Newspoll: 51-49 to Labor

After the last result gave Labor its biggest lead of any poll since the election of the Abbott government, the latest fortnightly Newspoll has come in closer to trend.

GhostWhoVotes relates that the latest Newspoll has Labor’s lead at 51-49 after a blowout to 54-46 a fortnight ago, from primary votes of 41% for the Coalition (up two), 35% for Labor (down four) and 11% for the Greens (up one). More to follow.

UPDATE: The Australian’s report, which just maybe reads excessive political import into what’s actually statistical noise. Although it could indeed be telling that Bill Shorten’s ratings have again gone down despite a better set of numbers for Labor on voting intention.

UPDATE 2: Leader ratings have Tony Abbott up two on approval to 38% and down two on disapproval to 50%, while Bill Shorten is down two to 33% and up four to 43%. Tony Abbott makes a solid gain on preferred prime minister, his lead out from 38-37 to 42-36.

UPDATE 3 (Essential Research): Essential Research is 50-50, after the Coalition hit the lead 51-49 last week. The Coalition is down two on the primary vote to 42%, while Labor and the Greens are steady on 38% and 8%, and the Palmer United Party up one to 4%. The monthly personal ratings have Bill Shorten up two on approval to 32% and up five on disapproval to 39%, Tony Abbott down one to 40% and steady on 47%, and Abbott’s lead as preferred prime minister down from 40-30 to 39-33. A question on Qantas shows respondents react negatively to the words “jobs being sent offshore”, 62% pressing the “disapprove” button despite the qualification of it happening improving the airline’s “profitability and long-term success”, while only 25% opted for approve. Fifty-nine per cent think foreign ownership would be bad for Australian jobs and 46% bad for the economy, versus 16% and 24% good. However, it would be thought good for Qantas profits by a margin of 48-19, and good for air travellers by 30-25.

UPDATE 4 (Morgan): The latest Morgan poll, conducted over the last two weekends from a sample of 2903 by face-to-face and SMS surveying, has a bounce in Labor’s lead from 50.5-49.5 to 53.5-46.5 on respondent-allocated preferences, which is a slightly more moderate 50.5-49.5 to 52.5-47.5 on previous election preferences. The Coalition is down 1.5% on the primary vote to 39.5%, Labor is up 1.5% to 37%, the Greens are up 1.5% to 12%, and the Palmer United Party is up half a point to 4%. Morgan has taken to including state breakdowns on two-party preferred, the latest set having Labor ahead 55-45 in New South Wales, 57-43 in Victoria and 51.5-48.5 in Queensland, while the Coalition leads 54.5-45.5 in Western Australia, 52.5-47.5 in South Australia and 52.5-47.5 in Tasmania.

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.

1,524 comments on “Newspoll: 51-49 to Labor”

Comments Page 1 of 31
1 2 … 31

  1. We now have Labor being declared to have lost support although the 54 was never real in the first place (and nor was the 48 from Nielsen). 51 is simply a more accurate reading. If Nielsen gets 51 those covering Nielsen will talk about how much Labor has gained.

    My aggregate is 51.2 to Labor.

    Shorten’s approval is down so I’m expecting he’ll have his first negative netsat.

  2. If I had to bring down a budget in WA, with Colon dumping millions into his hole in the ground, I might have a breakdown too. How bad is the budget going to be?

  6. Kevin

    [We now have Labor being declared to have lost support although the 54 was never real in the first place (and nor was the 48 from Nielsen). 51 is simply a more accurate reading. If Nielsen gets 51 those covering Nielsen will talk about how much Labor has gained.]

    How can pollsters say one poll figure “was never real” and presumably other polls figures are “real”, or are other polls sometimes “real”?

    I honestgly do not know how polls can be “real?, “never real”, and “sometimes real”, though I havr nrver seen a pollster say “sometimes” real.

  8. swamprat@6

    Kevin

    We now have Labor being declared to have lost support although the 54 was never real in the first place (and nor was the 48 from Nielsen). 51 is simply a more accurate reading. If Nielsen gets 51 those covering Nielsen will talk about how much Labor has gained.


    How can pollsters say one poll figure “was never real” and presumably other polls figures are “real”, or are other polls sometimes “real”?

    I honestgly do not know how polls can be “real?, “never real”, and “sometimes real”, though I havr nrver seen a pollster say “sometimes” real.

    Pollsters don’t (usually) say this; polling analysts say it!

    The point is that by combining polls from different companies we can tell when a reading is not representative. If Labor were really up 54-46 when last Newspoll was taken then why did no other pollster get this? If Labor were really behind 48-52 when last Nielsen was taken then why did no other pollster get this? It just happens that when polling companies do their sampling sometimes they are unlucky and get an inaccurate result – this is inevitable from time to time when you use random sampling. Thus while Newspoll’s sample happened to have a 54-46 split we can be highly confident that this was not the true state of Labor’s lead in public opinion when that poll was taken.

  9. This poll seems within the range we might expect. It will be interesting to see if there is any change in the PPM figures.

  10. Slips into “Everything/ModLib” mode …

    Well strictly speaking, William, the word “colon” can also mean “a settler from the metropole”. It isn’t necessarily associated with faecal evacuation.

    Sorry, but after all that half-considered nonsense earlier from ML earlier on the purview of the principles informing 18C I felt inspired to do lampoon.

  11. davidwh@7

    The last Newspoll was taken after a shocking week for the Coalition

    Quite often a week will be declared great for one side or shocking for the other by the commentariat and the polls will show no movement. And at the same time as that Newspoll there was a fortnightly Morgan showing negligible movement.

    Public opinion generally just does not change by multiple points in a week. Sometimes it will but such cases are very rare.

  12. William,

    [I don’t think I’m going to allow the “Colon” nickname.]

    You sound undecided, which begs the question ‘what considerations are yet to be made’?

  15. I really know very little about polls and stuff, but Kevin seems to, and I’m adopting his position.

    A tad over 51 to Labor.

  16. [If people don’t take the hint, I’ll start sounding decided.]

    Ah, yes. Self-regulation with the threat of intervention.

    I like.

  18. Polls take samples and then extrapolate that to mean the whole community is going to vote in a particular way. Sometimes despite the pollsters best efforts the sample they gather is not quite a representative as is it supposed to be so it gives a result that may be out by a bit. If that is the only poll in a period it might be hard to show that to be the case but where there are a number of polls it can be compared with it can be determined a poll was not really as accurate as normal. If there were 10 polls taken in a close period of time and 9 gave very similar results while one was quite different it would indicate something was off in that poll. There is very little chance of it being an accurate poll.
    Also by comparing the results of various pollsters it can be determined if a pollsters methodology makes their results favor one party or another party.
    People like William here at Pollbludger and Kevin Bonham study polls and produce weight polls in a way of getting a figure that measures them all together and further increases the accuracy of the polling. The more data, provided it is used correctly, the better. If a poll is published that differs largely from an aggregate without any reason then it can be determined to be highly unlikely to be accurate.
    So if it is determined that the poll is inaccurate then it can be said that there was never really that level of support for the parties in the community so the poll was not a real representation of the voting intentions of the community.

  21. davidwh@20

    Agree Kevin but that week was as bad as they come.

    I don’t think anything in the Coalition’s term thus far compares remotely with Labor’s week from hell in February last year.

    Curious to see what Abbott’s netsat will be. Thus far the relationship between Newspoll 2PP and Abbott’s netsat has been ridiculously tight and if that continues he will be at -5 give or take not much.

  24. Just listen to what Kevin says

    If people aren’t paying attention to the aggregates by now and are still going off individual poll sets (like pretty much every news organisation except for the AFR), they’re just plain stupid!

  25. ReachTel came out with 53/47 so Newspoll wasn’t the only poll with Labor leading by a big margin. I still think the Nielsen is the poll with a result going most against the likely position and trend. All the polling is really showing there is a general lack of confidence and committment around.

  26. ReachTel came out with 53/47 so Newspoll wasn’t the only poll with Labor leading by a big margin. I still think the Nielsen is the poll with a result going most against the likely position and trend. All the polling is really showing there is a general lack of confidence and committment around.

  29. I’m really not caring about Shorten’s rating anymore because I’m starting to think the “Generation Blue” effect is becoming an issue (if that is what is happening, the Coalition are in mighty big trouble)

  31. The issues that have dominated politics since the election would normally have been drowned out by the honeymoon period. I reckon the Coalition’s problems have got to be either demographic or the leader or maybe both. Time will tell I guess …

  36. A anon treasury source has indicated to me that they were in the process of preparing to sell everything that could possibly be sold. We will see how much they were exaggerating.

  37. The good thing with ReachTEL is they publish primaries to a decimal point. They round the 2PP to a whole number but that doesn’t mean we have to do the same. I got 52.7 to Labor from the primaries for that 53-47 ReachTEL.

    The question with the Nielsen is whether there was really a bit of a wobble downwards from Labor’s 51-ish lead (which the Nielsen exaggerated) or whether there was not really a wobble and just one that seemed to be there because of the Nielsen data.

  38. [#Newspoll Shorten: Approve 33 (-2) Disapprove 43 (+4) #auspol]

    Shorten is an appalling easy fellow to dislike, it is fortunate this statistic means almost nothing at all.

  44. zoidlord@43

    @Kevin/41

    How many for Abbott?

    Since he became PM he has been negative four times out of seven. (As opposition leader he was negative 86 times out of 97, including the last 75 in a row.)

  47. Opposition leaders are almost never popular, it goes with the job. They are always seen as weak, negative and carping, because that’s what oppositions do. It doesn’t matter. If the voters want to put a government out, they will do so. All that matters is that the opposition leader is seen as safe and competent.

  48. zoidlord@46

    @Kevin/44

    Is having a bad stats in Opposition Leader, lead to overall bad stats in Goverment?

    Impossible to say because normally it leads to that Opposition Leader never making government in the first place. Abbott is really the only federal exception to this. At state level though Kennett had some very bad ratings as opposition leader but good ratings for a while as Premier.

  50. Psephos@47

    Opposition leaders are almost never popular, it goes with the job.

    Opposition leaders are not much less likely to be popular than prime ministers.

    Median netsat for prime ministers 1986-2014 = -1
    For opposition leaders = -4.

    Hewson, Howard (2), Beazley, Latham, Rudd – all these LOpps had long periods of good ratings.

Comments Page 1 of 31
1 2 … 31

Leave a Reply

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