Prime Minister Stephen Harper announces the induction of legislation in Mississauga on Thursday, April 3, 2014 to create a Canadian Victims Bill of Rights.Nathan Denette/The Canadian Press
A new poll suggests the Conservatives hold a five-point national lead over the Liberals. Is this the first sign of a dramatic change in Canadians' political mood, or is the poll just an outlier?
The latest survey conducted online by Angus Reid Global, interviewing 1,505 Canadians on April 14 and 15, found the Conservatives to have the support of 34 per cent of likely voters, compared to 29 per cent for the Liberals and 27 per cent for the New Democrats.
These are rather surprising results, as the Liberals have only trailed in three of the last 49 polls conducted since Justin Trudeau became party leader in April 2013. The Conservatives have not enjoyed a lead of this size in any poll since before Mr. Trudeau's leadership victory. So what is going on?
Eligible vs. likely voters
It is important to note that the five-point edge awarded to the Conservatives in the Angus Reid poll is among likely voters, rather than all eligible voters. Among eligible voters, that lead shrinks to two points with the Conservatives at 32 per cent, the Liberals at 30 per cent, and the New Democrats at 26 per cent.
The method used by a polling firm to determine who is a likely voter can have a significant influence on the numbers, as was explained here. In previous polls by Angus Reid, the Conservatives have performed markedly better among likely voters. In the company's last poll from early March, the Liberals held a one-point edge among these voters. In a poll from mid-February, the Conservatives were up by one point. In this context, the Conservative result of 34 per cent is not significantly higher than the 31 and 32 per cent registered in those two previous polls. Similarly, 29 per cent for the Liberals is not greatly different from the 32 and 31 per cent the party managed among likely voters in Angus Reid's last surveys.
The Ontario shift
One noteworthy aspect of this Angus Reid poll is the shift in voting intentions in Ontario. Among eligible voters (an earlier version of this article was based on likely voters only for the province, before Angus Reid uploaded the results among eligible voters by region to its website), the Conservatives had the support of 41 per cent, compared to 28 per cent for the Liberals and 23 per cent for the New Democrats. That represented a nine-point gain for the Tories since March and an eight-point drop for the Liberals.
A 13-point lead for the Conservatives in Ontario is certainly unusual in the present context, as it has not been achieved by the Tories in any poll since January 2013, nor has the party managed 40 per cent in any single poll in the province since then. By contrast, the Liberals have been registered at 40 per cent or more in Ontario in four of the previous five polls.
The Conservative national lead in the Angus Reid poll is primarily due to the results in Ontario. The sample size for the province (unreported by Angus Reid) would likely have been around 500 people, which in a probabilistic sample carries a margin of error of about plus or minus 4.4 points. That has the potential to reduce the gap between the Tories and Liberals to a more recognizable four points.
But the field dates of the poll may be significant. The survey was carried out shortly after the death of Jim Flaherty, which may have triggered a temporary outpouring of sympathy. It may not be a coincidence, then, that the surge in support experienced by the Conservatives took place almost exclusively in Ontario.
An outlier or a sign of things to come?
Considering that calculations for the voting intentions of likely voters can vary from firm to firm and are based on assumptions that may or may not turn out to be correct on election day, it is more worthwhile to look at the shifts among eligible voters. This is especially the case since the next election is 18 months away (during a campaign, the importance of estimating the support of likely voters is paramount).
In this regard, the Angus Reid poll especially stands out. The shift in voting intentions among eligible voters is significant: the Conservatives were up four points over the February and March polls, while the Liberals were down three points.
At 30 per cent nationwide, the Liberals are lower than they have been in any poll by any polling company since an Abacus Data survey of Aug. 30 to Sept. 4, 2013, which pegged the party at 29 per cent. Prior to that, in only one other poll (again by Abacus) after Mr. Trudeau's leadership victory were the Liberals at less than 30 per cent.
At 32 per cent among all eligible voters, the Conservatives are at their highest level of support in any poll by any polling company since the fall. The Tories' polling numbers improved slightly between the end of September and the end of October, when the Senate scandal exploded again. At that time, three of five polls had the party at either 31 or 32 per cent. Following the re-emergence of the scandal, the Conservatives did not manage 30 per cent in 19 consecutive polls of all eligible voters until this new survey by Angus Reid.
The Angus Reid poll is thus clearly out of step with what other polls have been showing of late. That does not mean that the poll is wrong or an outlier. As explained here, what seems like an outlier poll can turn out to be the first snapshot of a changing electorate. But until other polls emerge to confirm or refute the findings of Angus Reid, caution should be exercised.
Éric Grenier writes about politics and polls at ThreeHundredEight.com.