Tuesday, March 22, 2011

Narrowing national gap, widening Ontario gap in new Nanos poll

The latest CTV/Globe/Nanos poll shows very little change in the topline numbers, and even a downward trend for the Conservatives in the West. But Tory growth in Ontario puts Stephen Harper on the doorstep of a majority government.

Compared to Nanos's last poll conducted in mid-February, the Conservatives have slipped 1.1 points to 38.6%, while the Liberals are up a point to 27.6%. That's a gap of 11 points rather than February's 13, but this variation is still within the margin of error of the two polls.

The New Democrats are up one point to 19.9%, their highest result from any pollster for some time, while the Bloc Québécois is up 0.2 points nationally to 10.1%.

The Greens are down 1.1 points to 3.8%, a very low result for them, even by Nanos's standards. Note that, unlike other pollsters, Nanos does not prompt any of the parties in their surveys, and that by now the poll is one week old. This telephone poll has a sampling margin of error of 2.8 points, and 22% of respondents were undecided (up from 18.8%).

Ontario is the big story of this poll, though the shifts are within the margin of error. Nevertheless, the Conservatives are up 4.2 points to 43.2%, making this the sixth straight poll putting the Conservatives at more than 40% in the province. The Liberals are down 1.9 points to 30.9%, while the NDP is down 0.4 points to 23%. Aside from the recent Léger poll, Nanos is the only pollster to have recently put the New Democrats at 20% or more in Ontario. The Greens are down 1.8 points to 3%.

This poll is part of a recognizable in the province. Along with recent Conservative strength, Nanos helps to confirm Liberal weakness. This is the eighth consecutive poll pegging the Liberals at less than 35% in Ontario, and if this trend would hold throughout an election it would likely mean a majority for the Conservatives.

The Bloc Québécois leads with 39.3% in Quebec, up two points from mid-February. The Liberals follow with 24.7% (+0.3), while the Conservatives are down 0.6 points to 19.6% and the NDP is down 2.9 points to 13.9%, still a good number for them. Give or take a few points, this is generally where the parties were in Quebec at the end of the 2008 election, but the slight increases in Liberal and New Democratic support mean seat gains, as noted below.

The Conservatives have dropped 7.1 points in British Columbia, and now lead with 37.5%. The New Democrats are up 8.4 points to 29.7%, while the Liberals are down 2.2 points to 24%. The Greens are up 0.9 points to 8.8%. Despite these wild variations, none of these are statistically significant.

Both the Conservatives and Liberals have picked up 4.9 points in Atlantic Canada, putting them at 42% and 36.9%, respectively. The NDP is down 2.7 points to 17.7%.

Finally, Nanos still clumps Alberta, Saskatchewan, and Manitoba together. The Conservatives have dropped 9.1 points in the three provinces, and now stand at 55.5%. The Liberals are up seven points to 25%, and the NDP is up 4.2 points to 16.9%.

With the results of this poll only, ThreeHundredEight's model projects 19 seats for the Conservatives in British Columbia, 57 in Ontario, 11 in Quebec, and 15 in Atlantic Canada. I've awarded them the 50 seats in Alberta, Saskatchewan, and Manitoba that my official projection currently gives them. In all, that is 153 seats, two shy of a majority and six more than I projected for Nanos's last poll.

The Liberals win six seats in British Columbia, two in the Prairies, 31 in Ontario, 17 in Quebec, and 13 in Atlantic Canada for a total of 70. That is a drop of 14 seats from last time.

The Bloc wins 45 seats in Quebec, dropping two seats from their current standing in the House of Commons due primarily to local factors. It is also a four seat drop from the last Nanos projection.

The New Democrats win 11 seats in British Columbia, four in the Prairies, 18 in Ontario, two in Quebec, and four in Atlantic Canada for a total of 40, an increase of 12 from last time.

Nanos also looked at the leadership numbers for each of the party leaders, grading them on trust, competence, and their vision for Canada. Combining them determines Nanos's Leadership Index Score. Since mid-February, Stephen Harper has dropped 16.1 points on the LIS, and now stands at 82.8. He lost much of that, not on trust as you might suspect, but on competence.

Jack Layton is second with an LIS of 51.4, an increase of almost eight points, while Michael Ignatieff is up almost three points to 39.7.

As for the top issue for Canadians, it is still healthcare at 29%, an increase of six points since February. Jobs and the economy is second at 18% (-2), while education is third with 9% (+4). Of course, both education and healthcare are provincial concerns.

When it comes to what motivates Canadians to cast their ballot in favour of one party or another, 48% said they make their decision based on the parties' policies. Another 20% said that the leader is the most important factor, while only 12% based their decision on the local candidate. Another 10% said their voting behaviour is based on who they traditionally vote for. None of these numbers had significant regional variations, but they definitely show the importance of the national campaigns - it determines how 68% of Canadians vote (combining platform and leaders).

Two other questions Nanos asked are of interest. On the F-35 fighter jets, only 27% agreed that they should be purchased now, while 68% said that "now was not a good time". That opinion is even shared by 56% of Conservative voters.

But the question is less than perfect. Nanos mentions that the F-35 jets could cost as much as $30 billion, the number quoted by Kevin Page of the PBO. That's fine, but Nanos did not spell out that the $30 billion is spread out over several decades. Listening to the Nanos question, respondents might have thought that the $30 billion will need to be spent right now.

Finally, Nanos asked who Canadians trust on economic policy. The Conservatives lead with 30%, but the Liberals are not far behind at 21%. The NDP is at 16%.

This poll has a litle bit of something for everyone. For the Bloc, they look to hold on to their vote. For the NDP, they have the potential to reach 20%. For the Liberals, their Quebec number is good and Harper's lustre appears to be dulling a little bit. For the Conservatives, this is as good as a majority.

But I suspect that the Liberals have already made up their mind to vote down the government, while the Bloc is just waiting to see whether the $2.2 billion for the HST is in the budget. It probably won't be, which means the NDP is on the hook. The Conservatives have held out a small olive branch, but the NDP hasn't yet given any impression as to which way they are leaning. The budget, slated for 16h00 in Ottawa, will be one to watch.

There is a Harris-Decima poll lurking around out there, but it hasn't been posted on the Harris-Decima site just yet. It will probably be posted sometime today, so look for a report on the poll here tomorrow.

29 comments:

  1. "Nanos asked who Canadians trust on economic policy. The Conservatives lead with 30%, but the Liberals are not far behind at 21%. The NDP is at 16%."

    I should point out that this question about which party people trust on the economy is traditionally one where the NDP fares very badly and they have historically been in single digits on this question. To be at 16% on economic management (people with no opinion are not divvied up on this question the way they are on the vote question) is actually very high for the NDP and suggests that they are inoculating themselves on what had been a major weakness for them.

    BTW: I still have to think that if the NDP vote in Ontario actually were to rise from 18% to 23% and the Liberal vote were to fall from 34% to 30% - we would be looking at more than just one NDP pick-up from the Liberals.

    ReplyDelete
  2. There is an interesting trend emerging across the polling reports. Using population sample sizes less than 2000, almost always puts conservative support into the upper 30s low 40s. In contrast, virtually every recent poll using a larger sample size (>2000) has them hovering around the low 30s. (e.g. See todays Harris-Decima showing conservatives at 34%, with a >2000 sample size). Accuracy and confidence interval validity almost always increases with greater sample sizes. I would be interested to see a correlational analysis of fluctuating conservative support and polling sample size.

    ReplyDelete
  3. I have them close in Beaches - East York and Oshawa, and to a lesser extent Davenport, with this poll. But no cigar.

    ReplyDelete
  4. But the question is less than perfect. Nanos mentions that the F-35 jets could cost as much as $30 billion, the number quoted by Kevin Page of the PBO. That's fine, but Nanos did not spell out that the $30 billion is spread out over several decades. Listening to the Nanos question, respondents might have thought that the $30 billion will need to be spent right now.

    Thanks for stating this.

    This is a blatant example of asking a question that you want answered a certain way. If there was some sort of professional organization that pollsters were required to register or comply with Mr. Nanos would have some explaining to do.

    The other question that would not stand up to any standards is his National concern poll:

    It offers as choices: Health Care, Jobs/Economy, Debt/Deficit, Education and Environment.

    It does not mention Democratic Deficit--- The issue that is most likely to trigger an election.

    I can understand that Democratic Deficit is not a traditional issue that concerns anyone but it has become the main grievance of the Liberals and to a lesser extent the NDP and Bloc. It has been the main political concern since the 2008 election. To leave it off the poll asking about voters concerns???

    ReplyDelete
  5. "The other question that would not stand up to any standards is his National concern poll:

    It offers as choices: Health Care, Jobs/Economy, Debt/Deficit, Education and Environment."

    Actually, you are WRONG. The Nanos polls poses an OPEN-ENDED question on national concerns. No options are read at all!

    ReplyDelete
  6. All recent polls are indicating that Conservative support in Canada is in the 43% range and is always hight teens in Quebec. Liberal support in Canada is in the 28% range. Quebec, while they obviously continue to be represented in the HOC have and will continue to support a Quebec first party. Their perspective is essentially irrelevant in Canadian politics. The Conservative lead over the Liberals therefore in Canada is more like 15%. Outside of Toronto, that lead is more like 20%. This chasm between the Conservatives and Liberals in Canada is growing and it seems inconceivable that the Liberals can benefit from forcing an election at this time unless their objective is to lose badly and reboot.

    GP

    ReplyDelete
  7. My Canada INCLUDES Quebec. i find it quite offensive that you are using the terms "Canada" and "Quebec" as if they were mutually exclusive!

    ReplyDelete
  8. "On the doorstep of a majority government."

    How many times have I heard that line before in the last half-decade? Fifty? A hundred? One thousand?

    Nothing has changed. Nothing! All of the parties would be nuts to push/drop/kick a writ out right now. All of them.

    ReplyDelete
  9. GP said "All recent polls are indicating that Conservative support in Canada is in the 43% range and is always hight teens in Quebec."

    "All" recent polls? Actually only one, from Ipsos-Reid, using a 2 day poll sample of about 1000 people, prior to the recent spate of scandals. Harris Decima now pegs them at 34%, as does EKOs, who both use vastly larger sample sizes.

    In any case, Kim Campbell entered the 1993 campaign averaging a 10-15% lead over the opposition parties. We all know how that turned out.

    ReplyDelete
  10. Thanks DL

    Thank you DL. I stand corrected.


    There is no one out of 1216 randomly selected Canadians are concerned with the Democratic Deficit.

    Really?


    And the Liberals are ready to force and election over this ?

    And this is not the banner headline News of the poll:

    Liberals threaten Election over issue that Zero percent of Canadians are concerned with

    ReplyDelete
  11. Goaltender Interference22 March, 2011 13:15

    When it comes to what motivates Canadians to cast their ballot in favour of one party or another, 48% said they make their decision based on the parties' policies. Another 20% said that the leader is the most important factor, while only 12% based their decision on the local candidate.

    That's the most convincing reason why the current first-past-the-post system needs to be changed. When 9 out of 10 people vote primarily for a reason other than which candidate they prefer as local MP, why is this the only thing that is decided by their vote?

    If most people want to vote for which leader or party they want in power, they should be asked this directly and their vote should decide this directly. There are dozens of alternative voting methods each with advantages and drawbacks, but anything is better than forcing people to vote for local candidates they don't care about in order to have an indirect influence over who will actually govern.

    ReplyDelete
  12. Goaltender - that the people aren't voting based on the local candidate is not a good reason to have the mechanics of the election recognise that.

    What outcomes does our electoral system produce? That's what we should be looking at. That the means by which those outcomes are produces is counter-intuitive shouldn't matter.

    I like the FPTP system because it tends to produce stable governments, and the people have the ability to replace those they dislike. That they do that by voting for local candidates is immaterial.

    ReplyDelete
  13. Anonymous, I'd suggest you consider three points.

    First, sample size, beyond a certain point, doesn't really matter all that much. A survey with a sample of 2000 people isn't more precise than survey with a sample of 1000, it simply has a smaller (but not much smaller) margin of error). Statistically, each poll is equally accurate, but the larger poll is more precise.

    Second, you might want to consider where Harris Decima had the two parties going into the 2008 election. That survey, which came out the week before the 2008 election was called, had the Liberals a 34 percent AHEAD (though not be a statistically significant amount) of the Tories at 33. So right now, the single most favourable poll for the Liberals not only has them running behind Stephen Harper, but running a good 6.5% behind Stephane Dion. That's not a great way to start a campaign.

    And, while the Kim Campbell comparison may give you some comfort, do you know of a new rival conservative party (to say nothing of a new Quebec based party) that's going to pop out of the woodwork in the next 6 weeks and take almost half of the Tory's votes? Because, if not, that comparison isn't all that comforting.

    ReplyDelete
  14. BC,

    In fairness, I'm willing to bet that no one said they were concerned with people bayonetting puppies. I'm equally willing to bet that, if the Liberals could prove that Stephen Harper bayonets puppies, it would be a great election campaign for them.

    Those sort of polls (especially the unprompted ones) are more or less useless. I mean, think about, someone calls you up after dinner and asks you to name the most important issue off the top of your head, what are you going to say? Proposed technical amenments to the income tax act? You're going to name one of the "big" generic issues: health care, the economy, education, the environment and the deficit. Heck, two of those issues (health care and education) aren't even under the Federal jurisdiction, so you really wonder if they're issues of NATIONAL concern (I mean, it's not like voters are going after Harper because Johny can't read, they'll be taking that up with Mr. Mcguinty).

    ReplyDelete
  15. Ira,

    I like the FPTP system because it's products 144 straight years of more or less competent government in Canada (and, significantly longer than that in Britain), which, at the end of the day, is what we want from our governments.

    I think the onus is on anyone advocating an alternative election system, to explain (a) what's wrong with our current system, such that it needs to be changed and (b) why they think that the altnertive will produce better governments.

    ReplyDelete
  16. I don't like any system other than FPTP because they allow hard core racist parties to get elected.

    I like FPTP because it keeps the fringe parties out of the house.

    No racists, no green types, no European style democracy in Canada thank you very much !

    ReplyDelete
  17. Peter - the only person I have seen say anything about an ABC campaign is you. Here. Nowhere else have I seen any such thing - certainly not in the mainstream media.

    ReplyDelete
  18. Well it looks as if we're all about to get the chance to take part in the only poll that counts. Unless someone blinks and that's probably not going to happen.
    And God help me, I'm looking forward to it even though I have no idea what the outcome will be.
    Gentlemen,and women, go to your corners and come out fighting.

    ReplyDelete
  19. Yep Pink

    Iggy - NO
    Gilles - NO
    Jack - NO

    It's on !!

    Ira - ABC has just started to appear. Be prepared.

    ReplyDelete
  20. You said the other day that ABC was appearing in the MSM. But it's not.

    You're wishcasting.

    I fail to see why anyone dislikes Stephen Harper this much. He has no personality, so that can't be it, and his governance has been so middle-of-the-road as to be hardly worthy of comment.

    ReplyDelete
  21. An ABC campaign would suit me fine - it would mean that the coalition is out in the open and people would have a clear choice.

    I don't think that Peter and DL and John would care for the results of that kind of campaign though.

    I was wrong about the NDP supporting the budget and I might be wrong about people voting for a coalition, but I don't think so.

    ReplyDelete
  22. Guess what Earl??

    Your record shows just how far you are out of reality !!

    The public is not as stupid or programmed as you think.

    This election will be a blast !!

    ReplyDelete
  23. ABC ?

    That's rather silly.

    The dirty secret of the coalition parties is that a certain percentage of their voters have CONSERVATIVES as their second choice.

    You wanna shake those branches go ahead.

    Some fruit might land in our lap.

    ReplyDelete
  24. ABC. Anyone But Coalition? I agree wholeheartedly.

    ReplyDelete
  25. Anon, you mean ??

    Conservative Reform Alliance Party - A Coalition

    ReplyDelete
  26. Earl: An ABC campaign would suit me fine - it would mean that the coalition is out in the open and people would have a clear choice.

    I don't think that Peter and DL and John would care for the results of that kind of campaign though.


    I'm against an ABC campaign, but not because I'm against a coalition or clearly stating the possibility of a post-election coalition--far from it. I just believe that people should vote for what they want, not against the candidate or party hit by the most mudballs.

    I'm also less sure that the results would be unappealing to non-Tories. It's a mistake to think that the visceral Conservative reaction to the word "coalition" is shared by the rest of the political spectrum in this country. Or for that matter, by Tories across the pond.

    I won't be commenting here much for the next while. My attentions will be directed to the voters in my riding. Work before play.

    ReplyDelete
  27. Goaltender Interference23 March, 2011 22:53

    Ira,
    What outcomes does our electoral system produce? That's what we should be looking at.

    OK, let's look at what FPTP produces: when I voted in 1998, the PQ formed a majority government even though it came second to the Liberals. Then I moved to BC, where the NDP had a majority government even though it also came second in the previous election. The whole point of democracy is to let the ruler be chosen by the number of votes they get. Any system in which you can win even though you came in second in votes is by definition a bad system.

    If you want majority governments, fine-- have some sort of PR where the winner gets a seat bonus, or just automatically allocate a majority of seats to the party that gets the most votes. At least that way the voters could choose who the majority government will be. That drawing some squiggly lines onto a map of Canada until they form 308 unevenly populated districts happens to occasionally result in majority governments is hardly a reason to keep doing it.

    ReplyDelete

COMMENT MODERATION POLICY - Please be respectful when commenting. If choosing to remain anonymous, please sign your comment with some sort of pseudonym to avoid confusion. Please do not use any derogatory terms for fellow commenters, parties, or politicians. Inflammatory and overly partisan comments will not be posted. Please keep discussion on topic.