rabble blogs are the personal pages of some of Canada's most insightful progressive activists and commentators. All opinions belong to the writer; however, writers are expected to adhere to our guidelines. We welcome new bloggers -- contact us for details.

Calgary mayoral election: Four polls, one horse race and an element of monkey business

Please chip in to support more articles like this. Support rabble.ca today for as little as $1 per month!

Bob Hawkesworth

Forget what you've read in the mainstream media. Some of it, anyway. The race to become mayor of Calgary is close to being a dead heat. It is unlikely, however, that it is a three-way race as the Calgary media claims.

Notwithstanding an anomalous poll that purports to show the third-place candidate, business teacher Naheed Nenshi, as a second-place contender, the real battle continues to be between former CTV anchor Barb Higgins and angry right-wing alderman and former front-runner Ric McIver.

But just as the media declared McIver to be the winner months before the mayoral race heated up and anyone really knew what the heck voters were thinking, it has been unable to resist the allure of the story suggested by just one of four recent polls, in which Nenshi's support seemed suddenly to rocket into contention.

Alas, the media is too enamoured of this dramatic come-from-behind yarn -- and too ignorant of the vagaries of polling public opinion -- to produce a story that is genuinely useful to voters. Nevertheless, the tale as it is now being reported does contain a strong element of monkey business that could affect the results of the race, likely in McIver's favour.

So let's consider the four most recent Calgary mayoral election opinion polls in the order they became public:

After hearing the media declare McIver -- "the toast of conservative Calgary" -- to be a shoo-in for months, a random sample of voters responded to a Return on Insight telephone survey of 501 Calgarians on Oct. 1 and 2.

Reported Oct. 5, the ROI poll showed Higgins, a more moderate conservative not affiliated with any political party, within striking distance of McIver, who had positioned himself far to the right with the backing and advice of the federal Conservatives. The ROI poll indicated McIver was leading Higgins by only a statistically insignificant 3 per cent in overall support.

The ROI poll showed McIver with 31 per cent support, Higgins with 28 per cent, and Nenshi with 16 per cent. It showed the other 12 candidates fading fast -- each with less than five per cent of the vote.

Next came results from an Ipsos-Reid poll, taken between Oct. 5 and 6, which could be read to suggest an Anybody But McIver trend was developing among Calgary voters. That poll, first reported on Tuesday, Oct. 12, still showed a statistically insignificant 3-per-cent difference, but this time Higgins' favour. That survey showed Higgins at 37 per cent, McIver at 34 per cent, and Nenshi moving upward to 21 per cent. Like the ROI poll, the Ipsos-Reid survey was conducted over the telephone, a methodology professional pollsters believe is superior.

Also Tuesday, an automated telephone survey conducted by IVRnet from a sample of 829 respondents willing to answer by pushing numbers on their handsets -- somewhat weaker methodology -- showed Higgins in the lead at 33.8 per cent, McIver second at 24.1 per cent and Nenshi third at 17.3 per cent.

The next day, another poll, this one conducted by Leger Marketing, showed McIver in the lead with 33.3 per cent, Nenshi unexpectedly in second place with 30.1 per cent and Higgins third with 29.6 per cent. That poll had a sample of 500 and was taken over the phone between Oct. 6 and 11.

Also on Wednesday, former NDP MLA and Calgary alderman Bob Hawkesworth, who had been polling around 3 to 4 per cent, gave up and dropped out of the race, endorsing Higgins as the best alternative.

Regardless of what else was happening, or whatever else other polls were showing, the media simply could not resist the story told by the divergent Leger survey, that of a dramatic three-way race. They reported it uncritically -- as journalists usually do with a good tale, no matter how unlikely.

So what's wrong with this, you may ask, if the ROI, Ipsos and Leger polls all used similarly reliable live-polling techniques?

The answer is that that the Leger poll -- which produced the "best" story from a journalistic perspective -- also had methodological flaws the others didn't, all of which seem to have gone unreported by the media.

First, it was conducted over a long weekend, traditionally a time disdained by professional pollsters because such timing encourages dubious results. Second, it contained a huge sampling of voters under 35 (probably because Mom was busy basting the turkey), fully a third of the respondents, more than double what it should have had. Leger tried to finesse this by applying heavy weighting to its results to account for the disproportionate numbers of young voters.

Regardless of the technical stuff, from a common sense perspective, the Leger results are hard to accept -- though barring a fifth, unexpected, poll, we will have to wait for Monday night to know just how misleading it was.

Which leaves us where, exactly? Certainly the results of the Leger survey are mischievous in the sense they confuse the Anybody But McIver vote about where it should go. After all, McIver's core voters come from groups more likely to actually get out and cast ballots -- older, wealthier, Conservative party faithful versus a younger, less well off, more progressive demographic skewed toward women and students.

This may be enough to tilt the results toward the polarizing candidacy of McIver to the delight of right-wing electoral schemers everywhere.

Certainly, if that is not to happen, the ABM vote is going to have to make a decision about which way to go within the next 24 hours, and then actually trouble itself to get out to vote.

For voters on the left, Hawkesworth's endorsement of Higgins' candidacy should make the decision about how to vote a little easier.

This post also appears on David Climenhaga's blog, Alberta Diary.

Thank you for reading this story…

More people are reading rabble.ca than ever and unlike many news organizations, we have never put up a paywall – at rabble we’ve always believed in making our reporting and analysis free to all, while striving to make it sustainable as well. Media isn’t free to produce. rabble’s total budget is likely less than what big corporate media spend on photocopying (we kid you not!) and we do not have any major foundation, sponsor or angel investor. Our main supporters are people and organizations -- like you. This is why we need your help. You are what keep us sustainable.

rabble.ca has staked its existence on you. We live or die on community support -- your support! We get hundreds of thousands of visitors and we believe in them. We believe in you. We believe people will put in what they can for the greater good. We call that sustainable.

So what is the easy answer for us? Depend on a community of visitors who care passionately about media that amplifies the voices of people struggling for change and justice. It really is that simple. When the people who visit rabble care enough to contribute a bit then it works for everyone.

And so we’re asking you if you could make a donation, right now, to help us carry forward on our mission. Make a donation today.

Comments

We welcome your comments! rabble.ca embraces a pro-human rights, pro-feminist, anti-racist, queer-positive, anti-imperialist and pro-labour stance, and encourages discussions which develop progressive thought. Our full comment policy can be found here. Learn more about Disqus on rabble.ca and your privacy here. Please keep in mind:

Do

  • Tell the truth and avoid rumours.
  • Add context and background.
  • Report typos and logical fallacies.
  • Be respectful.
  • Respect copyright - link to articles.
  • Stay focused. Bring in-depth commentary to our discussion forum, babble.

Don't

  • Use oppressive/offensive language.
  • Libel or defame.
  • Bully or troll.
  • Post spam.
  • Engage trolls. Flag suspect activity instead.