Federal Election Polls

[cybercoma]

It's good news for the Liberals that they're falling in the polls?

I said it's good news that more past Liberal voters intend to continue voting Liberal than past voters of the other parties. Was it really that unclear?

[Smallc]

This poll, in comparison to past polls, is bad news for the Liberals, even if there are some glimmers in it. The Liberals were in a great position about 6 weeks ago. That's not so any longer.

[cybercoma]

This isn't much different from Forum's last poll. The Liberals are down 2 and the Conservatives are down 1. That's why they're closer by 1 and statistically tied. Really, every poll, including this one, still has the Liberals numerically. If you look at the trend, there's been movement, but no change in the relative positions of the parties for about a year now.

http://en.wikipedia.org/wiki/Opinion_polling_for_the_42nd_Canadian_federal_election#mediaviewer/File:Canada_polling_since_2011_election.png

[cybercoma]

The only party that has truly been gaining recently is Elizabeth May's party.

[Smallc]

The Liberals, a few months ago, in this very poll, were leading by 10 points. Now, they're in a statistical tie.

[Ash74]

The Ontario Liberal government is getting ready for another tax hike for us rich folks.(That is everybody that has not left the province) the Liberals will take a hit again just as it has been since they have taken over.

[Keepitsimple]

Everyone bashes the government between elections. Here in Ontario, the provincial Conservatives were miles ahead of the Liberals - and in spite of corruption that made Harper's Conservatives look like choir boys, the Liberals walked away with a majority. That said, it's getting more apparent with each passing month that the bloom is falling off the Trudeau rose......and if the Conservatives don't shoot themselves in the foot - the only way the Libs can reverse the trend is to have Trudeau actually say things of substance - and with his inability to speak off the cuff without a bunch of "ers" and "uhs", it'll be a tough road indeed.

[On Guard for Thee]

Everyone bashes the government between elections. Here in Ontario, the provincial Conservatives were miles ahead of the Liberals - and in spite of corruption that made Harper's Conservatives look like choir boys, the Liberals walked away with a majority. That said, it's getting more apparent with each passing month that the bloom is falling off the Trudeau rose......and if the Conservatives don't shoot themselves in the foot - the only way the Libs can reverse the trend is to have Trudeau actually say things of substance - and with his inability to speak off the cuff without a bunch of "ers" and "uhs", it'll be a tough road indeed.

I suspect JT has a little more strategy than you give him credit for, and it might just be like a page out of Chretien's book. As you say, let the Conservatives continue to shoot themselves in the foot, and then down the road a ways, "carpe diem" . I personally think JT does not too bad on his feet, and I'd rather listen to a few ers than Harpers totally scripted talking points about being "very clear" and then quite often saying absolutely nothing that is either clear, or meaningful.

[Thinkinoutsidethebox]

Polls are not very accurate and should not be taken seriously but we all know that already.

That being said I think polls should be shut down a month before election day because they do influence election results. https://www.press.umich.edu/pdf/0472099213-ch11.pdf https://www.dartmouth.edu/~chance/course/student_projects/polling.html http://www.crikey.com.au/2013/05/21/the-power-index-election-deciders-pollsters-at-10/

If you choose to vote you shouldn't be influenced by how a random number of other Canadians say they are going to vote.

[cybercoma]

LOL, your are 100% correct

Actually low response rates have no effect on the results as long as there is no common characteristics amongst those who did not respond, also known as non-respondent bias. Research has shown that there is very little bias in those who do not respond; therefore, those who do respond are representative of the target population (in this case voters). Moreover, weighting techniques give an even more representative picture.

Read more:

http://www.gallup.com/poll/7510/looking-closely-survey-response-rates.aspx

https://pprg.stanford.edu/wp-content/uploads/2007-TSMII-chapter-proof.pdf Top of p. 508

[eyeball]

If you choose to vote you shouldn't be influenced by how a random number of other Canadians say they are going to vote.

How is one to vote strategically without having an idea what other Canadians are planning on doing?

Maybe an app that voters could consult just prior to casting their vote would be even more effective a tool.

[Thinkinoutsidethebox]

How is one to vote strategically without having an idea what other Canadians are planning on doing?

Maybe an app that voters could consult just prior to casting their vote would be even more effective a tool.

What exactly is voting strategically? Is this a soccer game or your choice of who best to represent Canada?

I don't agree, these are just tools to lead the voters off on yet another wild goose chase.

May as well have a tally board strolling right in front of the voter's nose in the booth, that's an excellent time to make up your mind... Right?

[eyeball]

In Canada voting can often be a matter of voters wishing to avoid an undesirable outcome by voting against who they think would be the worst to represent Canada. Strategic voting is a natural consequence of FPTP and having up to date information would make it more effective.

In any case what could possibly be wrong with a voter having as much information at their disposal as possible? That's an ironic question given all the whining about low-information voters you hear these days.

Edited November 24, 2014 by eyeball

[Thinkinoutsidethebox]

Actually low response rates have no effect on the results as long as there is no common characteristics amongst those who did not respond, also known as non-respondent bias. Research has shown that there is very little bias in those who do not respond; therefore, those who do respond are representative of the target population (in this case voters). Moreover, weighting techniques give an even more representative picture.

Read more:

http://www.gallup.com/poll/7510/looking-closely-survey-response-rates.aspx

https://pprg.stanford.edu/wp-content/uploads/2007-TSMII-chapter-proof.pdf Top of p. 508

He is still correct

[Thinkinoutsidethebox]

In Canada voting can often be a matter of voters wishing to avoid an undesirable outcome by voting against who they think would be the worst to represent Canada. Strategic voting is a natural consequence of FPTP and having up to date information would make it more effective.

That's why you vote for the person you feel best to represent Canada, poling information has nothing to do with it.

In any case what could possibly be wrong with a voter having as much information at their disposal as possible? That's an ironic question given all the whining about low-information voters you hear these days.

Poling is not "information" it's an estimate of where the candidates stand in relation to each other. Information is what the candidates and their parties stand for and what they can do for you and Canada.

[On Guard for Thee]

How is one to vote strategically without having an idea what other Canadians are planning on doing?

Maybe an app that voters could consult just prior to casting their vote would be even more effective a tool.

Well that strategic voting can have it's drawbacks. Recall the last Quebec referendum? Apparently a lot of people did vote that way because, while they didn't want to split the country, they didn't want it to be an overwhelming loss for Quebec separatists. It damn near backfired.

[eyeball]

That's why you vote for the person you feel best to represent Canada, poling information has nothing to do with it.

I'm clearly talking about something else - voting against someone as opposed to for someone.

Poling is not "information" it's an estimate of where the candidates stand in relation to each other. Information is what the candidates and their parties stand for and what they can do for you and Canada.

I was talking about an app that allows voters to cast a vote strategically - the app would allow voters to inform other voters who they'ed voted for which is something a strategic voter wants to know before cast their vote. There's nothing that says a vote has to be a secret.

[eyeball]

Well that strategic voting can have it's drawbacks. Recall the last Quebec referendum? Apparently a lot of people did vote that way because, while they didn't want to split the country, they didn't want it to be an overwhelming loss for Quebec separatists. It damn near backfired.

Yeah well, that's just one more argument in favour of putting more timely information into the hands of voters.

[Thinkinoutsidethebox]

I'm clearly talking about something else - voting against someone as opposed to for someone.

Now I'm confused. Isn't voting for someone voting against everyone else?

I was talking about an app that allows voters to cast a vote strategically - the app would allow voters to inform other voters who they'ed voted for which is something a strategic voter wants to know before cast their vote. There's nothing that says a vote has to be a secret.

This sounds like sheep mentality, how does this put the best party in office? Your making voting sound like a game, do people place bets on this?

[On Guard for Thee]

Yeah well, that's just one more argument in favour of putting more timely information into the hands of voters.

I like the idea of getting rid of the FPTP process and allow us to vote in order of preference.

[eyeball]

Now I'm confused. Isn't voting for someone voting against everyone else?

When there's only two candidates sure, but even then I suppose you could indicate you voted against both by spoiling your ballot.

[jbg]

I said it's good news that more past Liberal voters intend to continue voting Liberal than past voters of the other parties. Was it really that unclear?

Isn't that ever the Canadian pattern, that people are "ancestral voters"? Of course that couldn't really apply to the CPC since the CPC as a merged party goes only back to 2003, and old-line Tory voters are often not CPC voters.

[jbg]

The Liberals, a few months ago, in this very poll, were leading by 10 points. Now, they're in a statistical tie.

Not a surprise. They coudln't do much worse for a leader, though the last two were no bell-ringers either.

[jbg]

I like the idea of getting rid of the FPTP process and allow us to vote in order of preference.

How many FPTP vs. rep-by-prop threads do we need?

[On Guard for Thee]

How many FPTP vs. rep-by-prop threads do we need?

I don't know. Let's take a vote.