Quote:
Originally Posted by Wormius
Hmm. Mainstreet Research. They have a good track record. A sample of 646 people? It's such a simple question, how could they only sample 646 people?
|
Your ROI on a poll with 2000 versus 600 is minimal. Much higher cost for small changes in the margin of error. That being said, Alberta is tough to poll completely accurately without a larger sample, as the donut ridings around Edmonton are included in rural in most polls but they act differently than most rural ridings.
That being said a +7 for Notley in Calgary is something to watch. What most polls seem to be showing is that Smith is stronger in rural Alberta than Kenney. That being said, I suspect she still loses a number of rural seats that are traditionally considered rural (Banff Kananskis, potentially the other Lethbridge riding, a number around Edmonton). But if the +7 is accurate she is likely rolling at +12 or so north of Glenmore. Those are numbers that she can work with. Give Smith Three Hills, Two Hills, all the hills. Win the cities and pick up a few rural seats.