Election 2015

[The Bread Guy]

I have not seen (but I haven't looked very hard) anything new from M Trudeau since he said (two months ago), "I'll end ISIS combat mission, restore relations with Iran."

I haven't seen any change/finessing of that - here's his latest refugee plank:

.... Trudeau said his party had already set a target of accepting 25,000 refugees from Syria and the region, but are “committed to doing more.”

“To doing more to help people in their camps who are worried about things, who are fleeing for their lives, who are living in terrible conditions.”

He stopped short of saying whether Canada should be more active in helping its anti-ISIS coalition partners in Europe deal with the border security issues brought on by this mass influx of hundreds of thousands of refugees, asylum seekers and migrants. He chose instead to criticize Harper and the Conservatives on international aid and cooperation ....

 

[George Wallace]

There is an article in the Ottawa Citizen headlined, "Canada has no military role to play in Syria, Iraq: Mulcair" which says that "Mulcair dismissed military action, specifically Canada’s current bombing campaign in Syria and Iraq, as a solution ..."

And with that, I have said:  Mulcair is not part of the solution, but part of the problem. 

 

[Edward Campbell]

And with that, I have said:  Mulcair is not part of the solution, but part of the problem.

Most Canadians do not put foreign policy anywhere near the top of their electoral priority list ... same as national defence and money to build symphony concert halls.

The stands taken by Messers Mulcair and Trudeau may be "part of the problem" for you, but I doubt they will have any measurable impact on the election outcome.

 

[s2184]

What I will do?

In regards to Election 2015, I will vote for Conservative if Liberal seems stronger by the time of the election. I will not vote for Conservative if it seems stronger at the election time. In that case I have to still decide to whom I am going to vote.  :

I agree with you that it cannot be all the government, but private sectors, and other entities can play significant roles to deliver best output for the nation.

Still undecided..  :facepalm:

May be I should vote for the known devil instead.

Honestly I lost interest in voting.  :-X

 

[George Wallace]

Most Canadians do not put foreign policy anywhere near the top of their electoral priority list ... same as national defence and money to build symphony concert halls.

The stands taken by Messers Mulcair and Trudeau may be "part of the problem" for you, but I doubt they will have any measurable impact on the election outcome.

Agreed.  I have little faith in the Canadian voter making truly informed decisions.  Some here are very enlightened and informed, but I feel that it will boil down to last minute 'gut feelings' when the majority of Canadians mark their "X" on their ballot.  Only the 'diehards' will vote along their Party lines.

 

[Edward Campbell]

Agreed.  I have little faith in the Canadian voter making truly informed decisions.  Some here are very enlightened and informed, but I feel that it will boil down to last minute 'gut feelings' when the majority of Canadians mark their "X" on their ballot.  Only the 'diehards' will vote along their Party lines.

Well, George, you and I will just have to agree to disagree. Even when, in the 1970s and 1990s, I disagreed, vehemently, with the policies and programmes of the governments the Canadian voters chose, I respected the choices themselves. I didn't think much of the quality of the people elected to lead our country but my faith in the basic good sense of the electors remained strong. I understood, in the 1970s, that Prime Minister Diefenbaker was a tired old man, more than just a spent force, and that Mr Stanfield, estimable man though he was, as a person, did not have the vision that Canadians wanted. Equally, in the 1990s, I knew than Ms Campbell and M Charest were not the right people to lead Canada. I thoroughly detested Prime Minister Trudeau, as a man and as a politician, but understood that Canadians elected him because he, more than anyone else, spoke to and about their hopes and fears and dreams. I was not alone is disliking Prime Minister Chrétien, but, for many Canadians, as British Prime Minister Margaret Thatcher so often said, "there was no alternative," he was the least unacceptable choice for a plurality of the people.

I contin ue to believe what I said more than two weeks ago:

The numbers are fun, the strategies and tactics are fascinating, but: there is some good news and some bad news.

The good news is that we have a pretty robust, open, generally fair democracy. We all, well most of who are qualified, get to vote ... if we choose. And the party that the greatest number of us select is very likely to form a government. It is very likely that 60%+ of us are going to wish, on 20 Oct 15, that someone else was forming the government, but c'est la vie, as they say. There's more good news: none of Prime Minister Harper, Opposition Leader Mulcair or M Trudeau are bad men; none are going to turn into a right wing despot or a left wing nut. Most of the couple of thousand Canadians who will stand for office are similar to the leaders, not bad, not really good, perhaps, but, by and large, acceptable.

Now the bad news: none of Prime Minister Harper, Opposition Leader Mulcair or M Trudeau are particularly good leaders for a country that must tack its way into strong strategic (international), domestic/social and economic headwinds. It's not going to be smooth sailing and I doubt that any of the three current party leaders is really who any of us wants.

I will not welcome an NDP government, if that's who we all choose. But I doubt it will do real, serious, long term harm to our country, despite my mistrust of the economic and fiscal motivations of M Mulcair's back-bench and base.

I am not really afraid of a government led by M Trudeau, if that's the party we select. Despite my reservations about his 'bottom', he can put together a pretty solid front bench. I doubt his government will do much real damage either.

I rather hope Prime Minister Harper's Conservatives are re-elected, despite my real, serious reservations about Prime Minister Harper as a leader. I believe the CPC has a good team and I hope that Prime Minister Harper, who I believe has been [size=12pt]tarnished, perhaps irreparably damaged by his own words and deeds, will resign and make way for a new, better leader. But the CPC may have run its course, for now; it may need a rest on the opposition benches benches while it reconsiders its aims and objectives for Canada. That is the normal and natural fate of all political parties. It is why our ramshackle, messy democracy is, always, better than even the best managed one-party state.

My  :2c:  because I think too many of us take all this too seriously ... Remember what Gloria Gaynor said: those political buggers will be back and we will survive, no matter which of 'em leads our country.[/size]

 

[Rocky Mountains]

Two more polls came out Saturday,  all of which had the conservatives in third, grits in second and dippers in first.

Forum

NDP 36

LPC 32

CPC 24

Léger

NDP 31

LPC 30

CPC 28

A couple of more long times to go, but same as in a hockey game, you can only watch and comment on the period you're in.

Léger looks more accurate though.

Actually, in the Forum/Red Star poll the Conservatives improved from last time and the Leger/Globe they only went down .8%.  Don't read anything into either of these polls.

 

[The Bread Guy]

Actually, in the Forum/Red Star poll the Conservatives improved from last time and the Leger/Globe they only went down .8%.  Don't read anything into either of these polls.

<pedant> Also remember the accuracy/margin of error of the poll results, too.  If the difference is at or less than the margin of error, it might as well be considered a tie.

For example, for the Forum polling, "Results based on the total sample are considered accurate +/- 3%, 19 times out of 20."  That means the difference between the NDP and Liberals puts them pretty well neck and neck - although the Tories are behind more than the margin of error in that snapshot.

For the Leger poll, "a random sample of 2,119 respondents would yield a margin of error of +/- 2.1%, 19 times out of 20," meaning there's not enough of a difference between the results to be anything more than "all three are tied".

Thus endeth Poll Reading 101  ;D</pedant>

 

[Oldgateboatdriver]

Pedant, Milnews, but also incorrect (which makes it worse  ;D).

The +/- error is not something you can apply as you wish to increase someone's number or decrease someone else (i.e. you can't say that the poll shows the NDP  just as likely at 33% [36-3] than the Libs at 35% (32+3).

The plus minus error figures are non-distributive mathematical model errors: this means that (if you pair it with the 19 times out of 20 factor, known in maths as the 95 percentile confidence factor) if you polled the same number of Canadians, at random, 20 different times, on that same day, you could find different results for each party that would fall within the figure released + or - 3%, and the last survey would bear no such resemblance with such distribution at all. However, the further you are in the outliers (i.e closer to the 3% difference, the least repeated the error - so one out of 20 polls may show the NDP at 3% lower or higher, but 10 out of twenty would have them between + /- 1%, and 15 would have them within + /- 2%. The value of the error itself being a statistical distribution.

Moreover, as I indicated, these error figures are "non-distributive", but the effect on the polling is a distributed effect (i.e. the distribution must always total 100% - Thus if a party goes up, somewhere else figures must come down. The statistical likelihood, however, that the NDP be down 3% while only the liberals be up the same 3% is probably in the order of one in a thousand chance (sorry, don't have my tables with me and don't remember the formula by hearth anymore).

/even more pedant off.

:irony:   

 

[The Bread Guy]

Pedant, Milnews, but also incorrect (which makes it worse  ;D).

The +/- error is not something you can apply as you wish to increase someone's number or decrease someone else (i.e. you can't say that the poll shows the NDP  just as likely at 33% [36-3] than the Libs at 35% (32+3).

The plus minus error figures are non-distributive mathematical model errors: this means that (if you pair it with the 19 times out of 20 factor, known in maths as the 95 percentile confidence factor) if you polled the same number of Canadians, at random, 20 different times, on that same day, you could find different results for each party that would fall within the figure released + or - 3%, and the last survey would bear no such resemblance with such distribution at all. However, the further you are in the outliers (i.e closer to the 3% difference, the least repeated the error - so one out of 20 polls may show the NDP at 3% lower or higher, but 10 out of twenty would have them between + /- 1%, and 15 would have them within + /- 2%. The value of the error itself being a statistical distribution.

Moreover, as I indicated, these error figures are "non-distributive", but the effect on the polling is a distributed effect (i.e. the distribution must always total 100% - Thus if a party goes up, somewhere else figures must come down. The statistical likelihood, however, that the NDP be down 3% while only the liberals be up the same 3% is probably in the order of one in a thousand chance (sorry, don't have my tables with me and don't remember the formula by hearth anymore).

/even more pedant off.

:irony: 

Holy crap - I'll have to re-check my ooooooooooooold notes, based on this (NOT Wikipedia  ;D):

.... Knowing that a survey found that 60% of people support Candidate X doesn’t tell us much without knowing the margin of error. If the margin of error is +/- 15%, we would expect the true population support for Candidate X to be between 45% and 75%. Though at first it would appear that the majority of people support Candidate X, the large margin of error casts doubt on this conclusion. If, however, we know that 60% of people support Candidate X with a margin of error or +/- 3% we would expect the true population support for Candidate X to be between 57% and 63%. We would have much more confidence in the conclusion that the majority of people support Candidate X ....

Thanks for the new info.

 

[Oldgateboatdriver]

Your quote from the U of Texas is not incorrect as a statement. If candidate X has 60% support with a 15% error, his true support will fall between 45% and 75% (except that 20th time out of twenty when it will fall outside that range). However, since the error is itself a statistical distribution, it is more likely that his true figure is found between say 55% and 65% than between either 45-50% or 70-75%.

That second degree (and I won't even get into the third degree of statistical levels - which exists and for some things such as dynamics, must be taken into account) of statistical analysis is usually ignored for more mundane pursuits (such as political polling) but must be taken into consideration for more advanced modelling such as cosmology, quantum mechanics and macro-econometric model of open economies.

It is sufficient to know that the error figure is an indication of how much trust you can have in the values derived by the polling. But it does not mean that you can say that a poll of candidates showing four or five points difference between candidates with a 3% error margin indicates that they are basically neck-to-neck.

 

[The Bread Guy]

How about we agree that you have to look closely at poll results, so the rest of the non-stats-geeks (unlike both of us) can get back to political wrestling?  ;D

 

[jollyjacktar]

How about we agree that you have to look closely at poll results, so the rest of the non-stats-geeks (unlike both of us) can get back to political wrestling?  ;D

Thanks, my head was starting to smoke...

 

[Kirkhill]

Thanks, my head was starting to smoke...

JJT - I prefer just to leave them all alone these days.  Just like I don't spend any time at the track and I let other folks worry about the ups and downs of the stock market.

One horse race is much like another..... ;D

 

[a_majoor]

Given the increasing unreliability of polls over the last several election cycles (Federally, Provincially and Municipally, as well as in many foreign elections), I tend to doubt anything being posted by any individual polling company, and have even more doubt about polls commissioned by sources such as the Toronto Star.

Of course weighted averages like the ones published by 308 are problematic as well, if the input is unreliable to begin with then the end result will also be unreliable (GI-GO). I suspect lots of Canadians (like many Americans) only tell the pollsters what they think the pollsters want to hear, rather than what they intend to do in the privacy of the ballot box.

 

[dapaterson]

From the "Need to do a better job vetting our candidates" department, may I present Jerry "Pissing in your coffee cup" Bance.

http://www.cbc.ca/m/news/politics/topstories/jerry-bance-marketplace-1.3217797

 

[ModlrMike]

From the "Need to do a better job vetting our candidates" department, may I present Jerry "Pissing in your coffee cup" Bance.

http://www.cbc.ca/m/news/politics/topstories/jerry-bance-marketplace-1.3217797

The PM needs to dump him quickly. That being said, if the CBC had the footage in 2012, why did they wait three years to release it?

 

[dapaterson]

CBC did release it in 2012 (http://www.cbc.ca/marketplace/episodes/2012-episodes/when-the-repairman-knocks).

 

[Altair]

Given the increasing unreliability of polls over the last several election cycles (Federally, Provincially and Municipally, as well as in many foreign elections), I tend to doubt anything being posted by any individual polling company, and have even more doubt about polls commissioned by sources such as the Toronto Star.

Of course weighted averages like the ones published by 308 are problematic as well, if the input is unreliable to begin with then the end result will also be unreliable (GI-GO). I suspect lots of Canadians (like many Americans) only tell the pollsters what they think the pollsters want to hear, rather than what they intend to do in the privacy of the ballot box.

I thought 308 was pretty close to the actual results in the last election using the weighted polls?

 

[Edward Campbell]

The science (mostly mathematics) behind polling, market research, is solid and reliable. Companies and organizations as diverse as Apple and the Zillow Group (a multi-billion dollar US real estate/housing company) use it, almost daily, to figure out what you want and need. Ditto e.g. the Government of Canada and Greenpeace or the Liberal Party and Loblaws. Figuring out what you want (and need) by way of a career or wearable technology or pizza is a lot less complicated than figuring out what you want in areas like vision and leadership or what will inspire fear or hope in your heart.

One problem with polling is that it alters its own results. If there are enough polls indicating that e.g. the NDP are trending "up," then it is very likely that many Canadians will decide that they, too, "like" the NDP ... until something else, new and shiny, comes along.

Good market research is time consuming, painstaking and expensive. The results are, very often, not what anyone expects or ~ and this really matters in politics ~ wants. A good polling firm using innovative techniques might well be fired if its (very accurate) results run counter to what the party or campaign leadership wants (needs?) to hear.