[Ottawa – February 23, 2011] – In an interesting but not entirely surprising movement, the race has tightened back to the locked in pattern of last year in one fell swoop. The Conservatives have bled nearly five points of support back to 32.4 and the Liberals have moved up more slightly to 27.3. But a 12.5 point lead is now 5.1 points, which is a huge and highly significant shift (well beyond mere “noise”).
Based on 2,811 cases, there has been a whiplash-like reversal to the last poll’s near Conservative majority. Therein may lie the explanation. Not only our poll, but three closely timed polls showed the Conservatives moving into near-majority territory. The Government may have broken its policy of “never” commenting on polls with our breaking poll from 2 weeks ago not because they really thought it was wrong, but because they knew that there is always a recoil effect when they are reported in majority territory. The cumulative weight of four highly consistent polls made it impossible to maintain that they weren’t in majority territory or closing in.
The demographics show that the same unhinged voters who had moved to the Conservatives (possibly over the ads) may have blanched at the prospect of a majority and jumped offside (women and Ontarians in particular). This sudden narrowing of the gap is a manifestation of an electorate which is paying more attention than some experts have claimed. After a moribund period of stasis, the electorate have become basically unhinged. For those who question the substantive and statistical significance of this dramatic change, there is about a one in one thousand chance that they are right and that this is mere random play.
Although the survey doesn’t have direct evidence on this point, it is possible that this recoil effect, which is a recurring pattern of the past five years, was magnified by the Bev Oda affair. In a familiar pattern of voter recoil and government missteps, the tantalizing majority of a fortnight ago has evaporated into the narrow rut advantage of the last year. Zap! You’ve been Odasized!
Click here for the full report: full_report_february_23_2011
Do the demographics, time of day, geographical dispersion, etc. of this poll match up with the previous poll?
Are the questions the same?
If any of the above are different, why? And how much of that variation could account for the different results?
Does the fact the media generally took a side and placed the Oda issue out of context mean the media has a lot of clout?
If you think you have a rogue result in any poll, how can you check that in a subsequent poll? Use same questions or new ones?
Thanks.
BT
Harper is a dictater, I cannot trust him although he tries to
appear squeekie clean.
Wow. The Green numbers look very inflated. I am from Nova Scotia and I can tell you the Green Party does not exist out here. They are not even on the radar. For them to be at 16.8% must be totally wrong. You probaly have to deflate that number by about 12% and you get the real figure.
All of these Ekos polls have the Greens much higher than any other polling firm in the country. In this case almost twice the other polls. Why?
this is a Ontario/Quebec poll, 1660 people from those 2 provinces and 700 from the rest of Canada, so do those provinces represent 70% of the Canadian population. also 70% of respondents have college, university or better, does that reflect the Canadian population?
According to Statscan, in 2010 Ontario and Quebec made up 62% of Canada’s total population. And, again according to Statscan, in 2007 62% of men and 68% of women had completed post-secondary education.