Author Topic: Federal Tories and Grits Locked in a Statistical tie – Turbulence for Tories in  (Read 1207 times)

0 Members and 1 Guest are viewing this topic.

Offline GAP

  • Semper Fi
  • Army.ca Legend
  • *****
  • 211,930
  • Rate Post
  • Posts: 12,944
Taking a look at the national numbers, the movement in voter preference is within the margin of accuracy for the poll. Check out Quebec – major turbulence for the Conservatives (down fourteen points). When the Conservatives focus on their five priorities their numbers move up but that the focus on Afghanistan, pulling out of Kyoto and warm relations with George Bush has noticeably eroded Conservative support in Quebec.

The main beneficiary of the Conservative drop has been the BQ (up eight points). This illustrates the appeal that the Harper-led Conservatives had among soft nationalists in Quebec.

It’s a little early for a victory lap for the Liberals. Quebec is the most volatile in support right now. The Conservatives need to hold and expand support in Quebec to stay in government, the Liberals need to win support to block the Conservatives.

The Liberal leadership will be critical to the fortunes of both the Conservatives and the Liberals. A divided Liberal party (or anything that has a whiff of inappropriate behaviour by a leadership campaign) will be good news for the Harper Conservatives. If the Liberals come out united and have a leader who can have some sort of appeal to Quebecers – they will be back in the game.

NEW! SES Research Blog!
I have launched my own blog “Nik on the Numbers” a new interactive blog where you can share your views, rate the opinions of others, and ask me questions about this poll or any other issue. Check it out today at www.nikonthenumbers.com.
 
  Methodology
Polling between November 5th and 9th, 2006 (Random Telephone Survey of Canadians, 18 years of age and older). The statistics of committed voters for the current wave is accurate to within 3.3 percentage points, plus or minus, 19 times out of 20. The margin of accuracy will be wider for sub samples. Percentages may not add up to 100 due to rounding.
 
 
  Canada (N=884, MoE ± 3.3%, 19 times out of 20)
For those parties you would consider voting for federally, could you please rank your top two current local preferences? (Committed Voters Only) The change from the previous wave is in parenthesis.

National
Conservative Party – 34% (-2)
Liberal – 32% (+2)
NDP – 16% (-2)
BQ – 13%(+2)
Green Party – 5% (No change)

Quebec (N=224, MoE ±6.6, 19 times out of 20)
BQ – 50% (+8)
Liberal – 25% (+3)
Conservative Party – 12% (-14)
NDP – 10% (+2)
Green Party – 4% (+1)

The detailed tables with the regional sub-tabs and methodology are posted on our website at: http://www.sesresearch.com

Feel free to forward this e-mail. Any use of the poll should identify the source as the “SES Research National Survey.”
 
Two things are infinite: the universe and human stupidity; and I´m not so sure about the universe

Offline Thucydides

  • Army.ca Legend
  • *****
  • 194,765
  • Rate Post
  • Posts: 13,702
  • Freespeecher
An observation about another poll:

http://kitchenerconservative.blogspot.com/2006/11/decima-manipulating-polls.html

Quote
Decima Manipulating Polls?

With the recent release of a Decima poll showing the Liberals ahead by 2 points, I took a look at the numbers and something strange stands out with the way the last two polls where conducted.

Angus-Reid has a poll tracker that shows polls from different companies. So I went through all the polls from Decima since the last election and something seemed strange to me. I almost get the impression that Decima manipulated the numbers the get a desired result. Look for yourself and be the judge.

Links: (1st, 2nd)

1st:

Conservatives: 31%, Liberals: 28%, NDP: 18%, BQ: 9%, Green: 9%

Source: Decima Research / Canadian PressMethodology: Telephone interviews with 1,026 Canadian adults, conducted from Nov. 2 to Nov. 5, 2006. Margin of error is 3.1 per cent.

2nd:

Conservatives: 31%, Liberals: 33%, NDP: 15%, BQ: 10%, Green: 10%

Source: Decima Research / Canadian PressMethodology: Telephone interviews with 1,126 Canadian adults, conducted from Nov. 9 to Nov. 13, 2006. Margin of error is 3.1 per cent.

First, notice the discrepancy, the 1st poll adds up to 95% and the 2nd adds up to 99% assuming the the decimal points lopped off. there is a 4% difference between votes cast and shown. Why?

Second, the jump in points for the Liberals doesn't have a corresponding reduction for others. NDP went down 3%, but the Bloc and Greens went up 1% each, and the Conservatives stayed the same. There is a discrepancy of a missing 4% from the last poll. Why?

Third, why suddenly poll 100 more people that before, other pollster use similar numbers of respondents. Why?

It seem like something funny is going on here. Just food for thought.

Of course there are other ways to manipulate polls: "push polls" with leading questions; selecting the polling sample through time of day or area code or other demographic "loadings", or even (as was related by a poster on Army.ca) disconnecting people if they are giving the "wrong" answers.

I would take a poll like that with a very large grain of salt.
Dagny, this is not a battle over material goods. It's a moral crisis, the greatest the world has ever faced and the last. Our age is the climax of centuries of evil. We must put an end to it, once and for all, or perish - we, the men of the mind. It was our own guilt. We produced the wealth of the world - but we let our enemies write its moral code.