Australian Federal Polls

[Smid]

I think the previous "dead woman walking" thread is a bit presumptive and also a bit disrespectful, so thought I might create this thread for posting poll updates. If the boardbashi disapproves, it could always be moved over to the IGD board.

Latest Newspoll was out today.

The poll was conducted from January to March - State sample sizes were 700 to 1,692. Tasmania was excluded from separate figures because the sample size was too small but included in the national total. The national sample size was 5,741. Margin of error was 1.3% nationally and 3.7% for the state with the smallest sample.

Primary vote at the election:
Party/National/NSW/Vic/Qld/SA/WA/Male/Female/18-34/35-49/50+/Capitals/Non-Capitals
Labor/38.0/37.7/42.8/33.6/40.7/31.2/34/38/34/35/38/40.1/34.8
Coalition/43.6/44.1/39.6/47.4/40.2/50.6/45/42/35/45/48/42.1/45.9
Greens/11.8/10.7/12.7/10.9/12.0/13.1/14/14/22/14/9/12.8/10.3
Others/6.6/7.5/4.9/8.1/7.1/5.1/7/6/9/6/5/5.0/9.0

Primary vote in the latest poll:
Labor/32/31/34/30/34/31/30/34/33/31/32/32/32
Coalition/45/45/44/47/39/48/46/43/37/45/50/44/46
Greens/11/11/13/9/11/10/11/12/17/13/6/13/9
Others/12/13/9/14/16/11/13/11/13/11/12/11/13

Labor 2PP (table doesn't include gender or age breakdown):
Election/50.1/49.5/55.3/44.9/53.2/43.6/52.5/46.6
Poll/46/46/50/42/51/44/47/45

Results of polls taken between the election and the latest poll were also included but I didn't upload them here.

[Smid]

Team blue 2PP would be 54% nationally.

[Smid]

Newspoll out today. Bracketed number reflects change from the election.

Primary Vote
Labor 27 (-11)
Coalition 51 (+7.4)
Greens 11 (-0.
Others 11 (+4.4)

Two Party Preferred
Labor 41 (-9.9)
Coalition 59

[Smid]

News this morning was using the headline "worst result for Labor in eleven months" and the political commentator reiterated that and mentioned that eleven months ago, Labor's primary vote was one point below this at 26, and was Labor's worst-ever poll result.

[Smid]

Latest Newspoll (11-13 May) released today. 1141 respondents, weighted to reflect population dispersion.

First bracketed number is the election result. Second bracketed number is change since last poll.

Primary Vote
Labor (E=38%) 30 (+3)
Coalition (E=43.6%) 45 (-6)
Greens (E=11.8%) 12 (+1)
Others (E=6.6%) 13 (+2)

Two Party Preferred
Labor (E=50.1%) 45 (+4)
Coalition (E=49.9%) 55 (-4)

Virtually no change in leadership ratings. Big table of numbers relating to the budget.

[Smid]

What happened? I know that was a bit of a high, but still. Or did Swan leak some sort of pre-budget goodie?

Could be a budget bounce, or could be a bit of statistical noise. The Nielson poll yesterday declared "no budget bounce" - but I think their last poll was higher for Labor than Newspoll's last poll.

The time series figures published in the paper are (so election, and then every two weeks, starting from late February this year):
(dates are: Election - [2012 Feb 23-26, Mar 9-11, Mar 23-25, Apr 13-15, Apr 27-29, May 11-13])

Labor: 38,          35, 31, 28, 29, 27, 30.
Coalition: 43.6,          45, 43, 47, 48, 51, 45
Greens: 11.8,          11, 12, 11, 12, 11, 12
Others: 6.6,          9, 14, 14, 11, 11, 13

So as far as this year's Newspoll goes, the numbers are pretty close to the low-end of middle for Labor, low-end of middle for Coalition, normal for the Greens, and high end for others (presumably driven by Katter's Party).

The Age-Neilson poll published on Saturday was:

Labor: 28
Coalition: 49
Greens: 12
Independent: 5
Family First: 3
Other: 4

[Smid]

That would be on a uniform swing. I think Emmrrson would hang on, I know his seat pretty well. Even in the Queensland election, Woodridge stayed Labor (I think the safest Labor seat? Or is that Inala?). Regardless, voters in the parts of his seat West of the highway would really struggle to vote Liberal, especially without the Can-do Campbell factor.

Swan has lost his seat before, but the redistribution took some more Liberal parts out, I think. Still, I believe no state seats in his electorate went Labor at the Queensland election a couple of months ago, so he could have some difficulties. Of course, the Queensland Premier (now, LNP Leader then) had massive cult-like personal popularity, which no other leader currently attracts, so that could weigh in his favour.

Garrett, hmmm, I think most state seats in his electorate stayed Labor last state election, Coogee, which went marginally Liberal, is shared with Wentworth. I think Polnut knows that whole area better than me, but I suspect Garrett may hang on.

The other seats, I don't know them well enough. If any seat in WA resists the swing, it will be Perth and/or Fremantle. Perhaps Smith could hang on, perhaps his seat will swing. I wouldn't want to predict that one either way.

[Smid]

Latest Newspoll is out. Labor has risen for a second polling period, off their lows of a month or so ago (pre-budget). This has been mainly due to a slide in "Others" - primary vote, the Coalition is up 2, Labor up 1, Greens steady and Others down 3. I suspected that the last poll had overstated Labor's support by a percent or so and that this poll would see it ease off very slightly, and that the media would therefore over-emphasise any slide but clearly I was wrong because Labor's up again.

I'm not sure what is contributing to the slide in others... I think their rise was driven predominantly by the rise of Katter's party in Queensland (I'm sure that a correlation could be proven or disproven fairly easily by looking at the date in the surge to others and the date Katter announced he was forming a party, but I'm not going to spend that much time looking for it... I think it's probably true and I'll leave it at that). Anyway, if that's the case, perhaps the Katter party failing to win seats in the Queensland election outside his own federal electorate may be acting as something of a dampener on their support. Just speculation on my part, really.

Anyway, here are the polling numbers:

Primary Vote
Labor 32 (+2%)
Coalition 46 (+1%)
Greens 12 (nc)
Others 10 (-3%)

Two Party Preferred
Labor 46 - 54 Coalition

1152 respondents, +/- 3% MOE, weighted according to population distribution.

[Smid]

54-46 is heckuva more realistic 2PP number than 59-41. I'd say 1996 rather than 1983 will be the template come E-Day.

Absolutely more realistic, although there have been some unrealistic election results of late - Queensland and NSW both had 2PP results of >10% for the governing party - I think Queensland was 63%, and NSW was higher. Prior to that, I thought the 2002 Victorian result was an anomoly, where Labor received 57%. Of course, the larger the electoral population, the harder it is to get a margin like that... federal Labor is nowhere near as unpopular in Victoria and South Australia (and presumably Tasmania, but the sample size is too small for Newspoll to release their figures there and likewise for the ACT). Labor's support in those states and the ACT will make it harder for the nation-wide 2PP to blow out to the levels experienced in Queensland and NSW.

[Smid]

I missed last fortnight's poll, but Newspoll is out again today. first bracketed number is the election result, second bracketed number is change since last poll.

Primary Vote
Labor (38%) 30% (-1)
Coalition (43.6%) 46 (+2)
Greens (11.8%) 12 (-2)
Others (6.6%) 12 (+1)

Two Party Preferred
Labor 45 vs 55 Coalition (Labor down 1 since last poll, election was Labor 50.1 vs 49.9 Coalition).

1146 respondents, MOE +/-3%.

[Smid]

Newspoll and Nielson are both out today - Newspoll also has released their quarterly "polls by state and demographic breakdown" (see the first poll in this thread for last quarter's). Both are good polling firms, so I normally take a look at both to check they're roughly the same and then really just follow Newspoll. For the past couple of months at least, Newspoll seems to be consistently a point or two higher for Labor, but both are broadly in agreement. Anyway, I'll post the Newspoll here soon.

[Smid]

The poll was conducted from April to June - State sample sizes were 845 to 2,014. Tasmania was excluded from separate figures because the sample size was too small but included in the national total. The national sample size was 6,884. Margin of error was 1.2% nationally and 3.4% for the state with the smallest sample.

Primary vote at the election

Party:                    Labor     Coalition     Greens     Others
Nation-wide            38.0        43.6            11.8          6.6
NSW                       37.7         44.1            10.7          7.5
Vic                          42.8         39.6            12.7          4.9
Qld                         33.6         47.4            10.9          8.1
SA                          40.7         40.2            12.0          7.1
WA                         31.2         50.6            13.1          5.1
Male                       34            45               14             7
Female                   38            42               14             6
18-34                     34            35               22             9
35-49                     35            45               14             6
50+                        38            48               9               5
State Capitals        40.1         42.1            12.8          5.0
Non-Capitals          34.8         45.9            10.3          9.0


Primary vote in the latest poll

Party:                    Labor     Coalition     Greens     Others
Nation-wide             30           47               12            11
NSW                        29           47               12             12
Vic                           34           42               15              9
Qld                          22           54               11             13
SA                           33           45               10             12
WA                          32           48               11              9
Male                        28           49               11             12
Female                    31           44               14             11
State Capitals         30           46               14             10
Non-Capitals           28           49               10             13


Labor vs Coalition 2PP
                              Election                Poll
Nation-wide         50.1-49.9             44-56
NSW                     49.5-50.5             44-56
Vic                        55.3-44.7             51-49
Qld                       44.9-55.1             35-65
SA                        53.2-46.8             47-53
WA                       43.6-56.4             45-55
State Capitals      52.5-47.5             46-54
Non-Capitals        46.6-53.4             42-58

[Smid]

So 1975 all over again when I plug this into Green's 2010 swingometer, but ultimately I think the 2PP margin will be 5-7 points. So every QLD ALP MP goes down, including the Cabinet troika. Talk about Portillo moments, unless you guys call them Howard moments over there.

I agree with you that the swing reported in the poll is probably larger than swing if an election were actually held. Additionally, it's important to note that this poll was taken over the past few months and Labor was doing worse at the start of that period, compared to the end of that period, so that may have over-stated the swing slightly. That said, the Nielson poll out today had Labor's primary vote at 28%, compared to the Coalition's at 48%, and a 2PP of 58-42 in favour of the Coalition. All of Nielson's 1,400 respondants were polled this weekend just gone (Thursday night to Saturday night), so the same argument can't be used there (of it being a long-term poll).

[Smid]

As I've talked about on the chat, Labor have a natural floor of somewhere between 40-45 seats, unless there is a particular event. Add the fact they're the current government, and the floor is about 50-55. which would still be a wipeout, but please don't start salivating over 25 or 30 members. Federal politics is not state politics, and VIC, SA, TAS, and the ACT are not NSW and QLD.

What may be a genuine possiblity though, and one that would be incredibly harmful to Labor going forward, is a truly awful result in the senate. I could very easily see the coalition getting three across the board, and maybe a fourth or at least a non-labor/greens fourth in a couple of states.

Agree with much of what you've said here, particularly about the state divide. I'd expect the polls to tighten closer to the election, but I also expected that in the NSW and Qld state elections. Highly unlikely for the Coalition to win a majority in the Senate (I think I'd say impossible, but can't remember how many seats the Greens and Labor won last election and therefore will carry over into the life of the new parliament) so I wouldn't worry about that if I were you. If Labor's serious about confronting the Greens, as they have implied over the past few days, then Labor will have to agree to the Coalition bill abolishing the Carbon Tax because a Double Dissolution results in lower quotas for the Senate (since there are 12 positions to be elected, not 6), which will make it easier for the Greens to hang onto seats at Labor's expense.

[Smid]

Latest Newspoll is out today. As before, first bracketted number is the election result, second bracketted number is the change since the last poll.

Primary Vote
Labor          (38%) 31% (+1%)
Coalition     (43.6%) 38% (+2%)
Greens        (11.8%) 11% (-1%)
Others        (6.6%) 10% (-2%)

Two Party Preferred
Labor (50.1%) 44% (-1%) vs Coalition (49.9%) 56% (+1%)

1141 Respondents, MoE +/-3%

[Smid]

Here are my predictions.

The leaders will not change by the time for the next federal election, so both Gillard and Abbott will still be in place.
The 2PP margin will be something like 54-46.
Kevin Rudd will stand down at this election if he is not picked as party leader, likely putting Griffith at risk to the Liberals.
Labor will be reduced to 2 seats in Tasmania, Lyons and Franklin.
Labor will have 0-2 seats in Western Australia, with Brand gone.
Labor will lose Lingiari in the Northern Territory.
Labor will lose 2PP everywhere except Victoria, ACT, and possibly Tasmania.
Bill Shorten will get the leadership if he wants it after the election, but there will be a 50/50 chance he won't take it since Abbott will likely be in for at least 2-3 terms. If he does, his deputy will be either Greg Combet or Tanya Plibersek.

What are your thoughts/feedback on these? I've only been following Australian politics well for about six months, so these may not be the most accurate guesses.

Lingiari is a very big call, but that said, I was predicting Indiana for Obama back in about March 2008, so it's probably not a bigger call than that. I would be very surprised if it changed hands, however.

Lyons and Frankin retention for Labor isn't too out there... Bass is I think almost certain to change, and it's not outlandish to think Braddon might also. The real question, then, is Denison. A few seat polls recently suggest he's probably doing well. I don't think he'll get Liberal preferences, but if he's doing as well on primary vote as recent polls suggest, he'll probably finish first and get enough preferences (even if everyone preferenced against him, he'd probably pick up enough leakages to fall across the line, but I think at the very least, the Greens will preference him and he'll just get there). One of the recent polls had him first and the Liberals second - if that's the case, Labor would really need to hope for Greens preferences and other independent preferences to pull in front of the Liberals, and then to pick up Liberal preferences to pull in front of Wilkie, whereas I think it's more likely that if it's 1. Wilkie, 2. Liberals, he'll almost certainly win with preferences from the Greens and Labor. That said, Labor might preference the Liberals, because a Liberal MP would probably only hold the seat for a single term, and then Labor could win the seat back, whereas Wilkie serving a second term would be harder to dislodge at the following election.

Brand is also a big call, but not an unrealistic one. I think Stephen Smith will retain Perth, so I don't think Labor will be reduced to 0 seats in WA. I could be wrong, though. The tougher question for WA is what happens in O'Connor (where the WA National ran against an incumbent Liberal, won, and announced he would sit on the crossbenches, not as part of the Coalition - I assume this means that there will be a Liberal candidate against him this election).

[Smid]

New Newspoll out. I'll wait for Smid to post the full thing but 2PP is 56/44 and primary 46-28.

Indeed it is. Here are the numbers:

As before, first bracketted number is the election result, second bracketted number is the change since the last poll.

Primary Vote
Labor          (38%) 28% (-3%)
Coalition     (43.6%) 46% (-2%)
Greens        (11.8%) 11% (-)
Others        (6.6%) 15% (+5%)

Two Party Preferred
Labor (50.1%) 44% (-) vs Coalition (49.9%) 56% (-)

1158 Respondents, MoE +/-3%

[Smid]

Any chance Gillard continues as Labor leader if the defeat is narrow enough?

There will be a by-election in Lalor after the election.

[Smid]

The latest Newspoll is out

As before, first bracketted number is the election result, second bracketted number is the change since the last poll.

Primary Vote
Labor          (38%) 33% (+5%)
Coalition     (43.6%) 45% (-1%)
Greens        (11.8%) 10% (-1%)
Others        (6.6%) 12% (-3%)

Two Party Preferred
Labor (50.1%) 46% (+2%) vs Coalition (49.9%) 54% (-2%)

1141 Respondents, MoE +/-3%

[Smid]

Dead cat bounce or did Lab have some good news recently? Biggest primary bump we've seen in a while.

Labor didn't so much have good news, but rather no news... it's Winter Recess and the Olympics have taken most media coverage. With the focus off the government, there seems to have been something of a recovery. I don't like including leadership ratings here because they don't change the primary vote/2PP polls, but I might mention them because they seem to back up this "no criticisim" period - the PM's satisfied level has stayed the same, and her dissatisfied level has dropped 3%. The Opposition Leader's satisfied level has increased 2% and his dissatisfied level has dropped 5%, however the Opposition Leader is attracting 2% less support on the question of preferred PM, although he's lost that support to "Uncommitted" rather than directly to the PM. When the Olympics conclude and the focus is turned back on the Government, I suspect things may change a little once more.

[Smid]

The latest Newspoll is out, taken 17-19 August.

As before, first bracketted number is the election result, second bracketted number is the change since the last poll.

Primary Vote
Labor          (38%) 35% (+2%)
Coalition     (43.6%) 45% (nc)
Greens        (11.8%) 11% (+1%)
Others        (6.6%) 9% (-3%)

Two Party Preferred
Labor (50.1%) 47% (+1%) vs Coalition (49.9%) 53% (-1%)

1129 Respondents, MoE +/-3%
9% uncommitted and 2% refused were excluded from the sample.

[Smid]

Summer bounce, right?

There are a few theories out there... But I don't think so (or Winter bounce, for that matter, since our seasons are reversed).

Firstly, the Prime Minister backed down on asylum seeker policy. She's implemented one of the three planks of the Coalition's policy, as recommended by the panel of experts she appointed six or so weeks ago. There were news stories praising her for breaking the deadlock, there were news stories condemning her for not having acted sooner and for dismantling this policy when Labor first came to office, and there have been news stories condemning her for being too harsh by going back to Howard Government policy - so the plus one to the Greens might have been disgruntled Labor voters adopting that third position, but it seems that generally the electorate is viewing the policy backdown as a positive for Labor.

The other reason being cited is state-level factors, particularly in Queensland. The government is considerably decreasing the number of public sector jobs in the bureaucracy (not front-line workers like police, nurses and teachers). A non-partisan friend of mine who used to be in the public service up there, and with a number of friends in the public service up there, and whose company has several departments as a client, was saying that if you look at the jobs figures under the previous government, the public service was used as a way of hiding the true level of unemployment, and that the public service is too bloated and that when he was visiting a client a couple of weeks ago, there were numerous examples that he could see of people with nothing to do. That said, he also believes that the cuts go too far. I think that's probably an unbiased assessment - that it needed to be done, but this is seen as too many jobs.

This is reflected in a ReachTEL poll commissioned by Channel Nine and on which Channel Nine News reported last night. ReachTEL does push-button-response style polling, which gives them higher response numbers (since they're not paying a call centre). They do quite a bit of single electorate polling for this reason (and are typically the only public polling company that does single electorate polling, because they're the only ones who use that style of polling and therefore the only company where it's economical to do that). They haven't been around long enough for me to have an opinion on their accuracy. The poll shows the LNP dropping by 12% compared to the election result, and Labor increasing by 10%. On the question "The Queensland Premier Campbell Newman is implementing a number of initiatives to reduce the states debt. How do you feel about the new initiatives?" 49.3% responded with "Gone too far."

That said, these numbers seem to have come off "Others" rather than the Coalition. Perhaps Coalition supporters don't oppose what's being done, but there were some who were upset at Federal Labor, who had parked their vote with "others" and are now returning because of Queensland State Government public sector cuts? Of course, I don't know what's causing it and the poll doesn't reveal state figures, but that's one of the theories I heard discussed on the news this morning, and it was also mentioned last weekend (as in, a week ago, and before this ReachTEL poll) on The Insiders.

[Smid]

Hmm. Once Parliament resumes there should be another uptick. I expect the final 2PP margin to be no more than 53-54.

I agree. Elections in Australia are apparently never won by huge margins, even if a party has huge leads during the campaign.

In the past 63 years, representing 25 federal elections, the highest nation-wide Two Party Preferred vote received was 56.9% by the Coalition in 1966, followed by 55.7% for the Coalition in 1975. The highest 2PP result for Labor was 53.23% in 1983. This gives a range of results of just over 10%. Even in 1996, the Coalition only obtained 53.63%, and in 2004, 52.74%.

Obviously, state-wide ranges vary more considerably - even disregarding the ACT (for being so small), with its 22.3% range of 2PP results. Tasmania is over 17% and Victoria is over 16% from best to worst result in 2PP.

Furthermore, although there was a 2.58% swing to the Coalition nationally in 2010, this was by no means uniform. There were swings to Labor in Tasmania (4.41%), Victoria (1.04%) and South Australia (0.78%). Indeed, despite a knife-edge result nationally (second narrowest margin since 1949 - only slightly wider than 1990), Victoria and Tasmania recorded their strongest result for Labor in that time.

In terms of swing from one election to the next, the largest swing was recorded in 1975 (7.1% to the Coalition), with the second-largest in 1969 (7.1% to Labor - interestingly, the election before they formed government). In recent years, the largest swings were in 2007 (5.44% to Labor), 1996 (5.07% to Coalition) and 1998 (4.61% to Labor). Of the 25 elections, only 6 have experienced a swing to the incumbent government, while 18 elections have swung against the governing party (and swing isn't available for the 1949 election, obviously).

EDIT: It's not immediately obvious, but I was saying all that to support your "never won by huge margins" comment.

[Smid]

The latest Newspoll is out, taken 31 August - 2 September.

As before, first bracketted number is the election result, second bracketted number is the change since the last poll.

Primary Vote
Labor          (38%) 33% (-2%)
Coalition     (43.6%) 46% (+1%)
Greens        (11.8%) 8% (-3%)
Others        (6.6%) 11% (+2%)

Two Party Preferred
Labor (50.1%) 45% (-2%) vs Coalition (49.9%) 55% (+2%)

1151 Respondents, MoE +/-3%
7% uncommitted and 2% refused were excluded from the sample.

[Smid]

The latest Newspoll is out, taken 14-16 September.

As before, first bracketted number is the election result, second bracketted number is the change since the last poll.

Primary Vote
Labor          (38%) 36% (+3%)
Coalition     (43.6%) 41% (-5%)
Greens        (11.8%) 12% (+4%)
Others        (6.6%) 11% (-2%)

Two Party Preferred
Labor (50.1%) 50% (+5%) vs Coalition (49.9%) 50% (-5%)

1,166 respondents, MoE +/- 3%
6% uncommitted and 2% refused were excluded from the sample.