The Daily Exchange: Do election campaigns actually matter?

For years political scientists claimed that election campaigns don't matter – that is, that campaigns have no impact on election results.

Their position defied experience and common sense. After all, if campaigns do not matter, then how do we explain mid-campaign vote swings, such as the Liberals' collapse at the feet of Jack Layton's NDP (2011 federal election) or the unexpected defeat of David Peterson (1990 provincial election)?

In the face of evidence, even political scientists are coming around to the view that campaigns matter, at least sometimes.

It’s in this context that one should examine the recently released opinion polls. Anything can happen in a campaign, and frequently does.

How should the parties react? Not at all, to published polls.

The reason is that political parties don’t rely (or shouldn’t rely) on publicly-published poll results.

Each party has its own pollster, typically independent of the campaign to ensure that the leader and campaign management receive objective analysis. A party collects its own data, based on consistent survey questions and measured against benchmarks that are meaningful to the party, and relies on those results to determine and assess strategy.

Numbers from two surveys, using different designs, questions and methodologies, are not directly comparable.

Moreover, each party’s data should be fresher than results reported in the media. What’s reported in the news reflects a few days’ lag, while each provincial campaign manager will have access to data collected the night before.

That’s not to say the published polls are completely ignored. A party should look to the newspaper polls to ensure that its internal data make sense and significant shifts or trends are not being missed. (When internal polling differs markedly from external polling, a campaign must be satisfied it can account for the discrepancy.)

Of course, candidates and volunteers in the field, and reporters covering the campaign, don’t have access to a party’s internal polling. They must rely on what’s heard and read in the news. Consequently, published poll results can affect the “three M’s” of morale, media coverage and momentum.

I’ll continue the analysis tomorrow.

- Guy Giorno, former chief of staff to Mike Harris and Stephen Harper

 

Response from Erika Mozes, former senior adviser to George Smitherman and Gerard Kennedy:

Guy is right about one thing: campaigns do matter.

During campaigns, key events or moments – good, bad, and ugly – happen. It’s when “ordinary” Ontarians (as I defined in my last post) pay even closer attention to political party policy positions, and the quality and tone of their leaders. Importantly, they get information about all of this from the media, a fact I’ll focus on in this posting.

Notwithstanding the talk about social media, the most influential political reporting in Ontario is still done by reporters at “old media” outlets. Reporters – particularly print reporters – create and filter the majority of information we receive about politics and politicians.

Reporters love opinion polls – particularly the “horserace” polls that talk about party standings, rather than issues. Horserace polls can have a profound effect on campaigns, because momentum – real or perceived – interests the reporters who cover politics. That’s why polls, particularly bad ones, can take on a life of their own.

A story about a bad poll for “Party A” prompts reporters to ask Party A’s leader for their reaction to the poll instead of about policy. The eader’s answer results in another a story about the poll, then another story about “Party B’s" leader's take on it, then another about whether a particular tactic deployed by Party A failed or succeeded.

Politicians hate talking about this stuff. And once the process story spiral starts, it’s very difficult for a leader down in the polls to stop it and get back on message. Exhibit A is Tim Hudak, who hasn’t seen a story that didn’t include the phrases “foreign worker ” and “falling in the polls” for almost a week.

All the while, Ontarians are reading or watching media stories, and many alter their behaviour because of them.

With all this in mind, it’s clear that this week’s public polls are a gut-punch to Hudak’s Tories and a much-needed morale booster for the Ontario Liberals who, only a few months back, had been all but written off. To me, these polls confirm the projections of Ontario Liberal Party’s campaign pollsters: a third McGuinty mandate is in the making. And that’s why the Premier has a noticeable spring in his step.

That said, this campaign is far from over. The Ontario Liberals have to fight off distractions and continue talking about the things they are strong on, like getting results and pocketbook issues. We saw a great example of this with the Premier’s assault on the effect of Hudak’s plans on municipal property taxes today in Ottawa.

The Tories and NDP are quite capable of finding a way to change the channel to an issue they are strong on, and shift the momentum back. But for now, I’m certain that the Ontario Liberals are happy with where they are at this point in the campaign, and enjoying the benefits that accrue from being there.