Mind the gap

Written By: - Date published: 6:10 am, July 10th, 2008 - 16 comments
Categories: polls - Tags:

It’s worth remembering that polling gaps really do narrow when election campaigns start. We all remember the 2002 campaign. Labour reached 53% support in April; on Election Day in July, they polled 41%. The gap between labour and National had been blowing out to 30% in the polls but in the end it was less than 20%.

Take a look at what happened to Labor and the Coalition’s support in Australia last year. Once the campaign began the gap narrowed from over 20% to less than 5% on Election Day.

(source)
If National is only ahead of Labour by 5% on Election Day, John Key’s ambitions for his CV are toast.

[note the polls are evenly spaced on the graph but there were more frequent polls after the campaign proper started, so the narrowing of the gap is more sudden that it appears. Labor lost 16% of lead in just 40 days]  

16 comments on “Mind the gap ”

  1. Rob 1

    I guess you all live in hope but I don’t think so. This gap is there for a reason the public don’t believe this Government has listened to them and has ridden rough shod over them.

    This gap is a protest vote and people really want a change.

    If you get out of the Standard Tower and walk the streets and talk to people you will find it very hard to meet people who admit that they voted for Labour last time. Then any that you do talk to who said they did wont be this time.
    National will get in with a large majority a very large majority.

    The country desperately requires a economic vision and a plan for growth currently we have nothing thus our slide down the OECD ratings.

    Labour is a fine Weather Government when they can spend up big on their Social Policy’s, they are not a Government you want in when times get tough. Its time for a refreshing change.

  2. AndrewE 2

    Hmm…so Labour lost 12% in 3 months or so. What would that put them on now? Low 20s? 😉

  3. Aj 3

    When the Nats release their tax relief policy the gap will close. Unless it mets the public expectation of $50 a week for the average worker, and I don’t mean $50 spread over 3-6 years.

  4. The TNS Conversa poll (for NZIER on climate change) released on the 7th, also asked people who they would vote for.

    It showed National on 40%, Labour on 34%, Greens on 4%, NZF and Maori on 3%……and a HUGE 27% were undecided.

    That poll makes this election look like a close race right now.

    It is also the only poll that tells us how many are undecided.

    I sent ‘thestandard’ an e-mail about this poll. It got virtually no media coverage other than Fran O’Sullivan writing on the ETS / Climate change side of it.

    I blogged on it.

    http://truthseekernz.blogspot.com/2008/07/tns-convera-poll-indicates-close.html

  5. So Steve you see Helen Clark jumping across the Grand Canyon of the poll gap like Evel Knievel ? I admire your optimism.

  6. Thanks Steve, I’ll do something on that later. Hadn’t seen you email.

  7. Phil 7

    That might help you sleep at night, but ultimately is an irrelevant comparison.

    Two points you need to remember;

    Australia;
    The Australain system is still “first past the post” in every practical sense, so the motivations for why you might change your mind on who you vote for are different to the NZ case.

    I also hold the view that Labor ran a comparatively poor campaign.

    NZ 2002;
    When it became clear that National stood no chance of victory, a reasonable portion of traditional National voters switched camps to Labour, specifically aiming to push them over the 50% mark just to keep the greens out of power – whether or not that was a smart idea is a debate for another day…

    As election day neared, Labour didnt lose support back to National, they lost it to minor parties that were likely to support Labour. On the flip side, when the ’05 result was in doubt right up to election day, the minor parties were the big losers.

  8. Lew 8

    Phil: “The Australain system is still “first past the post’ in every practical sense”

    This is complete bollocks. There are huge differences between STV and FPP. Just because neither are proportional don’t make ’em the same, or even vaguely similar.

    L

  9. jaymam 9

    I would like to see ALL polling companies showing Undecided.
    Those that do not have no credibility.

    I’ve plotted the Conversa poll as a pie chart (ignoring Won’t Vote):
    http://i34.tinypic.com/21j9jdx.jpg

    Let’s see the same from all reputable polling companies.

  10. Lew 10

    jaymam: The reason for not doing so is the presumption that, all else being equal, undecided voters will split along the same lines as the remainder of the electorate. This is a fair baseline assumption, if not statistically valid or a good predictor of performance. However it tends to reinforce the plurality advantage on the basis that people like to back the winning team. I expect this is why you’re objecting it.

    L

  11. jaymam 11

    I suspect that the 27% Undecided will be deciding much closer to the election, depending on Party policies. There’s no logical reason why they should split along the same lines as the Decideds.

    ONE News Colmar Brunton Poll, Poll Method Summary states:

    “NOTE: The data does not take into account the effects of non-voting and therefore cannot be used to predict the outcome of an election.
    Undecided voters, non-voters and those who refused to answer are excluded from the data on party support. The results are therefore only indicative of trends in party support, and it would be misleading to report otherwise.”

  12. Daveski 12

    Actually, Steve I agree with you without being inconsistent with my other comments.

    At this stage, National looks like it will win but I would be surprised if the gap didn’t close.

    I agree entirely that there is a large element of a protest vote (natural given the economic situation and people’s perceptions about Labour). I also agree it’s smart for National NOT to push its policies until it is forced to as this will inevitably challenge some of those who want to send a protest vote to Labour.

    And before someone gets in, it is not undemocratic to hold off announcing your policies – just as it is not undemocratic for Labour to do things it hasn’t explicitly spelt out in policies.

    Another point that needs to be made is that any poll is only a snapshot at a point in time – and the only point in time that counts is some time in November.

    I would still be worried if I was in Labour’s camp, but I agree it ain’t over until the calorifically challenged wimmin gets up to yoddle.

  13. Phil 13

    Lew,

    Fair enough – my comment was poorly worded. Suffice to say that the act of voting for individual candidates is the primary determinant of the outcome of the Australian election.

    The factors that motivate you to change the way you intend to vote once the outcome becomes clear (clearer) are significantly different to a single party-based voting system.

  14. milo 14

    So what would it take to be a large gap then?

Links to post