Thursday, March 29, 2007

Federal Election Coming?

This Léger poll out today would give the Conservatives about 160 seats. There is, however, much volatility in the polls being done these days. A new Ipsos poll also out today has much weaker results for the Conservatives, with the suggestion they are "slipping out of majority territory" - all to be taken with a grain of salt.

I think there will be intensive polling and focus group activity over the Easter break and an election call soon thereafter.

Wednesday, March 28, 2007

More on Quebec Polls

A correspondent emailed today to point out this Angus Reid poll, released on the weekend, which I had not noticed before election day. It was relatively accurate, being closer on the major parties than the other polls. However, its total error of 9.3% placed it just behind Léger in terms of overall accuracy. Although closer to the totals of the major parties its estimates for the minor parties were further away from the mark than Léger.

Tuesday, March 27, 2007

Quebec Post-Mortem on Polls and My Model

The closing polls on the election were generally accurate but not completely so. CROP and Léger both underestimated the ADQ vote, and this is no doubt led to the sense of surprise on election night. Their error on the ADQ in both polls was greater than the poll's margin of error. Strategic Counsel did not have any individual results outside their margin of error but were not the most accurate overall.

When I aggregated the performance of the polls by adding up the absolute error between the poll result and the final percentage of the vote for each party, I obtained the following result:

Léger - 9.1%
Strategic Counsel - 12.3%
CROP - 12.8%

So the Léger poll achieved the greatest overall accuracy.

As for my own model when I applied the actual vote percentages from the election, I found I predicted the outcome accurately in 106 of the 125 constituencies, an 85% accuracy rate. I have done better than this, but I make a calculation based on previous voting patterns. When new patterns are established, as was clearly the case here, the method is less accurate.

The net seat distribution predicted was close to the actual:

PLQ - 45
ADQ - 41
PQ - 39

The minority government in Quebec produces a never before seen dynamic in a province that remains deeply divided about its place in Canada (there are sharp differences between the Liberals and the ADQ). There will be interesting days ahead.

Monday, March 26, 2007

Quebec Election - What's Next - 10:15 PM

At this point the CBC is calling it a Liberal minority and that seems right based on what I see.

Consequences:

1. The obvious first outcome will be the end of André Boisclair as PQ leader. However, Charest may not be far behind. He trails badly in his Sherbrooke constituency.

2. I have heard some say tonight that it is a good outcome for Harper. However, one possible effect could be to give a boost to the BQ. The poor result for the PQ was unexpected, but many nationalist Quebecers may want to keep the sovereignty option alive, and see Gilles Duceppe's BQ as a means to do so in the federal election coming soon.

3. Sometimes provincial elections can have paradoxical impacts. I suspect Harper would be comfortable with a Liberal minority, but very nervous having Dumont as a putative ally - a man who supported the YES side in 1995 and has been aggressively attacking social changes to accommodate minorities, a position that might have resonance in Quebec in this election but would be unpopular in the ethnic 905 ridings in Ontario where the Conservatives hope to make gains. Harper has to keep great distance from the ADQ even if he wants their voters.

4. The ADQ has a platform they obviously can't implement, so winning would not have been helpful to them.

5. It is important not to confuse seats with votes. The ADQ is up about 13 points a lot of it a protest vote, which is by definition soft. The ADQ will still have to work hard to keep it. The PQ is just 2 points behind. A reversal of fortune in the next election can't be ruled out.

Quebec Election - 9:20 PM

Always a surprise - the ballot box bonus went to the ADQ. The outcome remains close and very uncertain.

The model isn't working very well. It is based on the 2003 election results and the unprecedented upsurge by the ADQ in non-traditional areas of support is wreaking havoc with it. I am not surprised by that.

If I input the current vote percentages I get the following outcome:

PLQ - 42
PQ - 45
ADQ - 38

The quirkiness of the first past the post system is manifesting itself.

I kept telling myself that despite the obvious success of the ADQ it couldn't win. I should have known better - Quebec is always capable of producing the big surprise.

Quebec Election Day 6:45 PM Update

Update: On reflection I thought it made sense to average just the two Quebec polls: Léger and CROP and apply the model. The result was striking:

Liberal - 50

PQ - 50
ADQ - 25

A nominal tie: even closer than the earlier calculation but an outcome that would keep Charest in office. A late phone call suggested a closing tick back to the Liberals from the ADQ. However, given the three-way character of the race, in some cases this would simply deliver seats to the PQ. I also suggest reading this post by Harold Chorney.

Sunday, March 25, 2007

Quebec Election - Too Close to Call

I have averaged the final three polls from Léger, CROP and Strategic Counsel and applied my seat estimate model.

The results produce a PQ minority:

PQ - 56
Liberal - 43
ADQ - 26

However, if I simply push the margin of error to its outer limit on just the PQ and the Liberals, the results reverse, and we get a Liberal minority:

Liberal - 51
PQ - 47
ADQ - 27

I note also that the blog Democratic Space projects a Liberal minority:

Liberal - 55
PQ - 47
ADQ - 23

The differences in these projections represent very small vote share differences. The outcome is simply too close to be certain of the winner. There is added uncertainty due to the fact that the ADQ is clearly making gains in areas of Quebec outside its traditional base of support. Another imponderable is that, in past elections Liberal support on voting day exceeded its strength in the polls, but this did not happen in 2003.

The one certain conclusion is that no party will win a majority of the seats.

Thursday, March 22, 2007

New Quebec Polls - Bad News for Charest

Two new polls in have been released in Quebec, one from Léger and one from CROP. They are confined to the Quebec City region but strongly suggest that any prospect of a Liberal majority is gone. In fact the results are so good for the ADQ, which has about 40% in both polls, that if they could be projected province-wide we would no doubt be headed for an ADQ government. The regional character of the ADQ means that we can't do that, but they are headed for a significant result on Monday. The weekend polls will be most interesting.

The worst outcome for the ADQ, as a populist party, would be victory. It is clear that not only are they not ready to govern, they have also articulated a program that would be impossible to implement and would further divide an already fractured society if it could be.

Sunday, March 18, 2007

Quebec Election - Still Unclear

The polls taken after the Quebec election debate illustrate well the intrinsic difficulty of probing public perceptions: CROP and Strategic Counsel had Dumont as the winner, while Léger called it for Charest, the overall impression being quite muddled. As well, Strategic Counsel thought Boisclair benefited by exceeding expectations.

Those results are by their nature an unreliable guide to the outcome. TC is more inclined to credit polls that ask about behaviour, voting intention rather than perceptions. But things are equally unclear in the party preference surveys. The Liberals continue to lead in most voting intention polls, except for Strategic Counsel which has the PQ ahead, but among francophones, the PQ is narrowly ahead in all the polls to date. My estimate at the moment would suggest a PQ minority but that remains highly uncertain. The latest estimate by Democratic Space (a blogger and seat forecaster with a good record) gives the Liberals a one seat edge.

Curiously, the zeitgeist (judging from media coverage) is with Dumont and the ADQ. Perceptions of Mario Dumont and the ADQ in this SES poll, however, run consistently behind Charest and the Liberals.

I have always found Quebec politics extremely difficult to read and this election is no exception. It is clear that the ADQ has grown, but it has less money, weaker organization and less qualified candidates than its opponents. And it has a narrow geographic and socio-economic base - all indicators that it shouldn't win. But the campaign leaves the impression that Mario is on the verge of a significant breakthrough. He is certainly in the process of establishing a new vote pattern in 21st century Quebec, even while emulating the appeal of the old Union Nationale and the Créditistes.

The ADQ clearly damaged the PQ in 2003, and now it is hurting the Liberals in 2007. The Léger poll gave the Liberals just 33% compared to the 46% they won in 2003. A minority government continues to be the most likely prospect.

Tomorrow's federal budget is a key part of Liberal strategy, so there is much yet to be played out. And it even seems possible that a federal campaign could begin as early as tomorrow. As if the Quebec political scene was not already confused enough.

Tuesday, March 13, 2007

Quebec Leaders Debate

The debate, now playing in one corner of my computer screen, is almost over but my overall sense is that Charest has done well, Boisclair less well (calm, relaxed demeanour but prone to getting too technical), and Dumont has come over as too agitated, excited and defensive, clearly with major contradictions in his platform. We will see what the media types have to say.

But Charest is apparently behind. He was nominally in first place in the latest CROP poll, but third among francophones. My model says this poll would produce a weak PQ minority. In 2003 the Liberals picked up strongly among the undecided after the 2003 debate. They need to do it again.

Friday, March 09, 2007

The Pundit Class

Sometimes one runs across something in the press, which speaks the truth both eloquently and succinctly. Such is this excerpt from Rick Salutin's column in the March 9 edition of the Globe and Mail:

I hear the buzz ... in Ottawa is that Stéphane Dion is a dud. Makes me think of the buzz on previous Liberal leaders. In 1984, insiders said new leader John Turner had “the royal jelly.” He campaigned wretchedly and lost to Brian Mulroney. In 1988, they said he couldn't even control his own caucus. In that election, he brilliantly outperformed Mr. Mulroney and NDP leader Ed Broadbent, though he didn't win. In 1993, they called new Liberal leader Jean Chrétien “yesterday's man,” just before he won the first of three straight majorities. He didn't fall to the “refreshing” new face of Kim Campbell, who reduced her party to two seats. That same election, they said Preston Manning and his Reform Party had “peaked too soon” and would never amount to anything. In 2004, new Liberal leader Paul Martin had a “juggernaut” that couldn't be stopped. If you're a Liberal leader, you might prefer not to be anointed by these prophets.

(TC does think Salutin's point about Turner in '88 applies mainly to the TV debates.)

One could add almost endlessly to this: the pundit class devoted much of its space from 1978 to 1984 extolling John Turner's electoral virtues only to watch him start to unravel the moment he stepped back into the public arena; dismissed the importance of the Bloc Québecois in 1993; saw Stockwell Day as an unstoppable force in the summer of 2000 even after he arrived at a press conference in a wet suit demanding that MPs put in fewer days of work in Ottawa, and generally thinks the incumbent government will be re-elected (although they thought Chrétien would win only a minority in 2000). They dismissed Harper in 2004 and into the beginning of last year's election.

The Ottawa pundit class has made an abundance of errors in the past but seems unable to learn from them - a product of a perennially shallow outlook driven by such things as the most recent poll, regardless of sample size and usually described in the plural despite being singular. One can't always be wrong (like a stopped clock that is right twice a day), but when they occasionally develop original analysis that is insightful and anticipatory, it tends to seem like an accident. Their record over the years has been dreadful.

In the Saturday, March 11 edition of the Toronto Star this column by Thomas Walkom essentially makes the argument that it is too soon to write off Stéphane Dion. Walkom, however, is not based in Ottawa and not subject to its fevers. Hence his more judicious analysis. Well worth reading.

As for the fate of Dion, and Jack Layton who is also being dismissed by the pundit class: TC thinks it is way too soon to tell, but if there is an early election the Liberals are at a distinct disadvantage because they are still too early into the regime of their new leader to be ready organizationally and financially (something it should be noted that is not a problem for the NDP).

Monday, March 05, 2007

The Two Quebec Elections

Every time Quebec goes to the polls, there are always two elections underway at once - one among anglophones/allophones and one among francophones. The two contests are quite different. In the most recent Léger poll here is the linguistic breakdown.

Among anglo/allos (about 17% of Quebec's population):

Jean Charest's PLQ - 71%
André Boisclair's PQ- 12%
Mario Dumont's ADQ - 9%
Others - 7%

However, among francophones (about 83% of Quebec's population) a different story. You wouldn't know it from the media coverage but the PQ actually does have a small lead among francophone voters overall:

André Boisclair's PQ- 33%
Jean Charest's PLQ - 28%
Mario Dumont's ADQ - 28%
Others - 12% (including 6% for Quebec solidaire, which is drawing votes from the PQ)

When it is all added up the poll overall gives us the following outcome:

Jean Charest's PLQ - 36%
André Boisclair's PQ- 29%
Mario Dumont's ADQ - 25%
Others - 11%

My estimate is that this would probably produce a minority Liberal government, but when you have a three way race that evenly divided among francophones it makes the net outcome very unpredictable at this stage.

Saturday, March 03, 2007

New Federal Polls - Tory Majority?

I noticed yesterday quite a bit of blogging buzz about the new online poll out from Angus Reid's new polling outfit. Two other polls were also out this week, one from Ipsos Canada, Reid's former firm, and Decima.

Reid's new online poll would produce a Conservative majority of 161 seats (with no seat gains in Quebec) according to TC's calculations(Liberals would get 73, the NDP 25 and the Bloc 48), providing further encouragement, if any were needed, for Stephen Harper to trigger a spring election.

But there is a catch. The methodology of the Reid poll is a tad obscure. It is based on an "online panel ... recruited via an industry-leading process that incorporates a randomized, widespread invitation approach and a triple opt-in screening procedure". I think what these words mean is that they have made their best efforts to select a random sample over the internet. Nevertheless online polling is in its infancy, and while I think it will be a standard some day, it certainly isn't now.

There was an online poll conducted by Ipsos just prior to the last election. TC did a comprehensive analysis of the accuracy of all the polls taken at that time, and found that, although the Ipsos online poll had the largest sample size, it was the least accurate in predicting the national vote shares in the election (in fact, most of its errors were greater than its statistical margin of error). While overall it was fairly close, having an aggregated error of 9.6 it needs to be compared to SES, the closest to the mark, which had an error of 2.1 with a far smaller sample. The online poll did do better on its regional sub-samples, where it had larger numbers than the conventional telephone pollsters and thus a lower statistical margin of error.

My conclusion is that all online polls should be treated with extreme caution, and we really need to see confirmation in conventional polls before they should be taken as gospel.

The two other polls did reveal Conservative strength and Liberal weakness, but they both had the Conservatives at 36% compared to the online Reid poll's 40% and the Liberals averaging 29.5% compared to Reid's 26%. Interestingly, the largest error in the Ipsos online poll prior to the 2006 election came in its underestimate of the Liberal vote, reinforcing my inclination to view it skeptically.

Finally, looking at the online Reid poll and the 161 seats reinforces TC's view that it is almost (but not quite) impossible for the Conservatives (or the Liberals for that matter) to win a majority. And some of the numbers in the poll are just not believable. TC thinks that the results of the last election are usually a better predictor of what to expect next time than many polling numbers that appear in the interim and that certainly appears to be the case here in some regions. Just a slight deviation from these results would quickly reduce the Conservatives to less than a majority. However, it does look like an early election would likely produce another Harper government, simply because it doesn't appear that the Liberals have the money or the time to be properly prepared.