Friday, March 2, 2012

Liberals slip in EKOS polling

EKOS Research and released a new federal voting intentions poll today, showing that the Liberals have dropped more than three points since December and that the Conservatives and New Democrats are running neck-and-neck.
EKOS was last in the field December 14-21, and since then there has been very little change at the top of the table. The Conservatives are at 31.5% (+0.1) and trailed closely by the NDP at 29.2 (-0.3). The Liberals have seen the biggest shift, dropping 3.1 points to 21.7% support.

The Greens are up 2.2 points to 8.3% while the Bloc Québécois is down 0.7 points to 6% nationally. Support for other parties is up 1.8 points to 3.3%.

EKOS also makes a distinction between eligible voters and likely voters, which they define as people who voted in the May 2 election. It doesn't show major differences this time, however: 32.1% for the Tories, 28.9% for the NDP, and 22.3% for the Liberals. There are no regional breakdowns of likely voters in the EKOS report.

The Conservatives are leading in Ontario with 33.2% (-1.4) with the NDP at 28% (+2.6) and the Liberals at 26.8% (-5.1). The Tories are ahead in Alberta with 53.6% support (-0.8), trailed by the New Democrats (20.4%, +1.0) and the Liberals (13.1%, -3.9), and they also lead in the Prairies with 38.3% (-3.7), narrowly edging out the New Democrats at 34.8% (-3.9). The Liberals are well behind in Saskatchewan and Manitoba with 16.8% (-0.9).

The New Democrats lead in Quebec with 28.3%, down 6.1 points since the end of December. This seems to corroborate the slip in support that other polling firms have recorded in Quebec. The Bloc is running second in the province with 25%, down 2.4 points, while the Liberals stand at 19% (+0.3) and the Tories at 17.2% (-3.3).

The NDP also leads in British Columbia with 36.5% (+0.6). The Conservatives are nipping at their heels with 33.3% support (+4.2), while the Liberals are down 3.7 points to 17.4% in the province. Finally, the New Democrats are also ahead in Atlantic Canada with 34.1% (+6.2), with the Liberals in second at 28.7% (-4.6) and the Conservatives in third at 27.4% (-8.1). EKOS sampled 449 people in Atlantic Canada, the largest sample I have on record since that massive Angus-Reid poll conducted in February 2011. This EKOS poll means that the NDP has led in the region in six of the last seven polls - that's a definite trend.

With these levels of support and using the current 308-seat electoral map, the Conservatives win 127 seats. The New Democrats win 97 and the Liberals 67, giving them a total of 164 seats and enough to govern in tandem with a majority. The Bloc Québécois reaches official party status with 16 seats, while the Greens retain their one seat.

The Tories win 16 seats in British Columbia, 27 in Alberta, 13 in the Prairies, 52 in Ontario, eight in Quebec, 10 in Atlantic Canada, and one in the north.

The New Democrats win 15 seats in British Columbia, one in Alberta, 10 in the Prairies, 24 in Ontario, 38 in Quebec, eight in Atlantic Canada, and one in the north.

The Liberals win four seats in British Columbia, five in the Prairies, 30 in Ontario, 13 in Quebec, 14 in Atlantic Canada, and one in the north.

With 338 seats in the House, a rough estimate gives the Conservatives 143 seats, the New Democrats 105, the Liberals 72, the Bloc 17, and the Greens one. So, still enough for an NDP/Liberal majority. But the new seats to be created in Conservative-friendly provinces give them more of an edge - whereas they would hold 41.2% of the seats in the 308-seat House of Commons, that percentage would be increased to 42.3% in the 338-seat legislature. The NDP drops from 31.5% to 31.1% while the Liberals drop from 21.8% to 21.3%.

EKOS also broke down support by how people voted in May 2011, and the results are a little interesting. Generally, all parties but the Greens have retained over 2/3rds of their supporters. But all parties have also picked up a handful of supporters from their rivals, with no real pattern: New Democrats and Liberals to the Tories, Tories and New Democrats to the Liberals, Liberals and Tories to the NDP in roughly equal numbers. The only difference, perhaps, is that the Bloc seems to be doing best among NDP and Green supporters. It's all a little confusing.

What is also somewhat confused is the gap that currently exists between the Tories and the NDP. Is it the 2.3 points that EKOS finds among eligible voters, the 3.2 points it estimates among likely voters, or the eight to 10.5-point gap that the recent surveys from Forum and Nanos identified? It is difficult to argue that the gap is closer to EKOS's findings, as EKOS pegged it as 1.9 points back in December when other polls taken around the same time put it at between five and eight points. In other words, EKOS is showing general stability in voting intentions for the NDP and Conservatives, as other firms have shown. That is most likely what is going on.


  1. Its worth noting that if the NDP was back at 42% in Quebec, they would be leading the Tories across Canada.

    In Quebec it looks like all that has really happened since the last election is that a chunk of NDP voters shifted to Green, Other or undecided. No other party has actually picked up the slack.

    As usual Ekos has this ridiculously high 12% for Green and Other (compared to the 4% Greens and others got in the last election).

    1. You forgot the Liberals picked up quite a bit of slack as well - 5% is nothing to scoff at.

  2. I know I said i'm done with EKOS because in my view their performance has been getting steadily worse in federal elections.

    But they could help themselves out and try and fix those Green and Other numbers.

    Still way too high.

    Especially somewhere like Alberta where the 4% other is clearly Wildrose cross contamination.

  3. Eric, I think you misinterpreted the breakdowns of how people voted in 2011. It's not that 68% of identified Liberal supporters voted Liberal in the election, but 68% of people who voted Liberal in the election would vote Liberal today. It's a slightly confusing table, but it's the rows that total 100%, not the columns.

    So, according to Ekos, the Conservatives have maintained 76% of their supporters (losing 10% to the Liberals and 9% to the NDP). The NDP have maintained 72% of their supporters (losing 6% to the Conservatives and 12% to the Liberals). And the Liberals have only maintained 68% of their supporters, losing 7% to the Conservatives and 16% to the NDP). You would think that people who voted Liberal in 2011 would be die-hards, but that doesn't seem to be the case. The BQ, on the other hand, has a strong hold on those that did vote BQ in 2011.

    It might seem counter-intuitive that the Liberals had bigger percentage loses than gains, but they had a smaller base to start with. Gaining 10% from Conservatives and losing 16% to the NDP still puts them ahead overall.

    1. Yes, I think you're right. I originally wrote it that way then backtracked. I'll edit.

    2. You're right about Liberal die-hards. I expected the same on election day, but the post-election EKOS poll showed that something like half of Liberal voters changed their minds during the campaign. The Liberal floor is considerably lower than we thought it was.

  4. Feb. 21-28 Perhaps too soon to see full effects of election fraud investigation, closure of PEARL observatory & EC cuts.

  5. If the NDP can still get these kind of numbers with no leader, it's probably a good sign for them.

  6. No matter how hard the mainstream media tries to pull out all stops to promote their darling Bob Rae (there is nothing the media love more than a witty patrician) the public just doesn't buy it. The Liberals are as dead as ever.

  7. I do expect the Liberals to benefit at least somewhat from the robocall investigation. How much will be determined by how much actual wrongdoing is uncovered, and whose wrongdoing can be proven.

    This does strike me as the sort of thing the CPC would do, but I also think they would do it better than this.

  8. The Conservatives will drop based on some of the stuff that is going on now. I believe that their response to election fraud claims has been weak, and confused; unusual for them, and the more libertarian minded among their base will be upset with BillC-30. I wouldn't be surprised to see the NDP actually take the lead nationally if they elect a leader like Mulcair.

  9. And all of these polling results well before the Robocalls scandal struck !!

    One wonders if the polls were repeated today what the result would be ?

  10. Ah, actually the story broke on the 23rd or there abouts. But yes the effect on the general public and the scandal broadened and would not be felt until the week after this poll. I also believe Harper and Del Mastro's recent performance will negatively effect the polls in quite a big way. People are really disgusted.

  11. Re CPC problems.

    Now the National Post is bluntly saying that del Mastro and the party lied !! Check this out

  12. @ Peter,

    What else can you call it when they are proven wrong and admit it, then Harper sends Del Mastro and the party's spokesman out for the next two days to repeat the false claims.

    Its a National disgrace.

  13. Anonymous 19:51

    Completely agree !! Almost actionable IMO

  14. It sure is an interesting analysis. With the robotcall scandal, it should give an edge to the NDP in Ontario. The fall of the Liberal in this province came from the sponsorship scandal. Ontario's voters just doesn't seem to like openly corrupt government.

    And with the NDP chief election in 2 weeks, their new leader could probably stabilize a favourable situation for them. Topp or Mulcair can both achieve this. Without counting on the fact that Mulcair could rally the quebecois vote for him.

    In short, NDP are in for a nice Spring, as long as they capitalize on it. If partial election happen in the near future, it could be interesting.

    So yea, but for now, the robotcall scandal favour the Conservative, as it get out early in their mandate. A year is a long time in politic, 3 year is even longer.

  15. EKOS, snicker

    Lets just say that if graves had the tories at 25%... I still wouldn't rule out them winning a majority.

    NANOS is out this morning with new numbers.... well, old numbers.... they haven't changed much

  16. Eric, je sais que c'est une analyse détaillée que tu as fais pour chaque comté, mais je trouve ton article tire les conclusions un peu trop vite à mon avis.... Si le vote n'a pas été affecté dans les comtés ciblés par les appels, c'est peut-être qu'il y avait une stratégie derrière tout ca, vu que plusieurs de ces comtés étaient très serrés et ont été ciblés par plusieurs organisations de 'swap your vote' les efforts ont été mis dans ces comtés pour empêcher les électeurs libéraux et NPD d'aller voter et d'un autre côté de faire sortir le vote au maximum pour les partisans du parti conservateur, l'un ne va pas sans l'autre ici, logique non.... On n'empeche une catégorie de voteur (ceux qui ont été ciblés comme ne votant pas conservateur) et on cible aussi les conservateurs pour les faire sortir afin qu'ils votent dans un plus grand nombre et on est assuré de gagner le comté, le coté malicieux et bien orchestré de toute l'affaire dont ton analyse ne tient pas compte! Il ne faut pas tirer des conclusions avant la fin de cette longue enquête, parce que je suis sur que ce n'est fini et on va en apprendre de bien belles choses. Pire que les commandites, ca c'est sur, je suis certain que les conservateurs ne sont pas blanc comme neige avec leur bilan déja lourd.


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.