Further to the post on the latest Fairfax/IPSOS poll (blogged here Latest poll – the key is ‘undecided’) there has been an interesting comment at Kiwiblog fron Richard29 on the ‘undecided’:
The reporting of this poll on stuff had an interesting detail:
“Pollster Duncan Stuart said a breakdown of undecided voters suggested many were “soft” National supporters, who had started looking around.”
So basically an increasing number of people over time are getting bored of the current government that we’ve had for over 4 years, their lives haven’t really changed that much for the better, but they don’t really see the opposition as a workable government in waiting, so they are just kind of uncertain…
This also seems to gel with the differences between the Fairfax poll and the Morgan poll a few days ago. They are both on reasonably large samples and conducted at the same time, but the Fairfax poll has National at 45% and the Colmar Brunton poll has them at 49%.The big difference here is that the Colmar Brunton Poll has only 10% undecided and the Fairfax poll has 17.2% undecided. That is a massive difference and will be down to interview approach.
I’ve worked in and managed teams calling out on political surveys in NZ and Australia. The reality is that a whole bunch of people you speak to say they don’t know who they would vote for – they don’t feel that they have enough information to make a decision or they don’t want to give a ‘wrong’ answer in case they change their mind later.
Most people will give a preference if prompted with something like “So which party of the ones I listed, if you had to choose, who do you think you might prefer to vote for, even if it’s only a very small preference?” It was many years ago that I last conducted the Colmar Brunton Poll – but I’m fairly sure they employ a technique similar to this. That allows for picking up in changes of mood amongst the 20% or so of floating voters who are undecided.
The undecideds are a difficult group to get to grips with – on the one hand you don’t want to be reporting the preferences of people who are disengaged from the political process and may not show up on the day, on the other hand theirs are the votes which swing elections.
The last comments on what people base their decisions on are particularly interesting.
They are often making their decision not on the basis of a good understanding of the policies of all parties and their implications but based on broad perceptions of the political brand or personalities involved rathern than anything substantive:
- “John Key seems friendly and down to earth”,
- “Those Labour people seem to be quite angry and negative”,
- “That Goff guy has been around for years”,
- “The Greens seem young and energetic and that Maori girl co-leader has such a warm smile”,
- “Winston is a bit of a character but he did achieve a lot for people like me with the supergold card and pension increases”.
I have suspected that many people base their preferences on overall perceptions of the personalities rather than on the policy detail that parties seem to obsess over. And over time most people see past the lipstick and make their own judgement pigs despite all the PR.
Andrew
/ February 20, 2013I agree with Richard and Pete on this. How a poll deals with undecided voters can be important.
An additional point – how undecided voters respond to further prompting can differ at various points an election cycle. In a poll following the Christmas holiday period for example, when I believe everyday voters are less engaged in politics than usual, ‘don’t know’ can mean something quite different to what it means in a poll one week out from an election.
Pete George
/ February 20, 2013I can identify with that point.
For me in the past after Christmas mid term “don’t know’ would probably mean “I can’t be bothered thinking much about it and don’t have to right now”.
A week before the election I was often still evaluating possibilities and considering options.
And another point on don’t knows – quite possibly a higher proportion of them may not bother voting if they still ‘don’t know’.
Richard29
/ February 20, 2013Interesting point about those who rock up on the day to vote still unsure. I volunteered last election as a scrutineer/observer at my local election booth for my party of choice. This basically involved sitting in a chair for a few hours with a party rosette stuck to me and a big smile on my face for all the punters.The supposed role of observers is to prevent corruption, vote rigging and undue influence – this is not really an issue in NZ and I’m sure I wasn’t the only scrutineer for whom the experience basically just involved chatting to the scrutineers for the opposing parties (who were alll lovely people, none of them having devil horns that I could see) and lunchroom gossip with the Elections NZ facilitator people (one of whom happened to be an old work colleague, now stay at home mum).
Anyway – getting to my point… The messaging from the party and the local candidate and the messaging/instructions that the observers for the other parties were also given. Was that having a friendly looking party scrutineer beaming at them while they cast their vote can be just the push that undecided voters need to select Party X over Party Y.
As I said, all the parties seem to believe this, and at a conceptual level I can understand how that might work, but I have not seen a shred of evidence to substantiate it (and suspect it may just be a political old wives tale). For a start – I doubt very much that there are a large number of people who arrive with absolutely no preconception of who they will vote for, I just don’t know why they would bother. Secondly, I’m unconvinced that scrutineers make a difference – the supposed thought process just doesn’t seem that credible to me: “Hmmm wow, there are a lot of people here with Blue rosettes on and they all look very happy, I guess I should vote National”. Thirdly – I presume that any influencing effect would cancel itself out; at my polling place there were three observers – Red, Blue and Green, so my guess is that we effectively cancelled each other out and thus wasted all of our time.
Having said all that, the parties seem to all go to a lot of effort logistically to deliver their network of scrutineers on the day (the National Party person at our polling place got a lovely packed lunch delivered half way through the day, all wrapped up in a nice blue ribbon) so you would hope they are basing it on something other than a hunch, and it is the parties themselves that would have the data on which polling booths had observers and thus be able to determine whether this influenced the final count…?
Pete George
/ February 20, 2013I’d be surprised if scrutineers made much difference. As you say you would expect parties to cancel each other out, and surely the vast majority have made up their mind by then. And some of the few who haven’t won’t be swayed by a smile behind the desk.
But guess it’s playing for every possibility, as Waitakere showed it doesn’t take many votes to make a difference.
I’m also dubious about the benefit of election signs, between the canceling out effect and ignoring them it’s hard to see many swayed by a roadside sign.
And election meetings are mostly stacked with the already converted, I don’t see much gain from them either unless you can score hit from media coverage, but that’s usually well spread.
MSM is still the biggest influence, but it’s a very competitive environment and at the mercy of the whim of journos.