The infinite edition Polling Thread: Volume 5

On the one hand Scott says that depending on demographics some parties are more likely to have their supporters vote than others but then he uses a calculation that estimates this as being equal for all parties. In fact not only is this not true -- it is not even consistent for the parties from one election to the next. The commitment of supporters to actually get out and vote varies.But there is a much bigger more common misconception.

People assume that many of the people who are polled do not vote. I think the people who will say yes to a pollster interrupting their dinner to get their views are in fact very, very, likely to vote. The numbers are skewed not by the people who respond to the pollster and not vote but by the people who do not respond to the pollster but do vote anyway. Of the people you call and actually reach (not machines etc.) the refusal rate is probably somewhere around 90% meaning you get 9 refusals for every complete (and that is actually a very, very good rate).

So, rather than think about those who will be polled and not vote, think about those who vote and won't be polled- and why some people who vote also agree to be polled. Then you get some interesting answers. People who are likely to spend time with a pollster may be:

1) generous with their time to a stranger trying to do their job

2) feel strongly about their opinion and want to be counted

3) Interested in politics- perhaps even zealous

4) Perhaps naive- not burned by having had too many telemarketing calls

5) Open minded, curious-- open to pollsters and open to new alternative political views

6) Proud of their voting choice want to share it- vote for apparently unselfish reasons

This might reflect some parties more than others.

So willingness is a huge factor. So too is availability. This may trade off-- for example-- more Cons are available to be called but more of them refuse to be polled so the number comes out accurately. Perhaps NDP under-polls at times because while they are slightly more motivated they are significantly less available. Perhaps the Greens are disproportionately both motivated and available to poll.

For whatever reason- it is possible that the Green supporters will be over-counted in polls rather than be under-active in elections. The ones I have known have actually been more interested than average in politics, very likely to vote, activists. But they are also very likely to respond to a poll-- especially since they know that a point up or down can mean a lot to the impression of the party. So if most Greens who get a pollster call their home answer it then they will be over-represented in the polls.

I think many young people don't vote but it may even be that the Greens among them do vote and the others don't.

The reason it looks like this is that many people who hang up on pollsters actually do vote.  (Just not enough to bring our national suffrage rates above 60%).

So it may well be that it is not that the Greens can't get their people to the polls- it may be that they are over-represented in the sample of those willing to be polled and that in the general population there really are not that many of them and never were.

We have a false impression that many people have political opinions but do not vote. This is in fact unlikely. Most people who do not vote actually do not even think to support a party and certainly would not interrupt dinner to tell a pollster what they think.

Way too many people, somehow imagine that the greater commitment is to vote and the lesser is to tell a pollster-- but in fact it is the reverse. The real stats- the ones we don't talk about bear that out: Of those who get a call from a pollster less than 10% will provide an answer yet of those who have an opportunity to vote more than 50% actually do. The voting numbers are bad and that is making us look at polling stats backwards. The 10% who will talk to a pollster are very likely included in the 55-60% who will get out and vote.

In other words the party operatives have it wrong-- instead of focusing on declining suffrage rates of people with political opinions, a problem that may not exist, they might have to worry about the growing number of people who do not even care to have an opinion - the people if you led them by hand to the polling booth would have nothing to say. These people never paid attention, do not care, or think there is no difference. those are the people we need to reach if we want our vote to go up. Maybe that is what negative campaign people are understanding-- maybe they are not trying to move votes but to get the opposition to stay home. Look at it this way and it will turn your campaign strategies on their heads.

Hee-hee.

remind wrote:

madmax wrote:

think its time the polsters trimmed the fat of these ridiculously high numbers for the Green Party. It also has a deflating effect for those who do support the party and vote for them only to discover that that 15% polling number turned out to be 5% on Eday. That margin of error is MASSIVE. No polster should continuously put forth such erroneous data. There is enough empirical evidence to show that while they get the 4 parties (CPC, LPC, BQ, NDP) within the margins of error, the polsters are often out between 100% and 200% on green party numbers.

I agree they are putting out erroneous data, but why?

The data is the data. It isn't "erroneous". The polling companies are reporting what people tell them. There is a discrepancy between polling numbers and actual vote tallies but it isn't because of bad polling, it is because polling companies poll people who don't vote and mix their opinions in with people who will vote. This discrepancy affects the Green Party most of all because Green Party support is skewed towards youth, who as a group are the least likely to actually vote. If 24% of youth support the Green Party but only 1 out of 3 actually vote a 24% poll turns into an 8% vote.

Polling companies randomly call people up and as madmax said in the previous thread

Quote:

If an election were held today, who would you vote for? When prompted for an answer with choices, many randomly pick one. The truthful answer is they are not voting.

Only 59% of eligible voters voted in the last election. If you apply a 0.59 correction factor to compensate for this, and report non voters the last EKOS poll should be reported as:

Nobody 41%
Cons 22%
Liberanos 16%
NDP 10%
Greens 6%
Bloc 6%
__________________________________
One struggle, many fronts.

remind wrote:

OMG, it took me way too long to get that........

It's because you're too nice of a person, remind.

Ekos has a new poll out tonight.  They are reporting on Canadians' attitudes towards the environment, but you can infer the parties' standings from it as follows:

Cons - 36.4% | Lib - 26.7% | NDP - 16.9% | Bloc - 10.1% | Green - 9.9%

Nov 25 - Dec 1, N = 2412, MoE +/- 2.0

ottawaobserver wrote:

The Green Party is about to experience a retrenchment in their support, which their current understanding of campaigning as a slow building process won't allow them to understand very well

I just noticed this. Viz the highlighted part, could you explain what their [presumably] erroneous/misguided understanding is?

Layton making gains in Quebec: Poll

Layton is in second in most categories, Iggy is in 4th.

I wish the article would mention the actual party support numbers.

The poll the Globe is talking about is this one: http://www.ledevoir.com/politique/canada/278277/sondage-leger-marketing-...

I have the full break outs on my computer, but I can't remember where I got it from. At any rate there are the basic numbers...

I was just looking over the details of that poll and there are a couple of noteworthy factors.

1) 68% dissatisfied with the Conservative government to 26% satisfied. Cons may be at 20% but that's as much a mirage as the NDP numbers, they'll be lucky to hold what they have and I wouldn't read too much into the by-election win.

2) NDP numbers break down as 16% of Francos and 21% of Anglos. Liberals have 39% of Anglos. So don't get excited for the NDP to start poaching Montreal area Liberal ridings, what these numbers mean for us is that we'll start overtaking the Liberals in franco ridings. On the island in anglo ridings people have been voting Liberal all their life, we need much higher anglo numbers to be competitive in anglo ridings.

3)In this poll at least the regional numbers are 16% Montreal region, 19% Quebec City region (go figure) and 18% in the rest of Quebec. So notwithstanding the higher anglo numbers (they've always been higher) where we are actually picking up support is in francophone Quebec. This is held out by the Hochelaga result (it's on the island, but totally francophone, with a certain population of allophones)

I think this is really positive for Biovin in Gatineau and even Hull-Aylmer. Outside of those two, plus Outremont obviously and a couple of long-shot Montreal ridings with high francophone populations (Hochelaga, Jeanne-Le-Ber, etc.) we're not really in a position to win many other seats in Quebec.

I hope we can pull out at least two seats (fairly do-able I think) and as many as four this election. Where our real gains will come in this province (and this poll is showing the begining of how we get there) is when Duceppe leaves and the Bloc starts to fade. Then hopefully our progressive credentials can entice enough disenchanted Liberals and Bloquistes to start winning ridings we would never dream of now.

Good point OO, I assumed it was but looking at the break outs on the Le Devoir Poll it doesn't have the same questions. Well this is quite nice if that bump in the Le Devoir poll is held out by this seperate poll. In that case I agree about how annoying it is that there aren't numbers, unless the numbers in the article are all there is and they didn't ask voting intention. Which would be odd.

I'll go google that company see if I can find the original, I'll post if i find it...

thorin_bane wrote:

Funny not much polling being reported. I don't think the 2 headed beast likes the numbers they are seeing.

Because the media didn't pay for the poll. They tend to not give much play to polls they didn't pay for themselves.

The trust & competence numbers where 4 leaders are mentioned add up to 82.5% & 78.3%. The best PM numbers with 3 mentioned add up to 80.3%. I'm assuming the other 20% are Don't Know/Won't Say/Other & that Duceppe wasn't given as an option for Best PM.

Stockholm wrote:

The most recent poll in Quebec has the NDP at 17% and the Liberals and Conservatives at 20% each - that is a de facto three way tie. When you consider that the Liberals currently have 14 seats in Quebec and the Tories 11 - for the NDP to be within striking range is quite unprecendented. Its also clear that the Liberals only exist at all in Quebec because of support from non-francophones. Among francophone Quebecers they are now the fourth party and fading fast.

The thing is that the horse race numbers between the NDP, Conservatives & Liberals are meaningless to some extent. How many head-to-head Con-Lib, Lib-NDP & NDP-Con contests are there in Quebec? How many 3 or 4-way battles are there? The thing to look for is: what is the gap between each party & the BQ, & how does that compare to the last election.

Stockholm wrote:

The most recent poll in Quebec has the NDP at 17% and the Liberals and Conservatives at 20% each - that is a de facto three way tie. When you consider that the Liberals currently have 14 seats in Quebec and the Tories 11 - for the NDP to be within striking range is quite unprecendented. Its also clear that the Liberals only exist at all in Quebec because of support from non-francophones. Among francophone Quebecers they are now the fourth party and fading fast.

The Liberals are actually tied with the NDP in Francophone support in the poll above, and just a few points behind the Conservatives.

Although obviously that is not great.

The latest Ekos poll shows more Tory erosion with the NDP benefitting from Tory losses:

http://ekos.ca/admin/articles/cbc-2009-12-10.pdf

.

http://www.walrusmagazine.com/articles/2010.01-frontier-the-peoples-poll/

Interesting take on polls, which confirms my opinion that they are meaningless and are now used mostly as filler by news organizations too lazy or underfunded to go out and gather real news.

There's "youthful" energy in the green party? That's news to me. i was under the impression it was just May and her small band of 50-something acolytes and sycophants.

The Green party is certainly attractive to young people. But from what I have seen, there are is if anything considerably less weight to their actual participation, compared to younger people active in the NDP.

If any of the political parties put up a youthful face I suspect that they would improve their polling numbers-- if any actually combined that with policies on issues that matter to young people they might even get them to show up on election day. Unfortunately, youth-friendly politics without the youthful delivery may not register. Sad but the messenger has become more important than the message. Perhaps that is in part because people are lied to on a daily basis so they tune out and look for someone they think might tell them the truth-- For the NDP that means at least having the leader less scripted so he sounds like his responses are not canned as much as they do now. The party already has most of the right policies for younger people unlike most of the other parties. However, more work must be done on the unfortunately more important delivery aspect. The NDP could also have some of its more youthful members put in front of the national mic to discuss issues that matter to youth (Hello Niki?). A campaign specifically to engage youth would not be out of place.

An extremely important thing all parties have not learned-- it is no longer about not saying the right thing that matters most it is avoiding saying the wrong thing. By this I do not mean big controversial mistakes-- that's obvious. Rather I am referring to the boring change-the-channel stuff that leads politicians to be speaking to an empty room long before they say stuff people might have been interested in hearing. People today are bombarded with so much garbage they don't want, and so much they do want, that we all have become impatient editors-- too many "woke up, got out of bed had a coffee" posts and poeple stop following you on Twitter and don't want your posts on facebook. Too much irrelevant, petty, partisan crap and they stop listening. People used to talk about leaders being over exposed (and by this I do not mean Bob Rae with Rick Mercer) or underexposed. It is no longer a question of volume but of quality. The only way you get people to listen is to say something important when you pull out the megaphone. This means cutting out low quality communications at the same time as increasing the higher quality ones.

Sean in Ottawa wrote:

Unfortunately, youth-friendly politics without the youthful delivery may not register. Sad but the messenger has become more important than the message. Perhaps that is in part because people are lied to on a daily basis so they tune out and look for someone they think might tell them the truth-- For the NDP that means at least having the leader less scripted so he sounds like his responses are not canned as much as they do now. The party already has most of the right policies for younger people unlike most of the other parties. However, more work must be done on the unfortunately more important delivery aspect.

I happened to be in the area, so I went to see Layton talk to students at the University of Ottawa the first week of September, as part of his annual campus tour.  You would have been really impressed Sean.  Not a canned answer in the two hours, and in fact he spent very little time with the opening statement, spending most of the two hours taking questions and (more importantly) listening to what the young folks had to say.  He was completely spontaneous, very thoughtful, pitched his comments just right (not too obtuse, but not dumbed down), and demonstrated his command of a wide-ranging array of topics while making the links between then.  Not a scripted message anywhere to be found.  There was a full room and with the exception of some young Liberals who tried to trip him up (he handled them beautifully), everyone seemed to leave really impressed.  He went off to the pub with them afterwards (I went home).

Man if they could clone him and cover every campus every fall, we'd have a majority government within the decade !