EMRS: Liberal 55, Labor 22, Greens 18 in Tasmania

Another shocking opinion poll for Labor, this time the latest quarterly EMRS survey of Tasmanian state voting intention. Removing the undecided and including “leaners”, as other pollsters do but which EMRS publishes as “Table 3”, the poll has Labor down three points from an already disastrous result in May to 22 per cent, the Liberals up seven to 55 per cent and the Greens down four to 18 per cent (remembering that EMRS polls have traditionally been over-generous to the Greens). Since assuming the premiership in January, Lara Giddings preferred premier deficit over Opposition Leader Will Hodgman has progressed from 27-38 to 22-42 to 19-52.

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.

19 comments on “EMRS: Liberal 55, Labor 22, Greens 18 in Tasmania”

  1. I was polled by them last Thursday night. This is the second quarterly omnibus in a row in which they have polled me – which I would think is extremely unlikely. There is also an as yet unreleased federal component to the poll. The questions were asked (rough wording as I had to write it down as the questions were asked) in the following order…

    Do you know which electorate you are enrolled in?

    In a year from now do you expect your standard of living to improve, get worse or stay about the same?

    In a year from now do you expect opportunities for employment to improve, get worse or stay about the same?

    In a year from now do you expect the state of the state economy to improve, get worse or stay about the same?

    Can you please list social and economic issues of importance to you…

    If a state election were held today which party would you vote for?

    Who is your preferred Premier… Lara Giddings, Will Hodgman or Nick McKim?

    If a federal election for the House of Representatives were held today which party would you vote for?

    Do you support or oppose the Federal Government’s carbon tax?

    Strongly or somewhat?

    Do you support the Federal Government’s move to establish an immigration detention centre in Pontville to house detainees for up to 6 months?

    Strongly or somewhat?

    If a federal election for the seat of Bass were held today would you vote for Labor’s Geoff Lyons, Liberal Andrew Nikolic, the Green candidate or someone else?

    Geographic and Demographic weighting questions…

  2. It also really annoys me how the Tasmanian media unanimously reports the headline figures from Table 1, ignoring Table 3, and then try and compare the results to the previous state election or to national or inter-state polling figures which exclude the undecided. These comparisons should be made with Table 3. EMRS should help the media out by reporting Table 3 as their headline results in bold type if necessary.

  3. This doesn’t surprise me. Without knowing too much about the state’s politics, Tas Labor has probably had its day for a while, especially after the 2010 debacle. Maybe, if Abbott becomes PM, those numbers might narrow considerably. The one thing that works for Labor (and will probably ensure any Lib state government is brief) is that the Greens are a strong presence and do tend to favour ALP governments over Lib ones.

  4. I would like to know the reason for these 2, are they after Lyon’s personal vote?

    If a federal election for the House of Representatives were held today which party would you vote for?

    If a federal election for the seat of Bass were held today would you vote for Labor’s Geoff Lyons, Liberal Andrew Nikolic, the Green candidate or someone else?

  5. I provisionally reject this poll as unscientific on the basis of Adam’s evidence that voting intention was not the first significant question asked.

    We need to see adequate scientific standards in all opinion polls and the presence of contaminating lead-in questions – doubtless not intended to create skew but very likely to do so – is simply not good enough. I don’t doubt that the government’s plight is dire but a poll conducted in this way adds nothing to the already obvious.

    Tasmania needs a new pollster.

  6. I agree Kevin. The previous time they polled me for their last omnibus (again, what are the odds of that?) they asked a series of attitudinal and issues questions prior to voting intention as well. Here is my rough list of questions, in order, asked at the prior quarterly poll…

    EMRS questions; in order asked…

    Which age bracket are you in…?

    Do you think your households standard of living will improve, stay about the same, or get worse in coming months?
    Do you think the state of the Tasmanian economy will improve, stay about the same, or get worse in coming months?
    Do you think that job opportunities in Tasmania will improve, stay about the same, or get worse in coming months?

    What do you think are the most important social or economic issues facing Tasmania at the moment? (open ended question)

    South Australia has a container deposit scheme… would you support container deposit legislation in Tasmania? – strongly support, somewhat support, somewhat oppose or strongly oppose.

    Tasmania has 29 councils… do you think that is too many?
    There are (282?) councillors in Tasmania… do you think that is too many?
    Would you support council mergers?
    Do you think council mergers would save money?

    Voting intentions… if state election held tomorrow… Labor, Liberal or Green
    Preferred Premier… Hodgman, Giddings, McKim

    Do you have an Aurora Cable P.I… is it plugged in?

    Weighting questions…
    Which council area do you live in?
    What best describes your household circumstances?
    Employment situation…
    Household income…

  7. dovif

    I expect the Examiner has commissioned the additional Federal questions and that we can anticipate an exclusive story on the weekend.

  8. I agree, they need a new pollster. I am happy to apply for the job.. I just want to run some questions past you first:

    Q1: What state are you in (If respondents answer with one of the six states of Australia we can hang on to them… if they are from Craddock and respond:”Stoned off my gourd.. man!!” a polite phone click is correct etiquette here)

    Q2: How amazed are you at how bad democracies can actually get?
    Q3: Given that the ALP can’t govern properly anywhere, the Libs have their heads up their bot-bots and policies of the Greens show that they are not technically residing on the planet, how prettily do you think you could draw Smurfette on your next ballot?
    Q4: If Tasmania could be ruled by a psychic octopus in a despotic dictatorship, how much better do you think she could run the state?
    Q5: Do you think candidates might be more intelligent in this state if their parents hadn’t found their own first cousins so damn attractive?

    Is this a good start?

  9. Well there you go Glen, the Tas poll mirrors the results in the rest of the country.

    What was that I was saying about the damage to the ALP brand again?

  10. I knew it Mod Lib.

    I knew it…

    The Blue Tide is running all over Australia atm.

    Who have thought that a couple of years ago eh??

    Me thinks Federal Labor are adding a few % points to the Libs total in Tas too…

  11. This poll is obviously rigged
    Galaxy sample in QLD too small
    Neilsen a rogue
    JWS poor polling techniques
    Morgan Face to Face unreliable
    Newspoll biased
    The PBers refrain
    I think there is a pattern here
    Self delusion a la Crikey

  12. Stanny, I realise you’re probably doing your usual thing of chucking a bomb in and not hanging around to watch the demolition of the complete rubbish you’re talking, but here goes anyway.

    [This poll is obviously rigged]

    If the poll indeed did not put the voting intention up front, then anyone who is reasonably informed about the subject can tell you they’ve gone about it the wrong way. Who has asserted that the poll was “rigged” against Labor? Nobody. Adam Clarke made no intimation to that effect, and Kevin Bonham expressly said he doesn’t doubt Labor is in dire straits. Both of these people are clearly smarter than you are, as am I, and you should listen to us. A poll thus structured is unreliable – I make no judgement on whether it produces a lean towards Labor, a lean towards the Liberals, or simultaneous effects in either direction that cancel each other out. Based on EMRS’s track record, it would seem their problem is that they overstate the Greens.

    [Morgan Face to Face unreliable]

    Yeah – in favour of Labor, as everybody knows. But what the hell, throw it in anyway, even though it directly contradicts the point you think you’re making.

  13. There might be some vibe here in which strongly leftist or Labor-loyal posters routinely dismiss polls that are bad for Labor but I am not such and don’t do such so really I think stanny could have left this thread out of that little piece of sarcasm. For the record I don’t think Newspoll is biased at all – unlike the paper that reports on it. Indeed I have just used Newspoll as a basis for an analysis of leader unpopularity in the last 25 years that will be up on Tasmanian Times pretty soon.

    I believe the unsatisfactory methods of this EMRS poll would clearly have had the effect in the current political climate of focusing voter attention on the state’s economic problems and thereby inflating the Liberal vote. My suspicion as to the size of that inflation is at least a couple of points. Could be more. The open-ended question about “social and economic issues” is particularly troubling because that is causing the voter to think about policy matters that are often considered the domain of the major parties immediately before expressing their view. Likewise an open-ended question about the environment would have trained the voter to think about policy matters that are the preserve of the Greens and made them more likely to specify the Greens as their intended vote.

    The overestimation of the Green vote which EMRS is almost always guilty of occurs because they get an abnormally high proportion of undecideds and then redistribute those proportionally. However the electoral record is that the figure before the redistribution is a really good prediction and the Greens don’t really get any of the “undecided” vote. Same probably applies to other pollsters, but other pollsters don’t have anything like as many undecided responses.

    Anyway on TT I have served notice that I am not taking EMRS seriously until they fix the question order problem and make it publicly clear that they have done so.

  14. [There might be some vibe here in which strongly leftist or Labor-loyal posters routinely dismiss polls that are bad for Labor but I am not such and don’t do such so really I think stanny could have left this thread out of that little piece of sarcasm.]

    Yes, it would have been an almost reasonable comment on the main thread.

  15. As anticipated the Examiner will have exclusive Federal EMRS results tomorrow. This is likely to mean they requested to have questions asked on the ‘carbon tax’ and the Pontville detention centre prior to voting intentions – and it is very likely they knew this would skew the result. EMRS should not be complicit in this.

  16. For those still interested, The Examiner’s push polling finds the following federal results in Bass… Liberal 41%, Labor 27%, Green 14%, undecided 12%. Excluding undecided and redistributing the Green vote with an 80% preference flow to Labor, the Examiner reports the two-party result as 53.7-46.3 in the Liberal’s favour.

  17. Thanks Adam,

    Not that I would put too much stock in this poll, given the question order, but what about Braddon (and the others)?

Comments are closed.