Liberals well positioned for next election

gerryhatrick · December 4, 2006

Dion gets 55% approval from Canadians: poll
Updated Sun. Dec. 3 2006 11:03 PM ET

CTV.ca News Staff

The Liberals' choice of Stephane Dion as their next leader seems to have the initial stamp of approval from the Canadian public and has given the party a bump in popularity, a new poll finds.

snip

When asked which party they would vote for if an election were held today, the Liberals came out on top: (percentage-point change from an Oct. 12-15 poll in brackets):

* Liberal: 37 per cent (+5)

* Conservative: 31 per cent (-1)

* NDP: 14 per cent (-3)

* Bloc Quebecois: 11 per cent (unchanged)

* Green: 7 per cent (-2)

http://www.ctv.ca/servlet/ArticleNews/stor...?hub=TopStories

And he polls extremely well in Quebec. 62% said he was a good choice.

I think the CPC sabatoge might backfire. They hoped it would be Iggy, but Dion is probably a bigger nightmare for them than Rae would have been.

jdobbin · December 4, 2006

Some people here are already questioning the veracity of the poll although an Internet poll was linked as being accurate yesterday.

gerryhatrick · December 4, 2006

Some people here are already questioning the veracity of the poll although an Internet poll was linked as being accurate yesterday.

Internet polls are obviously crap.

Why are "some people here" questioning the veracity of this Strategic Council poll though?

jdobbin · December 4, 2006

Internet polls are obviously crap.
Why are "some people here" questioning the veracity of this Strategic Council poll though?

One person said it was all people from Quebec and Ontario.

punked · December 4, 2006

Cause it was 1000 people 1 third of them from quebec which also is given the highest error rate in the poll. Basically take this poll +/-10% from any party that is how acturate this is.

That was me it was more like 2/3 from quebec and ontario, they just assumed the rest of the country is static.

gerryhatrick · December 4, 2006

Cause it was 1000 people 1 third of them from quebec which also is given the highest error rate in the poll. Basically take this poll +/-10% from any party that is how acturate this is.
That was me it was more like 2/3 from quebec and ontario, they just assumed the rest of the country is static.

Not according to the story:

Technical notes:
* The poll sampled 1,000 Canadians on Dec. 3, 2006. The results are based on tracking among a proportionate national sample of Canadians aged 18 or older.

* The margin of error for the national poll was plus or minus 3.1 percentage points, 19 times out of 20.

* The margin of error for the Quebec subset, which had a sample size of 247, was plus or minus 6.3 percentage points, 19 times out of 20.

* The margin of error for the Ontario subset, which had a sample size of 379, was plus or minus 5.0 percentage points, 19 times out of 20.

By my calculation, that's 1/4 from Quebec (a little less, actually) not 1/3.

And a margin of error is existent in every poll. It doesn't invalidate the poll. Your "+-10%" comment is nonsense.

gc1765 · December 4, 2006

Cause it was 1000 people 1 third of them from quebec which also is given the highest error rate in the poll. Basically take this poll +/-10% from any party that is how acturate this is.
That was me it was more like 2/3 from quebec and ontario, they just assumed the rest of the country is static.

Where does it say that?

Nevermind. Thanks Gerry.

gerryhatrick · December 4, 2006

One person said it was all people from Quebec and Ontario.

Well that person is wrong and you're being willfully lazy about the poll by repeating it without looking for yourself.

punked · December 4, 2006

Cause it was 1000 people 1 third of them from quebec which also is given the highest error rate in the poll. Basically take this poll +/-10% from any party that is how acturate this is.

That was me it was more like 2/3 from quebec and ontario, they just assumed the rest of the country is static.

Not according to the story:

Technical notes:
* The poll sampled 1,000 Canadians on Dec. 3, 2006. The results are based on tracking among a proportionate national sample of Canadians aged 18 or older.

* The margin of error for the national poll was plus or minus 3.1 percentage points, 19 times out of 20.

* The margin of error for the Quebec subset, which had a sample size of 247, was plus or minus 6.3 percentage points, 19 times out of 20.

* The margin of error for the Ontario subset, which had a sample size of 379, was plus or minus 5.0 percentage points, 19 times out of 20.

By my calculation, that's 1/4 from Quebec (a little less, actually) not 1/3.

And a margin of error is existent in every poll. It doesn't invalidate the poll. Your "+-10%" comment is nonsense.

2/3 of the poll was from ontario and quebec with margin of error of 6.3 and 5 in Ontario pretty much the only provinces focused on. It is a flawed poll becuase it is assumed the rest of the country which recived 1/3 polling for 8 other provinces are static they aren't. In the only provinces they polled it is +/-5.8 margin of error that is the difference the liberals are winning by. It is just stuipd to take this seriously.

geoffrey · December 4, 2006

Absolutely. 2/3 of the poll did come from Ontario/Quebec. They only have a smidge over 50% of the population. The distribution of the respondants is ugly. Ontario is grossly over-represented.

gerryhatrick · December 4, 2006

Cause it was 1000 people 1 third of them from quebec which also is given the highest error rate in the poll. Basically take this poll +/-10% from any party that is how acturate this is.

That was me it was more like 2/3 from quebec and ontario, they just assumed the rest of the country is static.

Not according to the story:

Technical notes:
* The poll sampled 1,000 Canadians on Dec. 3, 2006. The results are based on tracking among a proportionate national sample of Canadians aged 18 or older.

* The margin of error for the national poll was plus or minus 3.1 percentage points, 19 times out of 20.

* The margin of error for the Quebec subset, which had a sample size of 247, was plus or minus 6.3 percentage points, 19 times out of 20.

* The margin of error for the Ontario subset, which had a sample size of 379, was plus or minus 5.0 percentage points, 19 times out of 20.

By my calculation, that's 1/4 from Quebec (a little less, actually) not 1/3.

And a margin of error is existent in every poll. It doesn't invalidate the poll. Your "+-10%" comment is nonsense.

2/3 of the poll was from ontario and quebec with margin of error of 6.3 and 5 in Ontario pretty much the only provinces focused on. It is a flawed poll becuase it is assumed the rest of the country which recived 1/3 polling for 8 other provinces are static they aren't. In the only provinces they polled it is +/-5.8 margin of error that is the difference the liberals are winning by. It is just stuipd to take this seriously.

You're changing your story now. But nevermind.

Just look at the national poll result.

37 - 31 % Liberals over CPC, MOE only 3.1%

Deal wit

jdobbin · December 4, 2006

Well that person is wrong and you're being willfully lazy about the poll by repeating it without looking for yourself.

I never said I believed it was all Ontario and Quebec people myself. I have no reason not to believe the polling company.

punked · December 4, 2006

Cause it was 1000 people 1 third of them from quebec which also is given the highest error rate in the poll. Basically take this poll +/-10% from any party that is how acturate this is.

That was me it was more like 2/3 from quebec and ontario, they just assumed the rest of the country is static.

Not according to the story:

Technical notes:
* The poll sampled 1,000 Canadians on Dec. 3, 2006. The results are based on tracking among a proportionate national sample of Canadians aged 18 or older.

* The margin of error for the national poll was plus or minus 3.1 percentage points, 19 times out of 20.

* The margin of error for the Quebec subset, which had a sample size of 247, was plus or minus 6.3 percentage points, 19 times out of 20.

* The margin of error for the Ontario subset, which had a sample size of 379, was plus or minus 5.0 percentage points, 19 times out of 20.

By my calculation, that's 1/4 from Quebec (a little less, actually) not 1/3.

And a margin of error is existent in every poll. It doesn't invalidate the poll. Your "+-10%" comment is nonsense.

2/3 of the poll was from ontario and quebec with margin of error of 6.3 and 5 in Ontario pretty much the only provinces focused on. It is a flawed poll becuase it is assumed the rest of the country which recived 1/3 polling for 8 other provinces are static they aren't. In the only provinces they polled it is +/-5.8 margin of error that is the difference the liberals are winning by. It is just stuipd to take this seriously.

You're changing your story now. But nevermind.

Just look at the national poll result.

37 - 31 % Liberals over CPC, MOE only 3.1%

Deal wit

I personally am a card carrying NDPer. I am also a major in Biolgy and math with several stats courses under my belt this poll is flawed. 3.1% isn;t the number that matters the number that matters here is those in Quebec and Ontario those are the only provinces to assume to have a change in their numbers, the Margin of error of 3.1% is over repersented because it is assumed there would be no change in the rest of Canada. Your looking at a real margin of error of 5-6% which is what the libs lead the CPC by. I am telling it is foolish to this seriously.

geoffrey · December 4, 2006

Even if the poll is crediable, it's still a statistical tie.

jdobbin · December 4, 2006

Even if the poll is crediable, it's still a statistical tie.

I think this close to the campaign it tells nothing accurately. It is like the poll that came out during the income trust changes. At first it said the Tories were doing fine and the next three polls showed slippage linked to the decision.

It will take a few weeks and several polls to tell what is happening.

gerryhatrick · December 4, 2006

I personally am a card carrying NDPer. I am also a major in Biolgy and math with several stats courses under my belt this poll is flawed.

I've taken stats courses too. So what. I'm sure Strategic Council has people twice as smart at it than you and me put together. I've pointed out your errors once and don't wish to argue the poll with you any longer.

punked · December 4, 2006

I personally am a card carrying NDPer. I am also a major in Biolgy and math with several stats courses under my belt this poll is flawed.

I've taken stats courses too. So what. I'm sure Strategic Council has people twice as smart at it than you and me put together. I've pointed out your errors once and don't wish to argue the poll with you any longer.

Well first of all it is the Strategic Cousel, next let me point this out they have a history of having margin of above 5% for all provinces but yet only 2-3 or three for the whole country so there is some fancy math going on there.

gerryhatrick · December 4, 2006

Well first of all it is the Strategic Cousel

I don't care how it's spelled, thx. That's not really anywhere near as important as getting the simple math of the poll wrong, as you did.

gc1765 · December 4, 2006

Well first of all it is the Strategic Cousel

I don't care how it's spelled, thx. That's not really anywhere near as important as getting the simple math of the poll wrong, as you did.

Actually, if spelling is the issue, you should tell punked that it's Strategic Counsel not Strategic Cousel.

But, I agree, spelling is not important.

Canadian Blue · December 4, 2006

I call BS, how can you give a guy a 55% approval rating when he's only been on the job for what 12 hours. Wait a few weeks to see what else happen's. One poll showed Harper would get 38% compared to Dion's 25%.

gerryhatrick · December 4, 2006

I call BS, how can you give a guy a 55% approval rating when he's only been on the job for what 12 hours. Wait a few weeks to see what else happen's. One poll showed Harper would get 38% compared to Dion's 25%.

You underestimate the power of first impressions. Dion has obviously made a good one, and they always linger.

And I believe the poll you refer to is one taken before the leader was picked, no? Obviously that would be expected.

geoffrey · December 4, 2006

I generally have a hard time getting a good first impression from someone that struggles with my language.

gc1765 · December 4, 2006

I generally have a hard time getting a good first impression from someone that struggles with my language.

So he has an accent. He speaks just fine. My french sucks, but I understand him perfectly well in English.

stignasty · December 4, 2006

Even if the poll is crediable, it's still a statistical tie.

Yeah, percentage polls never tell the story in a system like ours anyway. In '93 Reform had 18.5%, The PCs 16% and the BQ had 13.5. Seats were Reform 52, PC 2 and BQ 54. At best they're a rough gauge of voter trend.

mikedavid00 · December 4, 2006

Not so fast.

Canada is proven to have non reliable 'overnight' polling due to the makeup of our country:

-The poll sampled 1,000 Canadians on Dec. 3, 2006. The results are based on tracking among a proportionate national sample of Canadians aged 18 or older.

-The margin of error for the national poll was plus or minus 3.1 percentage points, 19 times out of 20.

-The margin of error for the Quebec subset, which had a sample size of 247, was plus or minus 6.3 percentage points, 19 times out of 20.

-The margin of error for the Ontario subset, which had a sample size of 379, was plus or minus 5.0 percentage points, 19 times out of 20.