Stuff/YouGov poll: Labour 41%, National 38%

Stuff have started political polling again, this time with YouGov, who are new to New Zealand polling. With no record to give any idea how they compare to other polls analysis of this poll should be even more cautious than normal (not that media or parties treat polls as they should).

  • Labour 41%
  • National 38%
  • Greens 8%
  • NZ First 8%
  • ACT 2%
  • Maori Party 1%
  • TOP 1%
  • Other 1%

The poll was conducted between 7 and 11 November by YouGov so events over the past two weeks are not reflected in these results, of particular note the revelations last week about a secretive foundation that handles party donations.

Labour and National shouldn’t be too worried bout this result. Greens will be happy. Winston Peters usually slams polls and had a major hissy fit against media last week, but should be relieved with the timing of this poll.

There’s a glimmer of hope there for ACT, who may benefit from David Seymour’s hard work and success over the End of Life Choice bill.

Stuff: Labour ahead while National dips below 40 in new Stuff poll

Labour and its coalition partners are riding high while National have dropped below 40 per cent support in a recent Stuff/YouGov poll.

That is a mediocre summary. National are down on other polls but have not dropped under YouGov polling, which is untested in New Zealand. Labour, Greens and NZ First together look strong, but that could have changed last week.

It is the first poll published by Stuff from YouGov, a global polling firm who run regular polls for The Australian, The Times, The Economist, and CBS News.

A spokesperson for National leader Simon Bridges said the poll did not match their own figures and was incorrect.

The poll isn’t incorrect, it is the results YouGov got. There could be a variety of reasons it differs from National’s own polling – and without publishing National’s polling it’s impossible to compare anyway, politicians are notorious for promoting their own polling (when it suits them) without showing any evidence or details.

Prime Minister Jacinda Ardern said it was “encouraging”, as it showed Labour building on its election result.

“It’s really encouraging to see all of the coalition parties up when we compare the numbers against the last election. We’ve taken on some big challenges but we’re making good progress — I’d like to think this poll reflects that,” Ardern said.

Ardern may like to think that but it’s also nonsense. It shows only what those who were polled thought 2-3 weeks ago.

Labour is widely seen to be making mediocre and disappointing progress. The poll is more likely to reflect the lack of progress Bridges is making with his negative, whiny dog whistle strategy.

Leaders’ favourable/unfavourable rating:

  • Jacinda Ardern +35%
  • Simon Bridges -37%

Winston Peters was about -23% (30% favourable, 53% unfavourable, but that was before last week’s Foundation/donation revelations.

That’s good for Ardern and bad for Bridges, but unsurprising.

The methodology for the YouGov poll is different to other political polls in New Zealand, which rely on phone-calling or a mix of phone calling and online responses. It is conducted entirely online by a panel of respondents, as other YouGov polls around the world are.

Certainly YouGov is untested in New Zealand, but Reid Research (for Newshub) have already been using part “online methods” (along with “Computer Assisted Telephone Interviewing”)

We need to see several YouGov poll results alongside the other public polls from Reid Research and Colmar Brunton, and at least one general election, before we can see how close to or distant from reality they are.

Campbell White, YouGov’s head of polling and public affairs for Asia Pacific, said online sampling was the best way to make sure a wide variety of people were polled.

“The reason is over time we are better able to represent the population online. Rather than just the people who answer their phone and don’t use call screening,” White said.

The sample has quotas, so various demographics are represented, and the figures are scientifically weighted to match the voting population.

Phone surveys also screen respondents to try to ensure they poll a representative sample of demographics are obtained.

I don’t know there is any research or evidence to show whether online only polling is any more accurate than other polling or not.

Mrgin of Error stated as +/-3.1% which is standard for 1005 respondents.

It’s good to see another public political poll, there has been a lack of polling over the last few years. YouGov results will add to the mix, but need to be viewed cautiously until they build a track record.