Guess what? A major polling firm has adopted this same methodology.
Ekos has recently conducted an IVR poll and found that had the election been held between September 13-15th, the Conservatives would have won 161 seats. The poll sample was of 2,848 decided voters-- a sample that is larger than usual.
Warren Kinsella says he's not impressed with the survey itself.
Not surprising. IVR polling is still new to Canada, but I believe it will be accepted.
Frank Graves, president of Ekos, says that the Liberal Party's fortunes are plunging due to the underwhelming "reception" of Stephane Dion.
He also says that the electorate is becoming more polarized. Perhaps this is true. If Canada is becoming more Conservative, as Stephen Harper said, that could be polarizing.
UPDATE Kady O'Malley is on the story.
Says CaffeineHangover in the comments:
As much as I’d love to slam Ekos for their ‘Robot polling’, their numbers aren’t terribly different from yesterday’s Decima numbers (38/27/16/9/8). They seem to be couched within statistical accuracy.
Another interesting comment...programming code as Liberal talking points:
Computer code for Liberal Partisans …
START
10 Poll comes out with bad news for Liberals
20 Nitpick the method used by the pollster
30 Remind everyone that its still early in the campaign
40 Offer up more ways to stop vote splitting
50 GO TO START
END
