Tuesday, January 4, 2011

CROP shows big Bloc and PQ leads in Quebec

CROP conducted a pre-Christmas poll in Quebec, reported on last week by La Presse. It doesn't show much change from their last poll taken in the second half of November.At the federal level, the Bloc Québécois is up one point to 40%, well ahead of the Conservatives. They're up three points to 20%, and are second in the province.

The New Democrats are next, steady at 19%, while the Liberals have dropped to fourth place, down one point to a mere 18%. It's a logjam among the federalist parties. Both the Conservatives and New Democrats should be pleased with this level of support, particularly the NDP.

The poll would result in 52 seats for the Bloc, 12 for the Liberals, nine for the Conservatives, and two for the New Democrats.

The regional concentration of support enjoyed by the Bloc (francophones), Liberals (Montreal), and Conservatives (Quebec City) squeeze out the NDP, who have no regional concentration.

At the provincial level, there is a bit more movement.The Parti Québécois is up three points to 41%, while the Liberals are up two points to 25%. It is still a huge lead for the PQ and a disastrous level of support for Jean Charest.

The ADQ is up one to 16%, while Québec Solidaire is steady at 12%. This is a very good level of support for both parties. QS seems to really be making some strides, while the ADQ is back up to respectable levels.

Pauline Marois is the best person for Premier for 29% of respondents, while 18% opted for Jean Charest. Both Amir Khadir (QS) and Gérard Deltell (ADQ) scored 13%.

The poll would result in 83 seats for the Parti Québécois, unchanged from my last CROP projection that was featured in The Globe and Mail. The Liberals are up one seat to 25, while the ADQ is unchanged at 15 seats. Québec Solidaire would win two seats, down one from the November projection.

Provincially, this poll does not raise any alarms. However, CROP does have the Liberals lower than Léger Marketing puts them, which is the opposite of how the two pollsters have traditionally trended.

The federal numbers are a bit more unusual. Both CROP and Léger have been polling very well for the NDP in Quebec lately, but other pollsters do not see Jack Layton's party at such a high level of support, nearly twice their 2008 electoral haul.

One could claim that, as CROP and Léger are Quebec-based pollsters who routinely sample 1,000 Quebecers, they are more reliable. I'm not so sure if we can leave it at that. Léger is one of my higher rated pollsters, but CROP is one of my lowest rated. They both use an opt-in online panel, which CROP itself admits is not ideal. Léger's panel appears to have been created randomly, giving it a bit more of a claim to accuracy. Methodology seems to be the decisive factor, as even Angus-Reid (which uses a similar opt-in system) tends to side more with CROP and Léger. Traditional methods, like those used by EKOS, Harris-Decima, Nanos, and Ipsos-Reid, yield very different results.

We'll have to wait until the next election to settle who is right.

20 comments:

  1. The Liberals are running FOURTH in Quebec? Granted, not by much, and it could always be a rogue poll, but still. Those sorts of polls keep up, we won't be seeing an election until the fall of 2012.

    Eric, I'm curious if you had any thoughts on the word-association poll that Abacus published over the weekend. I thought it was a clever way of getting a different perspective on public perceptions of the various political parties.

    Moreover, I was struck by the depth of the negative perception of the Liberals, both in terms of honesty and leadership. It's remarkable that, more than seven years and three Liberal leaders removed from the Chretien era, the Liberals continue to be seen as corrupt and dishonest. How often is it the case that the official opposition (especially one that has been in opposition for 5 years) is seen as more corrupt and dishonest than the government? It's even more peculiar given that the current leader wasn't even in the country (much less a member of the Liberal party) when the sponsorship scandal occurred and that a number of prominent Liberals are "new blood" (Hall-Findley, Kennedy, Rae, Trudeau, Leblanc).

    The poor perception of leadership is less surprising, given that neither Dion nor Iggy have been particularly impressive as opposition leaders (that's an admitedly partisan view, but one apparently shared by the Canadian public). Indeed, it may well be the lack of leadership that has resulted in the failure of the Liberals to turn the page on the sponsorship scandal. The Liberals haven't taken any bold steps to redefine themselves as something different from their predecessors (much less, on most substantive issues, from the Tories), so perhaps it shouldn't be too surprising that they haven't been able to shake the negative associations from the previous era.

    In any event, whatever the cause, given that public perception, the continuing poor performance of the Liberals in the polls isn't all that surprising and, until that perception starts to change, isn't likely to improve any time soon.

    ReplyDelete
  2. The most interesting thing I took from the wordles was how much more prominent Harper and Layton were in the Conservative and NDP clouds than Ignatieff was in the Liberal cloud.

    It says that the parties are really leader-driven, and that it's hurting the Liberals.

    ReplyDelete
  3. Where is the base of the 12% QS vote in Quebec? East-end Montreal?

    As the QS seems to be the most left-wing option in Quebec, do they take their vote mostly from the centre-left PQ?

    Or am I misreading the Quebec provincial scene?

    ReplyDelete
  4. Given Ignatieff's terrible popularity its probably a blessing that he's not as strongly associated with the Liberals as Harper is with the Conservatives.

    After all this time they clearly still have some remaining cache as "Canada's natural governing party" that holds them afloat in certain areas of Vancouver, Montreal, Toronto, and Atlantic Canada.

    ReplyDelete
  5. I really start to question when I see results like these. I simply can't find it in myself to trust online polls; the methodology or the sampling simply seems to yield results so wildly different from anything we've seen in traditional landline, and in the case of the UK, cell-phone based pollsters. If something seems out of place, then it probably is. Does no one else think its weird to find the Liberals running fourth in Quebec, while everyone else shows them running second?

    I don't know. After all, 2006 did show the Liberals can go very low in Quebec. But this doesn't seem right. Not right at all.

    ReplyDelete
  6. JQ,

    QS has a surprisingly wide-range of support. Sort of like the NDP - it will end up with the same result, 10% and two or three seats. The base of their support is in central/eastern Montreal, where Amir Khadir and Francoise David have done very well. David is almost a shoo-in to be elected next time. But you'll also see them doing well in Quebec City in polls. Will it hold in an election? Probably not, but I think they should double their support in the next election, at least.

    Yes, they do take primarily from the PQ, especially since QS is a sovereigntist party. It probably also takes a little from the left-wing of the Liberals, but it appears a lot of Liberal support is going to the Greens.

    If those two parties weren't around, we'd likely see the Liberals at 33% or so, the PQ at 49% or so, and the ADQ at 16% or so.

    ReplyDelete
  7. Online polls underrepresent the elderly, who vote strongly Liberal in Quebec. Charest is no doubt a distant second, but not by the margins shown in online polls. By the same token, while I agree Francoise David has an excellent chance of winning a seat if she runs again, QS will not get anything like the support that online polls would suggest.

    ReplyDelete
  8. GI said: "Online polls underrepresent the elderly, who vote strongly Liberal in Quebec."

    Which would be a good point if pollsters who use online polling (including CROP) didn't weight their data to reflect the underlying population(age, gender, etc.).

    ReplyDelete
  9. The problem being that perhaps older people who use the internet are somewhat different than the majority of older people.

    ReplyDelete
  10. In Quebec with the Liberals falling 6% from their 2008 level you have them losing 2 seats.

    I can understand this prediction that even though the CPC are only done 1.5% from 2008 they might lose 3 seats as they won some close seats last election.

    There were only 2 seats that the Liberals won by less than 14.9 % in 2008:

    In Brossard-La Prairie Alexandra Mendes won by 69 votes.


    In Papineau Justin Trudeau won by 1189 votes... 2.8% margin.

    This poll and your seat projection (2 ndp seats) also would mean that Cauchon is not likely to upset Mulcair in Outremount.


    The 3 CPC to lose seats would be
    Sylvie Boucher 4.2
    Denis Lebel 3.9
    Daniel Petit 11.9

    Lawrence Cannon hangs on to his seat despite a only 8.5% win as it was over a Liberal.

    It sure makes it more interesting when names are added to the number projections.

    Two front runners for the post-Ignatieff Liberal leadership are not likely to win their seats.

    ReplyDelete
  11. Eric said: "The problem being that perhaps older people who use the internet are somewhat different than the majority of older people."

    Which is a fair point. A priori, though, I'm not sure why we would believe that old people who support the Liberals are less likely to use the internet than old people who support the Conservatives/NDP/Bloc (or vice versa).

    Moreover, the same point could easily be made about telephone polling (are those people who pick up the phone and spend 5 mintues pushing buttons systematically different from those people who don't pick up the phone or hang up? Probably, but does that systematic difference lead to bias in political support?).

    ReplyDelete
  12. BC,

    In fairness, in 2008 Justin Trudeau was a newcomer (and a political rookie) challenging a Bloc incumbent. Even with a weaker overall Liberal performance, I'd bet on him winning that seat with a larger margin next time.

    ReplyDelete
  13. Carl,

    Yes, that is the debate. But virtually everyone has a telephone, while the internet still has not reached everyone, particular people in rural areas or who are elderly.

    In telephone polls, the people who can't be bothered to answer a survey aren't being represented, but in internet polls we have to add to that group the people who don't have the internet.

    ReplyDelete
  14. The older people on the internet and who opt into the surveys are probably almost certainly going to turn out to vote.

    The online universes are a better reflection of the people that do vote rather than the population as a whole.

    I think that is a reason why AR is so very much more accurate than their competitors in actual election predictions.

    How accurate was Crop the last time they were marked ( predictions versus actual in the 2008 election)?

    ReplyDelete
  15. Moreover, I was struck by the depth of the negative perception of the Liberals, both in terms of honesty and leadership. It's remarkable that, more than seven years and three Liberal leaders removed from the Chretien era, the Liberals continue to be seen as corrupt and dishonest.

    What steps have the Liberal party taken to clean up this breach of integrity?

    They look incredibly weak and shifty compared to Harper kicking a cabinet minister out of the party on a much more nebulous ethical infraction.

    The Liberals seem to be counting on time as being a cleansing agent. That will work as soon as all the Chretien era cronies (Kinsella, Donollo, Coddere, Jennings, Cauchon, etc) fully retire from politics. Chretien's legal gymnastics to clear his name that never was sullied are another reminder of the corruption. He doth protest way too much.

    The Liberals need to distance themselves from Chretien and acknowledge the fact that he was not a shining example of integrity the same way Harper moved away from Mulroney.

    ReplyDelete
  16. I'm not sure about the federal election, but CROP was the worst pollster during the 2008 provincial election in Quebec. I'm not sure why CROP is often called the "gold standard" by some journalists and commentators, they certainly are not.

    ReplyDelete
  17. BC said: "What steps have the Liberal party taken to clean up this breach of integrity?"

    Other than call the Gomery inquiry that ultimately cost Paul Martin his job?

    But I take your point, nothing is probably the right answer. There's been a failure of leadership in the Liberal party to drag it away from its past. The real problem is that neither Dion nor Iggy (or any of the alternative candidates for leader in either 2006 or 2009) were political heavyweights with enough clout on their own to reshape their party in their image. And the Liberals aren't sufficiently patient to give them time to develope as political heavyweights (in the way that politicians like Dalton McGuinty or Mike Harris grew into their roles) and put their own mark on their party.

    Moreover, Iggy and Dion have been hurt by facing a minority government. If Harper had won a majority in 2006 or 2008, that would have given the Liberals a good 4 years to shake things up and rebuild. Their problem is that, because the Tories might force an election at any time, and the Liberals just can't abide the prospect of losing an election, they don't want to risk the sort of open internal fighting that comes with rebuilding party.

    ReplyDelete
  18. Eric": in rural areas or who are elderly,

    Fair enough, but if pollster weight for regional residence and age (as Crop, I believe, does) it's really the same issue as for telephone polls - are the people who participate systematically difference in terms of political beliefs than those who don't.

    ReplyDelete
  19. Carl:

    To overgeneralize, older voters who have a conservative approach to technology also tend to be conservative in other areas of life, including political attitudes. They would be much more likely to vote Liberal than PQ or QS.

    ReplyDelete
  20. GI said: "To overgeneralize, older voters who have a conservative approach to technology also tend to be conservative in other areas of life, including political attitudes. They would be much more likely to vote Liberal than PQ or QS.

    Got any evidence to support that assertion?

    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.