We have another piece of evidence, by Léger Marketing, that the Conservative lead is around 5%. My raw polling average has it at 5.5%, but after adjusting for a likely Tory ballot box bump, it is 8.0%.
This poll has the Tories leading the Liberals by 11% in Ontario, which is consistent with the polling average. That's good news since Léger is usually pretty middle-of-the-road. The NDP is 8% behind the Conservatives there.
In Québec, this survey has the Bloc a little higher than others. Still, that's just 27%, 13% behind the NDP. In the Atlantic, it's a tight three-way race, though the Liberals trail the others slightly. Out West, there's nothing surprising.
We keep hearing about an NDP surge in BC. Léger still has them 10% behind the Conservatives, and the pre-adjustment polling average, 9%. While that's less than the 18% gap recorded in 2008, it's unlikely to cost the Tories more than 2-3 seats.
In the projection, the Bloc gets an NDP seat in Québec, the NDP gets a Tory seat in Ontario, and the Tories get a Grit seat in PEI, which gives:
CON - 153
NDP - 92
LIB - 49
BQ - 14
Update: Interestingly, according to this poll, still just 8% of Canadians think the next government will be led by Layton, while 66% think Harper will remain PM. This doesn't mean Canadians prefer Harper - he trails Layton by 4%, 30 to 34, on the best PM question.
15 comments:
Ekos says that his methodology (not using humans when asking questions) tries to compensate for people not telling the truth due to being embarrassed (The Reform party, for example, used to do better than expected).
Harper has done a lot of fearmongering this election. Do you think this explains some of the discrepancy between the Conservative poll numbers from different pollsters?
I don't think so. EKOS' methodology biases upward the numbers for the Greens and Others. If you reduce those and distribute the difference proportionally between the other parties, you get pretty close to the polling consensus.
True about the Greens. I guess you either have a cold machine that simply lists your choices or a warm human that only asks which way you'll vote. Neither method is perfect and both can influence your answer.
(At least that's how I'm assuming Ekos does it. I was actually polled by Nanos this election. A chick was on the other end of the line. I kept asking myself, "Do I really want to tell her this or that?" and felt I had to explain my choices.)
One thing I don't understand is the smaple size of Ontario(1,003) and Quebec(1,012) are almost the same but Ontario has 106 seats and Quebec has 75. How can this be arepresentative sample?
I was expecting polls from Abacus and EKOS this morning. Any news?
Anon: Léger probably oversamples Québec because that's the most important place for its reputation. I'm sure that they weigh their data accordingly so that the national figures aren't biased.
Skoblin: EKOS usually posts around noon - yesterday was an exception. Though I saw a tweet from Frank Graves about taking a break, so maybe they're holding off today.
I don't know about Abacus, but if I remember correctly, their polls have often come in the afternoon.
@Anonyme 10:52. Aux États-Unis aussi un échantillonnage représentatif dans un sondage est d'environ 1004 personnes. Il ne sert à rien d'augmenter le nombre de sondés si le sondage est bien fait. La population américaine est 10 fois supérieure à celle du Canada mais on n'a pas besoin de multiplier par 10 l'échantillon. Qulequ'un veut le lui traduire ?
LeadPipe: Je ne pense pas que Anonyme 10:52 disait que l'échantillonnage était trop petit en Ontario. Je pense qu'il se préoccupait plutôt que le poids du Québec dans les chiffres nationaux soit trop élevé en raison de l'échantillonnage accru au Québec.
Skoblin: No EKOS today. Are you sure Abacus is doing its update today?
I wonder if Ipsos, Angus Reid and Harris-Decima have said their final word, or if they're doing another survey now. It looks like Léger is done.
I'm guessing Forum, Innovative, Environics will release something new this weekend. So counting EKOS, Nanos and Abacus, we'll probably have at least 6 more national polls. I wonder if COMPAS is doing another survey, and whether the results will be just as crazy as their last one...
Is it just me or am I beginning to detect a late shift towards the Tories?
The only real shift I have noticed recently is today's Nanos - and I think they may be off the mark.
Andre: Too early to say. We only have 2 polls with data from yesterday, Nanos and Angus Reid. That 2% Tory increase could be just noise.
Besides, Nanos has the Tories dropping slightly in Ontario, so the trends are really unclear.
EW,
I believe I saw a twitter feed yesterday saying an Abacus poll would be released today - but no other confirmation.
My bad....Abacus will be released Sunday morning.
"Qulequ'un veut le lui traduire ?"
Here's a rough translation...
Leadpipe said:
@Anonyme 10:52. Even in the US a representative sample in a poll is about 1004 persons. There's no need to increase the number of those polled if the poll is properly done. The American population is ten times greater than that of Canada but there's no need to multiply the sample by ten.
Election Watcher said:
LeadPipe: I don't think that Anonyme 10:52 was saying that the sample was too small in Ontario. I think that he (she?) was more concerned that the weight of Quebec in the national numbers is too high because of the (sample size?) in Quebec.
Anyway that's the gist of it.
Post a Comment