11.27.2013 08:24 AM

Forum fabrication, falsehood, fiction?

Asked how his firm got the Brandon-Souris by-election wrong by THIRTY PERCENTAGE POINTS, here’s the whopper offered up by Forum’s president:

“I know people are going to say ‘Oh, your polls wrong, you don’t know what you’re doing,’” said Bozinoff. “But in my mind, the Liberals kind of dropped the ball and the Tories just, you know, beat them at their ground game and got their supporters out.”

Really? Seriously? Actually, Lorne, that’s a pile of horseshit.  Your “poll” was ostensibly measuring voter intention – not Get Out The Vote strength.  There’s a difference.  Even my dog Roxy could figure that one out.

This steaming turd of a quote has sufficiently motivated me to write an entire column about how Forum got it so wrong (as they have many times before).  Here are some of the questions I intend to ask them:

  • That poll would have the effect of suppressing Liberal vote in Brandon-Souris.  Did the Conservative Party, or any related entity, in any way subsidize that poll?
  • Do you give your polls to the media because you are aware that no media organizations will pay for them? Are you giving them away because someone else, whose identity you do not disclose, has paid for them?
  • The methodology you use, IVR (a.k.a. robo-polling), has been condemned by many reputable pollsters.  Why do you keep using it?
  • In metropolitan centres, where huge numbers of citizens no longer have land lines, how do you ensure you are reaching people who will give you a representative sample, and the right results?
  • You predicted a B.C. NDP majority, a Wildrose majority and a Parti Quebecois majority, among others.  Why will you not apologize for those errors?

There are lots of other questions, and I suspect my friends Wright and Bricker have a few.  Feel free to add yours in comments, so I can send them to Forum sometime today.

Thanks.

50 Comments

  1. SmallTownON says:

    I find the fact that the Star is reporting on this hysterical since they’ve routinely used Forum data in their stories in the past.

  2. Robin says:

    Polling and releasing polling survey results during a federal election or provincial election, including by-elections, should be outlawed since it is impossible for polling survey firms to accurately reveal the intentions of a representative sample of eligible voters now that most Canadians do not have land lines. Most Canadians have smart phones with call display and many young Canadians have pre-paid “text only” phones. With regards to dubious polling survey results, it is probably better to be uninformed than misinformed. It may be better for our democracy if Canadians make up their minds without being influenced by polling results that claim to know which direction “the herd” is headed.

  3. Mr. Bozinoff didn’t answer when I wrote him last July about his polling in the London West by-election (see e-mail below). Another poll, that aired on Sun News Network, had Gretzky at 7% in the riding. Excluding from polling a candidate with a poll result so clearly outside of the margin of error is – at least by the measure of my several university undergraduate and graduate courses in statistics – a practice that skews the results.

    Bozinoff kept excluding Gretzky who, on polling day, got 5.1% of the vote, beat the Green candidate, and was only about 10 or so points behind the Liberal.

    Oh, and – in particular – check out point number 3 of my e-mail.

    Paul

    From: Paul McKeever
    Subject: Skewed Forum Poll for London West provincial by-election (Ontario)? – exclusion of Al Gretzky (FP)
    Date: July 5, 2013 5:09:39 PM EDT
    To: toronto@forumresearch.com

    Attention: Forum Research – Head of London-West provincial by-election polling efforts

    Dear Sir/Madam:

    I am writing in response to news reports today of a poll conducted in respect of the London West provincial by-election. I am asking that Forum provide me with timely answers to my questions, below.

    1. A number of Freedom Party of Ontario’s members in London West were polled in connection with the poll results released today. They advise that, in Forum’s poll questionnaire, the available responses to the questions asked of those polled included as explicit responses: Ali Chahbar (PC), Ken Coran (Liberal), Peggy Sattler (NDP), Gary Brown (Green), and “Other”. Freedom Party of Ontario’s candidate in that riding, Al Gretzky, was not one of the listed names that poll participants could provide as a response: his name was deliberately excluded from the list of explicitly named candidates, such that those taking the poll had no opportunity to answer Al Gretzky (FP). In our view, Mr. Gretzky’s intentional exclusion from the list of explicitly-named candidates in the poll skewed the poll results. Mr. Gretzky is not a person lacking a record of electoral popularity. It is a matter of record that, in the 2006 federal election, Al Gretzky ran as the federal Conservative candidate in London West, and received over 21,000 votes. He lost to the Liberal candidate by fewer than 1,400 votes. None of the other candidates in the London West provincial by-election has come close to receiving that number of votes, ever, in any election. In point of fact, Mr. Brown – the Green Party candidate in London West – was included in the poll’s list of explicitly-named candidates for response purposes, yet he obtained fewer than 1,200 votes in the 2011 provincial election. Yet Mr. Gretzky was the only candidate excluded from the poll. So I ask you: in the poll’s questions, why was Al Gretzky (FP) not on of the explicitly named candidates that those taking the poll could select?

    2. Will Mr. Gretzky be included as an explicitly named candidate in the poll question in the next and future 2013 by-election polls for London West? If so, thank-you. If not, why not?

    3. Curiously, although the riding of London West has a population in excess of 100,000 people, two Freedom Party members – both of whose names would appear on federal Conservative databases – were polled, though fewer than 500 people were polled in the riding. The probability that two provincial FP/federal Conservative members would be polled in London West is fairly remote. Accordingly, I ask you: did Forum Research select the individuals polled at random, or was party affiliation a factor in the selection of those polled? If the latter, why, and would using party affiliation to identify those to be polled not result in a biasing of the poll’s results?

    Freedom Party anticipates speaking with the news media about today’s Forum Poll results over the next 24 to 48 hours. Accordingly, I would ask that answers to the above questions be provided in a timely manner. Should answers not be received, it will be our understanding that Forum Research refused to answer our questions. I trust, however, that you will find the information provided above – about Mr. Gretzky’s electoral record – to be useful in designing and improving Forum’s next London West by-election poll, and that Forum will be able to provide me with timely and positive responses to my questions above.

    Regards,

    Paul McKeever

  4. james Smith says:

    I was wondering when you were going to opine about this company & the cockamamie results of its shoddy work. You are likely on to something, as to the reason why the media use them. But even so, they are so often so very wrong why, even if free, would they continue to use them? I noticed that the story of Rob Ford’s continued support in Toronto is a result of Forum’s so called polling.

  5. smelter rat says:

    Brandon-Souris in a nutshell: 10,000 Con voters sat on their hands, 5000 dippers voted red. End of story.

    • Stewart McD says:

      ……and University of Brandon boys and girls got out and voted for Justin’s candidate — a cool punk rocker; but where was Justin when needed during the short strokes of the Liberal campaign in Brandon-Souris? If Justin had stayed in Brandon the Liberal candidate might have won. Somebody gave Justin bad advice to return to Montreal.

  6. dave says:

    1st thing that crossed my mind when I saw that poll spread was that it was meant to activate the Conservative voters.

    • J.W. says:

      And of course encourage Libs to stay home since it was clearly all wrapped up.

      • Stewart McD says:

        So why didn’t Justin stay in Brandon to counter the suspect poll and help the Liberal candidate win? Did Justin’s handlers believe the suspect poll and pulled him out back to Montreal? Massive gaffe if so.

  7. Paul says:

    Poll was ridiculous…media should know better than to rely on polls if they are not being paid for by the media outlet with well defined T&C’s, questions …great comment above ..no information is better than mis information … if it was meant to get vote out , given turnout , do you really think it worked…

  8. Lance says:

    I am addressing this to Lorne, because I suspect that he will be reading this thread. Don’t give me that bullshit that the Liberals “dropped the ball” and “didn’t get their supporters out”. Trudeau visited that riding something like three times during the by-election period. The Liberal Party knew it would be a coup if they could take it, and they fielded a popular and respectable candidate that had a viable shot at it, almost pulling it off. Don’t give me that bullshit that the Liberals “dropped the ball” and “didn’t get their supporters out”.

    • Stewart McD says:

      Perhaps Liberal strategists “dropped the ball” by believing the rosy Forum poll and pulled Justin out back to Montreal. Then the Harper letter showed up and paralyzed the Liberal campaign. If Justin stayed,,, just saying.

  9. patrick says:

    Polls should be banned from public use during elections. We are supposed to make our own minds.

  10. Lance says:

    I am at the point in life that I now realize that one gets what one pays for. This polling company offers their services freely? Well, well, well……………

  11. Al in Cranbrook says:

    I’m increasingly inclined to think polling, once a writ has been dropped for an election, should be banned. Can’t help but think it fosters/plays to a “herd mentality” amongst voters, and really serves nobody.

    But then what would the MSM have to talk/write/babble mindlessly and inanely about, eh?

    You know?

    Like, f’rinstance issues and policies???

    Stuff that actually matters???

    I know. Silly me.

    • Al in Cranbrook says:

      Thinking about this some more….

      What if nobody had any clue how an election was going to turn out?

      Would more people turn out to vote to make sure their guy wins?

      How many look at polling, perceive an already done deal, and stay home?

      Is there not a good argument that perhaps one of the main causes for decreasing voter turnout is indeed polling?

      There’s little, if any, mystery, nor more importantly, validation, left in the process, much of it negated by polling that lays out a probable conclusion before it even happens.

      • Stewart McD says:

        Polling during campaigning will still be done but it will be in-house party polling which I think is more accurate per riding. National polling numbers can be misleading and not indicative of ridings with split voting.

        Polling is expensive, and the Conservatives will be able to do more extensive polling than the Liberals and NDP. However, if the Liberals and NDP pooled their polling, they will get better information and can act accordingly,,, as a covert coalition. Perhaps legitimate polling firms might balk at pooled polling, but when money talks bullsh!t walks.

  12. Robin says:

    In the 2012 by-election in Calgary Centre, similar circumstances existing, the Conservative Party HQ interfered with the local nomination to ensure their preferred candidate won thereby infuriating local Conservatives. Some prominent Conservatives publicly announced their support for the Liberal candidate.

    The “preferred” candidate was not popular. The Liberals fielded a well-liked and respected conservationist lawyer; the Green Party fielded a well-known and popular author and conservation activist; and, the NDP fielded a respected social activist.

    Results compared to 2011 General Election: Green Party increased their vote by 2,201 as the NDP vote declined by 6,251 (many going to the Green Party) and the Liberal vote increased by 403 while the Conservative vote declined by 18,200; yet, the Conservative won by a margin of 1,167 ahead of the Liberal candidate (10,201 to 9,034 respectively). Note: the combined vote of the Green Party and NDP declined by 4,050 votes.

    The Liberals could have won in Calgary Centre but failed to to effectively identify and deliver the Liberal vote; they couldn’t “close the deal”.

    There are valuable lessons for the Liberal Party in both Brandon-Souris and Calgary Centre.

  13. Kelly says:

    I agree with Al in Cranbrook. No polls after the writ is dropped. Stop the horse race; it’s such a distraction.

    And on a side note, I think we should borrow Australia’s preferential voting system; it’s like an automatic run-off. You rank your choices in order. That way we get the most preferred representative for the electorate as a whole.

    • ChristineK says:

      All we need is two federal parties to make the voting choice easy and then we wouldn’t need any complex voting system.

      If we had a preferential voting system all that would happen is NDP and Liberal voters would rank each other and exclude the Conservatives. This would become the tyranny of the Left.

      The Centre-Left should get it’s shit together so we can have a convenient two-party system in Canada and no more whining about splitting the vote. The Greens are only for the neurotics.

      I sometimes hope the Conservatives trash the Liberals and NDP in 2015 and that would force them to merge into one viable party.

      • Kelly says:

        With all due respect, I think a two-party system is a sham democracy. My real preference would be for proportional representation but some argue it gives too much power to crazy fringe parties. Preferential vote is a good compromise. It might actually encourage parties to be more centrist and respect the views of non supporters — unlike the current government which is utterly dismissive — even contemptuous — of non supporters.

      • Al in Cranbrook says:

        You are absolutely right, “tyranny of the left”!

        Which, no doubt, is why the concept is so popular on the left.

        It’s a simple concept: You wanna win an election, you need to get the most votes, period!

        The process as it stands forces political parties to achieve the critical mass of “legitimacy”, as it should be. And Preston Manning proved it can be done.

        If people want to vote for the Doobies R Us Party, that’s their choice, it’s a free country. But they have no right to expect legitimacy (usually for their idiocy) to be handed to them on a silver platter.

        And, the very last damn thing Canada needs is the kind of utter gong-show coalitions that have become a fact of life in Europe. We can all see all too clearly how well that’s working out for them!

        Lastly while I’m at it, voting from one’s home computer is also a damn stupid idea, and a disaster looking for a place to happen! Who the hell said democracy was supposed to be easy and convenient??? If one can’t be bothered to pick their sorry ass up off the couch and get it to a polling station, that’s their problem, not the rest of country’s!!! I deeply suspect this concept is also popular with the left because it appeals in large part to the sort of constituency that tends to find actually turning up to vote too much of a bother…if you get my drift.

        • Patrick says:

          So true about democracy being better having to make an effort to get there – you know, when I first voted I had to walk 10 miles in the snow while pulling my car. My vote was at least worth two of those who lived next door.

    • Tim says:

      Yup. And also the ding on your income tax return if you fail to vote.

    • MikeM says:

      “a preferential voting system … would be a the tyranny of the Left”
      The left only wins in that case if they are the majority. Isn’t that how it should work?

  14. Tiger says:

    The PCs were polling at 14 points up in Ottawa South’s by-election, according to Forum:
    http://metronews.ca/news/ottawa/751084/wynne-stumps-as-ottawa-south-byelection-enters-final-week/
    http://en.wikipedia.org/wiki/Ottawa_South_(provincial_electoral_district)#Opinion_polls

    As you know, McGuinty’s guy pulled it out, and did so fairly comfortably.

    Wondering whether they ask their questions in an anti-incumbent sort of way, or if by-elections are just hard to poll.

  15. Brad says:

    Accusations are flying now in the wake of the Brandon loss. Some are saying that Justin should have been stumping in Brandon rather than retreating back to Quebec where Bourassa riding was a Liberal lock.

    Trudeau’s guy in Brandon started losing votes in the last week once Harper sent out a letter reminding voters of his pot talk to kids and repealing mandatory minimums on child predators.

    Perhaps Justin fled Brandon rather than face accountability for accusations in Harper’s letter.

    • smelter rat says:

      Total bullshit. Harper’s letter hurt the Cons more than it did the Liberals. Furthermore, accusations are not “flying”. One guy wrote a column. Take a fucking pill and get a grip.

      • Stewart McD says:

        So why did the leading Liberal candidate lose the election? Was it Justin’s premature withdrawal from Brandon? Was it Harper’s letter trashing Justin? Was it because the Forum poll was wrong and misleading causing Liberals to stay home in false confidence? Or all of the above? Liberal party members want to know.

        • Cameron Prymak says:

          No, the question is why did the Conservatives almost lose the ‘safe’ seat?

        • smelter rat says:

          I think there is some truth to the theory that JT should have visited some of the rural communities in Brandon-Souris. Poll by poll results aren’t available yet, but sis suspect the Cons did well rurally. They also did well in the advance poll, I’m told. That speaks to the Cons ability to get the grey hairs out on what was a pretty pleasant weekend.

    • Matt says:

      Justin was there 3 or 4 times wasn’t he?

      I would be interested to see the internal polling from the LPC and CPC for Brandon Souris.

      I’m guessing it was near bang on the results.

      • Stewart McD says:

        Then it was a Liberal party snafu? Brandon-Souris should have been Liberal. What happened?

        • Stewart McD says:

          So what was the factor that boosted the Liberal vote so greatly in Brandon-Souris,,,, and don’t say disaffection with Harper.

          Trudeau visited the riding 3 times while Harper only issued a meagre letter slagging Trudeau. Did the Harper letter tip the scales and the Trudeau letter (see below) was not significant.

          We will never know if a 4th Trudeau visit might have boosted Liberal fortunes and Dinsdale would have been the MP for Brandon-Souris now.

          I’m not happy that Trudeau was yanked out of Manitoba and returned to Montreal to revel in a safe Liberal seat in Bourassa.

        • smelter rat says:

          Bingo.

  16. Matt says:

    I mentioned this in another thread, but since this is a topic specific to Forum Research, I’ll repeat it.

    Forum did six polls in Brandon Souris. There were reports from one lady, a Liberal supporter, she received a call from Forum for ALL SIX of the polls.

    Another report said the mother (I think it was the mother) of Manitoba’s Progressive Conservative’s was called for 3 or 4 of the polls.

    Random sampling indeed.

    And his excuses remind me of Frank Graves at Ekos after the 2011 federal election. He said he had proof the Conservative’s were guilty of electoral fraud. What was his proof? His polling was wrong, and the only way he could have gotten it so wrong was Conservative electoral fraud.

  17. Matt says:

    Opps. Should be the mother of Manitoba’s Progressive Conservative’s leader.

  18. Matt says:

    Warren wrote:

    That poll would have the effect of suppressing Liberal vote in Brandon-Souris.  Did the Conservative Party, or any related entity, in any way subsidize that poll?”

    Would a polling company really be that stupid? Damage their long term reputation and future business prospects releasing a poll that is so dramatically wrong for a one time short tem financial gain?

  19. Simon Worden says:

    The dark truth is one segment of the powers-that-be have twisted polls into merely propagandic assets while a subsegment of them desperately wants to discredit polls in light of what they reveal globally – e.g:

    http://www.pewresearch.org/daily-number/stoning-adulterers/

    The truth is out there.

  20. Steven says:

    How about no polls between Writ and the end of Election Day?

    Period.

    • Ottawa Civil Servant says:

      How about freedom of speech? Freedom of the press?

      You want to restrict the information available to the public, for what purpose?

      This is the Left wing tendencies I see everywhere: when in doubt, restrict, regulate and circumscribe.

  21. Stewart McD says:

    Ooooops, here’s Justin’s response to Harper’s letter found on http://www.liberal.ca :

    Open Letter from Justin Trudeau to the Voters of Brandon—Souris
    Posted by Justin Trudeau on November 22, 2013

    I have had the pleasure and honour of being in Brandon three times in as many months. You have welcomed me into your community and your homes. It has been a terrific experience and I want to thank you for it.

    Monday’s by-election is an important moment. Your choice could not be clearer. Do you want a representative who will be your voice in Ottawa, or do you want someone who will be Mr. Harper’s voice in your community?

    If you want a strong, passionate voice for Brandon, please vote for Rolf Dinsdale.

    I believe that Canadians deserve leaders who tell the truth. Leaders who are open and honest, and who answer straight questions with straight answers.

    I am writing to you in that spirit. Many of you will have received in the mail malicious, negative and false attacks on me from the Conservative Party and – in an unprecedented move – Mr. Harper himself.

    Allow me to answer the charges raised in the letter directly.

    Our plan is to replace the failed Conservative drug policy with one that will make it harder for our kids to get drugs, and starve organized crime of the hundreds of millions of dollars they currently make from the marijuana trade.

    On crime, I believe that serious crimes deserve serious punishment. As for mandatory minimum sentences, for years our party has supported them for serious crimes, and continues to do so. We simply believe that the Conservatives have applied them across the board and are now wasting your money fighting losing battles in the courts, as they get overturned on constitutional grounds.

    That, my friends, is the truth. Sophie and I are proud, proud parents of two young kids. We have a third on the way in the spring. Anybody who suggests that I am anything other than fully committed to protecting my kids and yours from crime and exploitation needs to think hard about the way they practice politics.

    Mr. Harper and his party have brought politics to a new, nasty low in this country, and it is catching up with them. He wants you to elect a representative who will work for him, not for you. Let me give you a recent, stark example of what I mean.

    Two weeks ago in Ottawa, on the floor of the House of Commons, we witnessed an extraordinary event. We Liberals had just tabled a motion that would have compelled everyone involved in the Senate scandal currently engulfing the Prime Minister’s Office to testify, under oath, in front of the House ethics committee.

    One by one, every single Conservative M.P. voted to help the Prime Minister cover up his involvement, and that of his senior staff. We have since learned from the RCMP that this scandal reaches the highest levels of the Prime Minister’s Office, including his Chief of Staff and at least 12 of his closest advisors.

    There can be no doubt that a Conservative M.P. for Brandon—Souris would vote to protect Mr. Harper and his party’s interest, not your interest, the public interest.

    Mr. Harper long ago forgot the promise he made to you. You believed you were sending good, community-oriented, local Conservatives to be your voice in Ottawa. Instead, what you got back was nothing but Mr. Harper’s voice in your community. You have a chance to change that on Monday.

    Send Mr. Harper a message he can’t ignore. Send Rolf Dinsdale to be your strong, local voice in Ottawa.

    Justin Trudeau, M.P.
    Leader of the Liberal Party of Canada

    Good letter but not enough to counter Harper’s letter? Justin should have stayed in Brandon until voting day unless the Liberal party brass gave up on Dinsdale and whisked Justin back to Montreal to be with an assured winner. Sad, really.

  22. jack says:

    The big question is that after all these horriible results why does anyone report on this poll, or better yet, why shouldn’t they report on anyone’s poll that wants to do one using any method they wish. I suggest the Kinsella poll – lets see how much press you can get…..and of course just take educated guesses without calling anyone. YOu will still beat the Forum results.

    They need to release a complete sampling methodology that answers:

    where did they get their call list?
    do they know if the same number is included in the poll time after time?
    How do they account for “hit one and done” responses – i.e people who hear its a poll, push number 1 and hang up without even listening to the question or answers?
    is the sampling error number (accurate 19 times out of 20 indicating 95% probability) calculated solely off of the number of people contacted or is their a formula to eliminate the many types of errors than can exist in a robocall over an in person call?

    I took basic stats decades ago and one thing I know is that these results are full of it. There has to be something elkse at work here be it a deliberate bias, incompetence, flawed methods, or something. Being wrong badly maybe, just maybe could happen once but this many times? Something else is going on. And i am right 19 times out of 20.

Leave a Reply to Stewart McD Cancel reply

Your email address will not be published.