Talk:Opinion polling for the next New Zealand general election: Difference between revisions

Elections and Referendums List‑class

This article is within the scope of WikiProject Elections and Referendums, an ongoing effort to improve the quality of, expand upon and create new articles relating to elections, electoral reform and other aspects of democratic decision-making. For more information, visit our project page.Elections and ReferendumsWikipedia:WikiProject Elections and ReferendumsTemplate:WikiProject Elections and ReferendumsElections and Referendums articles List This article has been rated as List-class on Wikipedia's content assessment scale.

New Zealand: Politics List‑class Mid‑importance

This article is within the scope of WikiProject New Zealand, a collaborative effort to improve the coverage of New Zealand and New Zealand-related topics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.New ZealandWikipedia:WikiProject New ZealandTemplate:WikiProject New ZealandNew Zealand articles List This article has been rated as List-class on Wikipedia's content assessment scale. Mid This article has been rated as Mid-importance on the project's importance scale. This article is supported by the New Zealand politics task force (assessed as High-importance).

@Nixinova: I have no issue with adding the seat projections to the table in that way, however the discrepancy with the forecast section below that comes with excluding the assumption that the Maori Party retain their electorate is problematic. Especially when Colmar Brunton explicitly do make that assumption in their own projections.--Pokelova (talk) 07:50, 11 May 2021 (UTC)[reply]

I don't think the discrepancy is an issue. The forecasts section implies some level of reasoning is involved in working out what the results would be, while the seats column in the polling table doesn't make any assumptions and just works directly off what the poll results are.  Nixinova   T   C   08:07, 11 May 2021 (UTC)[reply]

But that seems to me to be a misrepresentation of how our electoral system works.--Pokelova (talk) 08:16, 11 May 2021 (UTC)[reply]

It either does make an assumption (no party bypasses the threshold through winning an electorate seat) or just willingly gives a potentially incorrect projection. --Gbuvn (talk) 13:48, 11 May 2021 (UTC)[reply]

Are there any sources showing where the numbers at the "Projected seats" column come from? If yes, ok. If not, the column should be removed since that's an interpretation of sources that is not reflected in the sources themselves, and thus, WP:SYNTH, which is not allowed. Impru20^talk 13:54, 11 May 2021 (UTC)[reply]

Any party could win any seat; including detailed analysis for every single poll seems pointless. Take those numbers, plug them into the seat calculator, that's what the column says.  Nixinova   T   C   19:32, 11 May 2021 (UTC)[reply]

I've changed the col heading from "projected seats" to "equivalent seats"; does that explain it better?  Nixinova   T   C   19:33, 11 May 2021 (UTC)[reply]

The Electoral Commission's seat calculator includes a column for number of electorates won, and while entering a "1" in that column may constitute an unprovable assumption, it seems to me that leaving it at "0" is also an unprovable assumption (as Gbuvn says), and not necessarily a better one. Is making such an assumption somehow more useful/informative than doing otherwise? That is, is there a reason why "Here's what Parliament would look like, assuming nobody relies on an electorate" is better than "Here's what Parliament would look like, assuming electorates stay the same?" They're both assumptions, and if I had to choose, the latter seems a better (albeit imperfect) fit for what has actually happened historically -- after all, we've never had an MMP election which didn't involve a sub-5% party winning an electorate. Either way, though, we're presenting readers with a seat count that we attribute to the Seat Calculator, but which involves feeding the Seat Calculator electorate numbers (zero or non-zero) which the poll being cited does not provide. I suppose there's no reason we can't present an seat count which assumes no electorate exceptions if we want, provided it's clear what we're doing -- but is it useful? Does it benefit readers to know how many seats the poll would translate to in the absence of a factor that, historically speaking, hasn't been absent? If we can't give a proper projection because that requires assumptions (in one direction or another) that we can't back up, why give a seat count at all? Personally, I think I'd prefer just to leave the poll numbers to say what they say, and simply advise readers that exact seat outcomes cannot be definitely predicted from national polling. -- Vardion (talk) 22:11, 16 May 2021 (UTC)[reply]

I share Vardion's thoughts and conclusions. Schwede66 19:45, 17 May 2021 (UTC)[reply]

I have been bold and removed the seat projection column altogether. It is very problematic: aside of the points raised above, it was coming to the point where sources themselves give their own projections (take the latest NCB poll, which gives a 59-36-12-11-2 allocation versus the 60-38-11-11-0 that was shown), yet those were disregarded in favour of the alternative projection! This is openly WP:SYNTH, if not outright WP:OR. Besides, the fact that pollsters may give their own seat allocations and that those are different from the projected seats means the method for projecting seats is untrustworthy. Impru20^talk 07:54, 27 May 2021 (UTC)[reply]

Kia Ora, I submitted a change that would allow any person that registers in a poll to appear on the preffered prime minister section. This was then deleted because they were not party leaders. The question that is generally asked is “who would you most like to be NZ’s current prime minister.” That has no mention of a party leader. All of the inclusions have polled higher than the included (polled highest at 1%) Winston peters. If people want Christopher Luxon or Chlöe Swarbrick etc. to be prime minister and that has registered in a poll then I believe that the corresponding polling number should be displayed. — Preceding unsigned comment added by Wikifan3431 (talk • contribs) 09:17, 8 July 2021 (UTC)[reply]

I think only people who appear regularly, or at least more than once, should be included. I also think people who never achieve a significant figure - perhaps 1% - should not be included. See Opinion polling for the 2020 New Zealand general election#Preferred prime minister for a table where there was an unwieldy number of people listed.-gadfium 09:32, 8 July 2021 (UTC)[reply]

I believe I was the one who added Shaw, Davidson, Baker, and Te Kahika, and I agree, with you that they could all honestly be removed. At the time my logic was trying to be fair but they probably aren't worth including any more since their polling didn't go anywhere unlike Seymour's.

As for Swarbrick, I am open to maybe keeping her if she continues to gain steam. --Pokelova (talk) 09:37, 8 July 2021 (UTC)[reply]

Most of the people added appeared a multitude of times. Though I we want to make those rules I think we should remove Peters. Maybe we could add a minor polling section for people that gain under 5%? --Wikifan3431 — Preceding undated comment added 09:48, 8 July 2021 (UTC)

A minor polling section for preferred prime minister would be deeply unnecessary. If we're going to include them at all, just include them in same table.--Pokelova (talk) 09:52, 8 July 2021 (UTC)[reply]

Does anyone object to what I have added to proffered PM? —-Wikifan3431 -— Preceding undated comment added 6:12, 12 September 2021 (UTC)

@Wikifan3431:, a lack of verbal objection is not the same as consensus. As I have said, I'm not necessarily opposed to including Swarbrick, but the problem with what you wrote in particular is that you don't define what it means to "register" in a poll. --Pokelova (talk) 08:49, 28 November 2021 (UTC)[reply]

What do you think the threshold to enter should be --Wikifan3431 (talk) 11:08, 28 November 2021 (UTC)[reply]

I've always said that I think it should be only party leaders, I even disagreed when Luxon was unhidden because he had yet to outpoll Collins, though it seems likely that he'll be leader now anyway.

But if we were to set it to include Swarbrick, the minimum would have to be 2%, but then we'd have to add Bridges, so we could do 2% AND be the highest polling member of their party. But let's wait and see if anyone else has anything to say.--Pokelova (talk) 10:26, 28 November 2021 (UTC)[reply]

I feel that party is regardless in preferred PM. I think 2% in at least 3 polls is fair (This would mean Swarbrick would have to wait for a bit to be included). --Wikifan3431 (talk) 10:15, 1 December 2021 (UTC)[reply]

Any other opinions on the preffered PM pollin?. --Wikifan3431 (talk) 6:56, 6 December 2021 (UTC)

Talbot Mills aren't cherry picked, they've been shown monthly, and details are given for e.g. sample sizes. Should they be put into the main section?  Nixinova   T   C   00:04, 14 October 2021 (UTC)[reply]

I was thinking about that. These aren't internal Labour polls. Worth consideration. --Pokelova (talk) 00:39, 14 October 2021 (UTC)[reply]

I agree with moving the regular Talbot polls into the main section. This doesn't apply to the polls for previous elections.-gadfium 00:59, 14 October 2021 (UTC)[reply]

The Curia polls are also the same, [1]. I think the private polling section should be entirely merged into the main section because they're not cherry picked unknowns.  Nixinova   T   C   00:18, 11 November 2021 (UTC)[reply]

I'm a bit more hesitant about Curia, they may not be internal National polls any more but they are commissioned by a plainly partisan group. --Pokelova (talk) 00:37, 11 November 2021 (UTC)[reply]

Curia is releasing their results now publicly [2] so they should go into the main section now.  Nixinova   T   C   23:55, 15 February 2022 (UTC)[reply]

All but one have tpu dot org articles so I've moved those up.  Nixinova   T   C   07:04, 16 February 2022 (UTC)[reply]

I think this is OK, but should be revisited if it appears that they go back to selectively releasing them. Without knowing who funds the TPU (the "full" poll results are for subscribers), still a bit less transparent than ideal. Limegreen (talk) 11:16, 16 February 2022 (UTC)[reply]