Home > Oz Politics > The result is in the mail

The result is in the mail

September 19th, 2017

We got our equal marriage survey forms in the mail yesterday, and posted them back today. From what I’ve seen, about half the forms were delivered last week and nearly all will be done by Friday. And I imagine, most people will either respond straight away or not at all. So, it was kind of strange to see the official campaigns being launched at the weekend, rather as if an ordinary election campaign started at lunchtime on election day*.

On the other hand, the results won’t be announced until November, and the ABS is working hard to prevent any release of partial information. That’s if the votes were kept under lock and key on election night and not counted until the last postals and absentees had come in.

In these circumstances, I’m hoping for the slow-motion version of an exit poll. Next week, any pollster so minded could survey people to ask if and how they voted. We wouldn’t have the problem, which affected pre-survey polling, of unpredictable turnout, so the results should be as accurate as an ordinary opinion poll (that is, a 95 per cent confidence interval of plus or minus 2-3 percentage points for a sample of 1-2000). Is anyone going to do this, I wonder?

* For byzantine funding reasons, the major parties now leave their election launch until the week before election day, when quite a few people have already voted. But this is taking it a step further.

Categories: Oz Politics Tags:
  1. Smith
    September 19th, 2017 at 13:50 | #1

    I suppose there are measures in place to stop people photocopying their form and voting multiple times.

  2. Graeme Orr
    September 19th, 2017 at 13:57 | #2

    Essential Polling today suggested 9% of folk had both received and returned ballots. Reporting 55-34 split of decided voters. And that reports of intention to actually vote skewed a little to ‘yes’ supporters. Younger folk reported being less likely to vote. So either some social desirability bias, or opposition is a bit soft (the latter seems counter-intuitive to me, but I may have misread the lumpen tolerant middle). Nb the sample size of sub cohorts in such a poll is not huge. http://www.essentialvision.com.au/category/essentialreport

    I’m glad there is no running tally of votes received. It’s not the US in the 19th century! That kind of election skews campaigning. That said, I heard the ABS will be doing tallying as they go. Presumably using different teams and ‘chinese walls’, to count different batches. To reduce the chance of any reliable vote totals leaking.

  3. John Quiggin
    September 19th, 2017 at 17:23 | #3

    @Graeme Orr

    Thanks for this. Results on climate change also heartening, for a question giving maximum wiggle room to deniers, 64-24 pro-science, rising to 72-18 among uni graduates.

    Do you believe that there is fairly conclusive evidence that climate change is happening and caused by human activity or do you believe that the evidence is still not in and we may just be witnessing a normal fluctuation in the earth’s climate which happens from time to time?

  4. Tim Macknay
    September 19th, 2017 at 17:24 | #4

    @Smith
    Each survey form has a unique bar code.

  5. Smith
    September 19th, 2017 at 17:29 | #5

    @Tim Macknay

    So they’ll all be machine-read to make sure everyone has voted just once. Will a machine be counting the votes too?

  6. Tim Macknay
    September 19th, 2017 at 17:35 | #6

    @Smith
    Interesting question. Since it’s an ABS survey rather than a genuine vote, I would presume that they will be mechanically counted, rather than manually as would be the case in a real election. The information on the ABS web site is also suggestive that they will be mechanically counted (e.g. mark ‘with a dark pen’).

  7. Smith
    September 19th, 2017 at 17:49 | #7

    @Tim Macknay

    More ammunition for the losing side to claim it’s all dodgy and the result should be ignored.

  8. Tim Macknay
    September 19th, 2017 at 18:43 | #8

    @Smith
    Indeed.

  9. John Goss
    September 19th, 2017 at 20:31 | #9

    Why does machine counting give ammunition to the claim it’s dodgy? Are you denigrating the machines we created?

  10. Robert Merkel
    September 19th, 2017 at 22:40 | #10

    An acquaintance of mine is observing the count and has provided some of the technical details of how the ABS is running it.

    The votes are being counted by OCR, and this aspect of it is frankly as slapdash as every other aspect of the survey.

  11. Smith
    September 20th, 2017 at 09:25 | #11

    @Robert Merkel

    What is the typical error rate with OCR?

  12. Ken Fabian
    September 20th, 2017 at 12:26 | #12

    And I imagine, most people will either respond straight away or not at all. So, it was kind of strange to see the official campaigns being launched at the weekend

    From polling I’ve seen the undecided vote is unlikely to change the outcome, even if a large portion could be swayed to vote No. The only point I can see to having this opinion poll is that it concedes enough to the Right faction that it will allow Turnbull to allow a conscience vote without any appearing to confront or directly oppose them; purely for internal LNP party purposes.

    Smith, detection of marks with a specific area of a page is not OCR – and it is quite reliable; they’ve been using it for multiple choice in school examinations for a long time.

  13. Robert Merkel
    September 20th, 2017 at 13:24 | #13

    @Smith

    As Ken has pointed out, the software is technically simple and potentially highly reliable.

    It’s just that the quality control being used to ensure that it’s doing the job correctly and not malfunctioning is less thorough than it should be. I’m not sure how much I can say here, but I am really not thrilled about this aspect.

  14. Moz of Yarramulla
    September 20th, 2017 at 14:05 | #14

    Smith :
    What is the typical error rate with OCR?

    I would expect that the two things “which box is marked” and “scan this barcode” would be very, very accurate. Both are areas where OCR actually shines, and these days even “find this box on the page” should be very accurate. Half an hour to set it up as a form in a modern scanner sounds reasonable to me, so a couple of weeks for a team of conslutants to do the same thing on a much, much larger scale would be generous I think.

    Opening the envelopes and piling up the forms would be the most expensive part I suspect, because it’s likely still cheaper to do that by hand.

  15. Smith
    September 20th, 2017 at 16:30 | #15

    @Moz of Yarramulla

    Piles of forms can go missing, as they did with the WA Senate election in 2013. I wonder if the ABS is going to reconcile total votes with total envelopes received.

  16. sunshine
    September 20th, 2017 at 17:13 | #16

    I think the no vote may be a bit higher than expected. There is alot of talk about free speech and protecting religious freedoms .That would be Christian freedoms of course . Muslim freedoms have been under siege for ages now- such hypocrisy, Christians think they can tell Muslims how to dress. . The government should have just declared homophobia the same as racism and said bad luck about your freedoms.

    Also I think the slippery slope argument is getting traction because it is a good one .But I think that it is a slope we should ,and will, be on .Gender fluidity here we come !

  17. Ken Fabian
    September 20th, 2017 at 21:24 | #17

    I think most people inclined to voting yes will have done so before the campaigning begins in earnest.

  18. poselequestion
    September 21st, 2017 at 13:03 | #18

    Ours arrived today. Just handling them nauseated me. To think that government in this country can waste so much money in abdicating it’s responsibilities and even worse, use the process to effectively stay in power.Can it get worse?

  19. Moz of Yarramulla
    September 22nd, 2017 at 09:03 | #19

    sunshine :
    I think the slippery slope argument is getting traction because it is a good one.

    I was thinking last night that the slippery slope argument is tricky for the side that has actual Nazis supporting it. They should be very careful to draw a line between what they’re arguing for and what their supporters are arguing for, and show that one cannot possibly lead to the other.

    http://www.huffingtonpost.com.au/2017/08/20/these-nazi-homophobic-posters-arent-just-vile-but-totally-false_a_23155271/

    But I think that it is a slope we should ,and will, be on .Gender fluidity here we come !

    We’ve always had gender fluidity, but now we talk about it more. Look at all the records of people being found to be trans only after they died, right back to the time of the first fleet. Before that they didn’t keep records but at least some countries had places for non-binary people.

  20. Stephen
    September 22nd, 2017 at 12:55 | #20

    “For byzantine funding reasons, the major parties now leave their election launch until the week before election day”. IMHO it is not really that byzantine.

    Before the campaign launch party leaders can claim back their travel expenses – ie the taxpayer foots the bill – because it’s parliamentary business. After the campaign launch such expenses are considered political, so the political parties have to pay them.

    Thus it’s in the parties’ financial interest to delay their campaign launches for as long as they plausibly can get away with it.

    Simples.

  21. sunshine
    September 23rd, 2017 at 10:38 | #21

    @Moz of Yarramulla

    Yes .Gender rigidity is a bad problem for all. Have a look at Julia Crispin s ‘Why I am not a Feminist’ if you like. A good account of the problems with modern feminism as feminism on Patriarchal terms ,individualistic and consumption based. Subsumed .

    It’s odd to hear the No mob complaining about violence given the long history of persecution ,bashing and murder by No voting types. People are driven to suicide by their violence.

  22. John Quiggin
    September 23rd, 2017 at 12:43 | #22

    @Stephen

    I was aware of this and stand by “byzantine”. Why would anyone define the start of the election by the date of a speech rather than by the issuing of the writs. The answer is that it’s an absurd loophole that stays in place because it suits the apparatchiks on both sides, at the expense of democracy.

  23. Moz of Yarramulla
    September 25th, 2017 at 08:43 | #23

    @sunshine

    It’s always tricky to talk about “the problem with” something broad. It’s very easy to latch onto some aspect of one part of a wide-reaching philosophy and decide that you don’t like the whole thing. You could equally say the the problem with modern Christianity is the death wish of the Dominionists, or with democracy is Donald Trump.

    My usual counter with feminism specifically is “are women property of men, yes or no”… oh, turns out you’re a feminist after all. Any woman who publishes a book is, by that fact alone, a feminist. There’s a spectrum, from “women shall stay in the home on pain of death” to “women should withdraw from patriarchy and form a separate, women-only society” (or another extreme “men and women are all humans and should have equal human rights as far as possible”… just as controversial, but in a different way).

  24. Moz of Yarramulla
    September 25th, 2017 at 12:09 | #24

    @sunshine
    Oh, and this Guardian review put me right off: https://www.theguardian.com/books/2017/mar/08/jessa-crispin-is-right-to-critique-modern-feminism-but-her-book-is-a-missed-opportunity

    This is particularly striking when Crispin’s presumably deep engagement with the broad spectrum of feminist theory is represented by little more than an acknowledgements list at the end of the book. After all, the objections Crispin makes to contemporary feminism are not new.

    The lack of theory, history and context represented in this work feels like a missed opportunity and makes it difficult to translate Crispin’s critique into a practical pathway forward. There’s no more than a passing mention of intersectionality in the book, for example – arguably the most prominent theory circulating in the contemporary feminist movement and a concept that has a complicated relationship with the often-atomising identity politics that pervades so much of the contemporary feminist activity that Crispin finds infuriating.

    I read that but didn’t link to it because it usually takes JQ hours (and hours, are we there yet? 😀 ) to approve posts with links and I wanted to get a reply in. But thinking about it, I reckon that commentary sums up my thoughts very well and deserves linking. From my perspective as an anarchist critic of both the modern nation-state and the capitalist system, looking at modern US feminism in an ahistorical, decontextualised way is almost entirely pointless. Like some economists attempts to critique the excesses of neoliberalism from a free market perspective… it kinda works, but it also misses most of the important points – still buying into the “we live in an economy, not a society”.

    I favour a politics of inclusion, partly tautologically (as a white man, other reform politcs tend to exclude me) and partly for the standard philosophical reasons (I’d like to try living in a radically more inclusive society). Although, and this might strike you as odd, there’s a wider variety of options for gender-specific dress in Lakemba than in most of Australia, thanks to the local Muslim majority (for example http://theislamicnews.com/men-dress-islam/ and both the garments pictured there are common where I live)

Comments are closed.