Making the news

What will tomorrow’s polls say? Depends on today’s coverage.

In last week’s election debate, NDP Leader Jack Layton got off the best (or most obnoxious) zinger of the night: “Where’s your platform?” he demanded of Conservative Leader Stephen Harper. “Under the sweater?” Like it or not, the point was well-made: in week four of the federal election campaign, the Tories had yet to release one. Major policy announcements were also lacking, as Harper struggled to reassure Canadians the economy was fundamentally sound. In the debate, as in the campaign, the Tory leader played it safe: his favoured tactic seemed to be “buckle down and smile” as opposition leaders piled on, says Stuart Soroka, co-director of the Media Observatory at the McGill Institute for the Study of Canada.

But if Harper plans to make any real gains in the Oct. 14 vote, this defensive strategy may not serve him well. According to the 2008 Federal Election Newspaper Analysis Project, a weekly election feature in Maclean’s, the PM failed to attract much more media attention in the fourth week of campaigning than he did in the first. This suggests he’s failed to build momentum in his campaign: shifts in election coverage can predict shifts in public opinion by several days, Soroka notes.

After the debate, “nobody was talking about Harper winning,” Soroka says. Whether or not debate performance matters, as we entered the final week before the election, a Tory majority was looking more elusive than ever.

Soroka’s team conducts an automated analysis of election coverage in seven newspapers: the National Post, Globe and Mail, Montreal Gazette, Toronto Star, Calgary Herald, Ottawa Citizen and Vancouver Sun. So far, “it’s been a remarkably flat campaign,” Soroka says. His data shows that, in week four, 41 per cent of all articles mentioned Harper first; compare that to week one, when 40 per cent did.

When it comes to getting headlines, though, the Liberal leader has done even worse. Last week, 12 per cent of articles mentioned Stéphane Dion first, slightly less than in week one, when 16 per cent did. (Their parties are closer: 34 per cent of articles mentioned the Conservatives first last week, while 21 per cent mentioned the Liberals.)

Dion’s tepid campaign did get a boost from the debates. On Oct. 1 (the day of the French-language debate), 42 per cent of articles mentioned Harper first, while just nine per cent mentioned Dion. By Oct. 3 (the day after the English debate), Dion had 17 per cent of first mentions. That didn’t last—on Oct. 4, the Liberal leader’s coverage plummeted to eight per cent, while Harper was at 44 per cent.

Immediately after the debates, Dion got better reviews in the dailies: his net tone (a score found by subtracting negative from positive newspaper mentions) jumped from -1.4 to 0.8 from Oct. 1 to 3, while Harper’s score saw a less dramatic increase from -0.7 to 0.1. Entering the final week of the campaign, Harper and Dion had the overall same net tone of -0.1.

In both 2004 and 2006, Harper’s campaign momentum stalled. Soroka’s data suggests it’s happening again. “There comes a time where holding steady isn’t enough,” he says. “At some point, somebody has to say something good about you—not just that you’ve survived so far.”