I'm not saying the poll is wrong, but I suspect it is that one rogue poll. You know how they say accurate 19 times out of 20, well this is that 1 time out of 20 that should be thrown out.
First off, they tell us that 902 people in an ONLINE poll answered the survey. Then they present the results as percentages, so you have absolutely no idea of how many they polled from Edmonton, Calgary and rural Alberta, and it is rural Alberta results that I have REAL problems with.
What this poll does is change the numbers dramatically, so huge a change that the Wildrose go from consistently leading by anywhere from 9% to 17%, to being statistically tied? For all we know, they only polled 5 rural voters.
Even if these are "true" results. Their own poll results contradict what they are saying.
Also mentioned is this little, teeny, tiny fact:
Non-random meaning biased. It's a biased poll, and all the lefty media are making a big deal out of the one and only poll that shows rural Albertans changing their votes back to the PC's. They had to stretch real far to find any reason at all that would change the vote that dramatically, talking about the "conscience rights" changing rural voters minds. Seriously? Rural voters are more likely to be in favour of "conscience rights" than urban dwellers. This would be a huge misstep if the PC's latch onto that statement.
This poll is that one rogue poll that shouldn't be considered too carefully until more polls come out.