Federal Election Polls

[CITIZEN_2015]

What you're trying to do is read a non-voters mind, when in reality you can't really say why any significant number didn't vote, or who they would have voted for if they had shown up. It's literally treating non-voters as some sort of an extra set of numbers to attack the winner with. If you don't vote, you have silenced yourself, and at that point I think it's deeply unfair that someone comes along and uses them deliberately as a cipher to impose their presumptions upon.

My assumption is that if there is a general election and a citizen doesn't take a few minutes to vote (unless it is in the middle of a horrible winter storm , not likely in the middle of October or sick with flu or something similar) then she or he doesn't agree with any (None) of the parties or policies put forward. So why 30 to 35% don't vote? All sick with flu that day?

Edited October 6, 2015 by CITIZEN_2015

[blueblood]

Trust me. I don't want to believe that the Tories are ahead, but I suspect that those IVR responses may be more truthful in this case. On the other hand, some people may just be frustrated with the robocalling and mash numbers to mess with the poll. Tough to say. Though I expect that it does correct for some desirability bias where a respondent may feel like they will be judged for voting Conservative.

I honestly think its wide open. Both the tories and the liberals campaigned well. I think the million dollar question will be if the tpp is a big enough sell for canadians to re elect the tories in spite of the electorate being sick of harper.

Trudeau has essentially ensured he gets another kick at the cat next time around barring an unforseen disaster.

[cybercoma]

There is also such a thing as an outlier, which may be a bell weather of a major change, or may just remain an anomaly.

Recent polls are really all over the place, and I think we are going to see a lot of movement between the Liberals and Tories now that it has become an effective two-way race.

I don't think it's necessarily an outlier though as there are a few different polls showing similar results. There's something bizarre going on with the methods, imho.

[cybercoma]

I mentioned earlier my issues with Nanos's methodology. I'm pretty sure we're seeing those problems exaggerated here. It's safe to say when 3 pollsters have the Tories 7 points up, that's probably the more correct poll than the one guy who has them 4.5 points down. For the life of me I can't understand how Nanos can be 11.5 points off the others, except to say that his nightly samples are too small and his process of stitching the samples together is not working.

[angrypenguin]

I don't think it's necessarily an outlier though as there are a few different polls showing similar results. There's something bizarre going on with the methods, imho.

Agreed. As much as I will disagree with you politically I do very much appreciate your insight.

Was just at 24 Sussex. I thought I could visit a part of it, turns out I couldn't.

[angrypenguin]

I can see that being true for some individuals, but not in sufficient numbers to explain the difference in the poll results. I think something else is going on, or maybe a couple of things. Maybe Conservative voters are more willing to answer the phone and stay on the line for the automated directions, precisely because they want the world to know they're voting Conservative. Undecided and "lefties" don't answer or hang up on the automation, where they wouldn't hang up on a voice.

I want people to vote conservative so I'm biased but I disagree with you. N= large so I doubt this.

There has to be another reason. These polls baffle me

Edited October 6, 2015 by angrypenguin

[ScottM]

Trudeau has essentially ensured he gets another kick at the cat next time around barring an unforseen disaster.

I think Trudeau was nearly playing with house money this time. Even with a highly disappointing result he's all but guaranteed to significantly build on the seat total from 2011. I have a hard time conceiving of even a "disaster" scenario where he's booted from the leadership. He'd almost have to revealed as a child abuser who pushes little old ladies into the street and kicks puppies.

I mentioned earlier my issues with Nanos's methodology. I'm pretty sure we're seeing those problems exaggerated here. It's safe to say when 3 pollsters have the Tories 7 points up, that's probably the more correct poll than the one guy who has them 4.5 points down. For the life of me I can't understand how Nanos can be 11.5 points off the others, except to say that his nightly samples are too small and his process of stitching the samples together is not working.

Not to mention the erratic swings we've seen from Nanos. The only reason I see to take them ahead of the others is that their data is more fresh. Even then, Ipsos shows a closer race than Nanos, giving me pause anyway.

[ScottM]

I was mistaken. Must have been thinking about the timeframe (2000) and tied the incident to our politics, rather than where it actually occurred, in the United States. Gore won the popular vote, but loss the US presidential election to Bush due to the way the electoral college votes.

All good. I was just trying to make sure we were on the same page. I never realized those elections were in the same calendar month. That was before I started following Canadian politics.

[Vancouver King]

Seems to be a consensus building here that Nanos stands alone with his Liberal lead. This implies Conservatives at 35% is likely closer and, therefore, adding in a given 3-5% for more efficiently getting out it's vote, etc. and we have another 4 years of Stephen Harper.

[CITIZEN_2015]

CBC latest poll tracker taking the average of all polls is likely the best and it says:

Liberals 32.6% -- 118 seats

conservatives 32.3& --- 126 seats

NDP 25% -- 92 seats

http://www.cbc.ca/news2/interactives/poll-tracker/2015/index.html

Still 12 days to go and a lot can change in 12 days.

Btw, EKOS just released a poll showing a 4.2% conservative lead down from 6.7% last week and down from 9.1% lead from two weeks ago by same pollster. If the 2.5% fall per week continues it will be almost neck and neck on October 19th, a percentage or so Liberals ahead. EKOS poll put NDP and conservatives tied in Quebec at 27.8%!!!!!!,

And btw, Nanos is not the only pollster putting Liberals ahead. Also Innovative Research and Leger Marketing recent polls places Liberals ahead.

Edited October 6, 2015 by CITIZEN_2015

[ToadBrother]

Seems to be a consensus building here that Nanos stands alone with his Liberal lead. This implies Conservatives at 35% is likely closer and, therefore, adding in a given 3-5% for more efficiently getting out it's vote, etc. and we have another 4 years of Stephen Harper.

Only if they get a high minority. I don't think even 140 to 150 seats is safe.

[ToadBrother]

I mentioned earlier my issues with Nanos's methodology. I'm pretty sure we're seeing those problems exaggerated here. It's safe to say when 3 pollsters have the Tories 7 points up, that's probably the more correct poll than the one guy who has them 4.5 points down. For the life of me I can't understand how Nanos can be 11.5 points off the others, except to say that his nightly samples are too small and his process of stitching the samples together is not working.

Eric at 308 does appear to have some issues with at least the Mainstreet poll, so I'm not willing to believe that the Tories are in some sort of home stretch quite yet.

[angrypenguin]

Eric at 308 does appear to have some issues with at least the Mainstreet poll, so I'm not willing to believe that the Tories are in some sort of home stretch quite yet.

Cite please. Id like to read his input into this. Thanks

[poochy]

Seems to be a consensus building here that Nanos stands alone with his Liberal lead. This implies Conservatives at 35% is likely closer and, therefore, adding in a given 3-5% for more efficiently getting out it's vote, etc. and we have another 4 years of Stephen Harper.

Still early, it pretty hard to beleive the new conservative lead.

[CITIZEN_2015]

Btw, anybody remembers which pollster (Nanos, Ekos, Ipsos, Leger, Angus Reid, Innovative Research, Mainstream Research,,,,,,,) was closest to the actual election results in 2011?

Edited October 6, 2015 by CITIZEN_2015

[angrypenguin]

Btw, anybody remembers which pollster (Nanos, Ekos, Ipsos, Leger, Angus Reid, Innovative Research, Mainstream Research,,,,,,,) was closest to the actual election results in 2011?

This is googleable. The past doesn't indicate the future though. Everyone under reported the Conservative % IIRC.

[ScottM]

Btw, anybody remembers which pollster (Nanos, Ekos, Ipsos, Leger, Angus Reid, Innovative Research, Mainstream Research,,,,,,,) was closest to the actual election results in 2011?

Here you go: http://www.threehundredeight.com/2011/05/ranking-pollsters.html

As angrypenguin said, everyone underestimated the Conservatives, save Compas which grossly overestimated. They were pretty close on the Liberals and a hair high on the NDP.

Nothing there from Mainstream, and a few stopped polling pretty early for whatever reason.

[ScottM]

This is googleable. The past doesn't indicate the future though. Everyone under reported the Conservative % IIRC.

Not definitively, but a company's track record is something to consider. All of the models incorporate that.

[CITIZEN_2015]

Here you go: http://www.threehundredeight.com/2011/05/ranking-pollsters.html

As angrypenguin said, everyone underestimated the Conservatives, save Compas which grossly overestimated. They were pretty close on the Liberals and a hair high on the NDP.

Nothing there from Mainstream, and a few stopped polling pretty early for whatever reason.

Thank you.

It appears Angus Reid was the best with average 1% error and EKOS was the worst. Somehow the latter doesn't surprise me though I was hoping nanos comes first, though they too did well with 1.4% error. If we exclude EKOS then they all were within 2% of each other. Nothing like now which is a 10% difference.

Edited October 6, 2015 by CITIZEN_2015

[SpankyMcFarland]

Seems to be a consensus building here that Nanos stands alone with his Liberal lead. This implies Conservatives at 35% is likely closer and, therefore, adding in a given 3-5% for more efficiently getting out it's vote, etc. and we have another 4 years of Stephen Harper.

I think he has to get a majority of seats in Parliament to stay.

[Triple M]

I don't see Harper staying on past the summer of 2017 if he wins. July 1 2017 would be an interesting last day of office

[dialamah]

I don't see Harper staying on past the summer of 2017 if he wins. July 1 2017 would be an interesting last day of office

That long?

[capricorn]

That long?

Plus, how can he take a walk in the snow in July?

[Shady]

I think it's just more difficult to poll people. They're not as open to answering the phone for unrecognized numbers, and they're less likely to be candid with the questions they're asked.

[angrypenguin]

I am dumbfounded. Nanos polls barely move The Signal, and yet the other two polls released today made the Liberals go UP in the Signal? http://signal.thestar.com/

I am at a loss for words. I'm not statistician but these last few polls have me so confused.