PHOTOS: Calgary Mayor Naheed Nenshi in the marbled hallways of the Alberta Legislature. He was reelected to his third term on Oct. 16. Below: Mainstreet Research pollster Quito Maggi (Photo: Mainstreet Research), Mainstreet Executive VP David Valentin (Photo: Twitter), Mount Royal University political science professor Duane Bratt (Photo: CBC), and Calgary pollster Janet Brown.

Postmedia, the newspaper chain that operates both the Calgary Herald and Calgary Sun, says it has put its relationship with Mainstreet Research “on hold” in the wake of the election-night revelation the Toronto-based pollster’s opinion surveys in the lead-up to Calgary’s mayoral election were spectacularly out to lunch.

Another factor in Postmedia’s decision may have been the reaction by officials of the Toronto-based polling company to criticism by Alberta academics and other pollsters of problems they identified in the polling data, such as young voters showing up as supporters of the conservative challenger.

What the Postmedia action really means, however, is not so clear.

A story published by the company’s newspapers on Saturday contains few details on what will actually happen while the relationship of about three years between the pollster and the newspaper company is “on hold,” or what Postmedia means by that phrase. Postmedia will await the results of Mainstreet’s assessment of what went wrong, said Gerry Nott, Postmedia’s vice-president of content, in a statement quoted by the Calgary Herald’s story.

So don’t hold your breath waiting for big changes, either in the relationship between the two companies or in Postmedia’s use of polling results in ways critics say interferes with the democratic process.

The backstory goes something like this:

Throughout the campaign leading up to the Calgary municipal election on Oct. 16, Mainstreet produced a series of public opinion surveys for the Herald and the Sun showing progressive-leaning Mayor Naheed Nenshi trailing his most serious conservative challenger, former Progressive Conservative Party president Bill Smith, by double digits.

In one case, a Mainstreet poll placed Mayor Nenshi 17 points behind Mr. Smith – which, if true, would have been slam-dunk territory for the conservative challenger.

As has been noted earlier in this space, Alberta conservatives badly wanted this outcome to advance their narrative progressive politics are kaput in Alberta, especially Calgary, and voters are swinging back toward the right.

In the same time frame, however, public polls by several other polling firms showed Mr. Nenshi in the lead, though by smaller margins – as did, we are told, unpublished private polling done some political campaigns.

This naturally raised suspicions among some of Mr. Nenshi’s supporters, who suggested Mainstreet was basically producing push polls so Postmedia could advance Mr. Smith’s campaign. As the headline on a CBC report described the argument, observers suspected Mainstream “co-ordinated polls to influence Calgary mayoral race.”

But it was the academics and other pollsters, not Mayor Nenshi’s campaign strategists, who raised red flags about Mainstreet’s results most vocally – and it was the academics and pollsters who felt the sting of the polling company’s response.

The controversy moved to social media after several commenters expressed their concerns to a Postmedia reporter covering one of the polls, and their criticisms didn’t make it into Postmedia’s next report. Political scientists Duane Bratt of Mount Royal University and Melanee Thomas of the University of Calgary took to Twitter to call out the Postmedia for omitting their concerns.

This prompted a response by Mainstreet Executive Vice-President David Valentin, who told radio station 660 News in Calgary the day before the election that when the results rolled in the next day, in the words of the reporter, “it will be payback time.”

“We’re going to be going through the case study and singling people out and showing what exactly it is they said and did,” Mr. Valentin told 660 News. “And I think anyone who comments to the media should expect that their comments are going to receive scrutiny after the fact. I think that’s fair.”

Critics of Mainstreet’s data and approach took that as a threat.

In the event, Mr. Nenshi’s victory by about seven points the next day settled questions about the accuracy of Mainstreet’s polling, if not the argument.

On Oct. 19, Mainstreet President Quito Maggi published an extremely long post on the company’s website apologizing for the poll results and describing what he referred to as an “unexplained error” as proof “that even a good sample can be wrong.” Mainsteet will launch “a root and branch review” of what happened, he vowed.

The same day, in a string of 15 Twitter posts, Dr. Bratt accused Mr. Maggi of “bullying and intimidation” and refused to accept an apology proffered by the pollster.

Respected Calgary-based pollster Janet Brown entered the fray the same day with a public Facebook post saying she was “flabbergasted that Postmedia has not reported on this or taken any action – even though it’s been well covered by 660, Metro, the Globe and Mail, and CBC.”

“A couple of brave souls who work at Postmedia have reached out to me ‘off the record’ on this,” she wrote. “But there has been complete silence from the many others I have interacted with over the years. I get it though. They are being bullied into silence by their employer just as I was bullied.”

On Saturday, Postmedia finally admitted it had egg on its face with the story about its relationship with Mainstreet being put on hold.

Whatever caused the wrong numbers, Ms. Brown told me in an email, “Calgarians deserved better than an election that was dominated by horserace polls, an overarching false narrative, and precious little discussion about the issues voters care about – taxes and economic stimulus.

And whatever Postmedia’s motives were in covering the election the way it did, it seemed to have had the effect of lighting a fire under Mr. Nenshi’s supporters, who got out to the polls in droves, and possibly also of lulling Mr. Smith’s campaign into believing it could cruise effortlessly to victory.

Join the Conversation

13 Comments

  1. Oddly, I have seen no mention of this either: The Marketing Research and Intelligence Association (MRIA) is a Canadian not-for-profit association representing all aspects of the market intelligence and survey research industry, including social research, competitive intelligence, data mining, insight, and knowledge management.

    As far as I can tell they are a legitimate association.

    Mainstreet had been a member. In 2016, after investigating a complaint, MRIA found that The Adjudication Panel found that Mainstreet Research violated
    1. Article 1 (Basic Principles)
    2. Article 2 (Honesty)
    3. Article 3 (Professional Responsibility) and
    4. Article 11 (Publishing Findings)

    They issued sanctions, but Mr. Maggi, Mr. Valentin and the company have not complied and as such are suspended.

    https://mria-arim.ca/sites/default/uploads/files/Mainstreet%20Research%20Sanction%20Posting.pdf

    Can we now close this chapter and recognize these folks as not legitimate?

    1. Right on!

      In my opinion Mainstreet has as much credibility as the Fraser (aka Koch) Institute, or Postmedia for that matter.

      This reminds me of the legal argument related to the fruit of a poisoned tree.

      When I first heard Mainstreet was associated with Postmedia, I immediately ignored their so-called scientific poll. A rotten tree yields a rotten fruit that always tastes bad.

    2. Thank you for posting this, David. Most of the people that comment on this blog have a liberal bias (including myself), so we are really just preaching to the converted. What you have done, however, is found an unbiased source, one that even Mainstreet respected enough to (presumably) pay membership dues, that is calling them out for misdeeds.

      How in heaven’s name did you find this when all sorts of other researchers (eg Nenshi’s campaign team) didn’t?

  2. As Maggi himself said. His methodology was shown to be accurate in most municipal elections in Alberta. Calgary was the glaring exception.
    So if the methodology is fine… that leaves only one thing to explain the error.
    NP and the Conservative establishment wanted Nenshi out.
    The methodological polling showed that Nenshi was losing.
    But the methodology was not to blame.
    So what was the problem?

    Maggi needs to explain how a methodology that worked well elsewhere failed so spectacularly in Calgary.

  3. Wow…That is a remarkable story. And kudos to DUane Bratt at Mount Royal and Melanee Thomas at U of C and others for calling out Postmedia. And thanks to Dr Bratt for his excellent twitter story fight back. This Postmedia boondoggle election strategy of bad pollsters and threats of retribution to those who make entirely reasonable critiques – that is not acceptable. And it is frightening. Are we going to see this same ruthless partisanship in our provincial elections. Is this the intensified PostMedia brand – the kind of newspapers that suit dictatorships driving home their single-minded points, platforms, and candidates. I’m very concerned about what the future holds now that we have failed and few newspapers in Canada and little independence from consolidated corporate interests. Congratulations to Mayor Nenshi on his hard fought victory and to his campaign volunteers. Look forward to his ongoing leadership.

  4. So Postmedia bullies Mainstreet into producing polls with the results they want, then get upset when the poll is wrong? That is bizarre.

    Also strange is that Mainstreet would put so much effort into defending their polls that they must have known were faulty.

  5. Postmedia should also sever ties with pollsters who utilize “unscientific” online polling techniques to survey respondents and stop publishing their results as fact. Whether it’s online polling on the carbon levy, election surveys or party support, those online polls are no more than idle entertainment.

    Unscientific online polls start with a badly selected sample. It’s not a sample of anybody; it is people who have volunteered/signed up to take online “internet” polls for incentives like prizes or money (i.e. ThinkHQ, Angus Reid). Respondents for online polls are not randomly selected because there is no complete list of voters’ IP internet addresses; therefore the poll is not probability-based and has no margin of error. Furthermore, there is no way such polls can take into account the views of non-internet users.

    Next time you see a published online internet poll from Postmedia with this caveat be very suspect of the results: “The survey utilizes a representative, but non-random sample, therefore margin of error is not applicable.”

  6. “Mainstreet will launch a ‘root and branch review’ of what happened, he (President Quito Maggi) vowed.”
    A good start would be the wooden heads in his polling department.

  7. I think someone involved with this dubious poll knew what exactly they were doing – trying to bolster support for Smith and encourage unsure voters to jump on the bandwagon. I am not sure if anyone at Postmedia was an accomplice to this, or just an unwitting dupe.

    Regardless, the newspapers have just as much or perhaps more egg on their face from this debacle. One might understand Mainstreet for over vigorously defending their own work, but journalists are supposed to be skeptical. Unfortunately for Postmedia, I think many of the skeptics have retired or been laid off and too often the papers now see their role mainly to just regurgitate what they have been given. They should have handled this poll, which others had pointed out the weakness in, much more cautiously.

    I suspect Postmedia will keep its distance from its business partner Mainstreet for a while. However, I wonder if their review will be inconclusive and is really just a way to buy time until the heat dies down. However, if they take this approach and do not deal with the fundamental problems, their credibility will continue to erode. The next time Postmedia publishes a poll with surprising results, will people believe them, or even pay attention – especially if that poll is from Mainstreet. We all know the story of the boy who cried wolf too many times.

    I think Mainstreet also did not help itself by its high handed and overly aggressive defense of its poll. If they had a bit more humility when they came out with the results, it would have made the climb down much easier for them. Now the tables have now been turned and it is Mainstreet that is receiving scrutiny, not those that questioned the polls. Hopefully, the pollsters will be able to come up with a credible explanation of how they got it wrong fairly soon.

    I think polling was easier in the land line era before the internet, as was running a daily newspaper. Perhaps both could eventually end up being causalities to advancing technologies, but it seems in this case as the saying goes, the truth was the first casualty.

  8. Mmm wasn’t too long ago Paul Godfrey and a bevy of Toronto media mavens were telling Calgarians how to vote and their house organs what to print and report… how well did that go? Same bs propaganda — different day/election.

  9. Mainstreet opinions can’t be trusted. To find out what people are really thinking you gotta venture into the sidestreets.

  10. Like others, I think the Postmedia/Mainstreet poll backfired on it’s backers who I strongly suspect wanted a specific outcome. Unfortunately they got the outcome they wanted in the published poll but certainly not in the elections.

    My guess is the election spurred on the Nenshi team and their supporters. The opposite for the Smith team.

    Wonder what they will do for the next Provincial election.

    We stopped bothering with the Calgary Herald several years ago after returning from a one year hiatus. We found that there was very little in it that could not be found in other media. Shame really but what do you expect when you run a newspaper that seems to be based on press service reports/reprints, real estate adverts disguised as articles, and too much other advertising.

Leave a comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.