Polling thread infinite edition: Volume 7

102 posts / 0 new
Last post
Stockholm

Its interesting that Layton has a clear lead as "best PM" in Quebec and the NDP is the only party that has gained significant ground in Quebec since the last election - and yet all the media coverage (as usual) tries to pretend that the NDP doesn't exist. Its as if the attitude is "damn that NDP getting high numbers in Quebec - it makes it so much  more difficult to write a story and ignore them".

nicky

 

In addition to the figures provided by Ottawaobservor, the LaPresse article also shows that the Liberals are at 24% and the Cons at 21% in Quebec. Given that the Bloq has 34% and the NDP 17% that would leave the others, including the Greens, at only 4%. (Appropos of other posts does anyone know if CROP prompts with party names?)

Assuming a constant 4% for other parties accross the regions that would put the NDP at 19% in Montreal, 17% in the Quebec City region and 16% in the rest of the provence. I expect the "others" might be a little higher than 4% in Montreal.

The NDP continues to be handicapped by a relatively even distribution of its vote accross the province, although Montreal is looking stronger. Unfortunately the Cons continue to dominate in the Quebec City region where they can expect to hold most of their seats with these poll results. I can forsee the situaion that the Cons can get 8-11 seats with the same percentage that would give the NDP only 0-2 seats.

Stockholm

CROP prompts party names along with leaders (i.e. The Liberals led by Michael Igantieff, the NDP lead by Jack Layton etc...). I think its no surprise that given Layton's great personal popularity in Quebec, the moment you mention his name along with the NDP - support tends to be a lot higher. Needless to say "le parti vert d'Elizabeth May" does not fare so well in this context.

Lou Arab Lou Arab's picture

Is this the first time Layton has led in the 'best PM' category in Quebec?

Unionist

Lou Arab wrote:

Is this the first time Layton has led in the 'best PM' category in Quebec?

I haven't got the sources in front of me, but I'm quite sure this has happened before. It's partly by default (Quebeckers hate Harper, they hated Dion, they're figuring out Iggy, and Gilles isn't in the running for PM), but partly because of Layton's sincerity, his social democracy, his ability to speak French - all good things here.

 

Stockholm

I also think Layton has been ahead before but only by maybe 1% - the 4% gap is the biggest I can remember

ottawaobserver

There's a new Harris-Decima poll out today as well ...

N=1,000, Jan 21-24, 2010

Cons-32%
Lib-31%
NDP-15%
Bloc-10%
Grn-10%

Allan Gregg interprets it as anger over prorogation. The undecided rate was not reported.

Debater

Hmm.  Interesting.  Another poll that shows a virtual tie between the Cons and the Libs.

It will be interesting to see if that lasts throughout the prorogation.

 

Sean in Ottawa

ottawaobserver wrote:
There's a new Harris-Decima poll out today as well ... N=1,000, Jan 21-24, 2010 Cons-32% Lib-31% NDP-15% Bloc-10% Grn-10% Allan Gregg interprets it as anger over prorogation. The undecided rate was not reported.

Now here is another point-- at times when the atmosphere is charged people are more likely to speak to a pollster when three weeks earlier they might have hung up. the demographic of potential respondents willing to talk changes. So do results.

Sean in Ottawa

Oh to get in to the issue of falsifying surveys-- this is huge. Interviewers can get paid more for completed surveys-- bonuses etc. As well they don't have to talk to people-- when the call connects they typically start to falsify-- first they hang up on the actual respondent that they would need to manage and then go through completing the survey-- this is unbelievably common and the one who falsifies can fill in a lot more surveys than the ones waiting for real respondents.

A good company will monitor heavily and avoid false surveys and delete all those of an interviewer who falsified. Others will monitor less and not delete bad surveys. This alone can make a huge difference. Those who falsify often random their picks in order not to show a pattern. This randomization will have all the parties more close in support. So survey one they falsify a Liberal, 2 a Conservative 3 a Green 4 an NDP etc. Have one interviewer do that in a shift and not get caught and your Green total could go from 4% to say 12% because for that interviewer those 4 parties will run 25% each.

The proper monitoring of falsifications alone can explain the difference. When I monitored calls I caught someone doing this on every shift. I got good at the signs and would tap in. Sorry to say but I fired more interviewers for falsification than anyone else when I was there and I was sent in to monitor when the results were of concern.

ottawaobserver

Thank you for the comprehensive answer, Sean.  I like what you do too.

ottawaobserver

Meanwhile, while Sean and I were having our kumbaya moment, Brian Topp managed to scare up more complete results from that CROP poll in his blogpost for the Globe online.

To recap (or re-CAPP, since that's what we should all be thinking about doing ;-)):

Best prime minister: Jack Layton 28 per cent; Stephen Harper 24; Michael Ignatieff 20.

Voting intensions (all of Quebec): Bloc 34 per cent; Liberals 24; Conservatives 21; NDP 17; Greens 4.

Voting intensions (Francophone): Bloc 40 per cent; Conservatives 20; Liberals 19; NDP 18.

I'll let him analyze it for us:

Brian Topp wrote:

It will be observed that the NDP is within one point of the Liberals among the critical Francophone vote, and only two points away from the Conservatives (all within the margin of error).

The bottom line: Prime Minister Stephen Harper and his government are not loved in Quebec. Or at least, Elvis is loved more. Michael Ignatieff trails all of his national opponents. At 34 per cent, the Bloc is underperforming. And the party with the most growth potential (looking at the leader and party numbers) is the Quebec slate headed by deputy NDP leader Tom Mulcair. Jack Layton has been working Quebec relentlessly since his first days as leader. It's paying off. He is also lucky in his federalist opponents, it would seem.

Pogo Pogo's picture

Again your theory works to explain the Green's missing support, but to be credible it needs to be measured against parties in similiar situations.  When you look at historical or provincial examples it doesn't hold water.

Sean in Ottawa

Pogo-- do you mean me? My theory? parties in similar situations? historical and provincial examples?

I can't argue with you because you have not actually said anything yet other than that you disagree. (I assume with me but you did not even dignify the theory you are attacking with identification). So I agree that you disagree based on something you couldn't be bothered explaining. Care to join the conversation with something real?

What is the point for people to take the time to work out and explain something-- so you can just come in and say nothing more than what you say does not hold water. After saying that it is pretty rude not to follow up. Want to put in a little effort?

Sean in Ottawa

Oh and my point is that uncommitted Green supporters do not explain the wild variations in exaggerations of Green voters. I did not know that I definitively said what did although I advanced several possibilities that at least logically hold more water than the idea that someone is committed enough to be in the 1% that talks to pollsters but can't get out and be in the 55-60% that actually vote. Rather I advanced several explanations to explain that the polls are not as accurate as people think. I also explained I thought it was likely a combination -- but if you have worked out some specific theory I have inadvertently said other than the above please share it with me. My preferred explanation is that pollsters are way less accurate than they would have us believe and that it is a combination of things that add up in one direction or another. Indeed there are so many variables that usually they can screw up wildly and cancel each other out but in the case of the Green party we are getting the perfect storm where all the errors are pointing in the same direction.

That said are you trying to say that the theory of Greens being extraordinarily committed when the pollster calls but uncommitted on election day  is somehow PROVEN by the fact that you don't think another theory or combination (for whatever reason) works for you? Logic does not work that way. One theory is not proven by disproving another unless there are only two possible explanations. My entire point was to refute the theory about the evaporating armies of Green supporters that I am questioning if they actually existed in the first place.

Please take the time to explain.

wage zombie

.

Sean in Ottawa

Sorry left the debate for a couple days as I was sick...

As for frustrations-- I think there is a lot at stake in the answers not just between parties in terms of strategies but internally for the NDP and the issues are complicated-- I'll get to that later. The point is we should all be forgiven for some frustration. What is leaving me feeling better is real interest at exploring these issues-- all of them-- rather than just the tired old Greens are less committed tripe we have been hearing.

Anyway -- yes Ottawa Observer I am referring to the gap between actual Green support and measured support. I think this is the more important gap since even the lowest of the pollsters I believe over-measures Green support frequently. And In spite of the difficulty in this thread I'd never give up on you because you contribute so much that I get a lot out of here.

The gaps between pollsters are relevant although less so. But here there are three gaps there as well: the gap between individual pollsters, the gap between those who probe and those who don't and the general inconsistency within the same pollsters findings unexplained by any change in actual  support. Of these three, I find the first and the third more interesting. The middle which is the most popular I find is almost baseless-- if it makes a difference it does not make much of a difference. Let's look at each.

The gaps among individual pollsters who do probe and historically among those who don't probe is significant-- and in fact wider than a simple comparison between those who probe and those who do not. I find the interpretation is also very partisan and there are things to be learned. For this reason I think we have to look at other explanations. OO in your post #40 you are on the right track but there are other differences in questions and prompts that you may want to consider. I have heard a few variants. The wording of the question is a huge variable -- with or without prompting. In some cases they ask you to think about a local candidate as you said. Others ask you which national party you would support - period. When they probe some do by party name but others say the Liberals led by Michael Ignatief, the NDP by Jack Layton etc. In fact that is the most common. Probing with the leaders' name will have a different result than probing with a party name only especially as some leaders are better known or have more or less negatives etc. Most of the pollsters, I think now probe with the leader's name. This is significant as it lengthens the question. As well the probing method is different-- some will not accept an answer till all have been probed equally, while others will take a response before the list is finished. Most will rotate the order of the probes others will not. On the probing issue I think the method of probing is far, far more significant than probing or not.

Then there is the sample management: some will take whomever is answering the phone. Demographically this makes a difference. Others will randomize by probing for whomever had the last birthday or soemthign like that. This is significant. Pollsters can get sloppy if you are not careful. There are different demographics to fill males are harder to fill than females younger harder than older. Poor sample management and prescreenign can lead to getting older Women first which are easy and then filling the hard to get groups getting younger men. Younger women and older men can be very underepresented. Same thing with regions. Urban is a hard to fill demographic so frequently people fill up suburban and rural with older females and then turn to young males for the urban. Of course urban younger females will be underepresentated as will be rural older males. Do you get my drift? In the end they will ahve the right number of males, the right number of females, the right number of young and the right number of older people but the sub demographics can be off a great deal. And that is the good pollsters. The poor ones may blow open all demos and not even get statistically viable samples in all categories just taking anyone to fill the quota. The degree of discipline used to keep demos filled separates the cheap from the more pricy pollsters.

Then there is the quality of the interviewers. Some interviewers read poorly, handle calls poorly or have a heavy accent. This is a variable. Some types more inclined to accept people who are different or struggling may be more likely to put up with or indulge or be more patient with such interviewers. Some interviewers intentionally bias or damage sample either to affect the result or to get more "completes". Some companies are more tolerant of this.

Also the timing of the poll makes a difference. If a pollster is busy they put their weak interviewers on the political polls (these are the easiest and shortest to do) if the pollster is busy they pull off the better interviewers and put them on their commercial polls and trainees go on the political polls. When the pollster is not busy they send the weak interviewers home and only the strong ones are there and they do the political polls.  This is partly what might explain wide variations for the same pollster using the same method-- the demographic of interviwers may have changed. Some pollsters will outsource the calls overseas. All these can make a difference to the results some people are more willing to talk to a Canadian or soemone who is more like them. People are more likely to talk to people who sound like them but less likely to those who don't - so a Chinese immigrant may be happy to speak to a Chiense interviewer but be unable to understand an interviewer whose first language is Arabic.

I find that the gap between pollsters who probe for Green support and those who don't is dwarfed by the above differences. As well many here have said that such and such a pollster probes and such and such a pollster does not. Often they are wrong. In fact some pollsters poll for different clients and sometimes are directed to probe and sometimes not. As I said the vast majority probe in one way or another. They do it for one reason: economics: if you probe you will get more completes. Question is how. (There is also the rare one that probes without party names-- "is there one you are leaning towards?") In almost all cases they will probe. As I said probing by adding the party leaders costs more and is another option to probing just party names.

Some pollsters in the intro say nothing about the poll and only say it is an opinion poll on politics to those who agree to participate. this is to reduce self selection by political junkies but it costs more because you do lose potential respondents who might have been interested. Those who let interviewers lead with the topic of the poll will fill up demos more easily and cheaply but the results will be skewed. Some pollsters only make more experienced interviewers remain that neutral but let juniors prescreen for those interested in politics...

Then as OO points out if you emphasize in your area-- people may want to vote for a viable local candidate but if you emphasize the national may opt either for a more viable national campaign (that might be different). As well the party leader is a variable-- some people vote for parties others vote based on the leader they like the best.

Some pollsters if you say none of the above will record that as refusal which is not counted, some as undecided which is counted and others will say but if you had to choose and then make them make a choice in order to fill the complete-- there I think the Greens might fare better and the reason is simple-- someone angry at all the others forced to making a decision is likely to pick the weakest because they are the least harmful and they really hater their other options. Those respondents really should have been allowed to say none of them but the pollster would then have to dial a hundred more numbers to find soemone else to talk to so they force a person. Some pollsters terminate calls when they know the person is drunk or not serious or clearly throwing answers that are meaningless others record them.

Now imagine all the variables I mentionned running off in different sometimes opposing directions and what you ahve is a mess-- one that cannot be relied on. Half the time they will even out but sometimes, as I argue in the case of Green support they can lead generally in the same direction and cause oversampling of one party.

Now let's look at the implications for a moment: if the viability of the local candidate is the issue then a party if it invests a ton of money in a no-hope riding they might actually change the impression and it become a possible win. But if the viability is not an issue then that investment would be a waste.

As for the specific differences between Ekos and Nanos to speculate on the above to their particular cases could lead to legal issues here so I won't run down a particular pollster but I am trying to say there are many other reasons than probing for a party. I can't account for the specific differences between these two pollsters as I have not worked for either of them.

Long but some people wanted more detail so here it is... Hope it helps.

And please no hard feelings OO. I like what you do!

Sean in Ottawa

Sorry in my long post I scrambled one of the sentences and delted the part that made it make sense-- here it is fixed (I edited it above as well)

 

Also the timing of the poll makes a difference. If a pollster is busy they put their weak interviewers on the political polls (these are the easiest and shortest to do) if the pollster is busy they pull off the better interviewers and put them on their commercial polls and trainees go on the political polls. When the pollster is not busy they send the weak interviewers home and only the strong ones are there and they do the political polls.

Stockholm

I was under the impression that most polls that we are seeing now are mostly omnibus surveys where a question on vote intention is tacked onto a survey that has questions on many other topics.

Debater

ottawaobserver wrote:

Meanwhile, while Sean and I were having our kumbaya moment, Brian Topp managed to scare up more complete results from that CROP poll in his blogpost for the Globe online.

To recap (or re-CAPP, since that's what we should all be thinking about doing ;-)):

Best prime minister: Jack Layton 28 per cent; Stephen Harper 24; Michael Ignatieff 20.

Voting intensions (all of Quebec): Bloc 34 per cent; Liberals 24; Conservatives 21; NDP 17; Greens 4.

Voting intensions (Francophone): Bloc 40 per cent; Conservatives 20; Liberals 19; NDP 18.

I'll let him analyze it for us:

Brian Topp wrote:

It will be observed that the NDP is within one point of the Liberals among the critical Francophone vote, and only two points away from the Conservatives (all within the margin of error).

The bottom line: Prime Minister Stephen Harper and his government are not loved in Quebec. Or at least, Elvis is loved more. Michael Ignatieff trails all of his national opponents. At 34 per cent, the Bloc is underperforming. And the party with the most growth potential (looking at the leader and party numbers) is the Quebec slate headed by deputy NDP leader Tom Mulcair. Jack Layton has been working Quebec relentlessly since his first days as leader. It's paying off. He is also lucky in his federalist opponents, it would seem.

It is important to remember that this is one poll on Quebec.  There have been several other polls lately which have had Quebec numbers and they have not had the NDP as high.

I realize it is always nice to zero in on the poll that is the most favourable while ignoring others, but it's important to look at all of them in context.  Wink

NorthReport

Actually Topp is correct to focus on the favourable CROP polling results for the NPD in Quebec, as it is fairly common knowledge that CROP, and perhaps Leger, are the only two polling firms with an accurate assessment of how the political winds are blowing in Quebec.   

ottawaobserver

Plus, they have a much larger sample size, contact the voters in their own language, and (as we learned up-thread) prompt with the leaders' names as well as the party name.

What context did you want to look at them in, Debater?

Pogo Pogo's picture

Sean in Ottawa wrote:

What is the point for people to take the time to work out and explain something-- so you can just come in and say nothing more than what you say does not hold water. After saying that it is pretty rude not to follow up. Want to put in a little effort?

Sorry for not going into great detail as I had already written in detail for post 45 and with no response.  I think my short post was pretty self explanatory, if you have a general theory prove it against more than one example.:

 

"The self-selection theory as I understand it is that because polls only count those willing to be polled and that highly political people are extremely overrepresented in this group core support is over represented in the polling data. 

This theory does provide a justification for Green polls/votes discrepancies.  However, I am not sure it works in other instances.  The NDP polling vs voting data doesn't show a wide shift.  In particular back in the 1990's when the NDP support was at its lowest levels one would anticipate numbers similiar to the Greens but as memory has it the polls/votes were not widely different.

My pet theory is based on the difference between the general and the particular  People want to vote for the Greens, but on election day they cannot bring them to voting for a last placed candidate that they know is not going to win.  Conversely with the NDP back in the 1990's they overcame lacklustre support for the NDP in general to vote for particular candidates who were in tight races."

wage zombie

I don't know if we can compare polling now to polling 15 years ago.  I suspect more people then were willing to answer polls.  Maybe not though.

What are the provincial examples?

Pogo Pogo's picture

Consider the BC provincial Conservative Party.  Pathetic at the ballot box pathetic in the polls.  By Sean's theory they should have higher support in the polls.  I think examples can be found across Canada.

The important thing about the theory is that it is a general theory of polling not a Green Party theory of polling.  It is critical then to test the theory against multiple examples.  Otherwise it is a just a consistent story that may or may not be true.

Debater

ottawaobserver wrote:

Plus, they have a much larger sample size, contact the voters in their own language, and (as we learned up-thread) prompt with the leaders' names as well as the party name.

What context did you want to look at them in, Debater?

In the context of also looking at the Ekos, Harris-Decima and Ipsos-Reid polls.

Sean in Ottawa

Pogo-- I gave multiple explanations for why small differences could exist. I also said in many cases they could cancel each other out unless they all went in one direction. All the examples of polling problems I have seen and heard myself so I know they happen. I did not make them up.

As for the BC Conservatives well maybe they were supported by older men but not older women and not younger men-- in my long post I illustrated how a poll could show that. Or possibly the BC poll was conducted very well so that there were not several different things going at the same time which is what I said it would take to severely bias the results. Indeed in the case of a provincial party there are almost never enough polls to make such assessments. I don't have enough specifics to argue a single provincial example.

Now let's look at your theory-- the single explanation theory and see how it stacks up using your logic-- against your example. You say people won't vote for the last party candidate but they will tell a pollster they will. Why doesn't the BC example fit that one? In fact ironically your example seems to support my theory that it takes several problems going in one direction to create the effect rather than the single uncommitted poll respondent theory due to the party being in last place.

I am advancing an explanation that is a perfect storm explanation where there are several different effects happening simultaniously that bias the results. Normally several of these will cancel each other out but in the Green case they act together. In this case I would not expect to see any one of my explanations in isolation bear out.

But in any case, Pogo, you should have come with more evidence than an assertion-- show us the polling data and final result for the BC cases otherwise you are not putting enough effort to even be in the debate at all. Funny you come to the table without any facts but talk about other people's arguments not holding water. At least mine were specific things I have seen. Can you at least look up the results that you so blithely assert? Can you not see why after my having spent the time at this your posts are annoying?

Sean in Ottawa

As far as the Quebec poll the NDP may or may not be quite that high but thi is not the first time we have seen this recently and there certainly appears to be progress for the NDP there. I do not take a single poll as that meaningful this is is more of confirmation of a trend.

The bad news for the NDP is that the party is more evenly spread out in the province than the other parties and lacks the strongholds. It would take one of two things for the NDP to take several seats: it would either have to pull substantially ahead of the other parties or develop some stronghold areas. Unfortunately the party seems to be running strong but behind everywhere. You can be the first place party overall but lose the race regionally in all cases and the NDP risks that. Unfortunately the party in Quebec is looking more like another example in favour of proportional representation.

The party will need a two pillar strategy in the province-- one to spread enough juice to deliver the higher popular vote so it does not look like a flash in the polls and second seriously target some seats to create enough close calls that it can develop a regional base for next time delivery of several seats. Get it wrong and the campaign will be a waste. This is a huge challenge as there may not be enough resources both to spread for an impressive popular vote and to concentrate to create chances at seats. Essentially that is the next nut to crack but it is sure better than having no interest at all.

Back to this poll-- the actual NDP numbers are less important than their distribution if you want seats and the distribution is the problem. The NDP needs to sell some star candidates beyond Mulcair -- but of course it needs to attract them with them and the party aware they will not win this time as their job is to try to create a more mountainous electoral landscape out of the flatlands we now have.

That said with this level of support the party is capable of the odd surprise but this is not enough support to guarantee company for Mulcair unless there is some concentration somewhere.

 

Stockholm

The latest ARG poll says Tories 33%, Libs 29% NDP 19%

ottawaobserver

See now Debater, *that's* the context I would want to add.  The Angus Reid survey was taken Jan 25-26, meaning only after the rallies, while EKOS was in the field from Jan 20-26, by the way (the CBC story, with links to the data tables, is here).

The full Angus Reid numbers are (N=1,005, MoE-3.1, Jan 25-26, taken from the Toronto Star story):

61% disagree with Harper's decision to suspend Parliament.

Cons-33%
Lib-29%
NDP-19%
BQ-10%
Grn-7%

The reported sub-samples are:

Cons - still on top in the West

Ontario: Cons-36%, Lib-35%, NDP-19%

Quebec: Bloc-42%, Lib-28%, NDP-14%, Cons-11%

ottawaobserver

Sean in Ottawa wrote:

That said with this level of support the party is capable of the odd surprise but this is not enough support to guarantee company for Mulcair unless there is some concentration somewhere. 

The best chance for another seat, after Gatineau of course, would be if incumbents step down in Montreal.  The next most likely to do so would be Francine Lalonde in La Pointe de l'Ile, and maybe Nicole Demers in Laval, for health reasons, and a few others who have been around a long time.  I would have thought Dion would have had enough, but he seems prepared to keep going, and still has leadership debt to pay off, although he has at least whittled it down substantially.  First-time incumbents would be the next most likely targets.

Caissa

Are there any Atlantic Canada numbers?

Sean in Ottawa

The general gist of all the polls seems to be that the Cons have taken a big hit and that all the other parties have benefitted-- some show the Liberals gaining more than the others but there is little agreement on that.

There is also no indication that thsi is lasting as this remains a current news story. The real information will be a month after parliament sists again can the Cons blow this off. I think they might be able to but this will be like any healed serious injury a weak spot that will make the next time blow more dangerous to them. If the next blow occurs before, during or causes the election then the Cons will be thrown out but we cannot assume that they cannot recover from this-- way too early.

Caissa

So although within the margin of error, the NDP is leading in Atlantic Canada.Wink

It's a virtual three way deadheat.

ottawaobserver

Caissa, here's the full Angus Reid report now (PDF).

ETA:  Surprise, surprise.  The Toronto Star somehow managed to overlook mentioning that Harper and Layton both have a 27% approval rating, while Ignatieff is back at 17%.  The leaders' momentum scores are Layton -5, Ignatieff -20, Harper -34.

Also, Caissa, the NDP LEADS in Atlantic Canada (33-31-31), but of course you won't read that in the Star either!

ottawaobserver

I think it's time for the party to invest some organizing resources in New Brunswick, myself.  With the new provincial leader getting decent coverage, and the selling of NB Power getting some huge traction as an issue, I bet the time is right to find some worthy candidates to invest in federally.

Pogo Pogo's picture

Sean I am left unconvinced.  Clearly the people who support the marginal parties in provincial politics whether it be the Conservatives in BC or the Liberals in Manitoba or the Christian Heritage in Ontario are some of the most committed supporters out there.  They have to be to keep these parties alive.  By your theory they should all poll way over their voting totals.  That you have to build into your model a number of Green specific demographic characteristics weakens the model in my view.  It is like you have the conclusion and you are looking for evidence to support it.

That said I just don't have time to hunt provincial data or CAP data (I assume they were also men at work).

Sean in Ottawa

Pogo you are ignoring what I wrote and bringing empty assertions. You are advancing nothing by way of evidence or facts. I am not doing your work for you. I'm done with you.

NorthReport

It's great to see the NDP hovering around 20% in the national poll.

Sean in Ottawa

Given where things have been yes, it is.

That said the numbers for Alberta, Saskatchewan and Manitoba remain dismal. Perhaps Saskatchewan should not be regarded fertile as it has really been a while since we have had good numbers there. Indeed if the party had historical numbers there the national number would be over 20%.

The Ontario numbers are also encouraging.

All that said-- I am very cautious about any one poll for the reasons I have stated in this thread and others.

So if we do not hang too much on this poll but instead look at broader contexts we can draw soem conclusions:

1 the NDP seems to be doing consistently well in Atlantic Canada-- I wish we could see a provincial breakdown-- I am sure the party will buy one. I suspect the party outside Godin's seat is still low in NB and PEI and that this are numbers from NL and NS. If they are concentrated the party may actually pick up a second in NL and 1-2 more in NS.

2 the NDP has carved out a place in Quebec (although I have said that this is still too low as it is not concentrated enough to earn the party more than an extra seat or so). The range seems to be between 12% and 20%-- interestingly this is a similar range for polls in Ontario so once the support starts to concentrate there may be a significant number of seats to come but this will take time.

3 Ontario remains all over the place-- this poll at 19% is pretty good and a good base to start from but this is the high end of the recent range and it is these Ontario numbers that have driven the poll up. I think there are some good reasons for this and it bodes well in terms of seat totals as there are a number of seats within range if the Ontario numbers go up-- also there are several close call seats in Ontario the party holds and needs to protect. But we need to be concerned about the volatility of the Ontario numbers. Clearly with the fluctuating numbers some serious money will need to be spent in Ontario in the next election once again.

4 I am tired of polls that show MB and SK together-- they are increasingly different electorates. Disappointing is the only way to describe these numbers. Likely incumbents are safe and maybe there is still room for one pickup.

5 Alberta is hard to read- if that 7% is in THE riding then we are no worse off but clearly there is nothing to build on from there as yet.

6 BC is another place where the polls seem to fluctuate wildly. This time the numbers are respectable but this is also a pile of tight races and the numbers fly all over the place at each poll.

Once again the key to keeping and gaining seats in the current election will be Ontario and BC and the NDP campaign will need to be heavily invested there. There is some moderate growth possible in Atlantic Canada and Quebec and serious work needs to be done in the Prairies.

The party also has some choices because it will need to make progress on two fronts-- one is to hold or improve on seat totals and the other is popular vote which the party needs to improve on in order to open more possibilities for the future. The parts of the country where popular vote gains may be possible and significant are not the same as the parts we are already competative and need to watch the splits for seats. There is no part of the country that can be ignored.

I do not dismiss the Green vote entirely. I believe that pollsters exaggerate it as I have already discussed at length. But also I am inclined to think it is not particularly fertile ground for NDP growth and would not invest much there:

1) I think the left of centre Greens are few and likely the most committed-- some are NDP activists who have been turned off

2) The overall Green vote is small and the pickings are therefore slim compared to other parties

3) The more right wing Green are all over the map but less likely to come to the NDP.

The NDP should continue to work hard on the environment because it is the right thing to do, its base wants it to do so and it may be a way to draw support from all the other parties but it is a mistake to think this will come from Green totals in large numbers. As well, some don't like May but she has a following (for as well as against)

There are a lot of people angry with the Conservatives who are not particularly political in a right-left way. I suspect that the NDP could move more Cons to their side in some races than Greens- not just because there are more of them.

Liberals have indicated little of how they are going to fight the next election-- for now they can get away with that. Soon however, they will have to choose to go to the left of Harper and open up some differences or run close to him trying to pull of angry Cons who may think there is little difference. I am not sure which would be a better strategy for Ignatief-- he may look to hypocritical to run to the left so he may have no choice. The NDP will have to adjust to whichever direction he goes. I would suggest avoiding the "they are all the same" rhetoric and instead go after specific proposals, records and pronouncements-- saying the Liberals and the Cons are the same may let the Cons off the hook when we should not. Campaigns where the Liberals are the main threat may have to be very different depending on where they are and local issues.

The BQ is also interesting and considered by many New Democrats as being in the way of substantial growth for the party in Quebec. However a weak BQ could manage to split the vote in places allowing a win with a low total vote. The party might want to look for some of those possible 4 way races. There are few real 4 way races in Quebec but those are the ones the NDP can win-- People have mentioned Gatineau but that could be one of them. One fear is that civil servants who have voted Conservative could go to the Liberals. It may be a good idea to sit down with the PSAC and see if there is any way of getting a message out there since the Liberals cannot be trusted for civil servants either. It is widely expected that the National Capital region is about to take a hit from either the Liberals or Conservatives.

So there is some thoughts-- not so much on this one poll but a wider perspective as I have little respect for single polls anyway.

ottawaobserver

A lot of the Quebec-side members of the PSAC supported the Bloc last time around, but with Daniel Paillé's recently announced economic plan (which involves cutting massive numbers of jobs in the federal public service), they've just made enemies of the Alliance.  I would expect that to clinch the Gatineau seat for us, and perhaps show some growth in Hull-Aylmer as well (of course, I would love to see Pierre Ducasse run again, but it may depend on the timing I gather).

Pogo Pogo's picture

Sean in Ottawa wrote:

Pogo you are ignoring what I wrote and bringing empty assertions. You are advancing nothing by way of evidence or facts. I am not doing your work for you. I'm done with you.

  It's your theory.

Sean in Ottawa

Ok against my better judgement I'll reply.

I am responsible for my theory-- or theories. I am not responsible for your argument against them which you have not made any attempt to back up.

If you use legal onus as an analogy it goes the same way. You bring a case you have to back it up but a theory that opposes it will shift that onus to the one presenting the opposing argument-- In that fashion the onus of proof (in civil law) shifts back and forth between the plaintiff and the defendant depending on whether they are arguing the claim which is to be proven by the plaintiff or individual opposing theories which have to be backed up by the defendant.

In this case I advanced a theory and backed it up with things I have seen first hand (if you like the analogy -not hearsay evidence). You have replied saying more or less it is wrong because it is wrong because you don't like it. You have to do better than that before I can respond.

I do not have to research the data on your argument against me just to refute it. That is illogical.

Debater

ottawaobserver wrote:

Sean in Ottawa wrote:

That said with this level of support the party is capable of the odd surprise but this is not enough support to guarantee company for Mulcair unless there is some concentration somewhere. 

The best chance for another seat, after Gatineau of course, would be if incumbents step down in Montreal.  The next most likely to do so would be Francine Lalonde in La Pointe de l'Ile, and maybe Nicole Demers in Laval, for health reasons, and a few others who have been around a long time.  I would have thought Dion would have had enough, but he seems prepared to keep going, and still has leadership debt to pay off, although he has at least whittled it down substantially.  First-time incumbents would be the next most likely targets.

The NDP is not likely to be in contention in any of those seats.  If they were to change hands, the Liberals would have the best chance at taking them away from the BQ.

adma

Debater wrote:
The NDP is not likely to be in contention in any of those seats.  If they were to change hands, the Liberals would have the best chance at taking them away from the BQ.

Laval, yes--it's more genuinely a technical Lib-BQ marginal.  La Pointe de l'Ile, not so sure, for Hochelaga-esque reasons.

For comparison's sake, the Liberals got 28.1% to the NDP's 12.5% in Laval, and 16% vs 12.9% in LPdl'I.

Stockholm

In other words La Pointe de L'Isle has all the signs of being another Hochelaga where the Liberals crash to third (or fourth) and the NDP becomes the new default non-BQ iption.

NorthReport

More opportunity for the NPD here as Harper's chief organizer in Quebec has quit.
Harper perd son organisatrice en chef au Québec

 

 

http://www.ledevoir.com/politique/canada/282156/harper-perd-son-organisatrice-en-chef-au-quebec

ottawaobserver

I know what everyone's saying about Laval, particularly Debater whose role it is to be a Liberal killjoy on any potential for NDP growth especially in Quebec.  But didn't Thomas Mulcair represent a Laval riding provincially?  One might be entitled to assume that he would still have some kind of organizational contacts there.

remind remind's picture

Why did Harper's organizer quit?

Debater

ottawaobserver wrote:

I know what everyone's saying about Laval, particularly Debater whose role it is to be a Liberal killjoy on any potential for NDP growth especially in Quebec.  But didn't Thomas Mulcair represent a Laval riding provincially?  One might be entitled to assume that he would still have some kind of organizational contacts there.

ottawaobserver, don't call me a Liberal killjoy, please.  I've asked you in the past not to label me, thank-you.

Let's just have a polite and objective discussion about the ridings.

The reason I mentioned those ridings are not likely to be won by the NDP and are likely to be won by the Liberals if the BQ were to lose them is because they are ridings (particularly Laval) where the Liberals have a much stronger presence than the NDP historically, and which were once held by the Liberals.  The Liberals also finished 2nd in these ridings recently.  Therefore, most objective observers in Quebec would come to the same conclusions.

Btw, Mulcair did represent Chomedy provincially, but I'm not sure why that should have much influence federally.  Mulcair has limited popularity and there is only so much he can do.  He is not the one who will be running there as he already is running in Outremont.

Pages

Topic locked