Morgan phone poll: 59-41

Possibly presaging a future pattern of early-week phone polls to tie us over until the Friday face-to-face, Roy Morgan brings us a phone survey of 611 respondents showing Labor leading the Coalition 59-41 on two-party preferred, and by 50 per cent to 36 per cent on the primary vote. Morgan alternated between phone surveys and larger-sample face-to-face polls until the end of August, after which it moved exclusively to face-to-face.

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.

495 comments on “Morgan phone poll: 59-41”

Comments Page 1 of 10
1 2 10
  1. “The phone polls tended to show somewhat bigger leads for Labor.”

    It’s interesting then that Morgan went on to say that the reason his face-to-face series has been recording a much higher Labor vote than other polls is because of an inherent Lib/Nat bias in relying on phone polls.

    “The Morgan Polls are showing higher levels of support for the ALP than most other polls. As this is of interest to all poll watchers it is worth noting that the Morgan Poll conducted ‘face-to-face’ includes those who don’t have telephones and those who only have mobile phones — generally an ALP skewed segment.

    “Results for telephone polls that exclude these people have an inherent L-NP bias. We believe this explains the difference between the Morgan telephone poll and the Morgan ‘face-to-face’ poll.”

    http://www.roymorgan.com/news/polls/2007/4221/

  2. Only 611 respondents does not inspire confidence. Any word on what the margin of error is?

    While accepting the above, it is certainly interesting that every outlier is skewing Labour’s way, with the range being from 55-61 TPP in favour of Labour.

    And can anyone give me any sort of meaningful reason why the polls will ‘inevitably’ narrow after the election is called. Is there historical precedent for this?

  3. Aussiemurf, 611 is around 4% MoE. The ACN conducted in the same period with 1404 would have about a 2.7% MoE, rounded to 3%. MoE decreases proportional to sqr rt of sample size, so you need 10,000 sampled for an MoE of 1%.

  4. Aussiesmurf – the error margin is 3.9%. As for whether the polls will narrow, Possum Comitatus covered this in his excellent post here where he examined the last five elections and concluded

    So The Narrowing is 2 from 5 at best, 1 from 5 that can be called certain, and where both of those occasions can be explained by the longer term trends and movement occurring in the polling series over the previous months.

  5. More of the same.

    I really don’t see what JWH has to gain by delaying firing the starting gun. Now that the anniversary of the last election is upon us, all political coverage will be couched in “when-will-he-call-it?” overtones. Everything he does from now (and, really has been for at least the last month, since APEC) will be seen through that prism. Reconvening parliament seems like a bad idea to me, not least because it takes all those marginal seat holders out of their areas for two weeks, allowing the ALP candidate to fill the local void.

    The Libs’ only perceived advantage in going back to Canberra is to drop a bucket on Rudd (and others) under parliamentary privilege, but given the specaular failure of the “character” wedge up till now, I can’t see that it will do them any good.

    For months we’ve been hearing that the formal start of the campaign would act as a “circuit breaker”, so why doesn’t he just get on and call it? The trouble for him is that there are now only three possible Saturdays, 17 & 24 Nov, and 1 Dec (maybe four, if you count 8 Dec). There is now such expectation about the coming election that the actual announcement may well end up having no impact whatsoever.

    On the previous thread, someone compared Howard to that last annoying guest at the party, who just won’t go home, and I think this feeling will start to catch on among voters the longer he leaves it. It could well be that he knows he is toast, and he is trying to experience every last little bit of being PM for as long as he can. Howard doesn’t seem to have much interest in anything other than politics, and suddenly not being PM any more will leave quite a void in his life.

  6. c s, if the estimate you want is near 50%, you can get a 95% CI using: 100/sqrt(n). With 611 sampled, that’s 100/sqrt(611) is approx 100/sqrt(625) = 100/25 = 4%.

  7. Lord D (#4 & #13) the confidence interval (MoE) is also affected by the size of the population you are sampling and how closely divided the responses are – eg a sample of 400 from a population of 1,000 will have a lower error margin than a sample of 400 from a population of 1,000,000, and there is a lower error margin if the result is 90%-10% than if it is 51%-49%.

  8. William, I just took a look at the difference between phone and face-to-face for the Labor 2PP vote. For all polls going back to May, the average for face-to-face is 58.22 and the average for the phone poll is 57.79.

    So, the difference is very marginal. Face-to-face produces a slightly higher Labor vote.

  9. Thanks to all for their links re ‘The Narrowing’ and also info on margin of error. Although I enjoy reading Michelle Grattan, it has annoyed me severely when she makes comments like “Both parties naturally expect the polls to narrow when the election is called”.

    Firstly, because she makes reference to what the parties expect without attribution.

    Secondly, because there is no statement by her as to her opinion as to whether this expectation is reasonable.

    Is there any thoughts as to what sort of legislation may be pursued if Parliament does sit next week?

  10. The polls are very consistent, which is really what you would expect. For many weeks now, nothing much has occurred that might change opinions.

    The most obvious thing taking place is this waiting game. Everyone has paused. The electorate will exercise patience for a while longer, but if too many more days pass, “the waiting” will start to become a dynamic force with its own momentum. It could engulf Howard completely if he is not careful.

  11. Canberra boy, I know that, but it seems to over-complicate it. Most of us are interested in the major parties’ vote, which are near 50%, and political opinion polls will always have a large population base, so the sample size to population ratio is too small to be relevant, even for single seats.

  12. The SMH is reporting that Howard’s planning to go to the Pacific Islands Forum next week. He wouldn’t do that in the first week of a campaign…which suggests that he’s planning to be away from parliament for its redundant final sitting.

    Has he gone mad?

  13. The narrowing theory seems to be driven by the theory that those voters who support Labor but how believe that the Coalition are better economic managers are somehow irrational soft or mushy types who will come to their senses eventually. But these voters could be quite rational, they just regard economic management as a less important issue than WorkChoices, health, environment etc. these issues are why they are supporting Labor. Why not identify ‘soft Liberal’ voters who prefer Labor on health & the environment?

  14. Yep, barbara, the govt was re-elected 9 Oct 2004, so, IMO, time’s up to call the election. This evidently isn’t Howard’s opinion, though, so expect a Dec 8 election. He’ll call it when he either thinks he can win, or he’s forced to. I don’t think the risk of being absolutely wiped is a factor; he wants to win, and will run the risk of Labor winning in utter carnage to do it.

  15. Barbara,

    Yes, it is 3 years today… I’m so sick and tired of waiting for an election – just call the damn thing so we can get over it already!!!

  16. If the poll is held on the latest practical day – say 1 December (surely 9 December would be a step too far) – with the shortest permissible campaign period, then Howard can wait until 28 October to call the election, 19 days from today. This is his timetable: 19 days to find something that might resonate in his favour. 19 days to try to inflict some damage on Rudd. 19 days to try to hack away at Labor’s primary support, then run as hard as he can.

    It’s not long. He’s had about 300 days so far, and gone nowhere. Surely, if he had something to run with, this late in the piece, he’d be running day and night.

    19 days has got to be the limit to call it on. 33 days to campaign. 19 + 33 = 52 days for Howard. I’m counting every one.

  17. Billy MacMahon waited and waited: 2 December, I think. I doubt that waiting made a scrap of difference in the end. He was always going to take a hiding.

  18. Are there not something like 1,000 new AWAs every month? AND as time goes by the prospect of an AWA gets closer for some. A reasonable number of fearful or upset punters in these two categories would spread the word into their immediate community.

    Combine this with the Union’s grass roots campaign Your Rights At Work it should mean a continual leaking of votes to Labor across the country.

    We also have the youth continuing to sign up which I gather equates to a higher Labor percentage.

    You would think that Howard could only recover meaningful ground is if there is some major event of the exact right category – space ships, War of Worlds. I don’t see that security issues will do it.

    The only reason the Govt would wait any longer is to get their smears and half invented stories [remember stripgate and the fabricated story from Downer’s office that Rudd touched someone] coordinated with some of those rather nasty journalists we know about. Or are they still searching for contributions!

    It would be beautiful justice if Howard tried one his famous election tricks and it backfired in a large way. Common Johny, gives us one of your Children Overboard/Tampa calls to fear and xenophobia.

  19. The phone polls tended to show somewhat bigger leads for Labor.

    Am I right about this? It’s not based on anything scientific, just a general impression I had.

    Other way round, isn’t it? F2F polls lead to bigger leads for Labor, as there are more people contacted through that method than can be contacted via phone, because traditional ALP voters are younger, and use more mobiles.

  20. If the Unflushable Turd is going to be OS next week and Parliament is due to sit I expect that the Dogs (Abbott & Costello et al) will be let loose on St Kev. Should make for enjoyable listening but I don’t see the sense in it. I know it has been only 3 years since the last election; it just seems like a decade ago.

  21. 36
    passthepopcorn Says:
    October 9th, 2007 at 3:22 pm

    planning shmanning…howard won’t leave australia, because he’s panicky.

    you think there might be a coup? wishful thinking, i’m afraid.

  22. yes, well, it must feel like that to him. seriously, when he’s in such an obviously dire position, he’d be hyperventilating at the thought of leaving the country. he needs the security of janette’s bosom. goodness, that made me feel quite queasy…

  23. # 38 – AM, Rudd has learned a very great deal from Peter Beattie. Accept that a mistake has been made, correct the mistake and move on. Far from being a “vote loser”, Rudd’s dressing down of McClelland just makes him look strong and presidential.

  24. Re #18 & #21 – I also looked at the variation between Morgan’s F2F and phone numbers for each month that the telephone polls have been done, and in some cases the F2F have higher Labor 2pp, while in other months the phone polls were higher. In all cases the margins of error overlap, both on monthly averages for the different techniques and between adjacent individual polls (even the 18-19-25-26 Aug F2F with 54.5% and the 21-22 Aug phone poll at 60% are consistent results!)

  25. Queensland, 1987. ABC Four Corners televises Chris Masters’ “The Moonlight State”. While Premier Joh Bjelke-Petersen is out of the state his Deputy (and acting Premier) Bill Gunn establishes the Royal Commission into Police corruption – the Fitzgerald Inquiry – which eventually led to the collapse of the Government.

    Would that this was a precedent for 2007.

    Can someone slap Mark Vaile and ask him to call the frickin election while Howard is OS please.

  26. AM at 38,

    I agree with Snake Boy – this issue is a flash in the pan sort of ‘mistake’ by Labor. They don’t look like they’re going to tear themselves apart over this – another instance of their phenomenal discipline this time around (when compared to previous campaigns).

    It was essential that Rudd correct the ‘mistake’ as soon as possible, however – already the SMH had written an editorial about the death penalty (commending Labor’s perceived position) and Sid Marris had written an opinion piece about it at The Oz (plus, of course, the atrocious front page headline). The issue will now be dead by the end of the week (at the latest).

    That said, contrary to what Snake Boy said, I’m not sure if Rudd gains anything out of it. It looks a bit “flip-flop” from Labor and it smacks of me-tooism. At best, it’s a net zero result for both parties.

  27. [What stage would it have to reach for the Governor-General to sack him?]

    I presume if he hasn’t asked to have parliament dissolved for a January 18th election the major general govenor general would just dissolve parliament for him.

    I don’t think that would count as a sacking.

  28. JWH is stalling for time so he can properly consider who he picks for the Prime Ministers XI cricket match with Sri Lanka early next year. He wants to lock in the team before Rudd takes office, so he can point to the teams victory as his last great act of public service.

  29. The death penalty issue worries me. I am an ardent opponent of the death penalty, but i could not stand another Howard goverment. It is a highly emotive issue. Please don’t let this be the the 2007 Tampa.

    I hope this is a non issue.

    Do you think the issue could win votes for Labor? My guess is that it would be 50/50 in terms of Australians who believe inthe death penalty.

    In the interests of staying on thread Morgan poll is a very small sample for a national poll but fun to look at!

Comments are closed.

Comments Page 1 of 10
1 2 10