The gap may have narrowed. But the federal election campaign still sees a disconnect in the polls between perceptions of leadership on the one hand and voting intentions on the other, especially where it relates to the Liberals.
Although public opinion surveys can measure voters' appreciation of the leaders' strengths and weaknesses, and in this campaign they have been quite consistent, they are an imperfect predictor of party choice on election day.
Conservative Leader Stephen Harper has outscored both Liberal Leader Stephane Dion and NDP Leader Jack Layton (along with the other two party leaders) by substantive amounts on the question of who would make the best prime minister, and in the other measures of leadership such as the Nanos Research ratings of trust, competency and vision.
Indeed, on Sept. 3 - just before the election call - Ipsos found that 50 per cent of Canadians believed Mr. Harper would make the best prime minister. Around the same time, Harris/Decima said 47 per cent "had a positive impression of him." And yet only between 35 and 40 per cent of those same voters are prepared to vote Conservative.
While we have seen the Conservative vote share edge towards the number needed for a majority of seats in the House of Commons, it is falling well short of a majority of the electorate.
In that Sept. 3 Ipsos poll, the NDP suffered from much the same disconnect. While 31 per cent said Mr. Layton would make the best prime minister, just 19 per cent said they would vote NDP. That gap has widened as the campaign has progressed.
For the Liberals, the leadership/party choice disconnect has been more fortuitous, at least until recently. Ipsos found only 20 per cent of the respondents said Mr. Dion would make the best prime minister, but 27 per cent would vote Liberal. Nanos had the Liberals and Conservatives in a statistical tie around 35 per cent each on writ's eve, although only 14 per cent said Mr. Dion would be best prime minister. (Nanos now has the Conservatives 9 points ahead.)
How can we square results showing a Canadian electorate that clearly favours Mr. Harper as prime minister, but won't vote for him at a level that will reliably make him prime minister? And how can there be an even greater gap for Mr. Layton?
The answer is, in short, is that voting is not only about leadership. This might seem obvious, but it is worth a reminder especially in this campaign where the apparent (or unapparent) leadership qualities of the major party leaders have had such prominence.
While it is obviously important, just how much of a role leadership plays in people's voting decisions is the subject of an ongoing dialogue among pollsters, academics and other political analysts.
Indeed, in a September 25th Ipsos poll, respondents said the most important factor in determining how they will vote is "the parties' positions on important issues" (62 per cent), not what they "think about the party leaders" (21 per cent) or their local candidates (17 per cent).
These results are consistent with what voting behaviour research tells us, at least insofar as they suggest that party leadership is just one — and rarely the most important — factor. (This research also shows, however, that "issues" are rarely very important to voting. Voters often have strong partisan attachments that effect on their vote, independent of both leaders and issues, although recent research suggests partisan attachment is weakening. )
Data in the Canadian Opinion Research Archive at Queen's University shows the current gaps between leader and party have lots of precedents. As much as 30 per cent of those preferring a certain party leader regularly vote for another party.
