Talk:2020 Queensland state election/Archive 1

Australian Institute for Progress Poll

Just a suggestion, but I don't feel like the recent voting intention poll is necessarily reliable nor has a neutrality of bias, as the Institute has several links to the LNP and its polling size are drawn from their own website. Personally, I feel like the poll shouldn't be up on the Wikipedia page, but I'm open to discussion. — Preceding unsigned comment added by Lachylan (talk • contribs) 10:00, 28 September 2020 (UTC)

FWIW, I don't really think it is appropriate to include any polling in advance of the election. All polling is biased based on who is polled, hoe they are asked the question, etc. Polling is not a reliable predictor of the outcome and the choice of which polls we include is a form of bias in Wikipedia. With great readership comes great responsibility; I think not including polls is the responsible choice. There's only one poll that counts, the vote itself. I think it's ok after the event to comment on whether the result reflected the polls. Kerry (talk) 11:25, 28 September 2020 (UTC)

Firstly, on the accuracy/reliable predictor question, the federal 2pp polling average in Australia within the final week of the election has come within +/- 2% of the final result every election since 1987 (introduction of Newspoll) bar 2019. State election polling for Queensland is harder to come by, but what final-week polling I can dig up has come within +/- 1.5% of the final result. Polls taken close to the election have been pretty accurate in Australia, especially if you throw out some of the polls with dodgy methodology (e.g. using respondent-allocated preferences), which I didn't do for the margins of error I calculated above - if I did they would likely shrink markedly.

Secondly, without making any comment on the reliability/biases, or lack thereof, of the AiP survey, I think even if Australian polling was absolutely horrid, there is definitely educational value in putting up a database of polling for readers. Unless there is obvious refusal to include polls showing certain outcomes, including polling is just a useful piece of information especially in elections which may not have the widespread coverage/polling aggregation of more widely covered elections. Including polls imo is no different from mentioning the results of meta-analyses of studies done on homeopathy on the Wikipedia Homeopathy page; it is simply information there for the reader to be informed, and as long as what we put up is a proper reflection of the research done, I don't believe this is a form of bias in Wikipedia. Arachnus64 (talk) 11:36, 3 October 2020 (UTC)

WP:CRYSTAL says "Predictions, speculation, forecasts and theories stated by reliable, expert sources or recognized entities in a field may be included, though editors should be aware of creating undue bias to any specific point-of-view." I feel the inclusion of any polls does that and it's particularly inappropriate in the leadup to an election. The fact that you want to include some polls yet exclude others further raises my concerns. Kerry (talk) 04:34, 5 October 2020 (UTC)

Neutral language and appropriate referencing when discussing in-progress elections

The article focuses excessively on the opinions and speculation of news media organisations and candidates. Outcomes of the 2020 State general election are a factual matter, it should not be necessary to rely so heavily on the opinion of sources. The introduction makes a number of assertions about election outcomes which are not factually correct, or have only become so since they were written.

When writing about an election currently underway, it is particularly important to keep an impartial tone and use appropriate sources. Given for this election, it can take nearly a fortnight after election day for outcomes to be declared, editors should be mindful not to present speculation as final election results.

As of 3 November 2020, with 75.7% of votes counted, the Labor Party will return to government for a third-term, led by incumbent premier Annastacia Palaszczuk.
— 2020 State general election 05:40, 4 November 2020 (UTC)

Will return is not neutral given only 2 of 93 electorates have had outcomes declared, as at 05:53, 4 November 2020 (UTC). At the time of writing, no outcomes had been declared by the Electoral Commission of Queensland. The reference given does not support the assertion that the Australian Labor Party will return to government with a majority of seats, instead clearly statying that ABC election analyst Antony Green projects Labor will win. This does not accurately reflect the source material, nor does it keep an impartial tone.

The Queensland Greens significantly increased their primary vote in inner Brisbane, winning South Brisbane from Labor MP Jackie Trad to become their second seat in the state legislature along with Maiwar, already held by Michael Berkman.
— 2020 State general election 05:40, 4 November 2020 (UTC)

The assertion that Queensland Greens had increased their primary (first preference) vote could be considered correct, as the Unofficial Preliminary Vote conducted by ECQ had indicated this. At the time this was written, and at present, it could not be said that South Brisbane ... [is] to become their second seat in the state legislature, as the results for this seat have not been declared. The source does not make this clear, however good research would have revealed so; even other articles by the same organisation to.

As the third-term for Labor, Palaszczuk has become the first woman to win three state elections in Australia.
— 2020 State general election 05:40, 4 November 2020 (UTC)

This statement was again premature, stating as fact something that was merely being speculated at the time it was written. The source used was written two days before results for her seat were declared, relying on preliminary results of unofficial counts conducted by ECQ. Neutral tone could have been preserved by opening the sentence with something like, "Widely expected to become the first-term for Labour, Palaszczuk is predicted to become the first woman...".

The counting of votes for this election is still ongoing. There are four stages to this counting; most votes will be counted in all of these. Only two of these, the Official First Preference Count and Official Distribution of Preferences Count, are considered official and count towards the outcome. Electorates will only have winning candidates declared after the final Official Distribution of Preferences Count, which has not yet started, unless the outcome is overwhelming and already mathematically certain from the Official First Preference Count.

Excessively presenting the speculation of sources, without adequately describing them as opinion, risks the appearance of partisan bias in the article. The factual accuracy is compromised by the insufficient differentiation of speculation and declared outcomes. Readers could come away from this article, reasonably believing that particular candidates have already been elected and seat distributions amongst parties finalised. ExoticViolet (talk) 04:45, 4 November 2020 (UTC)

Hey! Firstly, please make sure to sign your comments with ~~~~ when you post them, not afterwards :) Just to make sure I've gotten the general gist of what you're saying, it seems like you want to only use the officially declared statistics? If so, you're actually advocating for original research and is undue in terms of minimising the ABC's coverage, which is somewhat objectively the most detailed and accurate. Historically, Green's predictions and projections have been spot-on, and he's even received an AO for his election coverage and studies of election sciences. We reflect what the sources say, not what we believe the truth to be, and so far, not a single source has disagreed or deviated from Green's projections. In fact, they are actually emulated across the board on commercial media agencies like the Brisbane Times. Even blatantly biased 7 News is not equivocal on the fact that Labor's returning. If you want more evidence of it being universally accepted: ABC, SBS, Brisbane Times, The Guardian, New Daily. Can you provide a single source that says that Labor will not form government, that the Greens will not win either South Brisbane or Maiwar, or that Palaszczuk will lose her seat? ItsPugle (please ping on reply) 07:08, 4 November 2020 (UTC)

Also, at the very most, this is a factuality issue, not a neutrality issue per the fact every source out there corroborates what we're saying. ItsPugle (please ping on reply) 07:26, 4 November 2020 (UTC)

Agree with ItsPugle here, relying solely on official statistics are an over-reliance on a single primary source, when Wikipedia policy is to prioritise reliable secondary sources, which is what is happening here and has happened for every contemporary election article in Australia and internationally. To rely on the timetable of count completion and result declaration means Wikipedia editors would sit on their hands for four to six weeks while the electoral commission proceeds with their processes, seats are declared and writs are returned. Yes, that would be nice to be able to make sure everything is official and beyond doubt, but it doesn't serve anyone who uses Wikipedia as an information source to be so cautious as to not even summarise sensible analysis by experienced commentators and analysts like Antony Green. You are right that careful wording is necessary, but when even the most cautious and partisan outlets say that Labor has won another term in Queensland, and that it will almost certainly be in majority, it's not against policy to explain this (referenced of course). If the sources are cautious and the election is close, of course more care is required not to make assumptions, but that is not what is happening at this state election, or any other election article I've followed where coverage has been cautious and unbiased but timely and adaptable to rapidly-changing events. --Canley (talk) 08:01, 4 November 2020 (UTC)

If your concern is over the neutrality or accuracy of one or two sentences and the timing of some media sources used, then can I ask why you don't edit the text yourself to improve it, and include more recent sources? It seems to me that this would be a lot quicker and easier for all concerned than tagging the whole article for neutrality and posting lengthy critiques. --Canley (talk) 08:48, 4 November 2020 (UTC)

Oh OK, I see you did, my apologies! --Canley (talk) 09:01, 4 November 2020 (UTC)

@ExoticViolet: Just a word of warning to maybe not change wording, especially when your own discussion shows that there is a consensus against your proposals. While it's not large, little wording changes like "determines" to "considers" misrepresents the authority of Green and the fact that every source corroborates Green's determination. ItsPugle (please ping on reply) 09:30, 4 November 2020 (UTC)

References

Table and enrolment numbers

Hey! So there seems to be some confusion or misunderstand here about the table format and the electoral roll, so let's try and talk this out.

@LeoC12: Regarding your message here, in Australia, our electoral rolls are maintained AEC, not the ECQ or any other state/territory election commission. The AEC creates an electoral roll for each state and territory, which is then passed onto each state/territory (see the Commonwealth Electoral Act 1918 § 90, and why the ECQ website says to go the AEC to "Check your enrolment, enrol to vote or update your details"). This is now the second time that this has been misunderstood, so when this is restored, an {{Efn}} is probably warranted. In terms of the table format, that format is actually derived from the widely used {{Election box begin}} family - because the default form doesn't include seats, I've just gone ahead and done some quick changes to manually add a seat rows etc. I also don't know what you mean by "the standard format used across Australian election articles" - there's been no discussion at WikiProject Australia or WikiProject Elections & Referendums, that format is not used at 2015 Queensland state election or 2017 Queensland state election, and the only other article that I could find it at is 2020 Australian Capital Territory general election, when you changed the table format on the 1st.

In terms of why a different, more modern table is needed, the primary concern is about accessibility, but there are a few others:

1. The current table uses fixed width values for column widths. px measurements do not work well at all on mobile, making the table inaccessible to the majority of readers. The new table is based around em measurements, which are responsive, meaning they are relative to the device's scaling.
2. Terms like "IRV" is so unfamiliar and an overly technical specification, violating WP:TECHNICAL. IRV warrants discussion in Electoral systems of the Australian states and territories#Queensland, but is not necessary for inclusion in the title of a table, let alone in an article about a specific election, not elections in general.
3. "CV" is an ambiguous random abbreviation and is redundant: the expectation is that readers have a slight idea about the topic, which includes the fact that voting is compulsory in Australia. Also, both tables include the turnout in summary rows and it is also prominently placed in the infobox.
4. The summary rows are inconsistent with the design in the {{Election box ...}} family and most other election articles I've come across

I agree that a "Seats in doubt" row is probably worthwhile, but I don't think that in itself is enough to revert to an older, less accessible, and mis-designed table. ItsPugle (please ping on reply) 06:40, 4 November 2020 (UTC)

@ItsPugle: You are correct the Queensland roll is maintained by the AEC, but it's not necessary to mention it in the infobox, as clearly confusion is going to arise as it has (including by me), just mention the number of people on the roll with a reference. Also, what's the reason for rounding up to "3.4 million"? Is it because the number could change due to provisional enrolments?

In terms of the results table, I prefer the current version, sorry. I don't hate your proposal and there's some good ideas in there so I'm interested to hear what others think. My thoughts are:

• I don't think the party leader column is necessary – there are already enough arguments about who the leader is or whether there is one in the infoboxes. Am I correct this was just included to be similar to an electorate result table's candidate column?
• Fair point about accessibility and responsiveness of fixed pixel based column width, but there is nothing stopping the current table being adapted to use em measurements. The aforementioned leader column made the table so wide I don't think that was an improvement for either accessibility or mobile responsiveness.
• No objection to clarifying the abbreviations, although they are wikilinked to the relevant concepts so they are not completely opaque as to their meaning.
• A standard table format hasn't been discussed and decided before, with Template:Australian elections/Title row being used until 2018 when this version has been widely used. While not identical, it is far more in line with summary tables used for UK and New Zealand elections. I just don't see the need for a summary table to be the same as the results table for a district.
• My preference is to order by vote totals rather than seats won (although order by seats won in the infobox).

--Canley (talk) 07:38, 4 November 2020 (UTC)

@Canley: I'm perfectly happy to be done with the leader column, and to order parties by vote not seats, but I still think the newer version is a better option. And yeah, it was just a left-over from when I was developing the table based on the election box family of templates. Removing the leader column resolve any issue with the width of the column, and even then, we can always shrink the party column a tad. Because there hasn't been any actual consensus or discussion on tables, I don't think there's an issue there with using this new table.

Actually, on a bit of an investigation, this trend in a few election articles seems to be from LeoC12 going through and changing the stable table version to their preferred one, referring to their previous changes as evidence of a consensus: Special:Diff/986477920, Special:Diff/983948882, Special:Diff/974339258, Special:Diff/974338050, Special:Diff/982959508. While I absolutely believe they're trying to help, it comes across to me as though they're trying to make it look like there's some sort of implicit consensus, when it's actually just them changing table. The majority of the articles since 2018 actually seem to use the election box family more closely aligned with the new table I'm suggesting: 2018 Batman by-election, 2018 Darling Range state by-election, 2018 Fremantle by-election, 2018 Longman by-election, 2019 Cheltenham state by-election, 2019 Enfield state by-election, 2019 Tasmanian Legislative Council periodic election, 2020 Brisbane City Council election, 2020 Bundamba state by-election, 2020 City of Melbourne election, 2020 Tasmanian Legislative Council periodic election, the list goes on... ItsPugle (please ping on reply) 09:59, 4 November 2020 (UTC)

But those are all single-seat elections (mostly by-elections), so naturally they use the single electorate result table. The table for a general election are a summation of the results in multiple seats, so it's a completely different table for a different purpose and there is no need for them to adhere to the single-seat format. That there has been little or no objection to LeoC12's version of the table for at least five general elections (see 2018 Victorian state election, 2019 New South Wales state election, 2019 Australian federal election, 2020 Northern Territory general election, 2020 Australian Capital Territory general election, until now 2020 Queensland state election) over two years does imply some sort of consensus or at least tacit acceptance of that format. I don't mind the original template either but Leo's version looks more modern and uncluttered. --Canley (talk) 10:17, 4 November 2020 (UTC)

@Canley: Oh yeah, fair enough I guess. Still though, I think the principle of consistency can be maintained here by using similar styles. And for LeoC12's changes, they've all happened in the past two weeks, not over two years, and several weeks to years after the election where the previous tables were more than comfortably embraced. I agree that LeoC12's template is better looking than the previous, but it's incorrect for them to have tried to suggest there's some consensus (explicit or otherwise) in support of using that table format, and that at least in my logic, using the less-problematic table (with amendments) that I'm encouraging should be preferenced unless there's a consensus to overrule? ItsPugle (please ping on reply) 11:04, 4 November 2020 (UTC)

@ItsPugle: I don't know where you're getting "the past few two weeks" from as I was referring to this diff which was early December 2018 after the Victorian state election in November; then the day of the 2019 NSW election; then I added it myself for the 2019 federal election. NT 2020 I used the old template which LeoC12 replaced which was fine, and ACT 2020 I added a table in the new format which Leo corrected and fixed up. I was wrong though, it wasn't two years ago that this table style originated. In fact, this style was used for the 2016 federal election, the 2013 federal election, the 2010 election and, whoah, over 13 years ago at the 2007 federal election—these were at the time of the elections too, by various people, so they are not recent retrospective edits by a lone editor (although for 2004 and earlier the style was retrospective). All LeoC12 has done is to make several recent state general elections consistent with an established federal election format used for over 13 years by dozens of editors.

Anyway, happy to hear others' opinions and suggestions, maybe link to the userspace draft of your version, the diff, or include it on the talk page here so others know what the discussion is about. --Canley (talk) 12:38, 4 November 2020 (UTC)

@Canley: Do you mind if I collapse our convo here then, since it's largely just us (me) bickering about when the template started being used? I'm going to start a new thread in this section below with the different tables etc. ItsPugle (please ping on reply) 07:41, 5 November 2020 (UTC)

@ItsPugle: Sure, no problem! --Canley (talk) 07:45, 5 November 2020 (UTC)

Should short descriptions be used for Wikis similar to this one?

After reading WP:SHORTDESC I am in doubt as to whether I should have inserted a short description here as WP:SHORTDESC does not have specific guidelines as to whether to insert short descriptions for election Wikis but it is ambiguous as to whether this article needs a short description but I think this short description describes the article reasonably well though people may know what this article is about from the title alone. WP:SHORTDESC does state though that titles may sometimes be reason enough to not use a short description if readers know exactly what the Wiki is detailing. I see WP:SHORTDESC page is based on community consensus which I will try to garner here. Qwerty123M (talk) 02:18, 15 August 2023 (UTC)