Thursday, July 23, 2009

New Ekos Poll: 0.3% Conservative Lead

EKOS released a new poll today, taken between July 15 and July 21 and involving 3,158 interviews. This poll has a lot of great stuff, so I really want to go through it. First, the national result:

Conservatives - 32.8%
Liberals - 32.5%
New Democrats - 14.8%
Greens - 11.5%
Bloc Quebecois - 8.4%

Now this is a close race. The BBQ circuit has left Canadians more equally divided than ever. You can even split the country into three virtually equal groups: those who support the Conservatives, those who support the Liberals, and those who support an opposition party.

However, of note is that between July 17 and 20 the Liberals were topping out at 35% to 36% while the Tories dropped to below 30%.

This is a strong result for the Greens, but weak for the NDP and the Bloc. Demographically, the Conservatives hold the lead among males (35.7%), 45-64 year olds (37.2%), 65+ year olds (40.0%), those with a high school education (31.7%), a college education (37.5%), and in the cities of Vancouver (35.9%), Calgary (59.4%), and Ottawa (42.1%).

The Liberals lead among females (32.0%), 25-44 year olds (31.2%), those with a university education (38.2%), and in the cities of Toronto (48.2%), and Montreal (36.2%). The Liberal lead in Montreal is a change, and seems to be where the Bloc lost ground this week.

The Greens lead among those aged 25 or under with 26.9%.

Regionally, the Tories hold a lead in BC (35.8% to 25.0%) but the NDP are close behind the Liberals in third (23.6%). The Conservative lead in Alberta and the Prairies is large, and the Liberals have pulled ahead in Ontario (39.8% to 35.9%). The Greens place third in the largest province, with 12.2%. They edge out the NDP who had 12.0%, a disastrous result for them. In Quebec, the Bloc had a bad week with 32.7% support, followed by the Liberals at 30.8% and the Conservatives at 15.4%. The Greens had 10.2%, which seems like an outlier and a contributing factor in the Bloc dip. In Atlantic Canada, the Liberals are first (39.7%), followed by the NDP (28.2%).

This poll would result in the following seat totals:

Conservatives - 119
Liberals - 119
Bloc Quebecois - 46
New Democrats - 24

As close as can be. The Tories would hold 65 seats in the West and 54 in the East (Ontario, Quebec, and Atlantic Canada). The Liberals would have 19 in the West and 97 in the East, as well as the three in the north.

Now, the poll looked at two other factors. The first was "second choice". The option of "no second choice" was more than all parties, with 26.8%. Among the parties, the Liberals led, with 22.1%. The NDP was second at 19.7%, the Greens third at 14.1%, the Conservatives fourth at 13.2%, and the Bloc in fifth at 4.1%.

In other words, the ceiling for the various parties is 54.6% for the Liberals, 46.0% for the Conservatives, 34.5% for the NDP, 12.5% for the Bloc (49.1% in Quebec), and 25.6% for the Greens.

Regionally, the Liberals led the second choice category in all regions except Alberta, where the NDP was the favourite second choice.

Looking at it in terms of voters, Liberal supporters favour the NDP as their second choice (32.2%), with the Conservatives close behind (26.9%). Conservatives favour not to vote at all (40.2%), and then the Liberals (32.9%) and the Greens (12.8%). NDP voters would move to the Liberals (43.2%), not vote (17.9%), or go to the Greens (16.9%). Green voters would instead vote Liberal (30.4%) or NDP (25.2%). Bloc voters would head to the NDP (29.0%) or the Liberals (26.2%).

The second thing the poll looked at was whether Canadians favoured a minority or majority government, and which party they would want in power in either case. A majority Liberal government was the favourite option, with 26%. This was followed closely by a majority Conservative government (25%). A Liberal minority would have 15% support while a Conservative minority would have 9% support. Taking this question further, 41% of Canadians favour a Liberal government while 34% favour a Conservative government.

This means that virtually no one outside of Conservative supporters like the idea of a Conservative government.

It is also interesting to note that undecided voters were split down the middle between a Liberal or Conservative government, though the Liberals had a very small advantage.

Among NDP voters, a Liberal minority was the favourite option at 21%. Overall, 38% of NDP voters preferred the idea of a Liberal government to 16% who preferred a Conservative government. Bloc voters, too, preferred a Liberal minority (27%), and overall would choose a Liberal government (41%) over a Conservative government (17%). Greens voters liked the idea of a Liberal majority (18%), and preferred a Liberal government (34%) over a Conservative government (22%).

Regionally, British Columbia preferred a Conservative government (39% to 33%), as did Alberta (51% to 27%) and the Prairies (49% to 28%). Ontarians prefer a Liberal government (45% to 34%), as do Quebecers (46% to 24%) and Atlantic Canadians (41% to 29%).

All of this goes to show that the Liberals have the most room for growth while the Conservatives are not much below their maximum support potential. That the Liberals almost double the Conservatives in the second choice category, while more Canadians would rather vote NDP or Green than Tory if their first option was off the table, is bad news for Stephen Harper. But the Conservatives have had success in 2006 and 2008 by relying on their base, so perhaps this is not as much of a problem as it appears.

The Liberals need to reach out to supporters of other parties, since people seem to be receptive to them. But the NDP and the Greens also have room for growth. So what we have is the centre-left fighting for each other's votes while the Tories safely monopolize the right. This makes leading the Liberal Party a much greater challenge than leading the Conservatives. And since the Tories have the advantage in funding to boot, it makes it all the more difficult for Michael Ignatieff.

The weekly projection update will follow later this morning.


  1. The NDP is always going to get a bad result after a push poll like this. They asked first if you want a minorty or Majority Liberals, or Conservative government. This makes voters think their is really only two options. Then they have a real poll and the NDP and Bloc do bad. It isn't rocket science the support drops during push polls.

    I bet if they asked "would like a leader who lived in this country for most of his life and knows Canadians, or one who lived outside the country?" Then polled the Cons would have a huge lead. Push polling isn't hard to figure out and polling companies who want to be looked at as real polling companies should not do it.

  2. Do we know in what order the questions were asked?

  3. It doesn't say what order they were asked, but if they were asked in the order they are listed in the full report, the first question would be:

    If a federal election were held tomorrow, which party would you vote for?

  4. I would hope, and expect, that is what they did.

  5. Regarding the comment:

    "This means that virtually no one outside of Conservative supporters like the idea of a Conservative government."

    Are you saying this because 32.8% (vote intention) and 34% (government preference) are quite close together?

    But the 32.8% who currently intend to vote CPC is based on the smaller sample size of decided voters (2633).

    However, the 34% who prefer a Conservative government is based on the larger sample size that includes undecideds (3158).

    What this boils down to is that the 34% who prefer a Conservative government is composed of:
    - 24% (of 3158) who are Conservative supporters and prefer a Conservative government PLUS
    - 10% (of 3158) who are NOT Conservative supporters but still prefer a Conservative government

    In other words, almost 30% of the people who prefer a Conservative government do not intend to vote Conservative.

    This does not seem like "virtually no one" to me.

  6. You are assuming the undecideds won't vote for the Conservatives. Since undecided voters split almost evenly on whether the government should be Liberal or Conservative, and that the 32.8% can be assumed to represent 32.8% of Canadians (polls are designed to estimate support among all voters, not just decided voters) I don't see why my statement should be changed.

    The poll shows that about 33% of Canadians would vote for the Conservatives in an election. The poll also shows that about 34% of Canadians prefer a Conservative government.

    Those undecideds will have to vote for someone on election day, or not at all. Pollsters assume that undecideds will break in a proportion equal to decided voters, so this changes nothing.

  7. I am not assuming that.

    But even if we lay aside the debatable claim about undecideds, it is not merely about undecideds (although they are the reason for the difference in the sample size).

    Note in particular that 22% of Green voters prefer a CPC government, as do 17% of Bloc voters, 16% of NDP voters and 8% of Liberal voters.

    So, expanding on my earlier summary:
    the 34% who prefer a Conservative government is composed of:
    - 24% (of 3158) who are Conservative voters and prefer a Conservative government PLUS
    - 7.3% (of 3158) who plan to vote for one of the other four parties but still prefer a Conservative government PLUS
    - 2.3% (of 3158) are undecideds who prefer a Conservative government

    So, of decided voters who prefer a Conservative government 23% are people who intend to vote for different party.

    I would reiterate my claim that your statement about "virtually no one" is not a reasonable interpretation of the data.

  8. The sample sizes of the "government preference" poll do not match those of the national poll. If they did, the Liberals would have over 34% support rather than the 32.5%.

    When you do the math, what it turns out is that the 34% who prefer a Conservative government are made up of 76% Conservatives, 17% other parties, and only 7% of undecideds.

    Out of decided voters who want a Tory government, 82% come from the Conservatives and only 18% from other parties.

    With such a large majority, I'm comfortable in saying that virtually no one outside of Conservative Party supporters want a Tory government. Especially when you compare that to the Liberals, where only 67% of decided voters who want a Liberal goverment come from Liberal supports.

  9. "Out of decided voters who want a Tory government, 82% come from the Conservatives and only 18% from other parties."

    Where I got my number of 23% (rather than 18%):

    Note the raw numbers in the bottom section on page 10 of the detailed report.

    A total of 3012 people are listed as having answered the government preference question based on current vote intention.

    Those who preferred a Conservative government:
    88% of 871 Conservative voters = 766 people
    8% of 899 Liberal voters = 72 people
    16% of of 368 NDP voters = 59 people
    22% of 285 Green voters = 63 people
    17% of 210 Bloc voters = 36 people

    This is a total of 996 people decided voters who prefer a CPC government and 230 of them are people who intend to vote for a different party than the Tories.


  10. I have to give you that one, when I made the calculations I wrote a "6" rather than a "63" for the Greens.

    I'll take back my statement about Conservative supporters, but I still believe it to be significant that the amount of Canadians who want a Conservative government is virtually the same as those who intend to vote for the party.

  11. The difference in numbers still doesn't really change anything. The point that the conservatives don't have much room to grow remains a valid one and has been on display since the inception of the CPC over 5 years ago. In the last election the Conservatives didn't make any gains at all - rather they looked 'better' than the other parties who all lost support and saw their voters stay home instead of voting. This is why we still see Harper doing really pathetic things - he knows he has to keep the base happy because when they are happy, the come out to vote - offering the conservatives a plurality and nothing more.


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.

Note: Only a member of this blog may post a comment.