Newspoll: 53-47 to Labor

A fortnight of sound and fury ends with exactly the same set of voting intention numbers from Newspoll as last time.

After a week of post-Ipsos hype, The Australian reports the latest Newspoll finds absolutely no change whatsoever on voting intention since a fortnight ago: Labor’s two-party lead is at 53-47, and the primary votes are Coalition 37%, Labor 39%, Greens 9% and One Nation 5%. Scott Morrison is down one on approval to 42% and up three on disapproval to 48%, while Bill Shorten is down one to 35% and up two to 53%. Morrison’s lead as preferred prime minister is unchanged at 44-35. The poll was conducted Thursday to Sunday from a sample of 1582.

Author: William Bowe

William Bowe is a Perth-based election analyst and occasional teacher of political science. His blog, The Poll Bludger, has existed in one form or another since 2004, and is one of the most heavily trafficked websites on Australian politics.

1,194 comments on “Newspoll: 53-47 to Labor”

Comments Page 24 of 24
1 23 24
  1. Haven’t watched QANDA for years and regret that hour.
    Van Badham is a cartoon social warrior, reminded be of a left Alan Jones.
    Alex Hawke was polite until his bizarro rant about unions being a communist front.
    Butler was fairly polished but just trotted out well-worn lines.
    Peterson had a lot of pent-up hostility and was a pretty boring pub philosopher.
    McGregor was easily the most likeable and sensible.

  2. Don’t know who any of those are, but that Jordan guy was a jackass.

    And also the questioner who went “men kill themselves more than women, therefore men are oppressed”. I don’t see the public benefit in giving people who can’t do basic logic a free nationwide platform to air their nonsense.

  3. A question for William which he might/might not be able to answer:

    One Nation is polling 5.5% or thereabouts. As they have not had nation-wide (at least I don’t remember them doing so) candidates standing for seats in the HoR … how do you divide/estimate their preference flow? Are they automatically applies to the coalition – or a percentage of same?

    Also are we to assume they will not stand candidates in all 151 electorates? If so – are the 5.5% concentrate in more qld electorates, for instance? Would this effect the dynamic in that state?

  4. jenauthor I was wondering the same thing.

    Also, I’d really love to see statistics as to whether there are actually going to be more young people at this coming election as a result of the Marriage Equality vote, or if the enrolment caused by that vote is merely “brought forward”.

    If the former, the pollsters could truly be in for a surprise.

  5. Mavis…..it’s not very complicated. I object to the sectarian depictions of Catholics …and Muslims….and Jews….and so on. It’s very troubling. We have to stop vilifying each other. I have been very tiresome, I know. I object to racism and to sexism, among other things. I am not going to desist. I think idiotic, malicious, misleading and dehumanising tropes should be challenged. I am not religious – not at all – but I have seen what happens in people’s lives when religious prejudice and fears take hold. I think this is important. Make of it what you will.

  6. ar
    “I don’t see the public benefit in giving people who can’t do basic logic a free nationwide platform to air their nonsense.”
    But that’s what the whole show is about!!
    And the irony of Terri Butler pontificating about respect and polite discourse was hilarious, given her previous performance on QANDA which let to her having to settle a defamation case.

  7. One Nation is polling 5.5% or thereabouts. As they have not had nation-wide (at least I don’t remember them doing so) candidates standing for seats in the HoR … how do you divide/estimate their preference flow?

    Through an elaborate fudge based on the principle that a) Greens and “others” preferences will flow as they did last time, and b) your can calculate the One Nation flow from there using the residue from respondent-allocated preferences measures, which these days are provided only by Ipsos (vale ReachTEL national polling, apparently). I then split the difference between that result and the 2016 One Nation preference flow, which was basically 50-50. At the moment I have them breaking 53-47 to the Coalition, which is closer to 2016 preferences than what Newspoll seem to be doing, which looks more like 60-40.

    Also are we to assume they will not stand candidates in all 151 electorates?

    Most likely they will not.

    If so – are the 5.5% concentrate in more qld electorates, for instance? Would this effect the dynamic in that state?

    Pollsters are giving all respondents an opportunity to choose One Nation, regardless of which electorates they will run in, which nobody knows at present. So whatever pollsters are showing for them is probably an overestimate of what their actual national House of Representatives vote is likely to be.

  8. William do you have any insight into the issue I raised. Namely whether younger voters will be more highly enrolled this time around, or whether the enrolment surge generated by the ME vote has simply “brought forward” enrolments?

  9. One Nation only ran 15 candidates for the HoR in 2016, mostly in Qld. Probably a good strategy, considering the wacky candidates they attract. They concentrate on the Senate and it keeps the potential atrocities to a minimum.

  10. Briefly “I cannot see that there is any difference between ‘3 polls done close together’ and 1 poll that surveys the same number of respondents as the 3. The cumulative number of respondents is the same. The surveyed population is the same. The time frame is the same.”

    It has to do with whether the three polls are (somehow) independent. If they were, it would be correct to calculate the implied probability of the three being consistent with the actual value being 50:50 as the product of each poll’s individual implied probably (as someone did).

    The business about ‘3 polls done close together’ is an attempt to avoid causal linkage between the the three (since for example if poll 1 had a causal effect on poll 3 they would clearly not been independent; simultaneity does preclude direct causal linkage since simultaneous events cannot be directly casually linked). However this is not sufficient (nor in fact necessary) to ensure independence since the three may (and do in this instance) have an obvious joint cause (i.e. the actual voters’ intentions).

    Relevantly (and of great interest), Judea Pearl last year (with the assistance of a journalist or some such) published a popular (and excellent) book on his theory of causality, called “The Book of Why”. This is based on the theory he has developed over the past fifty years (previously he was the main inventor of Bayesian Networks). The popular book of course elides the Mathematics. It was (sort of*) thought that the only scientific way to establish causality was via prospective randomised controlled trials (RCT: the “gold standard). In fact Pearl shows methods for handling causality without requiring RCTs., including when “online” interventions are involved (his so called “do” operator) and even the analogue of causality in counter-factual speculation (appears to be correct but inherently less convincing…)

    * Karl Pearson, Ronald Fisher and various others promoted RCTs as it is relatively simple to ensure that they are done correctly. In fact Statistics always knew there were other ways (and there is now a flurry of statistics Professors emerita/emeritus publishing “Confessions” about things like the “conspiracy of the p-value etc.). Pearl’s book describes some of this history, including a statistician who lived to be about 100 and carried the flame of causality though the darkness of the Fisher-dominated period, as one might call it, and then out the other side.

  11. Briefly says… “we have to stop vilifying each other.” Do as I say, not as I do. Seriously, he just said that. Ha, ha, ha, ha!!!!!!

  12. William Bowe says:
    Tuesday, February 26, 2019 at 12:07 am
    I would be very surprise if the marriage equality referendum causes significant numbers of young people to vote who would not have voted otherwise.

    **********

    Do you think the jump on enrollments brought forward enrollments/updates that would have happened anyway in the leadup to the next election regardless of the plebiscite?

    (Edited)

  13. grimace….I was at an event with Albo this evening. He’d been campaigning today with Kim Travers (Pearce) and Melissa Teede (Canning). It sounds like things are going very well in both campaigns.

    The event was very upbeat….lots of people, and certainly not rent-a-crowd. A good cross-section.

  14. Do you think the jump on enrollments brought forward enrollments/updates that would have happened anyway in the leadup to the next election regardless of the plebiscite?

    Undoubtedly. And some of those who did enrol for same-sex marriage are still not going to vote at the election. I recall how excited some people got when automatic enrolment was introduced before the 2013 election, which was supposedly going to deliver Labor 1000 free votes in every electorate. It didn’t do anything of the sort. The enrolment rate went up, and turnout (i.e. votes cast divided by persons enrolled) went down.

  15. Mavia Davis (formerly Mavis Smith} @ <a

    I never found any of Barry Humphries' characters funny. They never made it to humour for me. They were one dimensional, insulting, boring caricatures of Australians who never existed, except in the mind of a culture-cringing fool.

    His 'humour' has neither wit, art nor intelligence and he should be consigned to the rubbish bin of theatrical history.

  16. Rossmcg says:
    Tuesday, February 26, 2019 at 12:41 am
    Briefly

    Did Albanese make a speech? Is Shorten under pressure?

    Lol

    He made a long speech, mostly about Perth infrastructure, the economy, the significance of Labor values and the qualities of WA Labor representatives in Canberra. He was good humoured and said nothing that could be taken as provocative or attention-seeking. He’s a thoroughly professional and serious politician. He came across as seasoned and friendly, and willing to meet and listen to people. I’m sure that Shorten is safe too.

  17. briefly says:
    Tuesday, February 26, 2019 at 12:36 am
    grimace….I was at an event with Albo this evening. He’d been campaigning today with Kim Travers (Pearce) and Melissa Teede (Canning). It sounds like things are going very well in both campaigns.

    The event was very upbeat….lots of people, and certainly not rent-a-crowd. A good cross-section.

    **********

    Kim’s field campaign is really pumping. From looking at the photos of door knocking campaigners on the Facebook pages of the 5 winnable seats here it seems hers is by far the biggest. Change the Rules is also active on the ground in Pearce and GetUp are mobilising.

    Aside from Porter’s bus, and some bin ads, his campaign has not been active yet. He’s a mile behind and will have difficulty catching up. He may well be planning on jumping ship to Curtin.

    Are you joining us door knocking on Saturday?

  18. Sue Lines seems very pleased with Kim’s progress. I have donated to Kim a couple of times….not a lot, but as much as I could put together at the time.

  19. Having doorknocked out my way for Kim, it’s clear the ground campaign is very well organised….good training, very up-beat, lots of volunteers….and great involvement from the candidate…

  20. There wont be a day goes by upto the next election when the Oz will do no story about Boats,Adani,Franking Credits or Negative Gearing. Todays edition doesnt disappoint.

    Stop Adani guide given to Labor MPs(Oz headline)

    12:00amJARED OWENS, MICHAEL McKenna

    Green activists gave MPs a playbook on overturning approvals for Adani’s coalmine in closed-door meetings, it is revealed.

  21. steve davis….I really think the LNP campaign for Labor when they run on most of those themes. They have nothing else, of course, so it’s understandable they might try. Every time they agitate in favour of Adani they shift votes to Labor. Every time they agitate on medevac/boats they shift votes to Indy Libs. They are cannibalising their own support.

  22. PuffyTMD
    says:
    Tuesday, February 26, 2019 at 12:48 am
    Mavia Davis (formerly Mavis Smith} @ <a
    I never found any of Barry Humphries' characters funny. They never made it to humour for me. They were one dimensional, insulting, boring caricatures of Australians who never existed, except in the mind of a culture-cringing fool.
    His 'humour' has neither wit, art nor intelligence and he should be consigned to the rubbish bin of theatrical history.
    ____________________________
    Well that's ridiculous, the relevance of those characters have obviously declined but Humphries at his best was hilarious. I would have liked to have seen his early Dadaist street work in Melbourne in the 50s. In one 'presentation' he dressed as a homeless bum, and to the shock of the crowd on Bourke street, he would pull out a silver service 3 course meal out of a rubbish bin and begin eating it complete with silver cutlery.

  23. steve davis says:
    Tuesday, February 26, 2019 at 1:10 am
    Briefly
    The phrase ‘as thick as shit’ comes to mind regarding the Oz.

    They seem to think voters are even more stupid than they are. I guess they can argue they’re trying to appeal to past-Lib-affiliated voters, on the premise that if they don’t shift to Labor the Libs win. But this overlooks the changes that have been occurring in the electorate, the environment and the economy…and their own hopelessly inadequate performance.

  24. A good cartoon would be a desert island with all the Oz journos shouting at everyone on the mainland but no one can hear them as the island is too far away and then it gradually submerges due to climate change.

  25. Boerwar from 9pm Monday on IQ tests:

    I got zero IQ for one test but then it was in a language I could not read.

    IQ tests do not contain any reading component, and they are only administered with extreme caution, if at all, on people who have not had adequate exposure (there is research suggesting school-aged students require at least 6 years of schooling in the new language to attain fluency) to the language the test is designed in. For people lacking the requisite language skills, if it is deemed necessary to evaluate their cognitive abilities, usually only the non-verbal/visuospatial aspects of the test would be administered. There is, however, an argument that ‘non-verbal’ tests are not purely non-verbal, as we e.g. use internalised language to assist with solving them.

    This was extreme, but IQ tests have difficulty with addressing cultural issues and with the different language experiences of testees.

    The most widely-used IQ tests today consist of verbal and visuospatial subtests in a 50-50 split. Separate index scores are calculated for the individual domains, but the overall (Full Scale) IQ remains the most valid predictor of scholastic achievement.

    It has since become evident that IQ tests also run into serious issues relating to gender. Girls have soft brains because they play with dolls, right? They are not so good at maths, right?

    Mathematical ability is not directly evaluated with the most commonly-used IQ tests today. There is an optional Arithmetic subtest, but it would only normally be administered as a substitute subtest if, for whatever reason, one of the other verbal subtests (yes, it comes under verbal reasoning ability) could not be adminisered, or was ‘spoiled’ (e.g. there was a major interruption during the administration of another verbal subtest).

    I found that it was quite easy to increase my ‘IQ’ by 20-30 points by learning the patterns of questions found in IQ tests and practicing the appropriate problem solving techniques.

    IQ is reported as a standard score (effectively, a z-score translated into a numerical score where 100 is the mean, and 15 is the standard deviation). It should always be reported with a confidence interval (usually 95%), as no standardised test will given an exact score. Depending on the age-group, the confidence interval could be +/- 7 or more. Usually it is skewed to regress to the mean, so e.g. if you obtain a high score, it may be +3/-7; and a low score example could yield a confidence interval of +7/-2.

    Naturally we are urged by the owners of IQ tests to only do the test once.

    Yes, because then practice effects may occur. Even if you don’t remember the specific test items, the strategies you used to answer the questions/solve the problems will likely remain familiar to you, and so there will often be some increase in the score if you undertake the test a second time.

    This sort of differential ability in different elements of IQ leads to interesting questions such as: which sorts or sets of questions should be weighted and how?

    Each subtest (normally you will undertake at least 8) is weighted equally when calculating the Full Scale score.

    I would be reasonably certain that my test IQ would have declined by 30 or 40 since then.

    That’s 2 or more standard deviations, which is extremely unlikely, unless perhaps you’ve had a major, serious neurological illness.

    The tests are standardised so that each age-group has a mean of 100 and a standard deviation of 15. From what I have seen (I am not a psychologist, but have worked alongside them) in the scoring manual, cognitive ability peaks in the early 20s and then gradually declines. So e.g. you may only need to get a raw score of 15 on one subtest to get a subtest scaled score of 10 (the mean) on one subtest if you’re in the 60-62 age-group, whereas you might have needed a raw score of 22 to get the same subtest scaled score if you were 24-25. (I’m pulling these numbers out of thin air; they don’t represent actual scores needed).

    Anyhoo, IMO, IQ tests are an interesting tool for personal learning with a grain of salt being the most useful tool.

    IQ tests are best, really, only at giving you an indication as to which broad category your cognitive ability falls into – i.e. are you average, well below average, or well above average (not the actual descriptors used). They also only supposedly measure your *potential* to benefit from formal education; they don’t actually measure mastery of skills such as reading, spelling, writing, or mathematical ability (achievement tests are designed for that).

  26. steve davis
    Sales gets a hammering.

    Sales defence will be that protests / objections she gets from both side proves her impartiality.

    It escapes her that it just proves her outright incompetence .. something both side of politics see.

    The ABC use the cover of devils advocate to hide their uselessness.

  27. Here’s a question: are conservative voters (particularly among the young) more likely to answer phone polls that progressive voters (who are probably more cynical about polling coys and busier doing stuff than conservative voters). If so, a polling coy could still get the right number of voters in the right demographic and the poll would still be off the mark.

Comments Page 24 of 24
1 23 24

Leave a Reply

Your email address will not be published. Required fields are marked *