Newspoll: 52-48 to Labor in Western Australia

A new Western Australian state poll from Newspoll chimes with this week’s ReachTEL in crediting Labor with a narrow lead, but marginal seat polls paint a grimmer picture for the Liberals.

The Australian has a Newspoll of state voting intention in Western Australia that has Labor leading 52-48, down from 54-46 in the previous poll published in May. On the primary vote, the Liberals and Nationals are steady on 40%, Labor is down one to 41%, and the Greens are down two to 9%. Colin Barnett’s personal ratings are now at dire levels, with approval down three to 28% and disapproval up three to 61%. Labor’s Mark McGowan is also down, although from a much higher base, with approval down five to 46% and disapproval up five to 33%. However, he has increased his already substantial lead as preferred premier from 46-32 to 47-29. Hat tip to GhostWhoVotes.

Some further local polling:

The West Australian reported on Tuesday that union-commissioned polls showed swings to Labor of 18% in the bellwether northern suburbs seat of Wanneroo and 12% in the regional city seat of Bunbury. In Wanneroo, Liberal member Paul Miles’ 61-39 winning result in 2013 had switched to a 57-43 lead in favour of Labor, with primary vote of Labor 43%, Liberal 33% and Greens 5%. In Bunbury, which retiring Liberal member John Castrilli has held since he won it from Labor in 2005, the Liberals’ post-redistribution margin of 11.9% was gone entirely, with primary votes of Liberal 34%, Labor 33%, Nationals 9% and Greens 7%, with two-party preferred at 50-50. The polls were conducted by ReachTEL for the Electrical Trades Union and Australian Services Union, respectively from samples of 657 and 529.

• I’ve been provided with results of polling from the Perth seats of Joondalup, Mount Lawley and Bicton, conducted from a sample of 200 per electorate for the Kimberley Like Nowhere Else Alliance. The questions mostly relate to marine park and environmental issues in the Kimberley, which were collectively deemed “quite important” in influencing voting behaviour by 42%, with 26% rating it more strongly and 32% less so. Voting intention was also gauged, but the value of the results is limited by the fact that no follow-up question was included to prompt the 20% undecided. Of the 482 who offered a response, 43% opted for Liberal or the Nationals, compared with 54.6% in these seats at the 2013 election; Labor was on 37%, up from 31.3%; and the Greens were on 14%, up from 9.9%. In two-party terms, this suggests a swing to Labor of around 9.5%, which is consistent with a statewide two-party preferred of around 52-48 in favour of Labor. The breakdowns suggested the Liberals would retain Joondalup and lose Mount Lawley with Bicton too close to call, but the samples at this level are too small to be instructive. The poll was conducted by phone and online from Patterson Market Research in September.

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.

8 comments on “Newspoll: 52-48 to Labor in Western Australia”

  1. That’s a new worst netsat for Barnett of -33. The lowest netsat a Premier has ever polled then gone on to win at the end of that term is -24 (Bob Carr).

    Only two Premiers who have polled netsats of -33 or worse were even allowed to make it to the next election, those being Anna Bligh and Lara Giddings (both were thrashed). Of others to do so, five resigned (Greiner at the behest of the crossbench) and Joh was sacked by his party.

    No Opposition Leader state or federal has led an incumbent Premier by more than 18 points in a two-way question since 1994. (Will Hodgman led Lara Giddings by 31 points in a three-way question in 2014.) Shorten led Abbott by exactly 18 points once.

    Of course, all this history is not necessarily predictive. The 2PP is still close-ish.

  2. Still a 10 seat hill to climb for Labor. Probably 4-5 of these are in the bag, but there might still be enough fat on other LNP seats to hang on. A seat like Morley is one Labor should have got back long ago. The local member is grumbling that he has few campaign funds even now. I notice the West has suggested the Libs will throw heaps into sandbagging seats in the hope this will get them over the line. I guess the point is, that despite all, the polls have not been good for the LNP for some time. I would suggest the previous big lead was over stated and there was likely a grumpy vote in the 54 for Labor. The thing is this current LNP government is an old government. It appears and acts as though it has had enough. I think the electorate senses it might be time for change but as we saw in SA old governments can and do hang on.

  3. Does the question of who commissions a poll (ie unions in this case) automatically cast a shadow on the results?
    Does a pollster alter their methodology to suit who commissioned the poll, with a view to giving a more favourable result?
    The pollster’s credibility would be tarnished if they did that I would imagine but is it possible?

  4. The ancient SA government had the benefit that the same power was not in power federally, and also the major benefit of SA’s electoral geography. Barnett’s has neither of these advantages.

    I wonder if the Greens to Labor flow at this election will remain as comparatively weak as in the past.

  5. henry @ #3 Thursday, November 3, 2016 at 11:52 am

    Does the question of who commissions a poll (ie unions in this case) automatically cast a shadow on the results?
    Does a pollster alter their methodology to suit who commissioned the poll, with a view to giving a more favourable result?
    The pollster’s credibility would be tarnished if they did that I would imagine but is it possible?

    One reason for caution about all commissioned polls is that if the commissioning source doesn’t like the result, they almost always won’t release it. So no matter how good and fair the pollster is there is always a risk of selection bias in terms of what actually gets out.

    I don’t see too much evidence of method-tweaking in commissioned polls as concerns the voting intention questions. One thing to keep an eye on is whether it uses respondent or last-election preferences, which seems not to have been stated in this case. The issue questions in commissioned polls are frequently extremely dodgy.

  6. Thanks KB and you are certainly right about the apparent tainting of the vote depending on who is in power in Canberra. Certainly in the Rudd-Gillard time not even the local Labor people wanted either of them over here while the the Government was more than happy to tar both local and Federal Labor with the same brush. Mind you, Charlie Court was happy to do it years ago. The Feds are always good for a bashing here regardless of who is in power in Canberra because many in the West have a chip on their shoulder the size of a mountain when it comes to those from the East. And your comment about SA is also salient as there seems to be some electoral imbalance between some of the seats in SA which seems to give some benefit to Labor – as I understand it.
    I still think Labor has quite a hill to climb to get the 10 seats – or more to govern.

  7. @Trico We want to be the “Wait Awhile” state when it comes to change, but not to “Wait Awhile” when we want our spare change. Barnett’s been the best at flaming that flame, almost monthly. Can’t remember Carpenter/Gallop being quite rabid about it, but they were naturally opposed to the Federal Howard government anyway.

    As to this poll, never fear, Stokes and Murdoch will get the Libs over the line, not to mention Gina will remember she’s a West Aussie and help out herself.

    Minority government most likely result, I’ll say +1 seat to Liberals.

Leave a Reply

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