Tonight is the big show: the English language debate at 9 Eastern/6 Pacific. The dynamics here are a little different than in French. The Conservatives are in a stronger position outside of Quebec, so Harper mainly has to stay out of trouble. I imagine we'll see the relaxed prime-ministerial sweater vest version of Stephen Harper again. All four leaders will be gunning for Harper again. I don't expect we'll see the NDP and Liberals squabble too much over who is the real opponent to Harper. The way the Conservatives are poised to break through in British Columbia means they both have to dump on Harper big time. Harper just has to weather the storm. It's pretty much impossible for an incumbent prime minister to "win" the debate; you win by not losing, which is about the best you can hope for.

A CanWest/Global National poll of francophone voters found that Dion won the debate, Duceppe finished second, and Harper third. The sample size (637) was fairly small, but that's still an interesting and somewhat surprising finding. I didn't think Harper was that bad, but perhaps the "sit back and look prime ministerial" approach didn't play as well in French as it did in translation.

I watched the French debate tonight (in translation; I'm an Albertan!) and thought it wasn't bad. I really liked the table: I thought it was a lot better than the stuffy podiums and made it feel more intimate and interactive. I usually hate the questions from "average" Canadians, but my favorite moment was actually where the leaders had to say something nice about the leader to their left. It was different and caught my attention.

In an hour and a half, the French language debate will be starting and it will be interesting to see how it unfolds. This really is the contest between Dion, Duceppe and Harper. Although the NDP is making some gains in Quebec, I don't see Mr. Layton as particularly relevant and Ms. May's struggles in French mean she'll be marginal for much of it.

As I've watched the Liberals sputter through this and the previous two elections, I've reached a conclusion: it would be good for the Liberals to lose this election. And I mean really lose it, as in a Conservative majority government. This is not a judgment about whether the Liberals would do a good job in government. That might or might not be the case. But for the Liberal party as a party, it would be healthy to have a break where they aren't in government and have no prospect of getting there for four years.

In my last post, I noted that Wilfird Laurier's Barry Kay had the Conservatives just a couple of seats away from a majority. His latest projections put the Conservatives right at the magic number of 155. This has led Gilles Duceppe to call for support for the Bloc in order to deny the Conservatives a majority. It will be interesting to see whether those calls have any resonance with the electorate. Stephen Harper has worked hard to put a moderate face on the Conservatives (often to the dismay of the more ideologically inclined members of his party).

Professor Barry Kay has posted his latest projections of how vote shifts will translate into seats and he has the Conservatives at 153 seats, tantalizingly close to that magic number of 155. It will be interesting to see if there is any fear of a Conservative majority out there and if that sparks a backlash against the party in the last half of the campaign. It will also be interesting to see whether Dion and Layton shift their campaigns from talking about forming a government to denying Harper a majority.

I had to take a little hiatus from blogging about the election to finish up a paper for a conference this coming weekend. In this paper, my co-author, Lisa Young of the University of Calgary, and I look at patterns of fundraising and spending since the adoption of the changes to Canada's electoral laws in 2004. You may remember that the government banned corporations and trade unions from donating to political parties, enriched the election expenses reimbursement given to qualifying parties (those who get at least 2% of the vote nationally or 5% of the vote in the districts in which they run candidates), made individual donations to poltiical parties more advantageous by strengthening the individual political contributions tax credit, and gave every qualifying party (2%/5%) $1.75 per vote per year. As the media have reported, the Conservatives have been raking in money like crazy.

Over the last decade or so, there's been a lot of speculation (and even some solid empirical research) about the impact of the Internet on the political process. Much of it has been about grandiose claims over whether the Internet is going to usher in a new golden era of citizen participation and engagement. Short answer: probably not. Jared Wesley's post on Facebook is right on the money.

Green leader Elizabeth May unveiled her party's platform yesterday and besides the expected range of environmental policies, there is a promise to bring in income-splitting. For those happy people who don't follow the nuances of tax policy, income-splitting would allow families to pool their income and report it jointly for tax purposes. If a person makes more than her spouse, she could transfer that money over to him, where it would be taxed at a lower rate. This would mean significant tax savings, expecially for families where one person does not earn any income.

One of the discussions I often end up having with my classes is how much local campaigns and candidates matter. There is a school of thought, best represented by the work of political scientists Ken Carty and Munroe Eagles, that argues that local campaigns can be very important and make a difference. Another school of thought suggests that in an age of electronic, media-intensive, leader-focused campaigns, local candidates matter very little. We often joke in class that one great way to settle this debate would be to convince a candidate not to campaign and see what happens. Unfortunately, we conclude, it would be difficult to convince someone to do this.

One of the challenges of interpreting public opinion polls and predicting election outcomes in Canada is the single member plurality electoral system that only loosely translates popular vote into seats. That's why models that try to predict seat totals are interesting. One of the oldest and most successful is done by the Laurier Institute for the Study of Public Opinion and Policy. Political scientist Barry Kay has developed a model that looks at regional shifts in party support as reported in polls and maps that on what we see at the district level in the previous election.

The election was pushed off the front pages by the crisis surrounding the financial industry in the United States, the subsequent carnage on the stock markets, and the implications for the global economy. What impact might all of this have on the election? Political science research into the economy and voting provides us with some clues about what to look for.

So, who won the first week? I'd have to go with the Conservatives. It certainly wasn't a perfect week for them: some overeager election staffers sent them off message and into damage control mode. However, Harper and the party dealt swiftly with pooping puffins and with e-mails about the political motivations of grieivng parents of dead Canadian soldiers. Furthermore, the party also had the most significant policy announcement of the week in the announcement of the 2001 end to Canada's mission in Afghanistan. I'd give them an A-.

All right, a disclaimer before we begin discussing the latest poll numbers from Decima-Harris and Nanos. There's a lot of trouble with how the media reports polls: they tend to do a horrible job of reporting margins of error on regional poll results, as Melanee Thomas detailed in her recent blog posting. Also, in a country with a heavily regionalized electorate like Canada, the national poll numbers don't mean a whole lot. If the Conservatives are up nationally because they the remaining 35% of voters in Alberta who haven't already succumbed to their charms have now decided to join the majority, it won't mean much for seats because the party can't win any more seats there. The key battlegrounds are British Columbia, Quebec and Ontario.