Friday, July 13, 2012

B.C. NDP maintains lead

Earlier this week, Angus-Reid released its latest numbers on the political situation in British Columbia. They show that Adrian Dix's NDP remains solidly in the lead, while the governing B.C. Liberals continue to struggle to hold-off the B.C. Conservatives.
Angus-Reid was last in the field in British Columbia May 7-9, and since then the New Democrats have slipped five points to 45%. They hold a solid lead over the Liberals, who are unchanged at 23% support. The Conservatives, meanwhile, are up three points to 22%.

The Greens jumped two points to 8% while 2% of British Columbians said they would vote for other parties or independent candidates.

Only the drop in NDP support appears to be statistically significant, if we assume a random sample from this online panel, though it is marginally so.

The New Democrats have the edge among men in British Columbia with 39% support to 29% for the Liberals, and lead by a huge amount among women: 51% to 21% for the Conservatives.

In Metro Vancouver, the New Democrats lead with 47% (-2) and are trailed by the Conservatives at 24% (+8) and the Liberals at 21% (-5). The gain by the Conservatives in Metro Vancouver is potentially significant, and undoubtedly a problem for the Liberals. The region is normally one of their strongest.

The New Democrats also lead on Vancouver Island with 46% (-6), followed by the Liberals at 26% (+4) and the Conservatives and Greens at 13% apiece.

In the Interior, the NDP sits at 41% (-7) and are trailed more closely by the Conservatives at 27% (+1) and the Liberals at 24% (+6). In the northern part of the province, the New Democrats lead with 48% (-13). The Liberals have made real gains with a 16-point jump to 29%, while the Conservatives are at 24% (+8).

It's no surprise that the New Democrats win a big majority with these numbers. They would win 70 seats with this level of support, while the Liberals would win nine (seven of them in Vancouver) and the Conservatives four (all of them in the Interior). Two independents would be elected as well.

Adrian Dix is clearly in a strong position with less than a year to go before the next election. He scored better than Christy Clark on every issue in Angus-Reid's polling: crime, health care, economy, environment, education, and federal-provincial relations. This is the sort of indicator that is horribly damaging to Clark's chances of winning.

Dix is far and away the most popular leader in British Columbia, with an approval rating of 48%. That compares quite favourably to Clark's 28% approval rating. His 38% disapproval rating is also much better than Clark's 64% disapproval. And this isn't a question of opinion yet to be formed about Dix - only 15% were not sure of their opinion of him.
John Cummins of the Conservatives is still relatively unknown with a "not sure" score of 28%, and of those who do have an opinion it isn't very good: 44% disapprove of him compared to 28% who approve. But the Liberals are in a sorry state if they are hoping to win a "who do people dislike less" competition with Cummins.

Dix is seen as the best person to be Premier by 26% of respondents, while 21% think none of the leaders would make a good premier. Clark scored 15% and Cummins 12%. After removing the "none of the aboves" and "don't knows", the respective results for the leaders are relatively close to their party's results, suggesting that none of the leaders are a particular drag (or particularly more popular) than their own party. This would seem to confirm the voting intentions result even more.

And those intentions are solidifying. Dix's NDP has held a lead of 20 points or so since March and has held a significant lead over the Liberals since November of last year. That makes nine months, roughly the amount of time between now and the 2013 election. If he has been able to maintain that lead for the last nine months, it stands to reason that he can hold it for another nine months. It won't be easy, of course, but the challenge facing Christy Clark is even greater.

39 comments:

  1. Didn't the NDP drop 5 points in this poll? I realize that's grasping at straws a bit lol, but dropping 5 points isn't ever a good sign.

    ReplyDelete
    Replies
    1. They did drop five points, but assuming a random sample that isn't a statistically significant drop. When a party is at a high level of support, it is harder to nail down the exact number. This makes the MOE higher.

      Delete
    2. MOE = ((p-p^2)/n)^0.5, which gave me 1.8%. What am I missing?

      Delete
    3. By my calculations, the margin of error for the difference between a party at 50% and 45% in two polls of about 800 respondents is +/- 6.8%. The formula I use looks nothing like yours!

      Delete
    4. http://en.wikipedia.org/wiki/Margin_of_error#Calculations_assuming_random_sampling

      Looking at the difference between the two polls is more correct than what I did, but that would still only give me an MOE in the range of ~2.5%.

      I should probably have mentioned that that's only 68% confidence level I'm referring... lol. So 5% is just barely at the 95% confidence level.

      Delete
    5. Are you accidentally assuming the two values are anti-correlated? Because that's what you would do when comparing two parties at 50% and 45% in the same poll, but it wouldn't be true here.

      Delete
    6. I'm just going by what's described in this paper:

      http://abcnews.go.com/images/PollingUnit/MOEFranklin.pdf

      Delete
    7. That's essentially the same formula I'm using. "3 Difference between two polls"

      That gives me an MOE of 4.9% though.

      Delete
    8. And now going through the formula I was using, I see that I had "q" as 1, rather than "1-p1".

      Thanks for catching the error, I will edit. I'll go through my other formulas to ensure other errors didn't slip-in.

      My apologies!

      Delete
    9. Lol no worries. My first calc was off too lol.

      Delete
  2. I wish they'd give us the number of undecideds too, but I guess that's not possible with an internet survey.

    ReplyDelete
  3. I don't think it will be difficult for Dix to hold onto his lead. BCers have clearly decided that they're done with the Liberals, and the policies the Conservatives are proposing simply aren't going to appeal to urban BC voters.

    I'm confident declaring that the BC NDP have the next election in the bag, and there's nothing the Liberals can do about it.

    ReplyDelete
    Replies
    1. Strongly agree.

      Anyone who watches the news around here knows that all the coverage has shifted to being very favourable to the NDP.

      The insiders are making peace with this reality. Don't expect any huge independent expenditures to save the BC Liberals out of fear of retribution from an NDP government in waiting.

      BC Conservatives probably won't do well in metro Vancouver or Victoria but they're going to supplant the Liberals everywhere else.

      A non-compete and coalition agreement with the Liberals centered around them running in urban ridings could be a future option to explore next election.

      Delete
    2. A non-compete would make sense now even, but it's not going to happen with Cummins as leader. He even had an NDP lawn sign last election (I would know, we're from the same riding).

      Funny thing is, it's not the HST (which he supports) or anything like that that makes him hate the BC Liberals. From what I've heard it's the treaties that have been signed with the First Nations.

      Delete
    3. Ryan people still remmeber Christy Clark heckling Brian Mulroney and her ex-husband's federal Liberal involvement.

      Both Cummins and Clark would need to go before any sort of arrangement can be reached between the Liberals/Conservatives.

      Besides this go around nothing is going to save the BC Liberals.

      Christy Clark should go out strong, take a year off, and then re-emerge to snag one of those new Vancouver seats on the federal scene.

      Ex-premiers become cabinet ministers.
      Pretty good consolation gig if there ever was one.

      Delete
    4. Ryan,

      Absolutely right about Cummins he is a vehement opponenet of the treaty process. From what I understand the BC Conservative party is mainly ex-Reformers who do not believ the Liberals or for that matter Mulroney and the PCs were conservative enough.. From what I have read by Norman Spector Cummins wants to supplant the BC Liberals so I would not expect any alliances with Christy Clark.

      Delete
    5. You say "ex-premiers become cabinet ministers", but the federal Conservatives wouldn't even let Grant Devine run for them in 2008.

      Delete
    6. Cummins only has a fringe of the Reformers with him anyways. The best talent (Strahls, Day, etc) is still firmly in the BC Liberals camp. All that Cummins has is a group of crazies that Harper to great care to sideline and muzzle.

      I agree that you'd need some third person to lead a coalition government - someone who's respected by both sides. Don't know who that is though. I'd suggest Chuck Strahl actually, but I doubt his fight with lung cancer would make it feasible. Still not happening with Cummins there though.

      Delete
    7. @Derek - The trouble with Cummins is he has no end game. The BC Liberals welcome conservatives and moderates alike. Is there a home for moderates in Cummins' party? I'd argue no, and without support from moderates the BC Conservatives won't defeat the NDP.

      Delete
    8. Ryan,

      I would suggest Cummins' end game is simply to beat the Liberal party. If the Conservatives win official opposition all pressure will be on the Liberals to fold into the Tories. Cummins doesn't need moderates at the moment because he will not win the next election. So rather than having an election about forming government his election will focus on creating the alternative to the NDP.

      Delete
    9. Right, but if the moderates all move over to the NDP or the Greens he's screwed. Centre-right isn't the only possible "coalition." If you get something like Vision Vancouver, Cummins still isn't going to see power. Ever.

      Delete
    10. I don't think Cummins' strategy is to gain power. In order to do so he would need to beat the NDP and the Liberals. He knows he can't beat the NDP this time around so why focus on the impossible? Cummins' goal is to shift the political spectrum to the right by replacing the Liberal party as the right-centre-right alternative. The danger for him is moderates moving en masse to the Liberal party thereby securing the Liberals' position as the centre-right government alternative.

      Cummins is trying to do provincially what Vision did in Vancouver; become the alternative by shifting the centre to the right much as Vision shifted the left to the centre. Cummins doesn't need 40% of the vote to become Opposition he needs 25% so he will create a campaign that distinguishes the Conservatives from both the Liberals and the NDP.

      In the future the Conservatives will undoubtedly need to gain the trust of the median voter in order to form government but, we are one or two election cycles removed from that becoming a real possibility.

      Delete
  4. Is there a riding by riding model I can look at? I'm just curious to see where those remaining Liberals and Conservatives are.

    Thank-you sir,
    Mr.Powell

    ReplyDelete
  5. I think this poll is more realistic than AR's last one. Other polsters have the NDP in the low to mid 40s so 45 is certainly reasonable. I certainly understand why Premier Clark is stubbornly sticking to her plan to wait until May for the next election, as anything can happen in politics. But I think anyone hoping the Libs can pull this one off is going to be disappointed.

    ReplyDelete
  6. The problem with this ARS poll is that the Conservatives are now in 2nd place in Metro Vancouver.

    There's a certain implausibility in that finding. Doesn't make any political sense whatsoever notwithstanding the by-election results in both Port Moody and Chilliwack where the BC Cons placed 3rd in both cases(Chilliwack representing Metro Vancouver's MOST Conservative riding).

    That's why I have reservations about both ARS and Forum polling here in BC (in terms of the ACTUAL numbers).

    Ipsos, Mustel, and Justason Research collectively have diff figures and seem to also corroborate each other.

    ReplyDelete
    Replies
    1. The samples are so small for the sub-regions that there is a major margin of error. The numbers aren't given, but since the whole poll is 800 people, we are looking at small numbers for any of the sub-regions.

      Delete
    2. The problem with that hypothsis is that Metro Vancouver typically always represents 50% of any provincial poll sample. In this case, that's ~ a 400 sample size.

      Anything under a 300 sample size is basically worthless in terms of accuracy (200 sample size for VI and BCI in this case).

      Delete
    3. What's going on with AR provincially anyways? They have such a good track record federally, yet they were so far out to lunch in Alberta...

      Delete
  7. The 5 point drop isn't significant when the party is at uber high levels of support. Just look at federal Conservative numbers poll by poll. Drops and upticks of 10 points between polls aren't uncommon.

    ReplyDelete
  8. Clark has been premier for 16 months and there never had been much public confidence in her government. A newly designated prime minister or premier should face the electorate within a year. This should be especially true when opinion polls consistently show wide-spread disapproval of a government from both sides of the political spectrum.

    If the B.C. Liberals win the next election, it will probably be one of the largest election surprises in recent Canadian history. The political survival of Dalton McGuinty and Alison Redford are not huge surprises, when comparing them with their opposition. The same could be said about Jean Charest potentially winning another election later this year.

    ReplyDelete
    Replies
    1. That's the big mistake. Clark wanted to (and campaigned on) going to the polls once the BC Liberals and the NDP each had new leaders. The party opposed it, and now we're reaping what we sowed. :(

      Delete
  9. New numbers from EKOS:

    NPD: 32.3%
    CPC: 30.2%
    LPC: 19.5%
    GPC: 10.0%
    BQ: 4.9%
    Other: 3.1%

    Undecided and ineligible: 13.1%

    In the analysis that EKOS has published, they seem very confident in that 10% number for the Greens, though they acknowledge that Green voters are the least likely to actually turn out and vote.

    ReplyDelete
    Replies
    1. Ekos says their confident in that Green number, but the reasons they give for their status as outlier don't hold any water.

      I suspect they don't know what the cause is.

      Delete
    2. How so Ira? I thought EKOS whole thing is they poll all voters and don't try to impose a likely voter screen between elections. They're trying to show what the preferences of the population at large is.

      Most other pollsters try to impose some sort of likely voter screen to get at how those preferences will translate to support at the ballot box.

      Two different stories...

      Delete
  10. Hi Eric, great site! I've followed for a long time but never commented.

    G&M ran an article about democratic reform today. How would single transferable ballot or instant run off affect the BC seat totals? Could the Lib-Cons win a coalition govt that way?

    ReplyDelete
    Replies
    1. IRV would likely give an even bigger NDP majority, though STV would make a Green-NDP coalition a real possibility...

      I think it would just make politics better though. Liberals would be trying to reach out to Greens for their second votes. The NDP would be reaching out to Conservatives. Not every issue is zero sum; sometimes we all can win. It'd be nice if our politics reflected that.

      Delete
  11. The only danger for the NDP is Liberal support is so low (and has been for a few months now)it may collapse altogether outside Vancouver. While, this would likely still result in a NDP majority a close race may create a new and unforseen dynamic.

    The Conservatives can now claim they are the "opposition to the NDP" in suburban Vancouver and the Interior. If the trend continues the Liberals may be pressured to run less than a full slate to increase the appeal and potential of "free enterprise" candidates.

    A large per centage of the vote in the North and Interior is a populist vote. If the Conservatives look able to win a swath of seats then some populist NDP & Liberal voters may switch to the Conservatives (we regularly witness this on the federal level). While I doubt the momentum would be sufficient to deny the NDP a majority this is BC and almost anything can happen in politics.

    ReplyDelete
    Replies
    1. The trouble with that is John Cummins himself. He's way too far out of the mainstream on social issues, the environment, First Nations... Quite frankly, he just comes of as a mean old man. He also supported the HST, which while a good thing in my books, probably won't help him with ex-BC Liberals who are still upset about it.

      If someone like Chuck Strahl was leading the BC Conservatives, then I could see the "free market coalition" potentially shifting over to the BC Conservatives like the Socreds did in the 1990s. I just don't see Cummins standing up to media scrutiny.

      Delete
  12. Rerired in BC17 July, 2012 11:14

    The part of this poll that doesn't make sense is that Dix would have an approval rating of 48%.

    I can understand that Clark can lose the election by disapproval but am rather shocked that people approve of Dix and only 38% disapprove.

    He is running ahead of the party. 48-45 Really? and the 45% that are voting Liberal and Cons don't all disapprove of him???? 38 disapprove and 45 plus 8 green voting against him and his party.

    If a Green voter approves of Dix why in the world would they be voting for the Green party?

    Either this poll has failed the internal sanity checks or this is an very interesting phenomenon .... Dix-amania ?

    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.