There are two polls out today. One shows a big Conservative lead, the other a statistical tie. Two guesses which one we'll be hearing the most about today, and the first one doesn't count.
I'm not going to bother disecting the polls. I suspect the truth is likely somewhere in the middle. I think however that Peter Donolo, the former communications aide to Jean Chretien and the spokesperson for the company with the Cons way ahead poll, makes a good point here:
Mr. Donolo said the Liberals may be having difficulty marking out their territory, given that they are forced to sometimes criticize the government and at other times prop it up.
“They are unable to define themselves strongly or convincingly, because of the need to constantly bob and weave to avoid an election,” he said.
I agree. I think it's getting harder to accept the argument that no one outside of Ottawa pays attention to abstentions and what not. It is beginning to resonate south of the Queensway. And I'm sure the NDP already has TV ads cut to remind everyone in the next election campaign.
It's also pretty unclear to many right now just what the Liberals stand for these days. When the powers that be decide what we stand for, hopefully they'll let me know. Send me a memo or something.
In the mean time, while as I said I think the truth is somewhere in the middle of these two polls, I think I know which way it's probably trending, and it's not in the way I'd like.
Does that mean I'm backing off my call for an election on the budget? Hells no. If we're going to reverse the trend we need to toughen-up, start defining outselves, start hitting back at Harper, and bring this government down. Recommend this Post on Progressive Bloggers
12 comments:
Wow - that's a pretty big difference for 2 polls, both from reputable companies (if not, perhaps, Nanos-worthy!), with the same sample size (1,000), and conducted on the same dates (Feb 14-17 - the Chronicle-Herald story you linked to does't give the dates the CP-Decima survey was taken, but this one one does).
I suppose they're *just barely* within each other if you stretch their margins of error to the breaking point, but still, the difference is interesting.
I agree the party should be staking out where it stands, but it follows that if there isn't a really good reason to call and election then it's in no one's interest.
Doing our their part to make Parliamnet work is a respectable goal.
Jason,
It is a very wide difference indeed. The Strategic Counsel poll is just so far out from everything else we'd seen I'd tend to discount it, but given the last week and a bit the Liberals have had, I wouldn't be surprised if things were trending in that direction.
Robert, making parliament work is a great idea, and I'm all for that, but the fact is we haven't been making parliament work. We've just been rolling over for the Cons. The opposition parties do have a majority, why haven't we worked together to pass amendments, to introduce legislation, to get things accomplished?
The answers for that are manifold. If parliament was working I wouldn't favour pulling the plug. But it's not working, and no one seems interested in making it work either.
There are plenty of good reasons to call an election. Number one is the economy, that's why Prentice et al. were hitting it so hard lately.
The opposition parties do have a majority, why haven't we worked together to pass amendments, to introduce legislation, to get things accomplished?
One theory I have on that, leaving aside legitimate policy & philosophical differences between the Libs and NDP (to say nothing of the BQ) is this:
When the Martin Libs were in government with a minority, it benefited both the CPC and the NDP, as well as the BQ, to go after them from different directions.
Obviously, the Tories aren't beloved by anyone else, but they aren't in the middle as much as the Liberals were/are.
I've said it before: the Liberals have traditionally benefited from being in between the Tories (in various forms) on the right and the NDP on the left, especially when one or both of them were weak. But when the Libs are being attacked on both sides by (relatively) strong parties, being in the middle gets them squeezed.
In this case, then, neither the LPC nor the NDP wants to be seen as building the other up, since they are to some extent competing for the same centre-left voters. That wasn't the case when Tories and NDPers were taking different hunks out of the Liberal hide.
That's only one theory, of course; there are probably other reasons why the oppo parties haven't teamed up more often (eg., nobody wants to be seen as being too close to the BQ, though the Tories, the Libs and the NDP have done so on various issues before).
I am curious whether there have been 3-way talks among the LPC, BQ and NDP in recent days. They have teamed up before, but it has been quiet on that cooperation front lately - at least, according to media reports.
Of course CTV is making a big deal out of this and makes no mention of any of the other recent polls that conflict with this one. You have to start questioning the competence and believability of these polling firms now. However, I agree the nervous nellies are starting to drag this party down. They need to take their pacifier's out of their mouths and gain some courage! This has just put me in a bad mood now.
Mr. Donolo's statements could also have been said of the Conservatives during PM Martin's minority government.
They blew hot and cold regarding an election, abstaining on confidence motions on more than one occasion when they thought an election was not in their interests.
In fact, I would say what we are seeing from the Liberals right now is a natural by-product of minority government. If the Official Opposition does not have to worry about triggering an election every time it votes it can be more forthcoming in establishing a more coherent position. Currently the Liberals do not have that luxury and they will not have it until an election so complaining about it does no good.
All we can ask is for the Liberals to be prepared for an election and to go for it when they deem the time is right.
Jason, two polling companies with two different polling methodologies and you see what you get. Very interesting to say the least.
Ottlib, how are the methodologies different in this case?
That's a serious Q., not a snarky one. They were both phone polls; I'm honestly curious as to how they were done differently otherwise?
''When the powers that be decide what we stand for, hopefully they'll let me know. Send me a memo or something.''
Shouldn't the grassroots of your party be deciding that?
Only 20 months left until the Oct 2009 election, and Libs still have not had a Policy Convention.
Libs are still letting Dion wing it, making up policy on the fly.
Waiting for a new leader?
jason:
I am not certain about the specifics of their methodologies but no two polling companies will conduct a poll using the exact same ones.
Conducting a poll or survey is more complicated than putting together a bunch of questions and then calling a 1000 people to ask those questions.
Some factors that can effect outcomes:
In which order do you ask the questions?
Decima and Nanos tend to ask the Party Support question right up front. Ipsos and SC will sometimes bury that question and if the questions that precede it are on issues that favour one party over the other they will bias the party support question.
I am not saying that is what SC did but I am saying that if they did the results would still be considered statistically valid.
What was the accepted response rate for the poll?
The polls always publish how many interviews they did for the poll but they generally do not state how many phone calls they had to make to achieve that number. Generally speaking the lower the response rate the higher the non-response error and less reliable the final estimates, regardless of what is stated as the MOE.
Note: Response rates for political polls are generally under 25%, which is a ridiculously low response rate for every other kind of market research survey but is acceptable for political polls.
What was the sample for the poll and how was it selected?
There are a whole raft of statistically valid methods of doing so that have different impacts on the final estimates.
What was the survey frame? From what source did the polling company choose the poll sample?
I can guarantee that the frame for each was different so that will impact the results.
What methods did they use to process the data that was collected for the poll?
Again I can virtually guarantee that they used different methods and error can be introduced at each step in processing polling data.
Those are just a few issues that come to mind with regard to why the estimates from two polls taken on the same days are so markedly different.
Of course, taken on their own both polls are statistically valid, but that does not mean that either one of them is an actual reflection of reality.
The order of the questions could be critical. The Liberals have the strongest pure brand, and there are many more people who "just don't like Harper" than there are who feel the same way about Dion. Therefore, there must be people who would vote Liberal even though, if asked about specific issues, they think the Conservatives would do better.
But consider the following dialog:
"Who will do best on issue Y?"
"Conservative"
"Who will do best on issue X?"
"Conservative"
"Who will do best on issue Z?"
"Conservative"
"Who would you vote for?"
"Liberal"
By asking issue questions first, the poll automatically increases Conservative support. The real question is whether the campaign will do that also.
Jason,
In this case, then, neither the LPC nor the NDP wants to be seen as building the other up, since they are to some extent competing for the same centre-left voters. That wasn't the case when Tories and NDPers were taking different hunks out of the Liberal hide.
I think you're on the right track. Neither the NDP or the Libs want to build the other up. The NDP want to make the Liberals irrelevant, so they don't want to cooperate. The Liberals can't seen to decide on anything even if they had a willing partner. The BQ may be willing to dance if it would embarrass Harper in Quebec, but everyone's leery of being seen as cozy with separatists.
The perfect example of a failed opportunity is one I've cited before. About a year ago, the Liberals moved a motion to tell NATO we're ending the Afghan combat mission in 2009. The NDP wanted the troops home now, so they voted it down with the Cons. If the Libs and Cons could have found compromise with each other and the BQ that motion or similar one could have passed, and if we even still had a Manley commission the context of the whole debate would have been much different. But everyone stuck to their narrow ideological positions, instead of working together, and Harper has been able to frame the debate himself.
liberazzi,
Of course CTV is making a big deal out of this and makes no mention of any of the other recent polls that conflict with this one.
CTV and the Globe sponsor the Stategic Counsel poll so its unsurprising, but they are doing a misservice by not putting it in context with other polling.
Ottlib,
If the Official Opposition does not have to worry about triggering an election every time it votes it can be more forthcoming in establishing a more coherent position.
There are ways of being coherent outside of confidence votes. Where's the amendments to legislation? Where's the private members bills? Where's the working with the other opposition parties, whom we have a fair bit of ideological common ground with, to force the government's hand?
And we can be much more coherent on confidence bills then we have been. Like the budget. One day it's you'd better meet our demands or we're voting it down, the next day it's well maybe if it sucks we'll still vote for it. We've been incoherent from day to day lately.
Liberazzi and Jeff, the Globe did say the following, in the story that Jeff linked to:
The poll, conducted by The Strategic Counsel for The Globe and Mail/CTV News, shows the widest Conservative lead of a number of recent surveys. Last week, the firm Ipsos-Reid had it 36-29 for the Tories, while Harris-Decima reported a 35-33 Tory margin in its poll released yesterday.
Post a Comment