- Liberals - 47%
- NDP - 36%
- Greens - 12%
- Others - 5%
On the surface, these numbers look and feel closer to other polls and what my gut says is happening in the election, but there are some areas of significant concern for me with the poll. Sacha Peter beat me to a punch with a lot of the issues. His take on bias within the poll based on what questions are asked and in what order is also of concern to me.
One thing I will note is that they have 84% of the people in their sample saying they are decided voters and will be voting. As I have noted before, there is a major error within the polling process when they get the sort of responses rates that indicate 84% voter turn out.
In this poll their total sample size was 601 eligible voters. Only about 370 of the responses should have said they are voting and of that 370, at least some of them should be planning on voting but do not know how they will be voting. They had 508 decided voters in the poll. Something is fundamentally wrong here.
Based on the lack of details for the poll, some question about potential bias, but a reasonable sample size, I am assigning this poll a value of 1 in my calculations, the same as Angus Reid.
I am also downgrading the De Dutch Burger poll to 0.2. I am trying to come up with a formula to weight the polling numbers that came out of the debate as to who won etc, but I am having trouble making anything work as a measure of party support.
So based on my current calculations I come up with:
- Liberals - 46.1%
- NDP - 36.5%
- Greens - 12.9%
- Others - 4.5%
Based on the above numbers, here is my projection of the election outcome and the ranges possible.
- Liberals - 65 (57 - 71)
- NDP - 20 (13 - 27)
- Greens - 0 (0-1)
- Independents - 0 (0-1)
As for the Greens, I still think Jane Sterk is a long shot to win in Esquimalt, there does not seem to be any other Greens that come close enough to matter in my calculations. Some of the Greens with real on the ground campaigns might be able to place a strong second this time around, I can think of about five strong Green campaigns this time around which is an increase from the last two elections.
We should see at least two more significant polls this week, both of which should be post debate. I do not believe the debate will have much impact on the polling numbers. Viewship was low and newspaper media coverage was in the Monday papers, the least read papers of the week.
No comments:
Post a Comment