Friday, March 28, 2014

"Tapping into the Pulse" is out!

I'm happy to announce that Tapping into the Pulse: Political public opinion polling in Canada, 2013 is now officially out and available for sale!

Please visit the section of the site for the eBook for information about ordering. Ordering via Gumroad is the recommended method - you can download all eBook formats as soon as you buy!

Those formats will work on virtually any eReader or device. So after ordering from Gumroad, you can read it on your laptop, your iPad, your Kindle, your iPhone, your Kobo - pretty much anything! You can even print it out if you're old-fashioned.

Tapping into the Pulse is a retrospective of federal and provincial politics in 2013 from the perspective of polls. It is a bit of a history book, but it also delves into polling issues, particularly concerning the misses in a few by-elections and the provincial election in British Columbia. The eBook also contains very detailed records of polling results in 2013 from almost all of Canada's major pollsters. An invaluable resource for political students and junkies alike!

If you do have a Kindle or Kobo and want to download the eBook directly via the Kobo or Kindle stores, that is an option as well!

Thanks again to the 200+ Kickstarter backers who made this project possible! If you enjoyed the book, please consider reviewing it on Amazon.

Thursday, March 27, 2014

What if the Orange Wave crested earlier?

As someone who politically came of age during the minority years after 2004, it is striking just how much the 2011 federal election transformed the political landscape in Canada. Prior to that election, and after the uniting of the right in 2003, the presence of the Bloc Québécois and its ability to take 40-50 seats off the table made cobbling together a majority government a difficult thing. But what would the elections between 1997 and 2008 have looked like if the NDP had made its breakthrough into Quebec earlier?

Before getting into this hypothetical exercise, a little business. Today is the last chance you have to pre-order Tapping into the Pulse: Political public opinion polling in Canada, 2013. Pre-ordering will save you 20% off the price of the eBook, which will be made available for sale via this site, Amazon, and Kobo tomorrow! 

A few caveats before we get into these hypothetical scenarios. Obviously, if the NDP was able to breakthrough into Quebec in 1997 the course of the party's history would be entirely different. But would a breakthrough in 1997 or 2000 under Alexa McDonough even be possible? If it had occurred before 2003, would Jack Layton still have been chosen as leader? How would the other parties have reacted to an NDP based primarily in Quebec? All of these questions are unanswerable, and will be left unanswered here.

The scenarios outlined below are based on the following premise: that the NDP takes the same proportion of the vote from the Bloc, Liberals, and Conservatives (or PCs and Reform/Canadian Alliance) that they did between the 2008 and 2011 elections. Would that happen exactly the same way in every election between 1997 and 2008, each with their own issues and dynamics? Of course not. But again, we're just doing a hypothetical exercise.

For the seats, I started with the UBC's electoral forecaster for 1997, but used a simpler system afterwards as that model does not handle huge surges by one party very well (in 2011, it would have given the NDP over 70 seats in Quebec). For the elections between 2000 and 2008, I merely penalized the parties by the same proportion as occurred between the 2008 and 2011 elections, and awarded the NDP the rest.

Results in the rest of the country are unchanged. Would that happen if the NDP was surging in Quebec? Probably not. But again...

What is the purpose of this exercise? Much of it is sheer curiosity, but I think it also shows just how important Quebec has been in the past and how it can still be important in the future.

The first scenario imagines that, after the referendum defeat of 1995, the Bloc Québécois is deemed no longer very useful and voters drift over to the NDP (note that in 1988 the NDP had 14% support in Quebec, better even than their 2008 performance under Layton, so it is not unthinkable). 

The NDP wins Quebec with 37% of the vote, while the Bloc Québécois drops to 30%, the Liberals to 20%, and the Progressive Conservatives to 10%. This gives the NDP 56 seats, leaving just 12 for the Bloc, five for the Liberals, and one for the PCs.

The NDP's surge in Quebec propels them to Official Opposition status ahead of the Reform Party, while the Bloc drops to fifth-party status behind the Tories. Jean Chrétien wins a minority government. His only viable route to getting legislation passed is the NDP, as the Tories are not strong enough to give them a majority of seats.

In 2000, we can either imagine that the NDP continues its dominance in Quebec, propping up the Liberals for three years, or that the 1997 breakthrough never occurs, and instead takes place for the first time in 2000. For continuity, we'll assume the former scenario.

The NDP takes 37% of the vote again, while the Bloc plummets to 23%. This barely puts them ahead of the Liberals, who win 22%, while the PCs take 17% of the vote. The NDP retains 56 seats, while the Liberals take advantage of the Bloc's tumble to win 13 seats. The Bloc loses enough support in rural Quebec to the PCs to give the party no more than a handful of seats, and it is reduced to four.

The Liberals are returned with another minority government of 149 seats, while the NDP just manages to hold onto the Official Opposition role with 69 seats. But they lose their influence in the government, as the Progressive Conservatives use their 13 seats to give Chrétien a workable majority government. Disappointed that she continues to struggle to breakthrough in the rest of the country, McDonough resigns as leader and is replaced by Layton, who emphasizes his Quebec roots in his successful leadership bid.

Disgusted that Joe Clark has teamed up with the Liberals, the PCs split apart. One faction officially joins the Liberals while the other group approaches the Canadian Alliance. An agreement is reached to form a new Conservative Party.

Layton manages to keep the New Democrats the first choice of Quebecers with 38% of the vote, but the Liberals have been boosted to 27% thanks to the PCs from Quebec having joined the party en masse. The Bloc takes 25% of the vote while the new Conservatives, seen as too close to the old Reform wing, manages just 9% of the vote. The NDP takes 54 seats, with the Liberals winning 18 and the Bloc taking three.

This gives Paul Martin a new minority government, but the Official Opposition role has gone to Stephen Harper's Conservatives. Determined to not make the same mistake Clark made, Harper refuses to co-operate with the Liberals under any circumstances, and the Liberals and NDP team up to form a majority government.

But the sponsorship scandal poisons the relationship between the Liberals and the heavily Quebec-based NDP, and the coalition falls apart and new elections are called in 2006.

The NDP takes 39% of the vote in another successful campaign, but the Bloc jumps to 30% as a result of the sponsorship scandal. The Liberals still take 20%, while the Conservatives drop again to just 7%. The NDP takes 60 seats, their best-ever result, while the Liberals are reduced to 11. For all their gains, the Bloc picks up just one seat.

Harper manages to form a minority government with 114 seats, as Layton refuses to join with the Liberals again. For two years, he instead works with the Conservatives to pass budgets, though they oppose everything else the government proposes until things come to a head in 2008, and the government falls.

The NDP takes 38% of the vote in Quebec and holds 59 seats, as the Liberals drop to just 12% and seven seats. The Conservatives, now in government, manage to increase their vote share (at the expense of the Liberals) to 19% and win five seats, while the Bloc slips slightly to 26%.

Harper forms another minority government with 138 seats, while Layton takes over the role of Official Opposition from Stéphane Dion, whom the Liberals selected in hopes of winning back the Quebec vote. Dion proposes that the Liberals and NDP join up again to form a majority government. Layton has a chance to become Prime Minister, but his support in the rest of Canada is still weak and his potential partners have just been dealt a terrific blow in Quebec. Does he accept?

The point of this exercise is not historical revisionism, but instead what these numbers tell us about the future. As the Conservatives won a majority government in 2011, one might be persuaded that all Quebecers did was replace a teal party with an orange party - it doesn't change the electoral calculations one bit if the Conservatives win a majority government completely outside Quebec. But that was the unusual event. In previous elections, the Conservatives had never been strong enough to win a majority government outside of Quebec. Recall that in 2006 and 2008 the Conservatives were kept to a minority despite winning 10 seats in Quebec. 

If the 40 to 50 seats the Bloc routinely won are now up for grabs by a national party, that changes things tremendously. Majority governments could have been cobbled together between 2004 and 2008 if the Bloc wasn't such an unacceptable dancing partner for a government. Barring a return to prominence for the Bloc, that is no longer a problem. What will that mean in 2015 and beyond?

Tuesday, March 25, 2014

Léger confirms Liberal lead as PQ drops among francophones

After two polling firms dropped into the campaign for the first time last week with numbers at odds with those recorded by CROP and Léger, the two Quebec-based pollsters tracking the race, many were looking forward to the next poll from CROP or Léger that would clear things up. This morning, the newest Léger poll for the Journal de Montréal confirmed what those polls had been showing: the Liberals have moved ahead of a faltering Parti Québécois.

The lead is not nearly as large as the 13-point edge Forum recorded on March 19, and so the projection has actually narrowed. The Liberals are now projected to have 41.2% support, or between 40% and 45% when assuming normal polling error. The Parti Québécois sits at 33.1%, or between 32% and 37%, while the Coalition Avenir Québec is at 14.5%, or between 13% and 16%. Québec Solidaire is projected to have 8.7% support, or between 8% and 9%.

In terms of seats, the big Liberal majority is now a small one. The party is projected to win 65 seats with these levels of support, against 55 for the PQ, three for the CAQ, and two for QS. The likely ranges lean heavily towards a Liberal victory (and mostly a majority one), but do envision the possibility of a PQ minority.

A few notes on the projection. It has been updated to reflect the finalized list of candidates. The estimated support levels of the Greens and the 'others' are now locked-in. These support levels are determined by the amount of support these parties and independent candidates averaged per riding in 2012, multiplied by the number of candidates they have nominated. As the Greens have less than a half-slate of candidates (44 out of 125), their polling levels will no longer be taken into account. The party averaged 1.9% support in each riding they had a candidate in 2012, so with such a reduced slate of candidates the party is now estimated to take 0.7% support. That may flicker by a tenth of a percentage point as the model rounds things up or down to reach 100%. The same goes for support for the other parties and independents. Option Nationale's position in the polls will be still tracked, as they have nominated 116 candidates.

Léger was last in the field between March 11-13, for Le Devoir and The Globe and Mail. Since that poll, which was conducted before the first debate, the Liberals have picked up three points to move into the lead with 40% support. The PQ dropped four points to 33%, while the CAQ was up one point to 15%. Québec Solidaire was unchanged at 9% support.

The drop in support for the PQ is outside the margin of error (or would be, if this were a probabilistic sample). It is also the lowest that Léger has pegged the PQ to be at since early December 2013, while 40% is the best Liberal score in any Léger poll in several years (CROP had the party at 40% in August 2013).

The poll confirms the findings of Ipsos Reid, which had the two parties at 37% and 32% before the debate with a much smaller sample. And, interestingly, the Léger numbers for the PQ and Liberals match exactly the Ipsos Reid estimation of support among those most likely to vote.

As for the Forum poll, Léger does agree with the support level of the PQ. But the Liberals were at 45% in that poll, which seemed high at the time and still seems to have been a bit of an outlier. But Forum did catch that the Liberals were on the upswing.

The source of the PQ's drop was among francophones, particularly those in the regions outside of Montreal and Quebec City. The PQ fell four points to 40% support among French-speaking Quebecers, while the Liberals were up to 30% and the CAQ to 17%. QS was steady at 10% support.

Léger did a massive poll, allowing them to break down provincial support region-by-region. I was able to obtain the results broken down by Léger's usual regions, in order to be able to compare apples-to-apples (and use the right data for the model).

In the Montreal area, the Liberals led with 45% to 32% for the PQ, 11% for the CAQ, and 8% for QS. The Liberals upticked, but not significantly so.

In Quebec City, the Liberals also had an uptick to 43%, while the PQ dropped to 24% and the CAQ remained relatively stable at 21%. QS was at 9% here.

In the regions of Quebec, the Liberals were up to 37% while the PQ dropped seven points to 33%, a significant decrease. The CAQ was at 18%, while QS was at 9%.

If we look at the regions reported by Léger, we get a good idea of where the parties stand. The Liberals led in the Capitale Nationale (42%), the Mauricie/Centre-du-Québec (35%), Estrie (38%), on the island of Montreal (50%), in the Outaouais (49%), in Chaudière-Appalaches (44%), in Laval (53%), and in Montérégie (37%). None of these are particularly surprising, though the strong numbers in Laval and Montérégie could spell trouble for more than a few PQ incumbents. The Liberals were also competitive in the Saguenay-Lac-Saint-Jean/Côte-Nord with 36%, suggesting Philippe Couillard may have a decent shot of election in his Roberval riding.

The PQ was ahead in the Bas-Saint-Laurent/Gaspésie (46%), the Saguenay-Lac-Saint-Jean/Côte-Nord (43%), Abitibi-Témiscamingue (47%), and Lanaudière/Laurentides (45%). Again, none of these are very surprising, though the PQ has definitely taken advantage of the CAQ's drop north of Montreal. The PQ was only a handful of points behind the Liberals in Mauricie/Centre-du-Québec (31%), Estrie (33%), and Montérégie (36%) so there will be plenty of close races.

The CAQ's only pocket of strength was in Chaudière-Appalaches, where the party registered 28%. That bodes well for a handful of their incumbents, but with no better than 23% support anywhere else only a handful of CAQ MNAs will survive. The party is tied with the Liberals for second in Lanaudière/Laurentides (where François Legault will have a hard time being re-elected), but trails the PQ by 22 points.

For Québec Solidaire, the island of Montreal was not even their best region (9%, the party had 12% here in 2012). They scored double-digits in the Bas-Saint-Laurent/Gaspésie (13%), Capitale Nationale (10%), Mauricie/Centre-du-Québec (10%), Estrie (13%), and Abitibi-Témiscamingue (13%), but with the small sample sizes this does not mean much.

Option Nationale did not score more than 1% anywhere, while the Greens were at 4% on the island of Montreal, where they actually do have a half-decent slate of candidates (numbers-wise, at least).

On who would make the best premier, Philippe Couillard jumped five points to 31%, his best showing ever, while Pauline Marois slipped to 25%. Legault was at 16%, while Françoise David was at 8%, no real change. Among francophones, Couillard was up to 23%, while Marois was still ahead at 30%.

The campaign has well and truly shifted. Earlier, Léger found that 52% of Quebecers thought the PQ was going to win the election, against just 25% who thought the Liberals would prevail. Now, 48% expect the Liberals to win, with just 33% thinking the PQ will come out ahead. The campaign has just been a disaster for the PQ.

Interestingly, Léger asked Quebecers whether they would prefer a PQ or Liberal government. While 48% said they preferred a Liberal government, 41% preferred a victory by the PQ. This is where strategic votes may come into play. On this score, however, the PQ might be slightly favoured. Among CAQ supporters, 51% preferred a Liberal victory against 34% who preferred a victory by the PQ. Among QS supporters, however, that split 62% to 21% in favour of the PQ. That means that, among CAQ and QS supporters, the PQ could potentially grow by 10.7 points, compared to 9.5 points for the Liberals. Narrowing the gap by one or two points could bring the Liberals down to minority territory.

But that is what the PQ appears to have been reduced to, after starting the campaign the odds-on favourite to win. There is still time to turn things around, but that time is running out. The Liberals were only out of power for 18 months after holding office for almost 10 years, and there may be a backlash if Quebecers decide they aren't ready to hand the reins of a majority government back to the Liberals for another four years. But the PQ has salted the earth to such a large degree over the last few weeks that perhaps only Legault and the CAQ stands to benefit. The campaign may not be over yet.

Friday, March 21, 2014

No wait, it's a landslide

For the second time in the election campaign in Quebec, a poll by Forum Research (this time reported by the Toronto Star, after the Montreal Gazette was scolded by the MRIA for the way the last Forum poll was contemptuously reported) tells us that all assumptions made about the state of the race in Quebec are wrong. Rather than a close race, the Liberals have soared ahead with a double digit lead. Philippe Couillard is, thus, on track to bettering Jean Charest's majority victory of 2008.

Accordingly, the projection has lurched violently to the advantage of the Liberals. The party now leads with 43.1% (or between 41% and 47%, assuming normal polling errors) against just 32.8% for the Parti Québécois (or between 32% and 36%). The Coalition Avenir Québec remains well behind with 13.3% (or between 12% and 14%) while Québec Solidaire has slipped to 7.8% (or between 7% and 8%).

These sorts of numbers give the Liberals a large majority government of 71 seats, with the PQ reduced to 50 and the CAQ to just two, putting them in a tie with QS. The likely ranges now point only to a Liberal victory, with between 61 and 76 seats while the PQ could win only between 46 and 59 seats.

Has the campaign really swung so heavily towards Couillard and the Liberals? The last time Forum reported, they gave the Liberals a two-point edge when both CROP and Léger saw the race as tied or leaning towards the PQ. Léger and CROP both subsequently recorded Liberal gains, so perhaps Forum will be a precursor to further Liberal gains again. But these numbers for the Liberals are rather extraordinary. The party has not been recorded at 45% support in any poll for many, many years. Has the PQ campaign really imploded to such a huge extent?

Considering the unusual nature of this poll in the context of other surveys, I will update the projection this weekend (something I would normally not do) if new surveys from CROP and/or Léger emerge that portray a different picture.

Forum was last in the field on the day of the campaign launch, on March 5. Since then, they have recorded a five point jump for the Liberals to 45% and a six point drop for the PQ to 32%. The CAQ was up one point to 13%, while Québec Solidaire was unchanged at 7%.

Since Forum does not report correctly weighted support levels by language (instead, they report the raw, unweighted, disproportionately distributed numbers) I have left them off the chart.

The poll was conducted on March 19 only. In my view, this is a problem and is one that Forum could easily avoid but continues to choose not to. The firm conducts its polls in one evening over a period of a few hours. What this means is that Forum polled 1,650 Quebecers who happened to be home between, say, 7 PM and 10 PM on Wednesday night.

The question must then be asked. Are Quebecers who are home between 7 and 10 on a Wednesday night representative of Quebecers as a whole? Are they markedly different from Quebecers home on Thursday nights or Tuesday nights, or Quebecers who were not home on this particular Wednesday? Luckily, the Montreal Canadiens were not playing on Wednesday, otherwise the poll might have been conducted among non-hockey fans only. The by-election miss in Brandon-Souris in November 2013 might have been partly caused by the Grey Cup taking place on the same night of the final poll.

That something as minor as this could even be a factor is an example of why polling should not be done on a single day. And with Forum conducting its poll in such a narrow time frame, call-backs are impossible. Telephone pollsters carry out call-backs to try to reach the people they missed when they first called in order to maintain the random nature of the sampling. Otherwise, no attempt is made to reach a representative sample of the population, just a sample of the people home on a particular night. And as Forum uses interactive voice response methodology, which has a response rate of between 1% and 3%, the need to conduct call backs is even greater.

So why not run the poll over three days, for example, instead of three hours? What is being sacrificed to get 'fresher' numbers for no particular reason? I can see the benefits of a flash poll after a big event, such as a debate, but not during an unremarkable period of the campaign.

The poll showed little real movement in Quebec City or the regions, but recorded a big swing in the Montreal area. There, the Liberals jumped seven points to 49%, while the PQ fell eight points to 30%. This was responsible for almost all of the shift in voting intentions province wide.

It wasn't the only big swing recorded. Pauline Marois's approval rating was virtually unchanged at 32%, but Couillard's increased by 10 points to 44%. François Legault's surged by 14 points to 48%, with his disapproval rating plummeting 17 points to 32%. That seems a little extreme.

I can't help but look at this poll with a good deal of skepticism, at least in terms of the precise numbers. The trendline seems consistent with other polls. But we can't definitively say whether the poll is on the mark or not until we get some corroborating numbers. Hopefully, we'll get those soon.

Thursday, March 20, 2014

Ipsos Reid gives Liberals advantage in new poll

A new poll by Ipsos Reid for CTV News emerged yesterday evening, painting a very different picture of the race in Quebec. The survey gave the Liberals 37% support against just 32% for the Parti Québécois. The message appears to be clear: the PQ's campaign is falling apart as the Liberals surge ahead. But can we really say that based on the information in this poll?

As this poll gives the Liberals a wider lead over the PQ than any survey has shown since December 2013 and includes a Liberal lead in the regions outside the two main cities, the projection has shifted quite dramatically. The Liberals are now projected to win 62 seats, or between 52 and 71 seats, while the PQ has dropped to 57 seats, or between 49 and 67 seats. This means that, if an election were held today, the Liberals would have a better chance of prevailing than the PQ, and also have a higher chance of a majority victory than the PQ. But, the race remains close enough that both parties could manage either a minority or majority victory.

A brief methodological note: this Ipsos Reid poll included numbers for committed voters, or those most likely to vote on election day. The projection has focused on those numbers rather than on the voting intentions of all decided eligible voters. For the regional results of the poll, a uniform swing based on the difference between the province wide support of decided and likely voters has been applied. Why have I focused on the likely voters? Because the purpose of the projection is to make a best estimate of likely outcomes. The outcome of the election will be decided by people who actually vote.

Ipsos Reid has not reported on provincial voting intentions in Quebec for several years. This is their first foray into the field, which makes it impossible to discern any trends. Discussion that the Ipsos Reid poll confirms the gains the Liberals have made is off the mark. We don't know what Ipsos Reid would have recorded earlier, so it is impossible to say whether the Liberals have made any gains or not. The poll does add to the evidence, however, that the Liberals are now ahead in voting intentions.

Well, sort of. This poll is the smallest one so far published in the campaign, with a sample of 810 Quebecers and 665 decided voters (the number of committed voters is even smaller, at 475). Technically, this means that the five-point lead recorded here for the Liberals is actually within the margin of error of a probabilistic sample of this size. That means that, while it is more likely that the Liberals are ahead, we cannot definitively say that they are. But when put in the context of the recent CROP poll, we are safe to say that the poll goes some way towards confirming the Liberal advantage.

The poll gave the Liberals 37% support against 32% for the PQ, their lowest result in any poll in four months (though, again, considering the sample size that puts it within range of the CROP poll that had the PQ at 36%). The CAQ followed in third with 16%, while Québec Solidaire was at 10%, Option Nationale was at 2%, and support for other parties was 3%.

Among those who said that nothing short of an emergency would prevent them from voting, the Liberals led with 40% to 33% for the PQ, 14% for the CAQ, and 9% for QS. I think this is a good attempt to measure the likelihood of voting, and the results are intuitive. For example, the likelihood of a respondent saying they'd vote increased as they got older and other polls have recorded that supporters of the CAQ and QS are less committed. These numbers would also seem to support the oft-repeated claim that the Liberals have a ballot box bonus (and they do, except when they don't, which is how most of these turnout models end up working).

The PQ led among francophones with 38%, followed by the Liberals at 29%, the CAQ at 18%, and QS at 12%. Compared to other polls, this is rather low for the PQ and high for both the CAQ and QS. Among non-francophones (or, as defined by Ipsos Reid, those who completed the survey in English), the Liberals had 80% against 6% apiece for the PQ and the CAQ and 2% for QS.

The Liberals led on the island of Montreal with 48%, compared to 25% for the PQ, 10% for QS, and 9% for the CAQ. Those numbers look very similar to those recorded by CROP. In the suburbs around Montreal, the PQ was ahead with 39% to 33% for the Liberals, 17% for the CAQ, and 9% for QS.

The numbers in the rest of the province are a little more unusual. In Quebec City, the PQ was narrowly ahead with 30% to 28% for the CAQ, 27% for the Liberals, and 8% for QS. Compared to other polls that have shown the Liberals well ahead and the CAQ collapsing, these numbers are certainly out of the ordinary. But the sample here is small: only 78 respondents, which in a probabilistic sample carries a margin of error of +/- 11%.

In the rest of the province, the Liberals were ahead with 37% to 32% for the PQ, 15% for the CAQ, and 11% for QS. The Liberals have not led in any poll in the regions since September 2013, but the last CROP poll did show the gap narrowing to just two points.

The Ipsos poll included the second choice question, finding similar results to Léger's estimations. A majority (58%) of Liberals chose the CAQ as their second choice, while PQ supporters were most likely (38%) to choose QS, and vice-versa (35%). Supporters of the CAQ still lean Liberal (46%), with just 21% choosing the PQ as their second choice.

On a series of leadership questions the responses aligned closely with voting intentions, with roughly 75% to 85% of supporters of each party choosing the leader of that party on questions relating to trust, the economy, values, etc. The one question where things did not match up so well with the province wide vote, however, was on who Quebecers would most like to have a beer with. On that score, François Legault came out ahead with 26% to 22% for Philippe Couillard, 19% for Françoise David, and 18% for Pauline Marois. If that is a proxy for likability, Legault might have the most to gain from tonight's debate.

(As an aside, I wonder if this question is gender biased, since 'having a beer' has more masculine connotations. Legault and Couillard together took 51% on this question among men, compared to 33% for Marois and David. Among women, Legault and Couillard took 45%, against 40% for Marois and David. By comparison, support for the PLQ and CAQ among men totaled 51% compared to 43% for the PQ and QS. Among women, the total support for the PLQ and CAQ was 54%, against 43% for the PQ and QS. In other words, women were less likely to want to have a beer with male leaders while men were less likely to want to have a beer with female leaders than their political support would have suggested.)

Ipsos polled on the question of sovereignty, finding that 37% (after distribution of undecideds) supported sovereignty-association (CAQ voters split 22/61 on this question). On outright independence, support dropped to 32% (PQ supporters split 67-13, QS 41-33, and CAQ 11-77 on this question). That is a little low, considering that both CROP and Léger no longer ask about sovereignty-association and have found support to be closer to 40%.

Finally, Ipsos also asked about the elephant in the campaign, Pierre Karl Péladeau. His entry apparently did very little to encourage people to vote for the PQ. Just 10% of CAQ and QS supporters said it made them more likely to vote PQ (much less than the 27% and 36%, respectively, who said it made them less likely). It did nothing to bring Liberal voters to the PQ.

But Quebecers do not necessarily dislike his entry into politics, as 46% said it was a good thing for Quebec, including 47% of CAQ supporters and even 36% of QS supporters.

Overall, what do we take from this poll? It has to be read with a good degree of context, considering the sample size (regionally, only on the island of Montreal does any party hold a definitive lead, but we don't need a poll to tell us that). When we look at committed voters, the results of the poll vary by only marginal degrees from the most recent surveys by CROP and Léger. The poll certainly suggests that the Liberals are having a good campaign, and that things have stagnated or worsened for the PQ. Is it a five or seven point lead? We'll have to wait for more polls to determine whether the campaign has really taken such a decisive turn.

Tuesday, March 18, 2014

Liberals lead in Quebec as CAQ collapses

For the first time this year, the Liberals have led in Quebec in a poll by either CROP or Léger. This particular survey, by CROP for La Presse, gives the Liberals a three-point edge over the Parti Québécois. But the PQ is holding steady - it is rather the Coalition Avenir Québec that is losing French-speaking voters to Philippe Couillard's Liberals.

The Liberals now lead in the vote projection, with 38.3% support to 36.3% for the PQ. Assuming normal polling error, the Liberals would be expected to get between 37% and 42% of the vote, compared to between 35% and 40% for the PQ. The CAQ continues to drop, now down to 13.4% (or between 12% and 14%) while Québec Solidaire continues to grow. They are now pegged at 9.6% (or between 9% and 10%).

But despite the Liberals' advantage in the popular vote, the PQ's enduring edge among francophones gives the party the edge in seats, with 62 to 57 for the Liberals. But that now puts the PQ in minority territory, though their likely seat range (56 to 70) straddles the line exactly. The Liberal range, at 50 to 66 seats, also puts them over the majority mark of 63 seats, though they are less likely to win a majority government than the PQ.

If those numbers look unusual, consider the results of the 1998 election. The Liberals won the popular vote by 0.7 points, but lost the election by 28 seats. Here, the margin is five seats with a two point advantage. And 1998 is increasingly looking like a good reference: Mario Dumont's ADQ took 12% of the vote in that election, a number that the CAQ could easily find itself with on Apr. 7 (there was no party like Québec Solidaire in 1998, however. Its closest forerunner took just 0.6% of the vote).

How much of an advantage do the Liberals need province wide in order to win a majority government? This depends greatly on the distribution of vote within Quebec and the amount of support the CAQ captures. But consider that in 2008, when the Liberals won a slim majority government of 66 seats, they beat out the PQ by 6.9 points with the ADQ at 16%. Roughly speaking, then, the Liberals can probably expect to win a majority government of their own when their lead grows to more than three or four points.

CROP and Léger have been juggling a few sponsors in this campaign. Léger first reported for Le Journal de Montréal and then this past weekend for Le Devoir and The Globe and Mail. CROP now has reported for both Radio-Canada and, this time, La Presse.

Compared to CROP's previous poll for Radio-Canada, conducted Mar. 5-8, the Liberals have picked up three points to lead with 39%. The PQ has held steady at 36%, while the CAQ has dropped four points to 13%. QS was up two points to 10%.

That drop for the CAQ was outside the margin of error, at least the margin that would normally apply to a probabilistic sample. Neither Léger nor CROP has ever had the CAQ at 13% before going back to the founding of the party.

The CAQ dropped most significantly among francophones, falling six points to just 14%. The Liberals were the ones who took advantage, picking up five points to reach 30% among this important demographic. The PQ hardly budged, leading with 43%, while QS was at 11%.

Among non-francophones, the Liberals fell eight points to 71%, with the CAQ holding steady at 10% and the PQ increasing to 9%. QS appears to have picked up the support the Liberals shed among these voters, increasing five points to 8%.

The Liberals lead on the island of Montreal with 44%, followed by the PQ at 34% and the CAQ and QS at 10% apiece (QS needs to be higher to be able to win a third seat). In the suburbs around Montreal, the Liberals increased to 39% while the PQ picked up eight points to hit 37% in the region. The CAQ saw its support cut in half, falling to 13%. QS was fourth with 9% in the region.

The Liberals also led in Quebec City, with 36% to 27% for the PQ and 23% for the CAQ. This generally confirms the two polls we have recently seen in the region from Léger. QS was up five points to 12%, oddly enough their best result in the province.

The PQ led in the regions of Quebec, with 38% to 36% for the Liberals. The CAQ had just 13%, followed by QS at 11%.

On who would make the best premier, Couillard moved ahead for the first time in any poll since December, with 27% to 26% for Pauline Marois. The two leaders traded a single point, though, and have been around this level for some time. François Legault was at 14% and Françoise David was at 7% on this question.

The CAQ has been in steady decline since the campaign began, the exact opposite of what occurred in 2012. The Liberals have taken full advantage, while the PQ's vote has been quite steady. If the CAQ continues to drop, the PQ could be in real trouble. But the dynamics in this campaign have been interestingly intertwined. Supporters of the CAQ went to the Liberals, seemingly to block a PQ victory. If a Liberal victory starts to look likely, will supporters of QS (and maybe some from the CAQ) start to move to the PQ?

Monday, March 17, 2014

Liberals gain among francophones, but PQ still in majority territory

A new poll by Léger for The Globe and Mail and Le Devoir confirmed that the Parti Québécois and Liberals have moved into a tie, as Philippe Couillard's party makes gains among francophone Quebecers. But the gap between the two parties among this demographic remains huge, giving the PQ enough support to have reasonable expectations of winning a majority government.

As the province wide results of the Léger poll hardly differed from the projection at the time, the numbers have barely moved. The PQ and Liberals are now tied with an estimated 36.9% support apiece, with likely support ranging between 36% and 41% for the PQ and 35% and 41% for the Liberals.

The CAQ stands at 14.5%, or between 13% and 16% support, while Québec Solidaire is estimated to have 8.6% support.

The seat count has moved slightly, with the PQ down one seat to 66, the Liberals up two to 54, and the CAQ down one to just three. The likely ranges give the PQ 58 to 73 seats, straddling the line of a majority government, while the Liberals should be able to win between 46 and 63 seats at this level of support. That makes a Liberal majority government possible, but not likely. The CAQ is now between two and four seats, while at its maximum level QS has moved into play in a third seat.

Most of the changes at the regional level have occurred around Montreal. The PQ dropped four points on the island, all of which was picked up by QS. In the suburbs around the island, the PQ picked up five points, three of them coming from the CAQ.

This is Léger's first campaign poll, having reported for Le Journal de Montréal just before the campaign began. The poll shows very little change since then at the provincial level, with the PQ unchanged at 37%, the Liberals up two points to 37%, and the CAQ down one point to 14%.

This is the third consecutive drop in Léger's polling for the CAQ, however. The party was at 19% in December 2013. The Liberals have been gaining, having been at 33% in January. This seems to be the story of the campaign so far, as CAQ voters drift over to the Liberals. The candidacy of Pierre Karl Péladeau, and the subsequent focus on sovereignty, may have pushed CAQ federalists over to the best anti-PQ option, whereas the PQ may have already made all the gains it can make.

Québec Solidaire was up one point to 9%, while the Greens, Option Nationale, and other parties had 1% apiece.

Among francophones, the PQ led with 44% while the Liberals picked up four points (which would be outside the theoretical margin of error of a probabilistic sample) to reach 27%. The CAQ fell to 15%, and QS was up to 10%. While that is a positive sign for the Liberals, they are still too far behind the PQ to have a good hope of winning a majority government.

The Liberals led among non-francophones with 74% support, against 10% for the CAQ and 9% for the PQ. The Liberals were also ahead in the Montreal metropolitan region with 40%, followed by the PQ at 34% and the CAQ at 12%, and led in Quebec City with 39% to 30% for the PQ and 20% for the CAQ.

In the rest of the province, the PQ was ahead with 40%, with the Liberals at 33% and the CAQ at 16%.

These regional results are quite poor for the CAQ, as they have no real region of strength. For the PQ, it gives them the ability to win quite a few seats in the Montreal suburbs and the francophone parts of the province outside the two metropolitan areas. The Liberals are competitive enough outside of Montreal to win their fair share as well, but as usual they have a lot of wasted votes on the island.

This poll included many other questions that allow us to dig down into the electorate. There are good signs in those questions for both the PQ and the Liberals, making it a poll for everybody! (Except the CAQ.)

Vote choice seems to be solidifying, with 69% saying their choice is definitive. That is up four points from Léger's pre-campaign poll. Another 20% say their choice will probably change, down two points.

The PQ has the most committed voters, with 79% (+5) saying their choice is definitive. The number sits at 72% for the Liberals (+2), 62% for QS (+14, a big jump), and 53% for the CAQ (+1). If that reflects turnout, the PQ would take 41% of the vote to 37% for the Liberals, 10% for the CAQ, and 8% for QS.

The PQ and Liberals seem unlikely to lose a lot of uncommitteds, so the second choice of their supporters is not too important. Nevertheless, 45% of Liberals (+9) would go to the CAQ while 40% of PQ supporters (+10) would go to QS. It seems that, even among the PQ and Liberals, the vote is polarizing.

But for the CAQ, who seem very likely to lose some uncommitteds, the Liberals have the most to gain. Fully 43% of CAQ voters say the Liberals are their second choice, with just 17% choosing the PQ. While the PQ does have the most to gain from wavering QS voters (35%, a gain of nine, select the PQ as their second choice) there aren't nearly as many.

On leadership, Pauline Marois is unchanged with 27% choosing her as the best person to be premier. Couillard was up slightly to 26%, while François Legault was down slightly to 14% and Françoise David up to 9%. Compared to support for their respective parties before removing undecideds, that signals that both Legault and David are slightly more popular than their parties, while Marois and Couillard are less popular.

But among francophones, Marois is the top choice for premier of 33%, with Couillard at just 18% and Legault at 16%. Among this important electorate, Marois and the PQ hold a distinct advantage.

Quebecers also think that Marois and the PQ have been running the best campaign: 31% said so, compared to 19% for Couillard and the Liberals and 10% for Legault and the CAQ (among francophones, those numbers were 36-15-11). And 40% of Quebecers (and 49% of francophones) think the PQ has presented the best team of new candidates, compared to 21% for the Liberals and 4% for the CAQ.

About half of Quebecers think the PQ will win the election, while one quarter think the Liberals will prevail. One good sign for the Liberals, however, is that a large majority of voters want to hear less about sovereignty and the charter of values, and more about the economy and healthcare. This suggests that Quebecers are tired of hearing about the two most important issues for the PQ in this campaign. But that is not necessarily negative - it could simply be that voters are decided on these issues and want to discuss something else. However, that would also mean the PQ has little more room for growth.

The Liberals are not particularly strong on this issue either. When asked who would be the best person to defend Canadian unity in a new referendum, Couillard had just 10% - behind Justin Trudeau (27%) and Thomas Mulcair (21%). It is difficult to see much enthusiasm among the PLQ voter pool.

Léger also asked about the opinions Quebecers had of some of the star candidates the parties have presented. The news was bad for most of them, as the proportion of Quebecers not knowing who they were was 83% or more for all but five of the 12 names listed. This included Andrés Fontecilla, co-spokesperson for Québec Solidaire and ostensibly the equal of Françoise David. Fully 89% of voters did not know who he was. Of those who did, opinion was split down the middle.

The big name on the list was, of course, Péladeau. Just 18% of Quebecers did not have an opinion of him, a very small number. Opinion of him was split, with 42% having a positive opinion and 40% having a negative opinion. This suggests he does not boost the PQ particularly well, since he is a polarizing figure.

The next best known candidate was Gaétan Barrette, who had an unsuccessful run for the CAQ in 2012 and is now trying his luck under the Liberal banner against Fatima Houda-Pepin in La Pinière. But just 16% had a positive opinion of him, against 48% who had a negative opinion. That is a rather bad split - take out the undecideds and it is 25/75. Does that make him a star, or a drag?

Martine Desjardins, a former student leader during the protests that dogged Jean Charest's government, is running for the PQ and has a relatively positive reputation: 31% have a good opinion of her, against 25% who have a bad opinion of her. Still, 45% did not have an opinion either way.

These were the only names listed where a majority of those polled knew who they were. Two others were recognized by more than one in five Quebecers: Loraine Pintal (actress, running for the PQ) and Diane Lamarre (former head of the order of pharmacists in Quebec, also running for the PQ). Opinion of them was also positive: 23% to 6% for Pintal and 16% to 5% for Lamarre. Altogether, this would seem to back up the polling on who has the best team of new candidates. But Péladeau has not nearly had the effect many expected to him have (including me), which is certainly a disappointment for the Parti Québécois. In fact, they may have had a better shot at winning without him.

The week will be dominated by preparations for Thursday's first leaders debate. With Marois and Couillard neck-and-neck, it will be a very important moment. Legault will have the opportunity to do something to turn his party around, as 15% of Quebecers said they are waiting for the debate to confirm their opinions. A small percentage, perhaps, but more than enough to radically change the course of the campaign.

Thursday, March 13, 2014

Post-PKP, Liberals gain in Quebec City

The first poll to emerge out of Quebec since the candidacy of Pierre Karl Péladeau was announced on Sunday was released this morning. It only looks at Quebec City, however, so we should exercise caution when drawing wider conclusions from the results. That being said, the poll does hint that Péladeau's effect may not be uniform across the province.

The poll has been added to the projection, though only for the Quebec City region itself. The effect is relatively small, but does boost the Liberals by one more seat at the expense of the CAQ. The Liberals are now projected to have between 36% and 42% support in the region, against 31% to 35% for the PQ and 19% to 22% for the CAQ. The Liberals would likely win between six and nine seats in the region at these levels of support, with the PQ winning two or three seats and the CAQ winning as many as two or as few as none.

In the chart below, I've included the latest poll from Léger (done for FM93) as well as the results of Léger's previous two polls in the Quebec City region.
The Léger poll gives the Liberals the lead in the region with 39%, a gain of seven points since Léger's pre-campaign poll. That gain is outside of the margin of error - or at least the margin of error of a probabilistic sample (Léger's polls, like those of CROP, are done online).

The Parti Québécois placed second with 32%, an insignificant gain of one point. The Coalition Avenir Québec, meanwhile, dropped five points to just 19%. That is within the theoretical margin of error, but only just, so it is likely real.

Québec Solidaire had 7% support, followed by Option Nationale, the Greens, and other parties with 1% support apiece. One in ten respondents were undecided or did not respond to this question.

This is the best number the Liberals have managed in Quebec City in polls by Léger and CROP since August 2013. The PQ, however, has been very stable. Polls by the two firms have pegged the party's support at between 30% and 35% since January (including the latest CROP). For the CAQ, this is their worst result in a non-Forum poll since October 2013 and an important drop, since earlier polls by Forum, CROP, and Léger conducted in the last few weeks all agreed that the party had between 22% and 24% support in the region.

In the context of the last three Léger polls, there is no clear trend for the Liberals. The PQ has gained, however, picking up eight points since January (which would be outside the theoretical margin of error). The CAQ has shed 10 points in just a few months.

A little more context: in the 2012 election, the CAQ took 38% support in the region, followed by 31% for the Liberals and 22% for the PQ. This means that the CAQ has dropped 19 points - a huge number. Eight of those points have gone to the Liberals and 10 have gone to the PQ. In this regard, the results of the poll are not as negative for the PQ as it would appear at first glance, though it makes them even worse for the CAQ.

But compared to the effect that the PQ hopes Péladeau will have on the race, the numbers might be a letdown. However, perhaps the results are not so surprising. If Péladeau sends the signal that the PQ can win the next election in a big fashion, that might encourage CAQ supporters to instead vote for the Liberals, who have the best shot of blocking the PQ from forming government. This is particularly the case in Quebec City, which is not nearly as pro-independence as other francophone regions in the province. The PQ may have maxed out the gains they could plausibly make in Quebec City already.

The question that the next set of province wide polls will help answer is whether the same effect will occur in the rest of the province. In francophone regions that have historically not been as opposed to sovereignty, Péladeau may give the PQ the kind of boost they are looking for. Nevertheless, this Léger poll suggests that he has the potential to be a polarizing figure. He may boost the PQ somewhat and solidify their support, but if he also encourages CAQ supporters who dislike the PQ more than the Liberals to cast their ballot for Philippe Couillard, his influence may not be as positive as Pauline Marois hopes. We'll soon find out.

Wednesday, March 12, 2014

February 2014 federal polling averages

There has been radio silence in federal polls since Feb. 19, when the last federal survey left the field. With the Quebec election in full swing, that is understandable. But despite polling being done in only the first two-thirds of the month, February still had six national and regional polls in the field interviewing a little over 7,000 Canadians. The result? Not much has changed since January.
The Liberals averaged 35% support in February, down 0.5 points from January. The party has now held a lead in national voting intentions for 11 consecutive months.

The Conservatives were up 0.3 points to 28.7%, while the New Democrats were down 0.1 point to 24.2%. The NDP has been very stable, posting an average of 23% or 24% in five consecutive months.

The Bloc Québécois was up one point to 5.9% and the Greens were down 0.8 points to 4.6%. Support for other parties stood at 1.6%.

Normally I would try to compare apples-to-apples by looking at the last time the firms in the field in February were active. But since one of those firms included Angus-Reid, which had previously reported in January 2013, it is not a worthwhile exercise.

Moving west to east, the race continues to be very close in British Columbia. The Liberals dropped 0.2 points to 30.2%, followed by the Conservatives at 30.1% (+1.2) and the NDP at 29.2% (+1.6). Doesn't get much closer than that. The Greens were down 1.4 points to 9.2%, their best result in any region of the country.

In Alberta, the Conservatives dropped for the third consecutive month to 46.8%. That represented a drop of 3.8 points since January and the worst result for the party in the province since May 2013. The Liberals were up 0.8 points to 28.4%, the first time the Liberals maintained more than 24% support in two consecutive months since before January 2009 (other spikes tended to be anomalous). The NDP was up 1.5 points to 16%, while the Greens were unchanged at 5.7% support.

The Conservatives were up 4.3 points in Saskatchewan and Manitoba to 39.9%, while the Liberals were up 2.9 points to 33.1%. The NDP put up their worst result since May 2013, dropping 6.4 points to 21.4%. The Greens were up one point to 4.2%.

The Liberals have been very steady in Ontario, with an average of either 37% or 38% support over the last five months. They were unchanged from January at 37.4% in February. The Conservatives dropped 1.2 points to 33.7%, while the NDP was up 1.8 points to 23.4%. The New Democrats have also been stable at 23% or 24% over the last five months. The Greens were down 0.3 points to 4.3%.

I know the monthly average tracking chart can be hard to read at the regional level, so each month I'll highlight one province or region with a magnified version. This month we'll look at Quebec, fittingly enough. You can click on the chart to see a larger version of it.

The Liberals continued to lead in the province with 32.8%, but that was a drop of 3.9 points from January. The New Democrats were down 0.2 points to 28.6%, but this marked the first time the NDP held more than 27% support in two consecutive months since before Justin Trudeau became Liberal leader. The Bloc Québécois was up 3.9 points to 22.4% (note that, if we assume every BQ voter also supports the Parti Québécois, Québec Solidaire, or Option Nationale, that leaves some 23% of Quebecers who support a sovereigntist party at the provincial level but a federalist party at the federal level). The Conservatives were up 0.2 points to 12.7%, a level of support they seem to be stuck at - the party has averaged 12% or 13% support for eight consecutive months. The Greens were unchanged at 3%.

In Atlantic Canada, the Liberals put up their best result since before January 2009, with a gain of five points to 57.5%. Both the Conservatives and New Democrats have dropped in three consecutive months, with the Tories down 0.6 points since January to 20.5% and the NDP down 3.7 points to 18.2%. The Greens were down 0.2 points to 3.2%.

With these levels of support, the Liberals would win 133 seats, a drop of three since the January projection. The Conservatives would win 121 seats (+1) while the New Democrats would take 73 seats (-5). The Bloc would win nine seats, a gain of seven, while the Greens would win two.

The Liberals lost seven seats in Quebec since January, but picked up two in Alberta and one apiece in the Prairies and Atlantic Canada. The Conservatives gained five seats in the Prairies, but lost one in both Atlantic Canada and Ontario and two in Alberta. The New Democrats were down six seats in the Prairies, but up one in Ontario.

Voting intentions have mostly settled in for the time being. But the election campaign in Quebec, particularly if the Parti Québécois wins a majority government, could shake things up considerably. The results in the pending Ontario election could also have some influence on the federal numbers. It will be interesting to see where things will stand at the end of the spring.

Friday, March 7, 2014

Everything we thought we knew was wrong?

It just wouldn't be an election in Canada anymore without an unusual poll from Forum Research raising eyebrows and over-turning the weight of evidence collected by other polling firms. Last night, Forum did it again in a new poll first reported by the Montreal Gazette.

The poll gave the Liberals 40% support against 38% for the Parti Québécois and 12% for the Coalition Avenir Québec. In the larger scheme of things, those results are not too far from the latest findings from CROP and Léger. But underneath the top line results, the numbers mark a serious departure from what the two Quebec-based firms have been finding to be the case for months. Or do they? More on that below.

The vote projection has moved rather dramatically as a result of this poll being added to the aggregate. The PQ has been bumped up just 0.8 points to 37.8%, but the Liberals have jumped 2.7 points to 37.6%. Much of that came from the CAQ, which fell 1.8 points to 13.6%.

It hasn't shifted the seat projection very much, however. The PQ has slipped one seat to 68, with the Liberals at 50 seats. The ranges have changed a bit more, with the PQ dropping from 62-81 seats to 61-78 and the Liberals increasing from 36-56 to 42-60. The likely ranges still give the PQ the victory, but a majority is less assured than it was on March 5. The CAQ's range has dropped from 5-7 seats to 1-5.

Now to the Forum poll, which was conducted on the first day of the campaign via interactive voice response. On the one hand, it is good to have a different methodology in the field. Both CROP and Léger use online panels, though CROP reverted to live-callers in the 2012 election campaign. We will have to see if they do so again. On the other hand, Forum's one-night flash polls are not the best way to go about collecting a sample. There's no time for call-backs, so instead of a poll of eligible voters in Quebec it is a poll of eligible voters in Quebec who happened to be home on Wednesday night. Are they different than those Quebecers home on Thursday nights?
As mentioned, the top line results are not too unusual. Though the Liberals are certainly higher than the 34% to 35% recently recorded, perhaps the start of the campaign has boosted them. The 38% for the PQ is well in line with other results, as is the 7% for Québec Solidaire. The 12% for the CAQ is low (Forum always seems to have the CAQ lower than other firms), but it is not much different from the 15% and 16% registered in the last polls by CROP and Léger.

But then we get to the linguistic breakdowns. Forum has the PQ at 41% among francophones, low compared to CROP and Léger but not unreasonable (that is where they were in polls conducted between October and January). But Forum has the Liberals at 36%. That is an incredible number - though almost exactly where Forum has the party in their last polls in September and October of last year. But just as those results were at odds with what CROP and Léger had at the time, they are again. According to those two firms, the Liberals have been at or below 25% among this demographic for the last two months. The last time either firm had the Liberals as high as 36% was years ago.

The non-francophone score is also quite high for the Liberals: 89%. But that is not an implausible number, as CROP has had the party that high in the last year.

It would seem that these numbers from Forum are unweighted, because otherwise they make no sense. If the Liberals are at 36% among francophones (who make up roughly 80% of the population) and 89% among non-francophones (who make up the remaining 20%), they would be at roughly 47% support province wide. The PQ's score would translate into about 34% support. This is likely due to the over-sampling of Quebec City, where the Liberals are at 46% support. This would have been done to build a useable sample. But it would seem that Forum did not re-weight their reported francophone numbers to take that into account (they would have to in order to calculate the provincial numbers, however). In other words, the number for the Liberals among francophones is not weighted correctly - that is the raw number, meaning the Liberals are NOT at 36% support among francophones.

Instead, the Liberals would seem to need to be around 28% for the overall numbers to make any sense, with the PQ around 47% and the CAQ around 16%. This would put Forum's numbers more in line with what other firms are showing.

The big difference, then, comes primarily in Quebec City. The numbers in Montreal and the regions are not dissimilar from those of the other firms, but the Liberals have not been put higher than 34% in any polls by CROP and Léger since December. The margin of error for a sample of 144, which is what Forum reports for the Quebec City region, is +/- 8%. So, it is possible that the Liberals are pushing the limits of that MOE in the provincial capital.

So, perhaps the poll is not as absurd as it appears on first glance. But these are things to keep in mind going forward: Forum's reported support for francophones seems unweighted, meaning the numbers are worthless if they are not regionally weighted. Focus must thus remain only on the regional and overall numbers if we are to take anything useful from the Forum polls.

Thursday, March 6, 2014

And if the polls change in Quebec...

Not one, not two, but at least three articles appeared in the last 24 hours reminding us that campaigns matter and that polls conducted at the start of a campaign are not necessarily reflective of the final result. Oh, and hey, those polls recently completely blew, didn't they? Except Nova Scotia, of course. But that is an ugly fact getting in the way of a narrative! Sometimes I feel that, 10 or 20 years from now, I'll still be reminded that Danielle Smith and Adrian Dix did not become the premiers of their respective provinces (though Smith still might). Maybe by then I'll also constantly need to refute the claim that pollsters don't call Moon bases. They do!

Obviously, the polls in Quebec will likely change. There are some 30 days of campaigning to be done, with one or more debates. Plenty of opportunity for a leader to fall on his or her face, a high-profile candidate to say something stupid, or the momentum of the campaign to escape the carefully planned agenda of each of the parties. Readers of this site know all about this possibility - I don't need to spell it out to you that simply because the site is projecting a majority for the Parti Québécois now means they will get one on April 7.

But can we try to estimate how the polls might move, if they do? No set of calculations based on anything plausible would have predicted a surge of the NDP by some 25-30 points in the 2011 federal election campaign in Quebec, so the range of possible outcomes in the province are virtually limitless. But let's try to do our best, using yesterday's Léger/JdM poll.

One of the questions asked by Léger was about which party each respondent would support if they were not able to vote for their first choice. The results were interesting, though largely intuitive.

The chart to the left shows how supporters of each of the four main parties would cast their ballot if they couldn't vote for the party they intended to support. The light gray band represents those who responded "others", "would not vote" or "I don't know", and those who refused to answer (just 1%).

Liberal supporters mostly went to the CAQ as their second choice (34%) or the Greens (13%). PQ voters moved over to Québec Solidaire (30%) or the CAQ (18%), while CAQ voters chose the Liberals (44%) and the PQ (19%) second. Supporters of Québec Solidaire primarily selected either the PQ (26%) or the Greens (16%) as their second choice.

The CAQ had the fewest number of respondents choosing one of the lesser-committed categories, suggesting they have fewer my-party-or-no-party supporters than either the PQ or the Liberals.

The question asked by Léger about how committed voters were to their party reflected this. While 74% of PQ voters and 70% of Liberal voters said their choice was definitive, just 52% of CAQ supporters and 48% of QS supporters said the same thing. About a third of CAQ voters said they would probably change their mind - compared to just 14% of PQ voters.

So let's use these two poll questions to estimate how the polls may move if various scenarios play out.

First, let's assume all those voters who said their choices were not definitive cast their ballot instead for the party they said was their second choice (or they stay home if they did not choose one). Recall that the overall results of the Léger poll were 37% for the PQ, 35% for the PLQ, 15% for the CAQ, and 8% for QS.

If all those uncommitted voters drifted, the results would not be much different:

35% - Parti Québécois
33% - Liberals
15% - Coalition Avenir Québec
9% - Québec Solidaire

With those lower numbers, however, the PQ would have a harder time cobbling together a majority government. But they'd likely still be leading by some 20 points among francophones, so it would be far from impossible. Québec Solidaire would have a good chance of electing a third or even fourth MNA.

Next, let's assume that the uncommitted supporters of the smaller parties (Québec Solidaire, the Greens, and Option Nationale) decide that the race is really between just the PQ, the PLQ, and the CAQ. Those supporters decide to vote for their second choices (but not other small parties) or stay home. Again, the results only change marginally, though QS would have difficultly holding their two seats:

40% - Parti Québécois
37% - Liberals
16% - Coalition Avenir Québec
4% - Québec Solidaire

Now, a PQ majority is far more likely. And with both the PQ and PLQ increasing their vote share, the CAQ would have even more trouble holding their seats.

That being the case, what if voters decide that the CAQ does not have a shot, and that they must support either the PQ or the PLQ? Uncommitted CAQ and smaller-party supporters vote for their second choices (only if they were the PQ or PLQ), or stay home. Now it gets a bit more interesting:

43% - Parti Québécois
42% - Liberals
8% - Coalition Avenir Québec
4% - Québec Solidaire

This is the sort of scenario that Philippe Couillard hopes plays out as the campaign goes on. If the CAQ vote was cut in half, that boosts the Liberals tremendously. And since most of the CAQ's support is among francophones, that is the electorate in which the Liberals would make their gains. This no longer makes a PQ majority as likely. Though it does increase the likelihood that the only parties in the National Assembly would be either the PQ or the PLQ.

But while Couillard may be hoping the campaign takes that kind of turn, he is the rookie leader. What if Liberal support collapses? All the uncommitted Liberals become disappointed with Couillard's performance, and cast a ballot for their second choices (if they are the PQ or CAQ) or stay home. Uncommitted supporters of the smaller parties decide that they should either vote for the PQ or CAQ or not bother.

44% - Parti Québécois
27% - Liberals
22% - Coalition Avenir Québec
4% - Québec Solidaire

Now a PQ majority is guaranteed. The CAQ would have a good shot at re-electing most of their incumbents, and the Liberals would be reduced significantly. But there aren't enough uncommitted Liberals to move the CAQ out of third place. There will need to be a very important shift in voting intentions and underlying support for the CAQ for François Legault to have a real shot at winning, or even placing second. The fundamentals simply don't point towards that happening.

But what if something more unexpected occurs? Pauline Marois has never been loved by voters. What if she does something that seriously turns voters off during the campaign, and support for the PQ collapses? In this scenario, uncommitted PQ and smaller-party voters cast a ballot for their second choices (if they are the PLQ or CAQ) or stay home.

42% - Liberals
31% - Parti Québécois
20% - Coalition Avenir Québec
4% - Québec Solidaire

The Liberals do have the potential to win if the CAQ vote collapses, but the PQ would stand to gain from such a collapse as well, making it still too close to call. Only if the PQ's vote collapses a great deal can the Liberals win (the results of this scenario are relatively close to those of the 2008 election). But the PQ still has a considerable base from which to work.

So based on the information we currently have to go on, a number of scenarios are reasonable: a large PQ majority, a toss-up majority/minority for either the PQ or the PLQ, or a PLQ majority. But a scenario that puts François Legault near the premier's office - or even the job of leader of the opposition - is currently not in the cards.

Wednesday, March 5, 2014

PQ begins election in majority territory

And we're off and running in Quebec's provincial election campaign, starting today and culminating with a vote on April 7, 2014. Pauline Marois and the Parti Québécois will be asking Quebecers to give them a majority government instead of the minority one they were handed less than two years ago. The polls suggest she starts the campaign with a good chance of getting her wish.

The Léger/Journal de Montréal poll out this morning goes a long way to confirm the sort of numbers that CROP had in February. The gap is smaller, at two points instead of five, but the PQ enjoys the same yawning margin over the Liberals among francophones, who decide elections in Quebec.

The current projection, based primarily on the Léger poll but also incorporating other polls going back to the beginning of the year, gives the Parti Québécois 37.5% support to 34.9% for the Liberals, 15.4% for the Coalition Avenir Québec, and 7.9% for Québec Solidaire. With these levels of support, and taking into account the potential for polling error, the PQ would likely win between 62 and 81 seats. With 63 seats required to win a majority, the polls point almost exclusively to a PQ majority government at the moment.

The Liberals would win between 36 and 56 seats, with the CAQ winning between five and seven and QS taking two. The precise projection - the result should normally fall closest to it - is 69 seats for the PQ, 49 for the PLQ, 5 for the CAQ, and 2 for QS.

Check out the Quebec projection page for the full details, including regional breakdown and riding-by-riding results. Tracking charts will be added as the campaign continues and new polls emerge.

The model is virtually unchanged from the one used in the Nova Scotia campaign, with one important difference. For that campaign, the minimum and maximum projections were based on the worst polling errors in recent years. In other words, the model was assuming the exact same sort of errors as in Alberta and British Columbia would take place to derive these maximum and minimum estimations.

That was a little crude, so instead this time the minimum and maximum ranges are meant to estimate 95% of potential results. These are calculated based on polling errors in the past, so it does take into account errors like in Alberta and B.C. But it leaves a little room for further error, which is always a possibility. Put simply, 19 times out of 20 the results of the election should fall within the projected minimum and maximum ranges. For a more precise estimation, see the chart below:
What the chart above shows is the likelihood of results falling between the various projected ranges. So, for example, if an election were held on March 3 (the last day of polling), there is a 75% chance that the result would be higher than the projected average. There is a 90% chance that the Greens and Option Nationale would get lower than the projected average. There is a 5% chance that the results will fall outside the minimum and maximums for any of the parties.

A full explanation of the methodology can be found here. A link is also omnipresent in the right-hand column.

Now, let's briefly take a look at the poll.
The poll shows no major change from Léger's last survey from mid-January. It gives the PQ 37% support against 35% for the Liberals and 15% for the CAQ.

Recall that the CROP poll which resulted in much spilled ink about a PQ majority had the split at 40% to 35%, so the Léger poll is only marginally and insignificantly different. Of most consequence, electorally speaking, is the francophone vote: 45% for the PQ against 23% for the Liberals. CROP had it at 47% to 24%, which was remarkable itself. So Léger has confirmed that the PQ has made big strides among francophones, and it could be a crushing advantage.

The poll had a few other tidbits that are worth paying attention to. The PQ and PLQ had the most committed voters, with 74% and 70%, respectively, saying their choice was definitive. For the CAQ and QS, only about half of their supporters had definitively made up their mind. That isn't surprising for a small party like QS, but it is incredibly worrisome for François Legault.

The Liberals have much to gain if those CAQ voters drift away: 44% said that the PLQ was their second choice, with only 19% saying the PQ was their second option. The PQ does have room to grow among supporters of QS and ON, but that is a much smaller voter pool.

On leadership, Marois topped Philippe Couillard with 27% to 25%. That is a gain for Couillard, which is good news for him. Less glowing, though, is that he scored just 17% among francophones - putting him in a tie with Legault. If Couillard struggles to make inroads among this demographic, the Liberals have no chance of winning. Quebecers seem to agree, since 46% of them think the PQ will prevail. Only 26% think the Liberals will win.

The Liberals should be doing better, though, as they score higher than the PQ on many issues: healthcare (29% to 24% for the PQ), jobs (33% to 25%), tackling the deficit (25% to 20%), and infrastructure (32% to 24%). These are all important issues in an election campaign. Perhaps the Liberals will be able to take advantage as the campaign unfolds, but the PQ has its own issues of strength: reasonable accommodations (33% to 21% for the PLQ), tackling corruption (26% to 18%), and protecting the French language (55% to 9%). It is obvious how the two parties will design their campaign strategies, with the Liberals focusing on issues of the head and the PQ on issues of the heart. 

Monday, March 3, 2014

Steep drop in support for Redford and Alberta Tories

New numbers from a Léger poll for the Calgary Herald and the Edmonton Journal show the Alberta Progressive Conservatives suffering a significant drop in support since the fall. But that is nothing compared to the fall in approval ratings experienced by Premier Alison Redford.

Léger was last in the field in mid-October, when the Tories and Wildrose were in a very tight race. But since then, the PCs have dropped by six points to just 25%, while Wildrose has increased by five points to 38%. These shifts would be statistically significant with a probabilistic sample of this size (the poll was done online, however).

The Liberals were down two points to 16%, while the New Democrats were up one point to 15%. Support for the Alberta Party stood at 3%, while 2% of Albertans said they would vote for another party. Of the entire sample, 24% was undecided.

These are horrible numbers for the Progressive Conservatives. They have now dropped in four consecutive Léger polls going back to January 2013, when the party was at 40% support. The score of 25% is the worst the Tories have managed in any poll since November 2009. Wildrose has been the major beneficiary, though it should be pointed out that since April 2013 the party has been polling in a relatively tight band of between 33% and 38% support.

The Tories dropped primarily in Edmonton, where they were down seven points to just 19% support. Wildrose was up seven points to 29%, followed by the NDP at 27%. The Liberals, down six points, were tied with the PCs for third with 19% support.

In Calgary, Wildrose led with 41%, while the Tories were down to 28% and the Liberals were steady at 18%. The NDP had 7% support in the city.

In the rest of Alberta, Wildrose increased to 44% and the Tories dropped to 29%, expanding the gap between the two parties by 10 more points. The Liberals and NDP were at 12% and 10%, respectively.

Due to a very inefficient vote in the two major cities, the PCs would be reduced to third party status with these levels of support. Wildrose would win a majority with 58 seats, virtually sweeping Calgary and the rural parts of the province. The Liberals would win 11 seats, 10 of them in the two main cities, while the NDP would win nine seats, almost all of them in Edmonton. The Tories would be reduced to just nine seats as well, generally divided between the three regions of Alberta.

(Yes, we all remember the results in 2012. But the model performed well when the actual numbers were plugged into it - in other words, if Léger's poll matched an election's results exactly, these seat estimates would turn out to be quite close to the actual result. A two-seat margin between the Liberals and PCs/NDP, however, is hardly decisive.)

The major problem for the Progressive Conservatives appears to be Redford herself. Her approval rating (Léger last inquired about this in September 2013) dropped 12 points to just 20%, among the worst numbers put up by a sitting Alberta premier in recent memory. Her disapproval rating was up 12 points to 64%, and just 35% of PC voters from 2012 said they approved of her performance. That is ghastly.

Danielle Smith of Wildrose had an approval rating of 39%, with her disapproval rating dropping four points to 33%. Among Wildrose voters, her approval rating was 85%.

Raj Sherman of the Liberals had an approval/disapproval rating of 29% to 28%, with Brian Mason of the NDP enjoying a 32% to 23% split (the best net rating of the four leaders). Among their 2012 voters, Sherman's approval rating was 68% and Mason's was 78%.

But Redford is not less popular than her party (among decided respondents, her approval rating was 24%) so it is unclear if her resignation would change things. It is clear, however, that she is a big reason for the dramatic decline in support for her party over the last year. In a little more than 12 months, she has lost almost two out of every five supporters. And the timing is certainly bad, as Smith and Wildrose move more towards the centre-right. Unless Redford can somehow turn things around, or if she is replaced by someone that can give the Tories new life (difficult after 43 years in office), the stars may be aligning for a change of government at the next election. And then, one assumes, we can expect several decades of Wildrose dominance. That is just Alberta's way.