Why is no-one questioning this?

I refer to my previous opinion piece (February 19, 2024) on the probole and/or honesty of political polls in this country.

For the record, I took to task David Farrar and Jordan Williams both co-founders and directors of the NZ Taxpayers Union, an organisation Farrar finally resigned from in July 2023. In Aotearoa all polls should be conducted and align to the Political Polling Code that is promulgated and published by the Research Association of New Zealand (RANZ).?

Remember that all rules around integrity and credibility in regard to polling are very much subject to the moral compass of the people running the poll and those publishing it. At no time does RANZ have the ability or the resources to properly research or force polling bureaus to conform to their rules. ?

As usual a new poll was published on March 8 this year and it identified a number of things on behalf of the Taxpayers Union-Curia poll, which was widely published in media outlets like Newshub and the NZ Herald.?

In the months leading up to our 2023 general election there is no accountability in the industry in regards to the polls, particularly by Curia, that have been published.?

In effect, Curia and the Taxpayers Union breached several rules with their published polling results. The breach of the standards pertains specifically to Curia Market Research Limited owned by David Farrar who is a member of RANZ.?

The standards in the RANZ Political Polling Code of 26 June, 2020, in violation are:-

1. The standard for reports the agency (i.e. Curia) must prepare a report suitable for publication with full results and methodology.?
2. The report must include the wording of the voting questions and all other information required by the RANZ Political Code.?
3. The report must be published on the conducting agency’s website.?

That did not occur.?

The RANZ Polling Code sets a standard for undecideds. The report must state the number and percentage of people polled that determines themselves to be undecided and/or refused. The Curia poll by Mr Farrar did not disclose this. The margin of error or other errors needs to be reported. The report must state the maximum margin of error for a simple random sample at the 95 per cent confidence level. That did not occur.?

The report must confirm the sample was weighted. That did not occur.?

Reports must use the actual wording in the poll; it must disclose the questions asked and any political questions asked before the principle voting behaviour question. The report must disclose if the questions were part of an omnibus survey. That did not occur.?

Collection method. Reports must identify whether one or other or both landline and mobile devices were used in the random survey and design. The report must disclose how a respondent to the poll is selected.?

Neither mobile phone, landline or, for that matter, how the respondents were selected have been identified.?

Avoid self-selection polls. Polling results must state that the poll conducted is a representative sample of New Zealanders.?

How can the perceived regulatory authority on behalf of pollsters in Aotearoa accept this nonsense and allow publication of it to occur??

In May 2023 the University of Auckland conducted its own report into polling and concluded? a lack of understanding into the mechanics of polling, alongside the methods used and the lack of transparency by pollsters, can impede the polling process and the result.?

How can basic journalists struggling for jobs and relativity allow this conduct to continue??

How do we know that Mr Farrar has not used a selective group of people to poll, such as, the 200,000 people they say they have at the Taxpayers Union?

A similar case in Australia occurred when in 2019 the Sydney Morning Herald and the Melbourne Age sacked their pollsters who were widely published in both publications, and in some cases, had their results printed on the front page.? After conducting several investigations they discovered their polling agency, uComms, was in fact owned by the Australian Council of Trade Unions. They also discovered almost all of the respondents being used for polling were heavily biased towards Labour and concluded, “The outcome of uComms polling is influenced by its ownership structure”.?

Polls do influence voter conduct and behaviour. Pollsters therefore must be held to account, particularly when they are overtly politically tainted.?

Can you imagine if David Farrar and Jordan Williams were Māori; they would be roasted alive for this conduct.?

-

John Tamihere is president of Te Pāti Māori. He is a former Labour Government Minister and chief executive of West Auckland urban Māori Authority Te Whānau o Waipareira and the Whānau Ora Commissioning Agency.