the perception of movement. They can also be used as weapons to affect waivering National MPs and I am sure there are more than a few of these right now.

Curia needs to get its polling methodology checked. Otherwise there will be a perception that it is being used by Taxpayer’s Union and other entities to drive public opinion and gather support for Act’s dog whistle Treaty Principles Bill.

Update:

And wouldn’t you know it as soon as I post this I come across two new examples.

Curia have been commissioned by another Atlas aligned entity, Hobson’s Pledge, to conduct polling on the Treaty Principles Bill.

I cannot see the polling methodology anywhere but the questions are clearly push polling. They include:

  • Do recipients support honouring the Treaty of Waitangi, provided it upholds fundamental human rights.
  • Does the Treaty mandate 50% Māori representation in Parliament.
  • What body should determine the principles of the Treaty of Waitangi, referendum, Parliament, the Waitangi Tribunal or the Judiciary?

The first question reinforces the rights claim that treaty rights are discriminatory. They may be but only in the way that baked in generational wealth is also discriminatory. The third question is a nonsence question and given attacks on the Waitangi Tribunal I am not surprised at the result.

The results are being presented by Act as support for the bill.

But when asked that question directly if they support the bill the result appears to be the opposite.

From OneNews:

More New Zealanders are opposed than support the Treaty Principles Bill according to a new 1News Verian poll, but a significant number say they don’t know enough about it.

The poll, which surveyed 1006 eligible voters and ran from 30 November to December 4, found that 23% supported the bill while 36% were opposed.

A slightly larger group – 39% – said they didn’t know enough about the bill, and 2% preferred not to say.

ACT have recently signed up an engagement officer and a communications officer. Get ready the country is facing 6 months of turmoil.

14 comments on “The unusual feature of Taxpayers Union’s latest Curia poll ”

  • Comments are now closed

  1. Barfly 1
    11 December 2024 at 10:09 am

    Curia needs to get its polling methodology checked. Otherwise there will be a perception that it is being used by Taxpayer’s Union and other entities to drive public opinion and gather support for Act’s dog whistle Treaty Principles Bill

    This is a feature not a bug

  2. tWig 2
    11 December 2024 at 10:20 am

    Repeating the link to Wiki aggregated NZ political polls.

    Here are all the current polls averaged since 2023, showing the actual trends across all parties, and across lw vs rw. This data gives a better idea of the statistical trend, taking into account poll bias and margins of error. It seems by far the best way to stop haggling about diverse pollsters and trend interpretations. I think would stop a lot of useless discussion and argument about my poll is better than yours, and identfy outliers that press unsupported hope into the hearts of the discouraged, and vv.

    You can also see for yourself where individual polls don't fit with others, suggesting bias.

    • mickysavage 2.1
      11 December 2024 at 11:32 am

      The trouble with aggregating the results is the outliers will still have an effect on the overall result.

      • tWig 2.1.1
        11 December 2024 at 4:23 pm

        But from the individual graph points for each poll, you can see for yourself easily by eye if one has a disproportionate effect, twitching the trendline up or down. Plus, you can check values for a particular poll in the table data against other polls near the same time.

        These kinds of stats are about confirming trends, not absolutes. And the wiki data confirms a negative trend in the past quarter for the government.

  3. Tiger Mountain 3
    11 December 2024 at 10:38 am

    Every dirty tactic employed on Brexit will likely be used by Act and Hobsons etc. here.

    These white supremacists and frazzled codgers need to be challenge every step of the way to put pressure on the PM and the Natzos.

  4. Muttonbird 4
    11 December 2024 at 10:42 am

    It really important to keep questioning, and amplifying that questioning, of Farrar's polling and, given his aggressive political activism, his suitability to operate as a pollster at all.

    The industry body, Research Association New Zealand doesn't think he's suitable upholding complaints and charging him with breaching the industry code of practice. Farrar is now outside any form of industry regulation and presumably will no longer adhere to industry standards.

    Unless people are made aware of it, Farrar will calculate publishing false or ill-gotten poll data for political gain is far more beneficial to him and his allies than any admonishment by his industry peers.

    Fools will keep swallowing whole Farrar's dodgy polls because of the increasingly well funded TPU/Curia brand. Lyin' Ryan Bridge fell over himself to gobble it up this morning.

    It’s really important to show the Curia brand is tainted and Farrar himself is conflicted at best and corrupt at worst.

    • tc 4.1
      11 December 2024 at 10:56 am

      "Fools will keep swallowing whole Farrar's dodgy polls…"

      Stuff and others think we’re all fools by publishing them. However when you publish the drivel Damian Grant comes up with on cue to support the cause it's hardly surprising.

      • tWig 4.1.1
        11 December 2024 at 4:26 pm

        A certain part of the population are like lemmings, following the mass of the rest off the cliff. These are the targets of polls that amplify propaganda from such groups as Hobson's Pledge (which is actually more about developing a Hobson's choice).

        • tc 4.1.1.1
          12 December 2024 at 1:38 pm

          Why polls should be banned in election cycles.

          Leemings swing govts in and out here and I recall someone in 2011 remarking key had it stiched up so no point in voting as poll after poll in granny etc had them winning comfortably.

          National scrapped in using dodgy john banks, a cup of tea etc…bradley ambrose did his job and look what happened.

  5. Reality 5
    11 December 2024 at 11:37 am

    Act is employing two new staffers to work on their propaganda. Interesting that all government departments are told to lay off staff but not the Act office. I imagine all ministers in the government will have a full quota of staff and none therefore will have to do dual roles.

    • tc 5.1
      12 December 2024 at 2:15 pm

      Seymour's threatened the media already so cant see them asking him the obvious questions about his pet ministry full of hand picked minions.

  6. Incognito 6
    11 December 2024 at 5:24 pm

    A biased poll is a bad poll because the results are unreliable.

    If there’s persistent bias, the poll methodology is wrong.

    Every data analyst knows that not all data is equal, and data needs to be checked and cleaned, if necessary (aka crap in, crap out) and assumptions need to be verified and met.

    So-called ‘outliers’ can force a trendline in one direction and this can be adjusted by applying a weighting to the data, i.e., not all data points are treated equally.

    The human eye-brain is very bad at discerning trends from noisy data and squiggly lines.

    Spin doctors and the likes know every trick in the book.

  7. Andrew R 7
    12 December 2024 at 10:45 am

    There are two parties to Te Tiriti. It is those two parties who have the right to discuss authoritatively what Te Tiriti means. Hobson's Pledge is not one of those two parties. The Curia poll commissioned by the Hobsons Pledge is straight out deceptive because it never presented those surveyed with the option 'the Treaty partners' for the question about determining Te Tiriti principles.

  8. lprent 8
    13 December 2024 at 1:04 pm

    What the political party poll and others from Curia recently) looks like is a sampling from the Taxpayers Union membership phone numbers. Here is my reasoning…

    What I find interesting (see below) is that the published poll results were for 1000 respondents. 800 by phone, and 200 by online panel. The selection phone numbers was from a set of 15,000.

    The number of decided voters on the vote questions was 908. There were 54 (5.4%) undecided voters and 25 (2.5%) who refused the vote question.

    There is, as usual no definition of what a "A random selection of 15,000 NZ phone numbers (landlines and mobiles)" means.

    From May this year

    Commerce Commission figures state that there were about 573,000 landline service connections in New Zealand, most of which use the Voice Over Internet Protocol with traditional handsets running through fibre.

    Of course that includes businesses who won't be targeted for polling. These days I'd say that the majority of extant landlines would be for the businesses and probably the elderly.

    Before my household dumped our landline earlier this year, the only people who called it were over the age of 78 (ie parents) and the occasional irritating robocall. The number was unlisted and has been since 1995, so we're talking scanning calls.

    Stats on 2023 census

    “Telephone use has dropped dramatically in recent years and is now around half the rate it was in 2018. In 2023, 31.0 percent of households said they had a landline, compared with 62.5 percent at the time of the 2018 Census,” Goodyear said.

    Of course most people these days don't list phone numbers and especially many VOIP generally don't by default. I just did a search for a eight people I know who still have landlines (I looked in my cell for recent landline calls). None are listed at https://whitepages.co.nz/

    Businesses list numbers. People usually don't in the modern world. You get spam calls.

    And of course, you generally don't have a public source for cellphone numbers. Which almost every adult has.

    So where do these phone numbers for polling come from? Well presumably some can be brought from telcos. But most likely they are culled from existing phone lists?

    Well after you read the widely different results of recent Curia polls, compared to others – one suggestion that has been made to me that makes sense.

    Who'd like to speculate that the phone numbers used in "The Taxpayers’ Union – Curia Poll…" now has a high proportion of phone numbers from the Taxpayers Union members.

    Now that could explain the recent weirdness of Curia polls.

    I may have to write a post along these lines. But feel free to write one and put it up if you're an author (or send in as a guest post). Can anyone point to anything that constrains Curia from getting a 'random' phone sample from a TU phone list?

    //——

    The waffle in the TU PR states that the poll

    … was conducted from Sunday 01 and Tuesday 03 December 2024. The median response was collected on Monday 02 December 2024.

    The target population is adults aged 18+ who live in New Zealand and are eligible and likely to vote. The sample population is adults aged 18+ who live in New Zealand and are eligible and likely to vote who are contactable on a landline or mobile phone or online panel. 1,000 respondents agreed to participate, 800 by phone and 200 by online panel. The number of decided voters on the vote questions was 908. There were 54 (5.4%) undecided voters and 25 (2.5%) who refused the vote question.

    A random selection of 15,000 NZ phone numbers (landlines and mobiles) and a random selection from the target population from up to three global online panels (that comply with ESOMAR guidelines for online research). If the call is to a landline, the person who is home and next has a birthday is asked to take part. Those who take part through an online panel are excluded from further polls on the same topic for six months.

    Multiple call-backs occurred to maximise the response rate. Those who said they were unlikely or very unlikely to vote were excluded.

    The questions are here