Newspoll: 53-47 to Coalition

From a glass-half-full Labor perspective, this week’s Newspoll shows an encouraging three-point lift on the primary vote. But it still corroborates the result of the most recent Nielsen survey in suggesting a solid win for the Coalition.

This week’s Newspoll finds Labor making a minor gain on two-party preferred, from 54-46 to 53-47. It comes despite a three-point improvement on the primary vote to 37%, with the Coalition and the Greens steady on 47% and 9%, and the balance coming off others. The disparity is down to a set of numbers last time which suggested the rounding to two-party preferred had favoured Labor. The personal ratings are the first in a while where Kevin Rudd’s position hasn’t deteriorated, although this may be due to a bad sample for Labor last week. His approval rating is up one to 36% and his disapproval down two to 52%, while Tony Abbott continues improving steadily with approval up one to 42% and disapproval down two to 49%. Rudd’s lead as preferred prime minister has increased slightly, from 43-41 to 44-40. The sample is back to its normal size of a little over 1100, after being bumped up to over 1600 last week to allow for fortnightly aggregated state-level results from decent sample sizes.

The weekly Morgan multi-mode poll has the Coalition leading 52.5-47.5 on previous election preferences and 51.5-48.5 on respondent-allocated, up half a point on both measures. On the primary vote, Labor is down two points to 34.5%, the Coalition is up half a point to 45% and the Greens are up 1.5% to 11%. Full details including state breakdowns here. Here too the poll is back to a normal sample size, of 3419, after blowing out to 4515 for some reason last week.

The addition of Newspoll and Morgan to BludgerTrack finds two-party preferred moving 0.2% in the Coalition’s direction, but Labor gaining one on the seat projection. This is down to the confounding pattern of strong numbers for Labor in state breakdowns for Queensland, the latest examples being a 37% primary vote from last week’s Newspoll aggregate, 39% from Saturday’s Nielsen and 36% from today’s Morgan, compared with 33.6% at the 2010 election. This is flatly contradicted by all seat-level polling, most notably Saturday’s large-sample Newspoll of the eight most marginal Liberal National Party seats, which had the Labor primary vote down 4.5% from 2010. Due to the probability that seats selected for such polling will not be representative of the state at large, the model can only use the statewide results. So while the BludgerTrack vote and seat numbers look broadly in line with expectations in the other states, I suggest the projection of a three-seat gain in Queensland be treated with considerable caution.

Finally, Adelaide’s Sunday Mail brought us a Galaxy poll of 586 respondents in Hindmarsh, Labor’s most marginal seat in South Australia, which had the result at 50-50, a swing to the Liberals of 6%. I’m not sure if this was a live interview poll like Galaxy normally does, or an automated one such as they did last week for the first time with marginal seat polling in Sydney and Victoria. UPDATE: The poll was automated. The primary votes were 41% for Labor (44.7% at the 2010 election), 44% for the Liberals (38.6%) and 10% for the Greens (12.2%).

UPDATE: Essential Research is still at 50-50, but Labor is down two on the primary vote to 38% with the Greens surging three points to 11%, and the Coalition down one to 43%. The poll also finds 60% of Coalition supporters saying they will “definitely not change my mind” against only 46% for Labor, while 17% of respondents rating it “quite possible I will change my mind” against 11% of Coalition. Since I started paying attention a few years ago, I have never seen the Coalition fail to do better than Labor on this measure. I’m not sure whether this is a Coalition/Labor thing, or if it’s to do with the fact that every election I’ve been observing has been a bad one for Labor. Kevin Rudd’s net approval has moved into negative territory since a fortnight ago, his approval down four to 41% and disapproval up two to 45%. Tony Abbott is steady on approval at 37% and up one on disapproval to 52%. Rudd’s lead as preferred prime minister has narrowed from 47-35 to 43-34. Two points of encouragement for Labor: their parental leave scheme is favoured over the Coalition’s by 35% to 24%, and respondents were slightly more inclined to believe Labor rather than the Coalition would be able to pay for its commitments without spending cuts (41% thinking it very unlikely the Coalition could do so against 35% for Labor).

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.

1,645 comments on “Newspoll: 53-47 to Coalition”

Comments Page 1 of 33
1 2 … 33

  6. [GhostWhoVotes ‏@GhostWhoVotes 3m
    #Newspoll Primary Votes: ALP 37 (+3) L/NP 47 (0) GRN 9 (0) #ausvotes]

    so where did the +3 come from? Others? or dodgy rounding by NewsPoll.

  8. Well thats a big pile of nothing so no change then

    [#Newspoll 2 Party Preferred: ALP 47 (+1) L/NP 53 (-1) #ausvotes]

    Polling for this election is pretty consistent, just like 2007.

  10. [Stephen ‏@TheAviator1992 3m
    If I recall correctly, last election Newspoll went 52-48, then 48-52, then 50-50 in the last week. I wonder if it will be so jumpy this time]

  11. That Coalition vote stubbornly way too high at 47%. 3% primary vote increase for Labor though-would that tend to indicate this would be closer to 52.5-47.5?

  15. So it is looking like an ’07 result rather than a ’96 result. Keating lost 54-46, Howard 53-47. Considering Abbott almost won in 2010 I would expect the ALP would have taken that earlier in the year

  17. [so where did the +3 come from? Others? or dodgy rounding by NewsPoll.]

    Others and rounding I bet, guess that last newspoll ALP PV of 34 really was too low.

  18. re wol kolla and the remarks about non-maths based degrees.

    oh, now I understand. Wal Kolla is one of those people with only the right-half of a standard human brain.

  19. I predict the last newspoll to show 55-45 to lnp before the election on past form. I think final result of 45.8 to 54.2 in the election

  20. Yes based on those numbers it would just be 53/47 but that works out on the 2010 preference allocations. Labor’s primary vote may have been a bit underdone last time.

  21. Bemused – Gillard led Rudd in last week’s Newspoll amongst ALP voters alone, but Newspoll does not have the link up on its site although it was reported on twitter

  25. Puffy

    The thing is many policy makers, financial professionals, H.R professionals, Managers etc have studied Social Science, Marketing, Human Resources, Finance, Accounting, Economics etc.

  29. [ALP 37 (+3) L/NP 47 (0) GRN 9 (0) ]
    Why would 3 in 100 Undecideds go to Labor? When, where and whom did they poll?

  36. We are a confusing bunch Zoidlord. Having said that I doubt Rudd is in trouble in Griffith based on 53/47 so we probably have to be cautious about those individual seat polls.

  37. So, I got polled by Morgan face-to-face yesterday.

    Now, it seems, they know my address, date of birth, telephone number, and who I voted for in the last election, and who I intend to vote for in future elections, not to mention how much I earn, how much I have in the bank, who I bank with, what car I drive, who that car is insured with, as well as 100 pages of other questions.

    Afterwards, I was left thinking “why did I tell them all of that?”.

  38. Libs haven’t even started their advertising campaign – expect at least another 1% to LNP by election time.

    I see ALP has delayed launch till teh last minute so they can continue to ferry journos around at taxpayer expense. Soem things never change.

  40. wal kolla well who is going to preserve our culture and heritage then and politics degrees are still useful in government roles. Arts and politics degrees may pay less than law, economics and business, maths. science and medicine but that does not mean they are not without use

  41. I am still struggling to compute how a 3 point rise in PV translates to 1 point 2PP if the other party’s PV is unchanged. Please explain.

  44. [sprocket_
    Posted Sunday, August 25, 2013 at 10:50 pm | PERMALINK
    the comeback. ModLibs projections looking fanciful.]

    Oh really?

    How does a Newspoll showing a 53-47 result make my projection of about 52.5-47.5 fanciful?

    Haha LOL 🙂

  45. [Polling was done over the weekend. So its probably the boat buyback that caused this.]

    I doubt it. One single relatively minor issue doesn’t bring back 3% in a day.

    It’s a correction from the low 34 and the rest is sampling.

Comments Page 1 of 33
1 2 … 33

Leave a Reply

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